Tools for working with electronics

Hand tools

Whether you are soldering a circuit, working on an enclosure for your project or repairing something, there are a number of tools to help you get the job done. Some typical tools appear in the image that follows:

Some of these tools-such as scissors, screwdrivers and tweezers-you may already have. Tweezers are handy for picking up small components as well as for guiding single wires into place on a circuit board. As for screwdrivers, they have more use than just for working with screws; some components (for e.g., a preset variable resistor) are adjusted using a screwdriver. Speaking of screwdrivers, you can buy screwdriver sets which contain bits for the main types of screws, especially the so-called security screws.

Even nail clippers are helpful to have as they can be useful for cutting off a small amount of a wire in an awkward place when it is not possible to do so using wire strippers. More specific electronics tools are an IC extractor for removing a microchip from its socket and a desoldering pump for sucking up solder when removing components from a circuit board.

Soldering iron

Components are held to circuit boards, each other and wires using solder, which as well as holding components into place, also provides an electrical connection. Solder melts when heated by a soldering iron but becomes hard again when cooled (in a few seconds once the soldering iron is removed). Soldering irons reach very high temperatures so should be handled with care!

Soldering is a very important skill to have to be able to make long lasting circuits and also to repair existing circuits. However, it does require plenty of practice and no doubt mistakes will be made now and then. Start with a cheap soldering iron with a stand (like in the photo below, right) and be sure to also get a desoldering pump which will let you suck up solder should, for example, too much flow.

To solder, first make sure the soldering iron's tip is clean and it's also a good idea to clean the circuit board too. Insert the component's leads into the holes provided by the circuit board, you may need to clip it down or bend its legs slightly to keep it into place. Apply a small amount of solder to the soldering iron's tip, which is known as tinning. Then, on the circuit board's underside, heat up one side of the lead with the soldering iron and then apply a small amount of solder at the other side. The solder should flow all around the lead in a few seconds which is when you can remove the soldering iron before soldering the other connections. It's important to wait a few seconds before removing the soldering iron otherwise the joint may end up bad.

In time, you may want to buy a soldering iron which has a feature to adjust the temperature of the iron. This way you can match the melting temperature of the solder more closely and you can lower the temperature when not using the soldering iron for extended periods; leaving the soldering iron idle for a long time can harm the tip.

As already mentioned, using a desoldering pump you can correct mistakes such as when too much solder flows, which can cause shorts. Using your soldering iron you need to heat up the solder until it melts, hold the desolder pump in your other hand (or someone else's) close to the solder and release the mechanism which will suck up the solder; you may need to repeat this a number of times if you need to remove all the solder. There is even a combination of a soldering iron and desolder pump known as a desoldering iron, but this is helpful when you need to only desolder components not when you're soldering components in a circuit as well; you should use a separate desoldering pump if you go wrong.

On the left in this picture above is a temperature controlled soldering iron, a dial is used to vary just how hot the iron gets. When it is not needed for a few minutes the temperature can be turned down as to put less stress on the tip and be less dangerous to touch. As soon as it is needed again to finish the job the temperature can be turned up again and it will only take a few moments to warm up unlike an ordinary soldering iron (as seen on the right) which can only be turned on and off, thus creating longer delays waiting for it to warm up.

Usually provided with a typical soldering iron stand is a sponge which should be soaked with distilled water for cleaning the tip, never use ordinary water from a tap because it can destroy the tip.

Going back to solder, in a lot of places only lead-free solder is available to buy, which is better for the environment. Unfortunately, lead-free solder requires a different melting temperature to previous types of solder and lead-free solder can be more harmful to soldering iron tips.

Below we have from the top a battery powered soldering iron that runs off three AA batteries (4.5V) and below it, a desoldering pump. Next is a solder dispenser tube and at the bottom in the middle is a tin of cleaning mixture which also applies a small amount of solder to the soldering iron tip at the same time. Finally, either side of the tin are several crocidle clips which are useful for keeping things in place, for making temporary electrical connections and to act as heatsinks to protect components from damage when soldering or desoldering them (diodes including LED's, transistors, IC's and the like).

Storage

As well as needing to store your tools you will need to also keep together your components, preferably in a box with compartments or drawers so that you can access particular components quickly. It can actually be quite difficult to keep track of all the components you have so you will need to be well organised. Label well boxes along with the individual compartments or drawers and if need be you could use a database to keep track of your components. A database is also helpful to be able to bring up information on a component, especially for IC's.

Circuit boards

Circuit boards, a.k.a Printed Circuit Boards (PCB's), are used to keep the components of a circuit together as well as forming part of the circuit since it carries the electrical current through the conductive tracks. A circuit board may be one sided, in that the components and connections are on one side only or double sided which has components and connections on both sides. There are also circuit boards that have more than two layers in which the additional layers are within the circuit board and carry connections between the components. A good example of a circuit board that has many layers is a computer mother board. Even hobbyist can make their own PCB's at home using various techniques although great care must be taken when creating PCB's as chemicals are used to etch away a copper coating to form the tracks.

As well as the circuit boards that are used in manufactured electronics, there are a number of different types of circuit board that are more useful for prototypes and home construction. Have a look at this photo:

On the left is a piece of breadboard (made from plastic nowadays instead of the wooden boards used once upon a time where the term breadboard originates from), a.k.a. prototype board. Breadboard has many holes across its surface in rows and columns (often numbered), connected together internally, to which you insert the components of a test circuit. Single stranded wire is used to act as jump leads to connect different parts of the circuit as needed (multi-stranded wire can be used but is not advised because of the risk of wire strands that can fall off and get stuck in the holes). You can cut up different lengths or buy ready cut pieces of wire.

Typical breadboard has 'channels' to accommodate IC's or other DIP (Dual In-line Package) components or for whatever situation a break is needed for a circuit. Another useful feature of many breadboards are the power rails which may be single or dual rows, and some times there will be power connectors fitted to the board. It should also be noted that some small breadboards have the ability to be slotted together as to form a much larger piece of breadboard suitable for bigger circuits.

Some components cannot be fitted into a breadboard, such as large IC's and components with thick connections but one workaround is to solder wires to the component which can then be inserted into the breadboard. However, breadboard is only suitable for low voltage circuits and low frequency signals.

I strongly advise that you do not buy cheap breadboard unless it has had very good reviews. Use of poorly made breadboard will result in difficulty in inserting components even causing components or wires to break. Even if you are able to construct a circuit on cheap breadboard you may end up spending a long time diagnosing your circuit only to find it wasn't working because of bad connections in the breadboard.

When you've finished testing you can then go ahead and make a more permanent version of the circuit (if need be), now that you are sure of what components you need and how to put it together. But breadboard can be used for more than just prototyping ideas, it is great for learning how electricity behaves by wiring simple circuits on the breadboard and measuring voltage, current and so on at specific points.

On the right of the photo above is a large board somewhat resembling breadboard but is intended for more permanent construction. Component leads are pushed through the holes and soldered on the other side. Circuits made in this way do not hold as well as when other techniques are used but are quite stable. This type of circuit arrangement is sometimes called nesting and certainly for manufactured products, is a thing of the past.

Next, we have stripboard (a.k.a. veroboard or matrix board) which has rows of copper tracks on one side that can be cut as required to break the circuit. After cutting a track it should be checked visually and then with a continuity tester or multimeter on ohms range to make sure the track really is broken.

A better version of the normal stripboard as shown in the photo above on the bottom has tracks already broken at regular intervals, designed especially for DIP IC's; this is sometimes known as tripad board. As for the prototype board seen above on the middle right, this has copper pads on every hole so you have much greater choice when soldering the circuit as there is no need to break any tracks.

Test equipment

When testing circuits and learning how they work you will need various test equipment and although some of them can be expensive they are worth the money in the long run. Always research well if you are going to pay out a lot of money to make sure the spec matches your needs and that the reviews are good.

One of the most basic but very helpful types of test equipment is the multimeter, which lets you measure voltage, current, resistance and so on. These are on sale cheaply when only basic functions are offered but shoot up in price when you want a more advanced version such as one that can measure capacitance, inductance and frequency. You can manage with a simple multimeter if need be as other features can be emulated, for example, if your multimeter lacks a continuity (a good connection) test, you can use a resistance test on the lowest range.

Multimeters are divided into two main types that will affect its cost and slightly, how it can be used; they are analogue and digital. Analogue multimeters use a form of display that makes use of a needle that moves across a scale which isn't as clear to read as a digital multimeter, and analogue multimeters tend to be more expensive. Digital multimeters on the other hand have an instantly recognisable (usually) LCD display and have a much higher input impedance (it will have little affect on the circuit under test).

That said, analogue and digital multimeters have other advantages and disadvantages. When you want to check for changes as they happen most digital multimeters are too slow to react in time, but with the analogue type you will be able to watch the needle move in time with the change in voltage, current, etc.

A basic digital multimeter, the Mastech M-830B, is shown on the left with the two test leads, coloured black for negative and red for positive. This multimeter didn't cost much and can measure voltage, current and resistance, and is able to test bipolar transistors and diodes.

The digital multimeter on the right has similar functions to the other one with the addition of an audible continuity tester and it also has a clamp at the top allowing for AC measurements to be made without connection to the wire from the circuit under test (a single wire passes through the clamp). Note that this meter can measure much high voltages than the yellow one but has less precision (for e.g. 1.5V will show as 1V).

A non-contact digital thermometer is helpful for checking how hot components are getting without actually touching the component, some of which can get very hot under normal use. However, some components only get hot when there is something wrong. With some IC's, for example, they will only get hot if there is a fault, such as reversed power supply connections or a shorted output.

Another important, but more expensive form of test equipment, is the oscilloscope. While you can use is to make similar measurements to that of a multimeter, an oscilloscope lets you see electrical waveforms. This way you can visually see changes as they happen, in real time. You can, for example, use an oscilloscope to look for voltage spikes or to check that a pulse is clean. An oscilloscope is also very much a learning tool in that you can discover the way voltage, current and so on change with time.

The early oscilloscopes were analogue and typically used a CRT to display the waveforms and it was actually quite straightforward to convert a basic CRT monitor into a simple oscilloscope. Modern oscilloscopes are usually digital and come in two main forms; those that work on their own (standalone) and those that require a computer to work. Standalone oscilloscopes are better where there is not enough space for a computer (but some have the option to be remote controlled by a computer) and typically have the ability to save waveforms to a flash drive and some can directly send a waveform to a connected printer. The oscilloscopes that need a computer to work have the big advantage of small size and direct interface with a computer for displaying and recording waveforms. It is also worth mentioning handheld oscilloscopes which are usually standalone but like their larger counterparts may have the option to be connected to a computer.

While an oscilloscope can be used to view digital signals it doesn't automatically interpret what is being displayed, unless it is a Mixed Signal Oscilloscope which is essentially an oscilloscope and logic analyzer combined. For advanced digital analysis a logic analyzer can be used and theses come in both standalone form and those that needed to be connected to a computer. Logic analyzers have a number of digital inputs that can be recorded, their timings displayed and the signals can be interpreted automatically if using a known protocol (e.g., I2C). This is very useful not only for testing but also for learning how something works. It could be that you are trying to work out how to interface with a display in a system which you have the pinout for but no other information. By connecting a logic analyzer it would be possible to see if the signals use a known format and if not you could work out what format is used by recording and comparing the signals at certain points (e.g., when a button is pressed which causes the display to be cleared).

Note that as logic analyzers are designed to read digital signals they will display waveforms much cleaner than an oscilloscope although an oscilloscope with a bandwidth limit option will 'clean' a signal. The point is that using a logic analyzer will only tell you half the story; you may still need to use an oscilloscope to get the bigger picture.

You will need to power circuits you are testing or building and you can end up with quite a collection of power supplies so you may want to consider buying a variable power supply which will let you select from a range of voltages and deliver current at a certain limit. Typically these power supplies have analog or digital displays which show the current voltage settings and the amount of current being drawn. Some have multiple output channels so that you can power multiple circuits at once and others have dual rail (it can output positive and negative voltages). Consider what voltage range you most likely will need (probably no more than 30V) and what current requirements your circuits will need (a couple of amps maximum would probably be enough). Building your own variable power supply is also an option so you may want to have a look at the Power supply circuits page.

Another form of power supply you may find useful is a fixed voltage power supply designed to be plugged into breadboard, such as the one below:

This power supply has four connector pins on the underside to be plugged into breadboard, with each pair outputting either 5V or 3.3V, typical voltage for logic chips. A short pin on each side selects between 5V and 3.3V and the output power can also be extracted from 8 pins on the top side of the circuit board using the supplied 8 connector pins. There is a power switch and power on LED indicator and the board can be either powered from USB or a DC barrel connector.

Now for reviews of products I own sorted by type of test equipment.

Component analysers

4070L LCR Meter

Most basic multimeters test voltage, current, resistance and some have a transistor and diode tester but many are without a capacitance or inductance test mode. As I only had a basic multimeter, the Mastech M-830B, I bought the 4070L LCR meter (there is no indication of the manufacturer), which tests inductors, capacitors and resistors as well as transistors too. You can see it below:

This multimeter has a 3 1/2 digit LCD with a backlight that can be turned on and off, a rotary switch to choose the mode but a separate on/off switch which saves having to turn the mode switch just to turn the meter off as with many other meters. The meter measures resistances from 0 to 2 gigaohms, capacitance from 2000pF to 200uF and inductance from 1mH to 20H.

I tested a 1K 5% resistor using the 4070L and it gave a reading of 990 ohms and my M-830B displayed 0.99 kilohms, so while both meters were very close and indicated a value within the 5% tolerance the 4070L was able to display an extra digit. Note that when measuring resistance using the 4070L the red lead has to be in the ohm socket and the black lead in the COM/LC+ socket.

With a 1nF 5% Polyester film capacitor the 4070L gave a reading of 1.03nF which is an acceptable value for the 5% tolerance. When taking capacitance readings using the 4070L the connected capacitor must not be charged and the black lead must be connected to the LC- connection while the red lead has to be plugged into the COM/LC+ socket.

In addition, I tested the 4070L with an inductor marked as 6.8mH and the meter responded with a reading of 8.45mH which, assuming 20% tolerance, would be an OK reading if you also take into account the meter's accuracy. For inductance measurements the test leads must be connected the same way as for when taking capacitance readings.

Lastly, a transistor test using a BC108 general purpose silicon NPN transistor. My M-830B reported 482 on its transistor test (hFE, current gain) and the 4070L answered with 480 but my DCA55 component analyser reported 442. This was probably because the 4070L tests the transistor using 3V @ 10uA but the DCA55 uses base current of 4.5mA but either way the values are within spec.

DCA55 semiconductor component analyser

There are times when you have a bunch of components that you are unsure exactly what they are or you cannot find information about the component such as the pinout. The Peak atlas DCA55 semiconductor component analyser is a handy device for testing and identifying a number of different semiconductor components, the full list of which is given below:

Bipolar transistors

Darlington transistors

Enchancement mode MOSFETs

Depletion mode MOSFETs

Junction FETs

low power triacs and thyristors

LEDs including bicolour and tricolour LEDs

Diodes and diode networks

I bought the DCA55 for about £40 from ebay and you can see it below testing a BFY51 transistor:

It is simple to use; connect up the component to any of the leads, turn the DCA55 on using the on/test button and after a few seconds the type of component will be displayed. Press the scroll/off button to move through each page of information. For example, when testing a transistor, the type, pinout, hfe and so on are viewable. You will find that often it's quicker to use the DCA55 to get the pinout of a component than look up the information.

It is vital when using the DCA55 that only unpowered components are connected otherwise a false reading may be given or the DCA55 could be damaged. The DCA55 uses a GP23A/MN21 12V battery which requires that the DCA55 be unscrewed in order to change the battery; it would have been better to make the battery easier to change. Another negative is that the DCA55 doesn't have a backlight or contrast adjustment.

I tested the DCA55 with a number of different transistors and diodes and the information displayed seemed correct. When testing an LED it flashes as no doubt when doing its tests the DCA55 takes a number of measurements on the component under test in different configurations.

Function Generators

Siglent SDG1005 Arbitrary Function Generator

Along with an oscilloscope, a function generator is a very useful piece of kit which generates different waveforms which can be used to test equipment and to learn how different waveforms can be useful. After much research I came across the Siglent SDG1005 which was appealing because of the number of features for the price and because I already owned a Siglent oscilloscope (SDS1052DL).

You can view the function generator in the photo below, underneath my oscilloscope, which is showing a sine wave generated by the SDG1005:

The 3.5" colour LCD shows the current settings such as frequency and amplitude although it would have been helpful if it had been more clearer which channel's settings are currently displayed. A single button allows selection of the two channels and there are context dependent function keys as well as a number of option buttons to select sweep, burst, utility, etc. There are buttons to quickly select a main waveform type and there is a number pad which makes entering numerical values easy. For slight adjustments the dial can be used along with the cursor keys to select a value to change (the cursor keys can also select menu options).

On the back of the unit are the following connectors: 10MHz in, sync out, modulation in and ext trig/gate/fsk/burst. There is also a USB for connecting to a PC and a ground connection. A second USB socket can be found on the front of the unit and was designed for plugging a flash drive for transferring waveforms that were designed on a PC and for saving configuration data. A GPIB and ethernet add-on are available but as far as I can tell they have to be specially requested.

Channel 2 doubles up as the input to the frequency counter which has to be selected manually; doing so automatically turns off the output on channel 2. The counter can measure freq, period, etc. from 100mHz to 200Mhz and by comparing the frequency of a signal reported by the SDG1005 to my oscilloscope I found them to be close. What I did have to do some digging to find was the range of voltage for the frequency counter but the manual does list them; basically +/- 5V.

I had a feeling that connecting the function generator to my PC wouldn't exactly be straightforward. Firstly, I downloaded the EasyWave software from the official site:

http://www.siglent.com/ENs/generator/SDG1000

(Firmware & Software section.)

But when I connected the function generator to my PC Windows couldn't install the driver. I had already installed NI-VISA for my Siglent oscilloscope so that shouldn't have been the problem. However, after downloading an older version of EasyWave and then directing Windows to the included driver although it complained about the unsigned driver it did install the driver. Now I was able to connect to the function generator using my PC

Features of the EasyWave software include sending your own waveforms to the function generator. Especially nice is being able to draw your own arbitrary waveforms (even better when using a Surface pro because of its pen). I did find it annoying that when I chose a standard waveform such as a square wave it was hard to see it in the window as the lines touched the top and bottom.

For testing it's a good idea to use an oscilloscope so I used my SDS1052DL which of course Siglent claims is ideal for connecting to the SDG1005. To connect the function generator to an oscilloscope you will need a 'BNC' lead which wasn't provided with the function generator by fortunately I already had a couple of suitable leads. I checked a number of different waveforms and varied the frequency and other values while observing the changes on my oscilloscope which behaved as expected.

The function generator will be very useful and so far I am happy with its performance and quietness (it has no fan unlike the higher frequency versions of Siglent's function generators). There are, however, a number of issues to mention: while my SDG1005 has the most recent firmware (as to date) installed there is a glitch in the built-in arbitrary waveform selection menu whereby pressing an arrow key causes some waveform names to change slightly.

Be sure to check out the following link to a forum with lots of info about the SDG1000 function generators:

http://www.eevblog.com/forum/testgear/the-sdg1000-and-sdg800-thread/

Logic analyzers

HP 1650A Logic Analyzer

What appealed to me most about the HP 1650A logic analyzer was its 80 channels which would be more suitable for testing my homemade computer than other logic analyzers I own that are limited to 32 channels. The HP 1650A is from 1987 and boasts 100 MHz timing, 25 MHz state analysis, glitch detection support, and 1K memory depth for all channels. It has a black & white 9 inch (approx. 23 cm) CRT in which to display menus, timing waveforms and state analysis. The unit measures roughly 42.5 cm x 19 cm x 35.5 cm.

When booted up, the HP 1650A will run through a series of tests and then attempt to load the O/S from the floppy drive which is a double-sided, double density drive. The floppy drive is therefore one of the most critical parts of the system and should it fail the unit becomes useless although replacement is possible (the model of he floppy drive in my unit is Sony mp-f52w-30), or even with an emulated (flash drive/SD card) floppy drive. I timed that from boot with the system disc inserted it takes about 55 seconds to be ready to be used. As well as loading the O/S the drive is also used for loading and saving acquired data and configuration information.

I did not have good luck when I received my HP 1650A as it would not read from the system disc so I took the unit apart, removed the floppy drive, found something loose inside and cleaned the drive. Once I had put the logic analyzer back together it read and loaded from the disc but the Select button (which has had its label rubbed off) would not work. As the system makes a clicking sound whenever you press a button that confirmed that indeed the Select button was faulty. So I opened up the unit again, cleaned the keypad, checked with my multimeter and having reassembled once more, the Select button was now responding. Then I found that channel 1 was not working and looking at the port connector I could see that the pin had broken off. It may seem like a lot of work for just one channel, but I took the unit apart, removed the main system board and attacked the IDC connector as to remove the plastic near the broken pin. I then soldered a pin from another connector to the broken one and bent the pin enough that the pod connector would fit; all channels are working now.

As usual when doing an initial test with a logic analyzer I used a 555 astable, which in the photo below you can see connected to channel 1. I checked with another logic analyzer and found the timing to be very close.

As you can see, at the front is the screen on the left, the floppy drive at the top right, at the lower right the hex keypad, menu buttons (format, trace, display and I/O), run, stop, don't care, and clear entry buttons. To the very right is the select key, selection knob and roll buttons (underneath the knob). As well as using the knob to change values you can also use the keypad to enter numbers which causes a pop-up menu to appear so additional options can be selected (e.g. unit of measurement). The knob allows for fast selection and in combination with the Roll buttons makes selecting data easier.

A case for storing the disc, manual and cables was screwed to the top but I removed that so I could stack things on top of the unit. A carry handle can be found at the front which tucks nicely underneath the unit and there are also feet that can be rotated to hold the unit upward for better viewing angle.

The ports at the back are: power socket and switch, external trigger out BNC, external trigger input BNC, intensity knob, RS-232 (for printer and programming), pod 1 data + j clock, pod 2 data + k clock, pod 3 data + l clock, pod 4 data + m clock, and pod 5 data + n clock. Ribbon cables were provided with the unit which are to be connected to pods that have removable leads; each lead has a signal and GND connection with the signal wires having 90K series resistance each; the pod also has a common ground. The ribbon cables have 40 connections which includes 2 +5V lines @600mA, as the cables can be connected to a preprocessor.

Inside the logic analyzer you will find the CRT, floppy drive, power supply, and main board which uses a 68000 CPU. When powered on the unit is quite loud as there is a fan at back where the PSU is. On the CRT circuit board near the fuse there is a connection labelled 'TP-1 Vid out' but I can find no mention of it in the service manual. 

The service manual can be found at:

http://www.qsl.net/n9zia/test/HP1650A_Service_Manual.pdf

The reference manual at:

http://www.qsl.net/n9zia/test/HP1650A_Reference_Manual.pdf

I'll talk a bit now about using the logic analyzer from playing with it but all the information is available in the reference manual linked above.

The pods can be defined as having TTL, ECL or a user defined threshold level; this is done on the Format screen. Once set up, I configured the trigger which is done on the Trace screen (press the Trace button). I set Trace mode to single so data is captured just once when triggered, Armed by to Run (that is the Run button is used to start capture but it can also be set to use a trigger from the BNC input), and Acquisition mode to Transitional. In the section Then find Edge I selected the bit to be used as the trigger-each bit is represented by a dot and bit 0 is on the right. Select a bit and use the Select button to toggle between falling edge, rising edge, falling/rising edge, or don't care (default). With that part sorted, I used the Display button to enter the Timing Waveforms screen. After pressing the Run button the system waited for the trigger and then captured the signal changes. Because of the slow 555 astable I used for testing I had to change the Time/div to actually see the signal transitions.

To go into more detail, on the Timing Waveforms screen the delay value is used to scroll through the waveform, and to the right of the delay box the current value under the marker is displayed (this can be set to show the value for a selected label). There are 2 markers, X and O only, and there are the following marker modes: Time, Patterns and Statistics. The markers can be moved and timing is shown between the 2 markers and the trigger point. Note that the ample period is fixed at 10ns for the timing mode. Each waveform for the signals can be turned on/off, deleted or inserted by selecting a pod number (e.g. Pod 1) and pressing Select. It is also possible to change the pod bit to be displayed by selecting it and pressing Select.

The logic analyzer supports 2 analyzers, each of which can be named, as you may want a timing and state machine; the pods can be assigned to the analyzers in different combinations. You can switch between the 2 analyzers from any of the screens or have mixed mode (both analyzers displayed at once as split screen). Signals can be grouped together in bus-like arrangements using the Format screen. There's a maximum of 20 labels but only 14 can be displayed for timing and 11 for state. Labels can also be inverted using the polarity option (+/-). Each label can have from 1 to 32 channels assigned to it. Mnemonics can be created containing a set bit pattern with up to 200 available.

A very handy feature is that multiple captures can be overlaid on one another using the Accumulate option on the Timing Waveforms screen as this can show if there are any variations from the same signal each time the capture is run.

To return to the System configuration screen from any of the other screens, select the very top label (e.g. Machine 1), press Select, highlight System and press Select again. The I/O pop-up menu has a number of utility options including disc operations which let you load, save, duplicate and more. I had tried to create a system disc using a PC, USB floppy drive and a special program but it always failed with an error. Even using the HP 1650A itself it would not format a disc which suggests the discs are at fault. Here is a link for creating an HP 1650A compatible disc using a PC:

http://sci.tech-archive.net/Archive/sci.electronics.repair/2004-11/0635.html

I like the advanced features of the HP 1650A and the floppy drive has the advantage of allowing the loading and saving of capture data and configuration settings. However, the floppy drive could also be its biggest downfall should the drive fail or if the system disc becomes unreadable before I can make a copy. But no doubt, having 80 channels as well as dedicated clock inputs makes the unit very useful for testing relatively slow (by today's standards) systems.

Leaptronix LA-100P Logic Analyser

Although I already owned an 8 channel Saleae logic analyzer clone I was struggling to find a fault with a simple computing I had been making, as the logic analyzer has too few channels to see enough signals at once to confirm what was happening. I figured that ideally I would need a logic analyzer with 32 channels but I found that even 16 channel logic analyzers were very expensive and the only 32 channel one I could find at a 'reasonable price' was from China and I didn't want to endure the long wait and import costs. After more research, however, I found the Leaptronix LA-100P 32-channel, 100MHz logic analyzer from TheDebugStore for £780. It was a risk as I could not find any reviews on it but considering a PC based 32-channel logic analyzer cost about £600 from the same store I thought it wise to pay out a bit more for a logic analyser that could run standalone as well as with a PC.

With a 6 inch colour LCD the LA-100P has 4 ports, each providing 8 channels and ground for monitoring signals from the circuit being tested. Rather than connecting to the ports directly you have to use the provided pods which have their own connectors for testing signals; these pods get warm during use. I imagine that the pods provide some form of protection to the LA-100P and perhaps a simple termination circuit too. Each test connector plugged into a pod has 10 wires with female end points; 8 channels and 2 ground wires (1 shorter than the other). Aside from the ground wires the channel wire colours match the signals on screen (a couple of the colours are off, however). The wires are in order from 1 to 8 and then the two grounds and written on the test connector you will find '12345678GG' to indicate the connections.

To see what the logic analyzer looks like, view the photo below:

As mentioned, there are 32 channels and the LA-100P can operate on either its internal clock or an external clock connected to channel 32. The channels can be renamed (not limited to just a few characters) and up to 16 channels can be grouped into a bus; each bus can be named. Note that if you press the Auto scale button your buses will be removed but channel names will be preserved.

You can set the logic level to one of three presets - TTL, CMOS or ECL - or you can manually set the voltage level yourself, Different types of trigger (high/low/rising edge, etc.) can be set up on each channel. For example, you may want to capture the data when a reset signal goes high.

The interface takes some getting used to and it doesn't help that the badly translated manual is confusing. There are a number of buttons for setting up the channels and trigger, there are 6 shortcut keys, a keypad for entering numbers and letters as well as for navigation and a dial for making selections. There have been times when I've wanted to use the dial to change something but instead you must use some other type of control. Unfortunately it can be quite slow to sift through captured data with the provided controls and you find yourself zooming in and out to find something but you can search for data on a bus or jump to a particular part of the captured data (known as a delay).

I'm used to using a logic analyzer that displays captured data on a PC but with the LA-100P you can actually see the signals as they change (of course, with fast changing signals you are not actually seeing them in real time as such). This is made possible using the Run/Stop button which is much like what you would find on a typical oscilloscope.

A number of onscreen cursors have been provided, which can be switched on and off, and are used to measure the time between changes in signals such as when one goes from high to low. While cursors are turned on the measurements are displayed between the various cursors ansd a button chooses between the cursors but it isn't obvious which one is selected for moving. While it would not be difficult to calculate, oddly there are no frequency measurements, something that even my cheap Saleae logic analyzer does.

One of the biggest disappointments of the LA-100P, something I would have thought to be standard on most logic analyzers, is that there are no built-in protocol analyzers. This means that if you are working with SPI, or something similar, you would have to decode the values manually. The newer version of the utility software (V2.06) does support protocol analyzers but I cannot get it to work with the LA-100P. It would have also have been useful if the LA-100P had a second USB port for saving waveforms to a flash drive. You can however save waveforms and the system setup (signal names, etc.) to internal memory in 1 of 5 slots; data can also be transferred to a PC.

Bundled with the LA-100P was a mains cable, so-called 'printer' USB cable, 4 data pods, 4 lead sets, software/driver CD (V1.0), 36 test probes (which attach to the test leads) and a printed manual in Chinese. Fortunately you can view an English version online:

http://www.leaptronix.com/English/PDF/LA100P_en(960321).pdf

Another disappointment with the LA-100P is the software which is very basic and although it runs on modern versions of Windows (e.g. Windows 7) the driver provided for the LA-100P is 32-bit so the driver won't install. To try out the software I had to run a virtual version of Windows XP (Windows XP Mode) on my Windows 7 desktop but for long term it would probably be better to obtain an old computer running XP.

If you do use Windows XP Mode, after installing the LA-100P software, connect the LA-100P to your computer via USB; on the LA-100P press Utility and select PC link, then launch the Logic Analyzer software. In the virtual machine click USB from the top and select to attach the LA-100P. If you want to save waveform images using the virtual machine you must enable integration features and then pick a drive to share (Tools->Settings) to make it easy to transfer files between the virtual machine and your actual PC. Note that to use integration features, XP must use a password for the account.

You can view the Logic Analyzer software below:

The area with the waveforms is basically what is on the LA-100P itself and the logic analyzer's screen will update in 'real time' as you interact with the virtual buttons of the software. As you can see from the image there are only basic functions; no importing/exporting for example, and scrolling is slower than using the controls on the LA-100P. There is a newer version of the software, V2.06, which can be downloaded from the Leaptronix website but it isn't compatible with the LA-100P.

In summary, although a specialist piece of test equipment with a 'high' number of input channels, the LA-100P is greatly lacking for the price. That said, it did greatly assist me in debugging my simple computer I was building and will continue to be of use as I now solder the computer. If the PC software had made up for the features missing from the actual unit I would think more highly of the LA-100P.

Saleae logic analyzer

A logic analyzer is very useful when you want to debug a digital circuit to find a fault or if you want to learn better how a particular digital circuit functions. However, while a logic analyzer will show you signals with clear, clean transitions which is good for analyzing, an oscilloscope will show how the signal actually looks and is more useful for determining electrical errors rather than logical errors. When considering buying a logic analyzer some important things to consider are: number of input channels, input bandwidth, memory depth, maximum sampling rate, and whether the logic analyzer works stand-alone or needs a computer to operate.

I got a Saleae logic analyzer for about £9 from ebay (although it appears to be a clone) which included a USB cable and some wires for connecting the logic analyzer's inputs to the circuit under test. The logic analyzer has 8 channels and has an LED that can be used as a logic level indicator for channel 1 (using inverted logic, that is, the LED is on for logic 0). As the logic analyzer uses a Cypress CY7C68013A microcontroller the channel inputs are 5V tolerant.

A photo of the logic analyzer can be seen below:

It has a connector on the right-hand side to access the channels along with two ground pins. There is an LED that lights up under the PWR label when connected to a PC and the logic level light is below the CH1 writing.

To try out the logic analyzer I first downloaded the software from http://www.saleae.com/logic under the downloads section; the software is available for a number of different O/S's. I then connected the logic analyzer to my computer via the USB cable; I used a 64-bit laptop running Windows 7 and I have also been able to get the logic analyzer running on Windows 10, Windows detected the device and installed the drivers but when I tried to capture data from the logic analyzer the computer froze until I removed the USB cable. It turns out that to be able to run at 24Mhz certain settings have to be changed so I switched to the lowest speed which did not affect the computer negatively. Note that it is a good idea to set the sample rate to at least 4 times the maximum frequency that a signal changes in a test circuit to ensure it is captured correctly.

As a simple test I connected a 555 astable running on about 5V to the logic analyzer's first channel and then captured the data stream. What I got was as expected; a digital square wave pulse with timing close to what I had calculated.

Below is a screenshot of the Logic program (V1.1.15) in which I have captured a number of signals while testing a computer I have been building:

On the left you have the channel numbers 0 to 7 (labelled as 1 to 8 on the actual logic analyzer) along with the channel names which can easily be edited by clicking on them, and the four selectable trigger conditions are to the right of the channel names. The captured waveforms are in the middle of the window with timing intervals at the top and to the very right are the measurement and analyzer sections. The measurements section will update as you move your mouse cursor across the waveforms and by clicking 'T1' or 'T2' you can set up cursors and get timing information based on their positions. In the analyzer section you can add 1 of 10 protocols to automatically decode the data that has been captured; the protocols that can be analyzed include CAN, I2C, 1-Wire and SPI.

Along the top left there are drop-down selections to choose the sampling rate from as little as one million to as high as 10 billion-this is the advantage of a logic analyzer that is connected to a PC, that it can use the computer's resources for big captures. The other selection is the frequency and there is a button to start/stop the capture or to start a simulation if no logic analyzer is connected.

While the software does the basics well and I've so far only had it crash on me a couple of times, certainly in the version I have it is lacking in a number of much needed features. While the channel count is low (just 8) it would have been helpful to be able to group a number of channels into a bus and then trigger should that bus contain a preset value.

Since the previously mentioned example, I've used the logic analyzer to help debug projects such as an FPGA board that I was interfacing with an LCD. Because the LCD wasn't working I was able to use the logic analyzer to monitor the control signals from the FPGA to the LCD and check the timing. Because of the limited number of channels, however, I've since had to buy another logic analyzer.

Thandar TA1000 logic analyzer

Although I already had a 32 channel logic analyzer, the Leaptronx LA-100P, I found it to be very lacking in features and so had looked into get another logic analyzer. I took a risk in getting the Thandar TA1000 for about £60 from ebay as I could find no information about it online but I'm pleased that I did buy it. The Thandar TA1000 has 32 channels and operates at 100MHz (8 channels only at 100MHz) and has 1K data storage depth. It supports both internal and external clock with glitch support and information is displayed on a 17 cm green only CRT but also has black & white composite out (which oddly will not work on my capture card or upscaler but does work on my TV). My Thandar TA1000 is running software V2.

Since I couldn't find a manual for the LA-100P I had to work out how to use the logic analyzer but it's fairly straightforward to use. I did find a scanned article online dated late 1988 which mentions the TA1000 and it mentions that it has CMOS battery backed RAM storage, and that it was priced £1,799. On a related note, Thandar are an offshoot of Sinclair Electronics.

Here is the TA1000 showing the output of a simple 555 astable:

There is a single power on switch on the front and the system boots quickly with responsive menus although you do get flickering when pressing buttons. Under the screen are 7 context sensitive shortcut keys and to the bottom right there are the clock and 4 port connectors which either connect to a standard pod (with connectors for 4 data ports and clock) such as the one in the photo above under the breadboard, or a 68000 CPU pod with a clip that first directly on to a standard DIP 68000 IC. On the right side are the cursor keys, menu keys (config, trigger, timing, list, utilities, options), and the alphanumeric keypad (for entering a letter that is not a hex value you press shift and then the letter you want) including RUN and STOP keys.

The unit is not too heavy with the main weight on one side where the CRT is and the fan isn't too distracting. The handle can be positioned to angle the unit upward; to rotate the handle you have to press in at the sides and then move the handle. At the back are the following connectors: RS232, centronics, IEEE-488, composite video, trigger out, and restart out.

I did a simple test using a 555 astable to generate a clock pulse of approx. 7.8KHz (as in photo above) with the 555 output connected to D0 of data pod 1. I set the clock to internal at 100KHz and glitch off and measured:

7.9 KHz

127us period

High 88us

Low 39us

Using my modern, USB logic analyzer I got the same values.

To trigger, I set word 1 to trigger when D0 goes high, on the trigger page which you get to by pressing the Trigger menu button. On the trigger screen you use the cursor keys to select the words for each data pod (there are multiple groups of words so there can be multiple triggers) and the shortcut keys to pick the characters (bit numbers). To set the trigger value you type type 1 for active high, 0 for active low, or x for don't care. Each group of words for a data pod can be changed between octal, hex and binary using the shortcut keys and it converts between the bases (sometimes it doesn't, however). You can also set up more advanced triggering such as sequences and clocks.

For the 555 test, I powered up the circuit and then pressed RUN and after the display updated I pressed the STOP. I then later worked out that by putting the mode to manual it would only trigger once when RUN has been pressed. I pressed the TIMING button to see the waveforms and to check the timing. There are two cursors, which are referred to as cursor and marker, selectable using the end shortcut key. Timing  is shown between the cursor and marker, and cursor and trigger. The arrow keys left/right can be used to move the currently selected cursor which moves very slowly but you can type in the required cursor position as a 4 digit value. To move the page up or down press the cursor up/down keys to scroll through all the data lines. On the right side of the signal lines the 0/1 value is shown for the current cursor position. There are also two shortcut keys to zoom in/out (change timing per division).

To edit the labels press the left shortcut key to turn edit on and then press cursor up/down to select a label. The shortcut keys let you set default labels, insert a label, delete the currently selected label, scale, and move to the next/previous label. E.g., if I delete a label, I can select Insert and then press next or previous until I get to the correct label. You can also have multiple versions of the same label by selecting a label and navigating to the same label as another using the Next/Prev shortcut keys or by pressing Insert and then using Next/Prev shortcut keys to get to the same label as another. There is also a SCALE line which has fixed markers that don't change as you alter the timing per division; you can add another SCALE marker as you would any other label.

While in edit mode you can press arrow right to get to another menu. SET MARKER moves the marker to the current cursor position. FIND: press FIND shortcut key and then use the arrow up/down keys to select a signal and use the keypad to type either 0, 1 or x (don't care) to set up a search. Every time you press EXECUTE FIND the cursor will jump to the position in memory that satisfies the search value. There are also options to jump to the trigger, cursor or marker using EXECUTE FIND.

On the config screen there are the options to choose between Auto and Manual using the shortcut keys. When set to Auto it will continually capture data for every trigger but on manual it  will capture just once when you press the RUN key. To set signal threshold: enter the trigger screen and use arrow key up to select a threshold option and arrow keys left/right to select the threshold for a data pod. Use the shortcut keys to increase/decrease the threshold or the TTL shortcut key to set to TTL level.

The Utilities page lets you set up RS-232, printer, GPIB (IEEE-488) - (not installed on my unit), and lets you save and load settings/acquisition data. There are 2 directories, A and B, each of which has 2 store locations and 8 set-up slots. You can name the store and set-up slots by selecting with the arrow keys and typing a label using the keypad. It does not show the currently selected character and you have to get used to pressing shift to enter letters from G onward. There are 7 menu options for loading & saving, shortcut keys for selecting what slot to use and the EXECUTE shortcut key must be pressed twice to load/save, which happens instantly.

To set up busses: go to the trigger screen and select to edit the labels. Type a letter (use shift for letters after F) and repeat the letters to group bits into a bus using that letter. When you type a different letter it will group the bits using that letter so each bus effectively has a single letter for its name. The system will warn if a bus is split across multiple data pod lines, which is helpful. After you set up the busses the Timing and List screens will reflect the labels. On the List screen, which brings up a listing of values for each location captured,  you can set up exactly the busses you want displayed and show different combinations by pressing the arrow key left/right when not in edit mode. To set the busses to display enter edit mode, use the arrow key up to select the labels, and arrow key left/right to select a label and type the letter representing the bus to display. You can also insert/delete busses, and in edit mode you can change the format between binary, octal, hex, decimal & ASCII when either heading is selected with arrow the keys left/right.

Inside the unit is the main power circuit, the high voltage circuit for the CRT, the CRT itself, the top logic board and the bottom logic board. Both logic board look to have all the IC's in sockets, perhaps to make servicing easier. Without taking everything to pieces I had a look at the top logic board which has some interesting chips:

3 ROMs; 2 with V2 written on them and the 3rd with V1.07 and 68K written on them. There are 3 sockets spare for additional ROM's.

Z80B CPU (6MHz).

cxk5864 8KB SRAM x4.

scn2674bc4n40 advanced video display controller.

Thurlby LA160 logic analyzer

While I already had a number of logic analyzers I noticed the Thurlby LA160 on ebay selling at a reasonable price with accessories (about £85) being sold by a charity, so I snapped it up. As well as the analyzer itself it came with the LE32 32 Channel Extender and LC-02, LC-03 and LC-05 data pods but no power cable, which is just a standard 3-pin IEC mains connector. I had done some research beforehand and learnt that the LA160 has an internal battery that tends to leak (please see the Repair section) and took that into consideration with regards to a potential repair.

The LA160 (also, LA-160) was originally made by UK based company Thurlby and there were two main versions of the logic analyzer, LA160A which has a 10MHz internal clock and LA160B the 20MHz version; the version is indicated at the rear. As well as the original versions of the LA160A and LA160B from Thurlby, RS Components (also UK based) produced their own versions, with part numbers 611-290 and 611-818 respectively. The RS branded version has 'RS' and the part number on the front near where it says 'Thurlby' and there is also writing above the Escape, Modify and + keys, ‘10MHz Logic Analyzer’ or ‘20MHz Logic Analyzer’ depending on the exact model. From looking at ebay listings, it appears that Interplex electronics did their own version for the US market as they are a US based company, using model number IE-1620, designed to run on 110V, as indicated by the markings under the power switch. On the IE-1620 it has written 'INTERPLEX ELECTRONICS IE1620 logic analyzer' below the display with no mention of Thurlby.

This Electronics & Wireless World June 1986 extract has prices for the LA160:

https://worldradiohistory.com/hd2/IDX-UK/Technology/Technology-All-Eras/Archive-Wireless-World-IDX/80s/Wireless-World-1986-06-OCR-Page-0082.pdf

It lists LA160A as £395 and LA160B as £495 (not including 15% VAT).

On the second page of Electronics & Wireless World Feb 1987 it lists the LA160B-PC for £620, which is the logic analyzer, cable and software for interfacing with a PC:

https://worldradiohistory.com/UK/Wireless-World/80s/Wireless-World-1987-02.pdf

The manual for the RS versions of the LA160 (but should still be useful for the other variations) can be viewed at:

https://archive.org/details/bitsavers_thurlbyLA1heetOperatingManual_273774

At the time of writing this, only the RS version was uploaded.

Here is a photo of my LA160B, which is an original Thurlby version:

The LA160 has at the front an 8-digit red 7-segment LED display, various buttons (keys) as part of the membrane keypad, and 24 inputs (16 data and 8 control) across x3 female DA-15 connectors labelled Data 15-8, Data 7-0, and Control inputs, also labelled 'A', 'B', and 'C' respectively. 

Here is the view from the back:

At the rear is the power switch, fuse holder, 3-pin mains in (as I have the UK version it's rated AC 240V 48-63Hz), oscilloscope BNC outputs trigger and vertical, and a DE-9 female aux. connector (for connecting to a printer, or PC, also used by the LE-32 extender module). Note that on my LA160 above the large label bottom right has 'B' stamped as expected but the 'fitted with' part does not have 'LR-64' stamped in the provided box even though the extended facilities ROM has been fitted (more about that further on). Provided on the underside is a metal bar that can be pulled out to angle the LA160 at a better viewing position.

When powered on, the LA160 should briefly display the model number (LA160A or LA160B) and the ROM version on the LED display before displaying 'ready', there may be also some 'decimal point' LEDs lit on the display, representing settings/modes,  which are listed under the display. An internal buzzer sounds when a key is pressed but there appears to be no way to disable it and another issue I've found is that because of the limited display it's not hard to mistake a '6' for a 'b' as there is only one segment difference between them.

The settings are retained even when the LA160 is powered off since it contains a rechargeable battery although the LA160 will need to be used frequently otherwise the battery will run down and the settings will be lost. Aqusition data is lost when the LA160 is powered off and is random when powered on, the captured data can be sent to a PC via RS232, originally a PC connection kit was sold for that purpose.

A very useful feature of the LA160 is its logic monitor mode, which shows in real time the state of the data inputs without needing to be triggered, in sync with the sample clock waveform. To access the logic monitor mode press Shift then Monitor but note that the oscilloscope display will not be available.

When using the internal clock to do an acquisition the clock period can be altered by first pressing the CL Period key and then using arrow key left/right to select one of the 15 options. On the LA160A the period defaults to 100ns (10MHz) and can be altered down to 800uS (1.25KHz) and on the LA160B the period defaults to 50ns (20MHz) and can be changed down to 400uS (2.5KHz).

The only video I could find online featuring the LA160 can be viewed below:

The video gives a brief overview of the LA160 and a look at the internals.

Even though the LA160 has a limited built-in display, Thurlby provided the means to connect an oscilloscope, allowing the data waveforms from either the Acquisition or Reference memory to be displayed, as you would expect from a logic analyzer with a CRT or LCD screen. Note that there is no scope output while data's been captured.

It doesn't explain much about how to connect an oscilloscope to an LA160 in the manual and I only found one image online of the LA160 connected to a scope:

http://www.bitsavers.org/test_equipment/thurlby/LA160/la160-withscope.jpg

However I will share with you how I got it working on my scope (SDS1052DL DSO):

To connect the LA160 to an oscilloscope you will need two coaxial leads terminated both ends with BNC connectors, with one lead running from the LA160's vertical output to the oscilloscope's channel 1 input, and the other from the LA160's trigger output to the scope's channel 2 input (or external trigger input). Both channels should be set to 1X, DC coupling, 100mV/div (for 8 channels only, 200mV/div for 16 channels), but have channel 1 only on, set its vertical volt position to -404mV, and you may want to turn bandwidth limit on to clean up the signal. Set the scope to trigger on channel 2, trigger type as negative pulse, trigger when pulse width less than 32uS, and trigger voltage to 68mV. Set the time base to 25uS and the trigger delay to 301uS. If you are using a digital scope then turn persist to 5 seconds and turn the waveform intensity down (I used 36% on my scope) otherwise you'll see a single waveform much brighter than the rest.

Unfortunately, despite trying many different settings, occasionally the waveform display drops out (both when showing 8 channels only and all 16 channels) and then is quickly redrawn so I'm not sure if there are better settings or there is a fault with the LA160 but you can always press run/stop on your scope to pause the screen.

We will now look at the scope waveforms but note that they are also downloadable from the bottom of this page.

When you power on the LA160 you should see the set-up waveform displayed on the scope as a means to centre the waveform, allowing you to make any adjustments on your scope as necessary:

Even when the set-up waveform is not being displayed it can be called up by pressing and holding the Scope Ref key; additionally the reference waveform can also be called up by pressing Escape, meaning that Scope Ref doesn't need to be held down.

To see acquisition data as waveforms on the scope, after doing an acquisition press the Data key. By pressing the 8/16 key you can cycle through displaying the high 8, low 8, and all 16 data inputs; the following screenshot shows the upper 8 channels:

Data input 15 is at the top and working down, data input 8 is at the bottom; the centre of the screen shows the values at cursor position 0 of the LA160 and pressing arrow left/right on the logic analyzer will cause the waveforms to shift accordingly. Notice how the channels are grouped into fours to help with reading the data, however, not all of the individual waveforms line up with the graticule lines, which can cause difficulty determining if a signal is at logic 0 or 1 if there is little activity (the Scope Ref key can help though). Another issue is that 51 values are supposed to be displayed for each data input, 25 before the value at cursor 0 and 25 after, but on my scope I was only able to get 36 to show (a time base limitation of my scope perhaps).

Here is an image showing all 16 data channels displayed:

Don't forget for 16 channel display to set your scope's channel 1 to 200mV/div. As with the 8 channel only display, when showing all 16 channels they are grouped into fours.

You may be intrigued how the LA160 outputs 8 or 16 channels of information to an oscilloscope and indeed it's quite clever. Here is what the signals look like that the LA160 outputs on its scope connectors when set to show 8 channels only (set to trigger on channel 1, trigger type rising edge):

The yellow signal is the scope vertical output of the LA160 and the blue is the trigger (I moved channel 1 up a bit - I forgot to set it back to 100mV/div). As can clearly be seen, the vertical is a staircase waveform (the voltage increases in steps, from 88 to 1.54mV) and there are 8 steps when the LA160 is showing 8 channels and 16 when showing 16 channels. The logic level for each memory cursor position of a channel (12uS in width) is superimposed on each vertical step (a bit difficult to see due to having to zoom out to show all the steps). The logic levels (0/1) for each channel is at different voltages to the others and thus a different position on the scope's screen. Due to the persistence of a CRT in an oscilloscope or the emulated version on a DSO and the trigger and time base settings it will look like all 8 (or 16) channels are on screen at once.

The trigger coincides with the start of each step and lasts 31uS, ranges from 40 to 504mV, and there is 1.25mS between each trigger, and a data value (logic 0/1) is.

Here is a zoom in showing a vertical step more clearly:

Let's now look at various LA160 accessories:

LC-02 variable threshold data input pod

The LC-02 data pod by Thurlby (RS model 612-467, Interplex electronics model IE1620-1) is recommended as one option for interfacing with the LA160 as connecting directly to its inputs may damage it should the input voltage limits be exceeded. It is similar to the LC-01 buffered input pod (RS model 611-307) but with variable input threshold. The specifications for the LC-02 are:

Working input range +/-15V.

Maximum continuous overvoltage +/-100V.

Impedance 1M/8pF.

The data pod has a switch to select the input threshold voltage between high, low, and TTL (+1.4), making it compatible with a range of logic level standards. When the switch is set to either high or low, the data pod's potentiometer selects the voltage. Unconnected inputs pull up to +4V.

On my LC-02 the pinout of the inputs aren't listed but I found online that the RS version does have the pinout on the data pod. Looking from the top with the pod facing you (left to right) the pinout is as follows:

GND

7

6

5

4

3

2

1

0

GND

Since the inputs are on a male connector it's easy to interface with, you can easily use 'DuPont' wire with female connector one end.

I only got one LC-02 with my LA160 but it allowed me to test the data inputs of the logic analyer eight inputs at a time. With power off to the LA160 I connected the LC-02 to the Data 15-8 connector. After power on, when 'Ready' was displayed, I pressed Shift then Monitor, causing 'FFFF h' to be displayed. Then, taking GND to each input to the LC-02 in turn, I checked the display updated accordingly. For example, for input 0 to the data pod, when taken low the display will show 'FEFF'. You can press Octal, Decimal or Binary to change how the values are displayed with Binary being especially useful. In Binary mode, logic 1 is represented by segment b only on, logic 0 is shown as segement C lit only. You can also press Mixed for binary for the upper byte and hex for the lower byte. I then tested the Data 7-0 inputs in the same way but by moving the data port to that port 7-0 (with the logic analyzer powered off) and checking the lower byte on the display. Note that you press the Escape key to come out of monitor mode.

The monitor mode is a simple way to see instantly the state of the inputs, provided they don't change too quickly, but typically you would want to capture the input states over a period of time by doing an acquisition. As a test, I used an Arduino to generate a digital square wave of 494Hz (of course you can use whatever signal generator device you want), with the digital pin carrying the signal connected to one of the inputs to the LC-02. I pressed the CL Period key and used the arrow key to change to 1uS and then pressed the single key to start the acquisition and FULL was quickly displayed. You can set up more complicated acquisition modes but for the simple test it wasn't necessary; if you find that pressing the Single key doesn't work then try doing a reset by holding Escape until the countdown ends but of course you will loose any changes you made to the configuration.

After the acquisition you can use the left/right arrow keys to move through the captured data but to move through the data quicker, you can press the Cursor key then Modify, type the required cursor position value using the numerical keys, then press Data to return to the data display at the chosen position. You can alsp hold the Cursor key to return to position 0. In hex mode, the 16-bit data value is shown on the left and the cursor position on the right. Looking through the captured data by changing the cursor position I saw that the data value changed from FF to FE, confirming that the signal change was recorded correctly.

LC-03 module

Also included with the LA160 when I bought it was a LC-03 module but there is no mention of it in the archived online manual. It has three cables (labelled as 'A', 'B', and 'C' on the PCB) that each terminate in a DA-15 male connectors that look like they interface with the data and control inputs of the LA160. The PCB is labelled as 'LC03’, has a small green sticker (perhaps a pass mark), x3 resistor arrays RP1 to RP3, x2 8-way male headers with labels, a DB-25 female connector (CON2), and a 24 pin female micro ribbon (so-called Centronics) connector (CON1). There is a single link component (effectively a short) 'LINK', which has 'DAV' written on one end and 'CLK' on the other; since the PCB is single sided it is necessary to use some kind of wire link on the component side where tracks can't otherwise cross.

A clue as to the intention of the module is that ‘IEC625’ is written above CON2; IEC625 is synonymous with IEEE 488 (GPIB/HP-IB) but it is CON1 (the 'Centronics' connector) which is the type of connector commonly used with GPIB, thus it's likely the module was used to debug GPIB equipment.

I have reverse engineered the LC-03 as there was no information about it online, the schematic follows but is also downloadable from the bottom of this page. I've tried to be as accurate as possible but of course there may still be errors.

D0 to D15 are the digital inputs to the LA160, which I've matched with the naming convention used in the service manual but have used GPIB signal names elsewhere. The DAV signal being connected to CLK makes sense since DAV signals valid data. Since they've provided various control signals on the two 8-pin headers, perhaps the intention was for wire wrapping so that various conditions could be set up using the GPIB signals.

The Centronics pinout (CON1) matches the GPIB standard, however, the PCB wrongly lists 'EOI' as 'EO1' but I have put the correct version on the schematic. Both CON1 and CON2 carry the same signals but with a different type of connector and pinout; it is unknown the purpose of the DB-25 connector (CON2), as commonly GPIB used the Centronics connector, however, it may have just been an alternate form of connector for convenience. Since GPIB typically uses stackable connectors, we can assume that the GPIB device under test could be connected to a computer to be controlled as normal but with the LC-03 also connected.

LE-32 channel extender module

The LE-32 sits under the LA160 (see photo below) and increases the number of data inputs from 16 to 32, making the logic analyzer more practical for use with a simple computer system (such as one based on a 6502 or Z80, for example). Like with the LA160, RS also made their own version of the LE-32, which has on the front top-left, 'RS Components Ltd.' and the model number '612-350'. On both versions on the front bottom-left it says 'Thurlby' and 'LE32 thirty two channel extender module'.

At the front of the LE-32 there is a male IDC 40-way connector labelled 'Data and Control' carrying 32 data and 6 control inputs for connecting an external module, such as the LC-05, but you could connect directly to the LE-32 using an IDE hard drive ribbon cable but it's recommended to provide some protection to the inputs. The LE-32 also has two individual female connectors for additional control inputs, one marked as TRE2, and the other CLE1, as well as three ribbon cables terminated each with a DA-15 male connector which plug into the 'A', 'B', and 'C' mating connectors on the front of the LA160. As with the LA160, the LE-32 also has a metal bar at the bottom to raise the unit up and at the rear a 15-way female D-type connector marked 'aux. outputs' and a matching male DE-9 which plugs into the mating connecting at the rear of the LA160 (which appears to act as a passthrough). At the rear of the LE-32 it will say whether it's been set for a LA160A/B, for example, on mine it says that it's set up for use with LA160B (with the 'A' option crossed out).

When using the LE-32 with an LA160 its oscilloscope output (see the main LA160 section for further information) still behaves as normal but switching between group A and B will show the data for the selected group only (that is there is still the limit of 16 channels maximum that can be displayed at once on the scope).

I used some simple tests to check that the LE-32 was working by making use of the LC-05 module (see the next section for more information about the module) for the convenience of the provided connections. With just the LE-32 connected to the LA160, and the LC-05 connected to the LE-32, when I pressed the Single key the display showed '0000 h'. Unlike with the LA160 on its own, in which its data inputs default to high, with the LE-32 connected they default to zero. Additionally, the LA160 no longer triggers just by pressing Single since it's waiting on the external clock, which must be provided via a circuit connected to the LE-32. We can see this by pressing the CL Source key, which results in 'Etn only' being displayed, as does pressing 'CL Period'.

The 32 channels are split into group A (the first 16) and group B (the other 16) if in 32 channel mode since the display can only show 16 bits at a time. You can switch between the two 16 channel groups, group A and group B, by pressing the shift key and then right arrow to switch to group B, the (B) LED will be lit. Or press the Shift key and then left arrow to switch to group A, the (A) LED will light.

As a test, I provided a clock pulse of 1MHz to the CLKA/CLKB input to the LC-05 using an Arduino and when I pressed the Single key, data was captured (all zeroes since I had nothing else connected to the LC-05). Note that when the LA160 is in LE-32 mode, the cursor only goes up to 499 since we double the amount of data is being stored (32 bits instead of 16).

We can enter monitor mode in the usual way even with the LE-32 connected and enabled but the display only displays the value for channel group A, even if you switch to group B and then enter monitor mode. However, monitor mode allowed me to easily tested all of the A inputs by taking each A input on the LC05 to GND (note that no clock input is needed in monitor mode). Because of the inverters on the LC-05, taking an input to GND results in logic 1, whereas a not connected input results in logic 0 (you can, however, set the LA160 to use negative logic). After that I tested all of the B inputs by pressing the Single key (which does require the clock input) by doing an acquisition, making sure that the captured data reflected that state of the B inputs.

Now for a more interesting test, let's view a 6502's reset process using the LE-32 and LC-05 configured for a 6502 but without RAM or ROM as to keep things as simple as possible. I used a Rockwell R6502 from 1984 that I had at hand but please check the datasheet for the version you have and adapt if necessary as different versions of the 6502 changed the behaviour of some pins or added additional pins that on the original weren't functional and the reset behaviour may be slightly different.

Place the 6502 into the LC-05 (either socket) and supply +5V to its power pin (pin 8), which you can get from the LC-05 using a test clip and wire running from one of the 74LS04 ICs (pin 14) or you can use an external power supply with a common GND connected to the LC-05. You will also need to connect the 6502's RDY (2), IRQ (4) and NMI (6) pins to +5V through 3.3K resistors, you may need to use breadboard to do so. You will need to also connect all of the 6502's data bus pins (26 to 33) to GND though 3.3K resistors, which will force a BRK instruction (opcode 0) when the CPU is fetching an instruction and address zero when reading an address value.

We need to provide a 1MHz square clock pulse to the 6502 clock input (pin 37) and a reset pulse lasting at least 2 clock cycles to the reset pin (40), for this I used an Arduino Uno but of course you can use an alternative circuit, just make sure there is a common GND between it and the LC-05. The Arduino code follows:

#include <digitalWriteFast.h>

//Outputs

const int CLK=9;      //Connect to 6502 clock input (pin 37)

const int RST=3;      //Connect to 6502 reset (pin 40)

const int RST_TRIG=4; //Briefly connect to GND to trigger reset pulse

const int ocr1aval=7;   //1MHz clock

void setup() {

  //Set up I/O

  pinMode(RST_TRIG, INPUT_PULLUP);

  pinMode(RST, OUTPUT);

  digitalWrite(RST, HIGH);

  pinMode(CLK, OUTPUT);

  //Set up clock

  TCCR1A=((1 << COM1A0));

  TCCR1B=((1 << WGM12)|(1 << CS10));

  TIMSK1=0;

  OCR1A=ocr1aval;

}

void loop() {

  delay(200);

  if (digitalRead(RST_TRIG)==LOW) {   

    digitalWriteFast(RST, LOW);

    delayMicroseconds(7);

    digitalWriteFast(RST, HIGH);

    delay(1000);

  }

}

The Arduino's digital pin 9 (CLK) is connected to the 6502's clock input and produces a 1MHz square pulse and the nice thing about the way the clock is produced, as seen toward the end of the setup() function, is that the CLK will continue to run independently, allowing up to run other code in the loop() function without having to update the pin.

For the reset pulse I'm using the digitalWriteFast library so that the Arduino switches its I/O much quicker than normal, in this case the RST digital pin (3) which is connected to the 6502 reset input. In the main loop, if RST_TRIG (digital pin 4) becomes low then we take TRIG low and after 7 microseconds we take it back high again, completing the reset pulse.  The delayMicroseconds() function isn't really accurate, especially at low values, so I used a value of 7 to be safe, which I measured with my oscilloscope results in RST being low for 5.88uS, equivalent of almost 6 6502 clock cycles, which satisfies the requirement of the reset pulse lasting a minimum of two clock cycles. Taking RST_TRIG low again will reset the CPU once more, as long as at least a second has passed since RST_TRIG went low, since I added a delay just to try to avoid a false retrigger.

The process for capturing the 6502 resetting is to power everything up and set the LA160 to use negative logic for both group A and B (select group A, press the Logic key, then Modify, then ‘-’, repeat for group B). Next, set the LA160 trigger word for group A only to be 0xFFFC (the 6502 reset vector low byte) by selecting group A if not already selected, then press the TR Word key, then Modify and type 0xFFFC (the group B trigger word should be 'don't care'). Press Single (you should see some activity on the display) then trigger a reset pulse (in my case the Arduino waits for an input pin to go low and then sends the reset pulse to the CPU). After a few seconds the display should show FULL and you will be able to view the acquired data, group A for address and group B for data and control.

In the following table you can see the data that was captured (I manually wrote down each value) with comments - please see the LC-05 section if you haven't already to view how the 6502 is connected to the LC-05. Note that the cursor is shown in the table as a decimal value but all other columns use hex values (0x). A negative cursor value allows us to see acquired data before the acquisition begins, which is very useful as it exposes what happens before the 6502 fetches the reset vector.

Group A is the CPU address and group B low byte is the CPU data, group B high byte is control, with the only bits of interest for this test being:

Bit 8 SYNC high when the CPU is fetching an opcode, represented as an 'S' in the SYNC column.

Bit 9 R/W high when the CPU is reading from memory, represented as an 'R' in the R/W column, low when the CPU is writing to memory, represented as a 'W'.

A 6502 doesn't immediately reset when its reset pin transitions from low to high, indeed, the 6502 datasheet states that it takes six clock cycles before it fetches the reset vector from memory locations 0xFFFC and 0xFFFD but we'll get into that more shortly. Jump to cursor position 0 and you'll see that the CPU fetches the reset vector low byte from memory but as we don't have any memory attached to the CPU and simply a bunch of resistors to force a zero, the CPU will read in zero. Similar at cursor position 1 it'll read in the reset vector high byte from 0xFFFD but again it will fetch zero.

The CPU will then jump to address zero (cursor position 2) and it will fetch the opcode 0x00, signified by the SYNC signal going high, but of course it will see a zero, which is the BRK instruction. At address 1 (cursor position 3) the CPU will fetch the so-called BRK signature byte, which it will see as zero. At cursor positions 4 and 5 respectively the CPU will store the program counter+2 high byte (0x00) and low byte (0x02) to the stack, which happens to be addresses 0x01C0 and 0x1BF but we don't have any memory attached so the values are discarded. Even though the CPU has been reset you'll notice that the stack address isn't 0x01FF as you may expect and that's why it should be initialised early in the reset routine as it otherwise will be a random value (at least on early 6502 versions).

At cursor position 6 the status register (value 0x36 - may be slightly different for you) is put on the stack at address 0x01BE and then at cursor position 7 and 8 respectively the CPU fetches the IRQ/BRK vector high byte and low bye, which it'll see as address 0x0000. This is a neat thing about the 6502 in that just forcing a zero on the data bus means we have created a loop that we can monitor as the 6502 will jump back to address zero (cursor position 9) and execute BRK again, with the only difference being that the stack pointer addresses will be different.

Now that we've looked at the CPU executing BRK after fetching the reset vector it's interesting to look at what happens in the negative cursor positions, starting at -5 in which the CPU appears to execute a BRK instruction but rather than ending with fetching the IRQ/BRK vector it gets the reset vector instead. We can also see that the stack pointer was decremented twice before the CPU jumped to the reset address so again it's good practice to set that stack pointer to a known value in your reset routine.

LC-05 buffered personality module

The LC-05 buffered personality module, or at least the version I have, is configured for testing a 6502 based system (there is also a Z80 version) by capturing data, address and control signals, and is designed to be connected to the LE-32 extender module. It can be seen below:

As can be seen top left of the PCB, there is a 'Thurlby' sticker and another, 'LC05 6502', and 'ISS 1' is printed on the board. There is also a 40-pin interface ribbon cable (CON 1 on the PCB) that connects to the LE-32, x2 40-pin DIL IC sockets SKT1 & SKT2 (connected in parallel, oddly SKT1 has golden pins, SKT2 silver), and x4 20-way male headers, the 2 furthest from the sockets have matching pinout for CON 1, and the 2 nearest to the sockets are connected like-for-like to the IC sockets. Various wires connect the pins of the 4 headers together (using wire wrapping technique) to configure how the LE-32 is connected to the CPU under test.

Below the second socket there is provision for a switch (possibly for a 4 position DIP switch, as is the case for the LC-04 module) and there are x2 4-way male headers which mirror the switch connections. There are a number of links and only one capacitor, C1, which appears to be 10uF, has been used and is located near CON1; ideally each IC should have its own local decoupling cap. There are x7 74LS04 hex inverter ICs (IC1 to IC7), all socketed (on my LC-05 they have date codes of '84, so likely the chips are original); the inverters, as well as inverting the inputs, provide protection and boost the signals and can easily be replaced if damaged since they are socketed.

The 6502 under test can either be inserted into one of the sockets or an LC-06 test clip assembly can be used, which has a 40-way DIL header plug one end to plug into the IC socket, and a 40-way IC test clip on the other to attach to the 6502 while in the circuit being tested. If the CPU is placed in one of the sockets, however, then an LC-06 or something similar will need to be connected to the other socket to provide signals to the 6502 not provided by the LC-05 (power, clock, reset, etc.).

I have reverse engineered the LC-05, since there's little info about it online, but the details should also be useful for the other buffered personality modules, as likely the same design was used for multiple versions of the LC-05. The schematic that I worked out can be seen below:

The schematic is also downloadable from the bottom of this page. While I've done my best to ensure all of the connections are correct there may still be errors so please double check should you use any of the information. From the circuit we can see that there are some unused inverters as there are 6 in each IC but 42 aren't required; there's no reason why you can't use them for your own purposes by soldering to the IC socket pins directly.

Since the LC-05 is essentially an extension of the LE-32 it has 32 data inputs A0 to A15 and B0 to B15, labelled on the x2 20-way headers. Also, on the headers are 2 GND pins, and 6 control inputs (CE0, TE0, ARM, TE1, CLKA and CLKB).

Here is the complete listing for how the LC-05 is configured for a 6502:

Note: NC=no connection between the socket pin and any of the pins of the 20-pin headers.

The A0 to A15 inputs to the LE-32 match the address inputs A0 to A15 of the 6502. B0 to B7 inputs to the LE-32 line up with the 6502 data bus bits D0 to D7. The remaining B inputs to the LE-32 (of those connected to the 6502) are wired to various 6502 control pins, and the 6502 phase 2 clock out is wired to the LE-32's CLKA/CLKB inputs. We can see that although the 6502 GND pins are connected to the LC-05 GND (pins 1 and 21), the 6502 VCC (pin 8) isn't, likely because the 6502 would be powered by the main circuit that was being debugged.

To use the LC-05 it needs to be connected to the LE-32 channel extender module (see the LE-32 section for more information) and once connected it's a good idea to check at the very minimum that the LC-05 is getting power by using a multimeter to measure the power pins of one of the ICs (7 and 14), which should be around 5V (I got 5.04V). Please see the 'LE-32 channel extender module' section for more information about using the LC-05.

Further information

Thanks to online archived versions of electronics magazines we can find mentions of the LA160, furthering our understanding of using the equipment, and giving a look at other people's views of the logic analyzer.

The Maplin Electronics Magazine March 1992 mentions the LA160 and can be viewed at:

https://worldradiohistory.com/UK/Mapelin/Maplin-Electronics-1992-03.pdf

The LA160 is talked about as part of an article called 'Microcomputing Testing & Fault Finding, starting on page 52. It's a little odd that the LA160 would be used as an example piece of test equipment at such a late date in the logic analyzer's life, however, on the plus side it would have been an established piece of kit.

On the top left of page 55 it shows a diagram of the LA160, LE32, LC05 and 'Einstein' (the computer by Tatung); later in the article they describe removing the Z80 CPU from an Einstein computer and plugging into the personality module which has its own Z80. Bottom right of the page shows the LA160 connected to an oscilloscope to produce a timing diagram display, with the vert output connected to Y1 (channel 1) of an oscilloscope and the trig output connected to the external trigger input, and the timebase set to 100uS/DIV. Throughout the article there are also a number of listings related to the LA160 and LE-32, bottom right of page 78 shows a diagram of the LA160 connected to a PC, and on page 79 there is a printout listing from the LA160 connected to a PC.

Repair

Please do not attempt repair unless you are experienced with working with mains powered equipment - always disconnect the mains before opening up the LA160.

The service manual for the LA160 can be found at:

http://www.bitsavers.org/test_equipment/thurlby/LA160/SINCLAIR_THURLBY_LA160_SERVICE.pdf

Unfortunately the schematics are quite hard to read and there only seems to be one version of the service manual that has been uploaded to multiple sites but nonetheless it's still a great help in repairing the logic analyzer as well as understanding how it works.

The LA160 uses a linear power supply which does add some weight to the system because of the heavy mains transformer but at least the power supply should be easier to repair than it it were a SMPS. Although there are no dreaded RIFA capacitors in the LA160, it's still worth checking for any leaky capacitors - in particular the electrolytic type - when servicing the logic analyzer. If the unit is dead there are two fuses that need checking, one accessible in the rear, the other on the PCB.

LA160 Teardown

If you have the LE-32 extender connected then first remove the cables connecting to LA160 (three cables at the front, one at the back), and then take out the 2 slot head screws ( which are very long), holding LE-32 to the LA160. On my LA160 (which had the LE-32 attached) there were no additional screws. Here is an image of the long screws:

Notice that the head of the top one is discoloured, possibly from the leaky battery in my LA160.

If you don't have the LE-32 connected then remove the four screws from the bottom of the LA160. Then lift up the top part of the case, with side-to-side movement; when putting back, you may find the top part fiddly to get back on.

Inside the LA160 you will find the smaller, upper board and the larger, lower, board underneath. Some ICs are socketed, so replacing those chips is an option should they be faulty. The following image shows the upper board which connects to the lower PCB with a ribbon cable, and two soldered wires (red and black) carrying the power rails:

We can see in the centre the 6502 CPU (IC3), 6522 Versatile Interface Adapter (IC2) below it, 4MHz crystal (X1) above the 6502, buzzer top right, and the 8 digit C.C. multiplexed LED displays bottom left. There are a couple of stickers on the PCB with possibly dates: 9/1/85 & 15/1/85 and since it was made in England that would be January '85. 

To remove the upper board first carefully lift the membrane keypad connector up - gently rock it back and forth, then take out the 4 screws in the corners - note that there are also cardboard washers. To completely isolate the upper board, disconnect the ribbon cable and desolder the red and black wires, but you can carefully work on the upper board with the ribbon cable and the two wires attached save disconnecting them.

A rechargeable battery normally stores the LA160's settings but the type of battery used tends to leak (as seen in the photo above) and as it's soldered to a PCB the acid can damage the circuit. If you're lucky the battery may not have leaked but it's still advisable to replace it even if it hasn't; after removing the battery, the PCB will need to be cleaned and the battery acid neutralized, and any damaged components or circuit traces replaced/repaired. 

In the next photo we can see a close-up of the system ROM (IC6), labelled 'LA160 21' since it's V21, and the ROM extended facilities ROM (IC11) below it, 'LR64 21', also V21:

Both ROMS are 2764 ICs, which are 8Kb x 8 UV ROMs, and below the bottom ROM is a D446D (IC5), which I couldn't find any information about online but I can confirm is the configuration settings RAM, backed up by the battery, and is a HM6116 in other versions of the LA160, and is a 2Kb x 8 CMOS SRAM chip. The leaky backup battery can also be seen bottom right. and a close-up of the leaky battery:

Even the screw was damaged, which was difficult to remove.

In the following photo you can see the leaky battery in my LA160 has even got into an IC socket (barely noticeable when the IC was in place):

If you have a suitable replacement rechargeable battery then that can be installed otherwise you can use a non-rechargeable battery, such as a CR2032, with series diode, which will prevent the LA160 trying to charge the battery. I was fortunate that the battery leaking has not done any non-cosmetic damage to my LA160 but I was only able to clean it up so much.

Here is a photo of the replacement battery I installed with the series diode in the red sleeving:

I will eventually replace the battery with a holder so that the battery can more easily be replaced when it runs down.

After changing the battery and powering on the LA160 it will prompt you with 'SET TYPE'; press 'A' if you have a LA160A or 'B' if you have a LA160B. If you press the wrong key, the only way to set the correct type is to either wait for the battery to go flat or to disconnect the battery. Even if you do set the wrong type the LA160 will still work but the timing will be off.

A photo showing toward the back of the lower board, which contains the power supply, as well as the connecting ribbon cable to the upper board, the oscilloscope outputs, mains in, mains fuse, power switch, aux. connector and mains transformer:

There are date codes on some ICs of '83, others '84. Annoyingly there are no component numbers on the PCBs on either side so I will mention certain components of interest along with their number but note that the numbering starts from one for both the upper and lower boards..

Another view of the lower PCB with the PCB fuse just peaking out toward the right of the transformer:

Notice in particular the two low profile TMM2018D-25 RAM chips (IC6 and IC7) adapted to fit the IC sockets, which I've seen from an online photo is not a one-off. Makes you wonder if they used whatever compatible chips that were available. Bottom right of the photo is the (unplugged) keypad connector and to the right of it a 20 MHz crystal. 

Here us another view of the lower board:

LE-32 Teardown

Remove the screws from the underside of the LE-32; if it was attached to an LA160 then likely there will be only two screws in opposite corners, then you can lift the top piece off (note that the piece goes on one way only). There is a single PCB inside which attaches to the front and back panels (which can easily be lifted up) with three nuts and right angle brackets (two at front, one at back), and the DE-9 aux output connector further holds the PCB in place.

The following photo shows part of the PCB from the front end, with the Data and Control connector toward the middle bottom, and the TRE2 and CLE1 connectors bottom right:

Note the label bottom left near the bracket which seems to indicate a date of '85, and the four resistor arrays (100R) near the Data and Control connector and a fifth top right partly behind the ribbon cable, which are possibly to provide some protection to the inputs. There are also four 74LS374 ICs, Octal D-Type Edge Triggered Flip-Flops with 3-State Outputs, so likely they latch in the data inputs, and the three ribbon cables that connect to the A, B, and C sockets on the LA160 are visible at the top of the photo.

The PCB is marked on the top and bottom as 'LE32-100 ISS 2' (my version - yours may be different), and it contains various logic ICs, x4  potentiometers, a switch with labels A & B (mine is set to B, yours should be set appropriately). 

Multimeters

A830L Multimeter

When I was looking to buy replacement test leads for my clamp multimeter I came across the A830L multimeter on ebay for just £4 and saw that the leads were similar. After buying the meter I found that the leads do work on the clamp meter and it was useful to have another multimeter as there are circumstances when you need to test multiple points in a circuit at the same time. As you would expect for £4 the meter is not exactly something you would call professional but I've seen similar meters for £10 and with less features.

As with the box the meter came in, there is no company name on the meter and if you look online the A830L is associated with different companies or branding. With the multimeter you get two test leads as to be expected, a paper manual, and a PP3 9V battery already inserted into the meter. The test leads have been poorly made and are very thin; they would likely break over time and I wouldn't trust them with high voltages. As for the manual, it has not been too poorly written but it does mention features-such as temperature measurement-that are not available on the meter. It would seem the manual covers features of a more advanced multimeter that is similar.

Have a look at the meter below:

The dimensions of the A830L are 138x69x31mm and the weight is around 170g so very portable and could fit in a pocket. The meter has a decent size LCD which is 3 1/2 digits and limited to 1999 counts, but does have a backlight that is activated by a button. There is also a hold button to freeze the current value, something that I wouldn't have expected on such a cheap device. A transistor tester capable of testing bipolar transistors is present and takes a rectangular form. There are three sockets for the test leads; a common, a 10A unfused for current measurement and a socket for measuring 200mA max (fused) as well as other forms of measurements. A stand can be accessed at the back and the battery is located at the back behind a cover fixed down by a single screw; or changing the fuse you would would need to take out two screws at the rear.

The types of measurements that can be made with the meter are; DC/AC voltage, DC current, transistor, diode, continuity (with buzzer) and resistance. Measurements types are made with a large rotary switch in the middle of the meter which includes an off position. The AC voltage range is especially limited but is the usual case for cheap meters.

For the price the meter is not bad but the poor quality of the test leads is not too encouraging but as a second (or third) multimeter for occasional use you can't complain too much.

Marksman 68276C Analogue Multitester

For my first analogue multimeter I looked for one that was cheap but had the basic functions such as measuring voltage and current. I came across a Marksman 68276C Analogue Multitester which cost about £5 from ebay. The unit itself is small, with dimensions 90x60x30 mm, and thus is very light. As perhaps to be expected, the included paper manual is in poor English with plenty of spelling mistakes.

Here is the meter:

As well as the meter, also included in the package was two test leads, the manual and a single AA battery, which is needed for checking resistances. To fit the battery, the meter's back has to be taken off by removing a screw and the battery only just fits in.

The meter measures DC volts (500V max), AC volts (500V max), DC current (250mA max; fused at 250mA), resistance (x1K), decibels (-20 to +22dB). Each option can be selected using a rotary selection switch in the middle of the meter which includes an 'off' selection. The reading is displayed on the analogue meter using the appropriate scale.

The claimed accuracy is '5 degrees' which I think is supposed to be 5% accuracy.

For measuring resistances there is the single x1K selection so you have to know roughly what the resistance value is. However, the resistance measurement is useful for quickly testing continuity. To make a resistance measurement you need to short the test leads and then alter the adjustment dial at the side of the meter until you get a zero reading, then you can do the actual resistance test.

There is no mention in the manual about the dB scale but it is commonly used to measure voltage ratio.

While the meter is cheap and probably best not used anywhere near the maximum voltages listed it is a convenient tool for making quick measurements and for learning how to read an analog meter.

Mercury MTTR01 true RMS multimeter

After having had managed for a long time with a basic multimeter I did some research into the best budget priced multimeter that would do the usual testing as well as frequency measurement and some form of USB connectivity. I came across and bought the Mercury MTTR01 true RMS multimeter from ebay for about £36 even though there was little information about it online. I feel that I did make a good choice and I'm very happy with it but there are a few little let downs that you can read about in the following paragraphs.

Here is a photo of the multimeter:

The multimeter weighs 420g, is 188 x 81 x 48mm, and has a strong, mainly orange case which fits quite comfortably in the hand. A large (64 x 43mm), 6000 count LCD with blue back light displays the current reading along with any necessary additional information. Six buttons under the LCD change between functions; to hold data or turn the back light on; to show the max/min reading; set the range; enter relative/USB mode; toggle between frequency and duty cycle display.

The round function switch near the centre lets you pick from the different modes:

AC/DC/combined voltage

Resistance/diode/continuity test

Capacitance

Frequency and duty cycle

Temperature

Transistor hFE

Current

There are two off positions, either side, so you don't have to turn the switch all the way one way to turn the meter off.

For inserting the test probes there are four sockets; the common, 400mA to 10A for current measurement, 400mA max for current measurement, and the socket for measuring anything other than current.

For connecting to a PC there is a mini USB socket at the top of the unit, not the bottom which the manual claims.

In the package you get: the meter, software CD, manual, mini USB lead, test leads pair, K type temperature probe and component adaptor (for connecting transistors etc to the meter). The meter uses a PP3 9V battery, which was already installed, but to change the battery (or either of the two fuses) you have to remove a couple of screws.

As well as having a stand the meter also has a strap with magnet for attaching the meter to a surface for hands-free operation.

I tested the meter's different modes, comparing the results with my other meter as well as the values I expected and I was satisfied the meter was operating correctly and accurately. A nice touch, although common to have on multimeters, is that the display shows a bargraph below the digits which lets you see changing values better than the numbers which take a very seconds to settle.

A few bad points to talk about, the first of which is the buzzer that sounds when pressing the buttons; it would be good to be able to turn the buzzer off as it can get annoying. The back light only stays on for a brief amount of time so it would have been helpful to have an option to change the time the back light stays on. As for the continuity test, it does not sound if the leads touch very briefly, so you have to be sure to hold them together for at least a second. When measuring high value capacitors the meter is slow to react; the manual does warn that it can take 30 seconds for a measurement to complete but that is for the maximum value. Lastly, the component adaptor could have been made better; you have to hold the component in while you take measurements.

The advantage of a multimeter that has a USB connection is that a PC can be used to log values, such as a changing voltage over time. The software, WH6000 Communication Program V1.5, supports O/S's from Windows 2000 to Windows 7 but the manual doesn't walk the user through installing the software. After I installed the software it failed to run but it was a simple matter of running it as admin.

Once connected to my PC I tested the meter's USB mode by pressing the Rel/USB button for 2 seconds (you do the same to exit USB mode). As the software was already open when I entered USB mode on the meter it selected the correct port. Interestingly, Device manager reveals that the multimeter is using a Silicon labs CP210x USB to UART bridge. Once you press the Connect button in the software a window with a digital and analogue style meter is updated to show the values on the actual multimeter and values are logged as a table and graph in the same window.

You can see the window below:

To be able to save the logged information you have to click the disconnect button which enables various data logging options such as for creating a new file, opening an existing file or saving to file. The log can be saved as xls, csv, rtf, htm and other formats. The graph can also be saved, in different formats (text, HTML, Excel, etc.).

As of 2018 the link to the meter's online manual now takes you to the manual for a different meter and the link provided in the manual to download the software is now broken.

It would have been very handy had the meter made use of a battery that could be recharged using the USB socket.

Vici VC97 Multimeter

When I was looking for a new multimeter I did consider the Vici VC97 which had very good reviews and although I ended up buying the Mercury MTTR01, I have now bought the VC97 as my original digital multimeter had finally died. From ebay I bought the VC97 for approximately £17 and while it doesn't have all the features of the MTTR01 it has some better points about it.

Like the MTTR01, the VC97 is auto range and has a large LCD but its count is limited to 4000 (6000 for the MTTR01) and there is no back light. The meter is powered by two AAA batteries which for me is a welcome change to the more usual PP3 (9V) type battery requirement; oddly the back of the meter references the use of a PP3 battery but at least the manual says to use X2 AAA batteries. The dimensions are 185x93x35mm and the weight is 290g so it is very portable and comes with a bright orange sheath which can hold the probes at the back. As to be expected, the meter has a flexible stand at the back.

You can see the meter below:

The multimeter can measure AC/DC volt, AC/DC current. resistance, capacitance, frequency/duty cycle, and temperature, and can test transistors and diodes. For measuring current, there are two main ranges: 500mA max and 20A max, available through two separate sockets, both fused appropriately. While using the meter I accidentally put too much current through the mA socket, was alerted to the over current by the meter's bleeping, and I suspected the fuse had blown. The manual implied that that it would be quite straightforward to replace the fuse but unfortunately it isn't that simple. After taking off the sheath, you have to remove four screws to get the back off and another two screws (located either side of the circuit board where the top buttons are located) so that the circuit board can be popped out, exposing the fuses.

A K-type thermocouple was included and is plugged into dedicated sockets on the meter rather than using the probes sockets such as on the MTTR01. The diode test option uses the same selection as for continuity and a circular socket is used for testing transistors which is very much standard for multimeters.

A large multi range switch selects the test mode to use and a single button marked DC/AC selects between DC and AC measurements. A reset button wakes up the meter and a relative button allows a reference value to be stored. A hold button will prevent the display from being updated and there is a range button for manual range selection. An annoying feature of the MTTR01 is the auto power off but at least with the VC97 the APO can be disabled or delayed. 

Oscilloscopes

Siglent SDS1052DL oscilloscope

When it came to building a high frequency oscillator for a project I found that my HPS10 oscilloscope had seemingly hit its bandwidth limit sooner that it should have. After some research I bought a Siglent SDS1052DL digital oscilloscope from Amazon for under £200. Not only does the SDS1052DL have a much higher bandwidth (50MHz/50GSa/s) than the HPS10 but it also has two channels and external trigger, a far better screen (7") and many other useful features such as being able to save waveforms to a USB flash drive or print directly to a printer, as well as the option to control the oscilloscope using a computer.

You can see the SDS1052DL oscilloscope below:

In the photo above the oscilloscope has its first channel connected to a Voltage-Controlled Oscillator circuit producing a frequency of about 2.4MHz. You can see the waveform is approximately square wave and in fact has the bandwidth limit option turned on to clean up the signal.

An example of making use of the oscilloscope's two channels was finding out the relationship between the two outputs of the VCO previously mentioned as the datasheet had not made clear how one output differed from the other. By connecting each output to one of the oscilloscope's channels I was able to see that one output was the inversion of the other.

The oscilloscope has an Auto button which when pressed sets itself up to best present the waveforms for either or both channels. There are manual controls which allow the adjustment of the waveforms' horizontal and vertical position. Cursors can be set up to measure time or voltage at any part of the waveform although the cursors are shared between the two channels.

As already mentioned, the scope can be connected to a computer using either the USB port or RS-232 port at the back of the unit. I was successful in getting the oscilloscope to work on my Surface pro 3 tablet via the USB connection. This is was I did to get it to work:

* Installed Easyscope software from http://www.siglent.eu/Downloads (make sure you install the right version)

* Connected the scope to my tablet and turned the scope on; make sure the oscilloscope back USB is set to USBTMC (utility menu). At this point, Windows failed to install the driver.

* Installed USBTMC driver from http://www.ni.com/nisearch/app/main/p/bot/no/ap/tech/lang/en/pg/1/sn/catnav:du,n8:3.25.123.1640,ssnav:ndr/ (use the correct link for your computer's O/S).

After a restart I then started the Easyscope software and added the Siglent SDS1052DL so that it could be accessed. You can read how to do this in the scope manual in the remote control section:

https://mediacdn.eu/m/media/wysiwyg/siglent/Downloads/Manuals/SDS1000DL_UserManual.pdf

Now, using the Easyscope software you can view the scope's waveforms with a user selected refresh rate that can go down to less than a second. A waveform can be saved (annoyingly you have to remember to take off the Auto refresh option whenever you want to save) or copied to the clipboard as can also the oscilloscope's own screen. Commands can be sent to the scope which are detailed in the manual linked below:

https://mediacdn.eu/m/media/wysiwyg/siglent/Downloads/Command/SDS1000_RemoteManual.pdf

Another cool feature is that you can virtually control the SDS1052DL by using virtual buttons that perform the same actions as the scope's actual buttons. However, I found the response to be laggy when using the fine controls so it would probably be best to stick to the oscilloscope's buttons. Certainly, one of the main benefits of connecting the scope to a PC is to more easily view, save and print waveforms.

Velleman HPS10 Personal Scope

My first oscilloscope was a Velleman HPS10 Personal Scope, a handheld digital oscilloscope that cost me about £70 from Maplin. Included with the scope was the manual and a test probe but no batteries or mains power supply.

The HPS10 is a good general purpose, single channel scope that can run off batteries (x5 AA) or a mains derived power supply. It has a 10MHz sampling rate with 2MHz analogue bandwidth as well as 5 different trigger modes and various display screens with different readings. The screen is a monochrome 3.5" LCD with a resolution of 128x64 pixels.

The oscilloscope can be set to auto mode to try to take the measurements itself and to align the waveform on screen. It can also be set to manual mode whereby the settings can be changed yourself which includes the use of markers to measure the voltage and timing at specific parts of the waveform. 

A photo of the scope is shown below which is connected to a standard 555 astable oscillator:

The oscilloscope has a built-in oscillator output in the battery compartment which is used to calibrate the oscilloscope probe. It was not practical to use the oscillator output for the photo so I instead used a 555 oscillator. The oscilloscope's auto mode was not helpful so I had to manually set up the timing and move the markers to measure the frequency and voltage of the signal. The readings I got were exactly 8Hz for the frequency and 8.63V as the highest point of the waveform. The values I used for the 555 astable should have given a frequency of approximately 6.8Hz but when you take into account the tolerances of the timing resistors as well as the capacitor, 8Hz is acceptable as being correct. Most curious is the maximum output voltage of 8.63V considering the 555 running on 9V should not be able to output that high voltage but the higher voltage may be a result of the 555 having no load (the oscilloscope probe was connected directly to the 555's output and ground).

The HPS10 has a number of problems, some of which are not home to the more expensive HPS40. A big problem with the HPS10 is that the LCD has no backlight making viewing difficult in low light. Also to note is that HPS10 is very slow when the timing is set so that a waveform repeats many times; the screen takes a few seconds to update. There is no way to save waveforms using the oscilloscope as unlike the HPS40, the HPS10 cannot be connected to a computer.

(The function generator is too deep to be able to place is above the oscilloscope.)

The SDG1005 is part of a family of function generators from Siglent but this particular model has a maximum output frequency of 5Mhz (some waveforms cannot output anywhere near that high). It has two channels (oddly, CH1 is on the right and CH2 is on the left) both of which can output either one of the built in basic waveforms (sine, square, ramp, pulse and noise) or one of the built-in arbitrary waveforms such as stair up/down, SNR or Tan; also capable of AM, PM and FM.

The unit came with a power cable with a non-UK plug and an adapter but fortunately the mains socket on the back of the function generator is a common so-called 'kettle' socket. There was also a paper manual and calibration certificate (both in Chinese), a CD containing the EasyWave software, a USB lead and a BNC cable with crocodile clips one end. You can view the English manual online at:

http://www.isotest.es/web/Soporte/siglent/manuales/siglent_sdg1000_user-manual.pdf

However, even in English the manual is still poorly written; more detail would have been very helpful.

Power supplies

CSI 3005EIII Multi-output DC regulated power supply

A power supply for testing is arguably one of the most important pieces of equipment to own since every circuit you will make or test will likely need to be powered in some way and it when testing it there are circumstances when it's better to use a well designed, reliable power supply than the one that came with the device under test. Fortunately you can pick up decent, variable power supplies relatively cheap but do of course read reviews and check the power supply meets your needs before buying. For me, I settled on a CSI 3005EIII power supply, which I bought from ebay for £190, because it actually contains 3 linear power supplies, 2 of which are adjustable 0-30V/5A and also the fixed 5V/3A, and features current limiting, over temperature and short circuit protection. All 3 power supplies output DC only and the 2 adjustable power supplies can operate either independently, or in series to give 0-60V/5A (which also allows for dual-rail power), or in parallel to give 0-30V/10A. For the product page which includes a link to the manual (at the bottom of the page) please visit CSI's site at:

http://www.circuitspecialists.eu/csi-3005eiii-bench-power-supply-0-60v-0-5a-linear-0-10a-0-30v-5v-fixed-output/

The PSU came with a paper instruction manual, UK mains lead, and a number of test leads with a banana connector on one end and crocodile clip the other end. The leads are: 2 standard banana to crocodile, one short banana to banana for series mode, and a banana to crocodile pair for parallel mode.

You can view the PSU below:

At the back you will find the mains input socket (rated 220-240VAC 50/60Hz) with integrated 5A fuse as well as air vents for the 2 fans, and additional air vents on the left and right sides, and on the top of the unit which is home to a handle as the unit is large (360 x 265 x 165mm) and very heavy. I was quick to remove the handle so I could stack equipment on top (avoiding the air vent holes). On the front are 4 displays (each 3 digits) for volts and amps of the 2 adjustable power supplies as well as 2 sets of current and voltage dials and also C.C. (Constant Current) and C.V (Constant Voltage) LED's. Two switches in the middle select between independent, series and parallel modes for the 2 adjustable power supplies which are split into master (labelled I) on the right (when looking from the front) and slave on the left (II). Note: when operating in series mode you can use the provided short cable between the + and - of the master and slave to provide a higher current path between the 2 power supplies than what the switch provide. In series mode the slave's + is connected to the master's - which forms the GND point should you need it when used as a dual-rail supply; the slave's - becomes the negative connection and the master's + the positive connection.

A large power button is located bottom left and to the right of it are two sets of output connections (+, GND and -) for the 2 adjustable power supplies and far the - and + outputs for the fixed 5V output; all output connectors will accept standard 4mm banana plugs. Note that the GND connections are actually connected to the unit's earth connection and may be useful for e.g. to connect an anti-static wrist strap.

Aside from operating the adjustable power supplies independently, in series or in parallel, they can also behave as either constant voltage (voltage stays the same but the current can alter), constant current (current will stay the same but the voltage may alter) or constant voltage with current limiting (current can be drawn up to a maximum which may cause the voltage to change). These modes will now be discussed:

Constant voltage

Default mode; C.V. light lit. Without a load connected adjust voltage with voltage dial and make sure current set to max. When connect a load the Amp display will reflect the current being drawn. Found the voltage that is output to be very close to my multi meter reading. 

Constant current

Without a load connected set the voltage to max (the voltage reading may be as high as 32V) and set current to min; C.C. LED will light. Connect the load (there will be no volt/amp reading) and rotate the current dial until you reach the required value; now there will be a reading on the volt and amp displays.

Constant voltage with current limit

With no load connected set the voltage low (0.1V works well) and set the current to min; C.C. LED will be lit. Connect the + and - outputs together (you can use the provided short cable). Rotate current dial to set current limit then disconnect the short; C.V. LED will be lit. Set the voltage with the voltage dial before connecting a load.

Now if you connect a load the C.C. LED will light, the amp display will show the current being drawn and the volt display will show the voltage. Note that the voltage may drop to make sure the previously set current limit isn't exceeded, however, when I set the limit to 100mA I was able to draw 110mA. To reset the current limit you need to power off and on again.

I'll now go through a number of tests I did to illustrate how the PSU behaves:

In constant voltage mode the voltage of the adjustable supply can be turned down to 0V which under no load gives -0.03V on my multimeter whereas with the voltage set to 0.1V and no load connected I get 0.14V on the multimeter.

I set one of the adjustable supplies to 2.5V in constant voltage mode without a load and got a reading of 2.52V on my multimeter and with a 1R/10W resistor connected although the display still showed 2.5V my multimeter showed 2.43V with a current draw of 2A. With the same resistor still connected I increased the voltage to 5V and found the voltage dropped to 4.87V (the PSU's display still showed 5V) at a current draw of 4A.

Next I hooked up my Zenfone 2 smartphone using a cable I made up to charge the phone from the fixed 5V output of the PSU. With the phone on and charging my multimeter reported that the voltage at the PSU's output connections was 5.07V (without the phone connected the voltage was 5.10V) at a current draw of 449mA.

Let's now look at some key positives and negatives I have in regards to the PSU:

Positives

Wide voltage range with high current capability.

Two adjustable supplies with 3 modes and fixed 5V supply.

Display indicators are bright and easily readable.

Relay clicking when reach certain voltage/current is reassuring (but would be annoying to other people).

Volt/amp adjust controls are quite precise (not difficult to adjust by 0.01 steps).

Negatives

Large and heavy.

Could do with extra digit for each volt/amp reading.

Would be helpful to have a current meter for fixed 5V output.

Can be quite slow to turn volt/amp controls to reach high/low values since you need to turn the dial many times (but could be viewed as a safety feature).

Would be useful to be able to turn off the adjustable supplies when operating independently at least for safety reasons.

Would have been useful to be able to fit a multimeter probe securely in the output sockets.

In summary, although cheap single output power supplies are readily available to buy online I had the specific need of a power supply that could operate in dual-rail mode for use with op-amps which many power supplies cannot do even though they appear to do so (the so-called GND connection is usually an earth connection). From my initial testing the power supply performs very well and has 2 adjustable outputs which can operate independently or together, which includes functioning as a dual-rail supply and having the fixed 5V output will be very handy for when working with logic circuits.

HW-140 variable power supply module

The HW-140 power supply module can accept a fixed DC input from 5.5 to 30V and output a variable DC voltage from 0.5 to 30V by means of a buck-boost converter and controller. Output current can be as high as 3A but can even go to 4A with a suitable heat sink attached to the voltage regulators; a suitable heat sink is sometimes old with the module. The module features an LCD with backlight (see photo below) which displays either the input or output voltage and the output current in amps or output power in watts.

At the left of the module is an 'IN' connector to wire in a DC voltage supply (such as from a wall adapter) and an 'OUT' socket on the right side to connect a load but you can also solder wires to the provided solder pads if you prefer. An 'IN/OUT' switch (bottom right) when pressed quickly toggles between displaying either the input or output voltage but when pressed for longer switches between showing either the output current or output power. Next to the IN/OUT switch is an 'ON/OFF' switch which when pressed quickly turns the output on or off (the 'ON' LED - above and to the right of the switch - will light when the output is on) but when pressed for longer sets whether the output is on or off by default when the module is powered off and on again.

When the 'CC' (constant current) LED (to the left of the ON LED) is off the module operates in constant voltage (CV) mode allowing the load to draw the current it requires and the output voltage to be set by adjusting the 'CV' trimmer. To get the module into CC mode you need to have the output on and while the output connections are shorted; you may hear a whine, which is normal and the module features short-circuit protection. You can now adjust the CC trimmer to set the required constant current and you should see the CC LED light when the current is set low enough; you cannot force higher than normal current to flow through the load for the same voltage setting. When you connect the load the CC LED should still be lit and the chosen current setting will be displayed although the voltage outputted may be less than what was set in CV mode since it lowers the output current by lowering the voltage.

Let's go through an example to better illustrate how CC mode works. In CV mode set the CV trimmer so the output voltage is as close to 5V as possible and connect a 3.3R/10W resistor to the output connector. Since I=V/R:

I=5/3.3=1.51A

So we expect about 1.5A to flow through the resistor (when connected to the module the value may not be exactly the same since the resistor is unlikely to be exactly 3.3R).

I chose a 10W resistor for good reason as:

P=VI

P=5x1.51=7.55W

Careful, the resistor will get hot!

Now we will enter CC mode so short the output and adjust the CC trimmer down to 1A and see that the CC LED will light. When you connect the resistor again you should see that about 1A flows but the voltage would have dropped to around 3.36V as to limit the current flow to 1A.

Programmers

TL866CS MiniPro EPROM programmer

As a project I've been building my own simple computer and it has been necessary to write my own code to a flash chip so I looked online for an EPROM programmer. I came across the TL866CS by autoelectric for about £30 on ebay and I honestly have to say it's one of the best tools I've ever bought even though I have only just begun to use it. It can read and write to over 13,000 chips of different types which includes EPROM's, flash IC's, SRAM chips, and microcontrollers. The programmer is USB powered and features a ZIP socket in which you simply drop in the chip to read/program and lock into place. Using the provided software (called 'MiniPro Programmer') you can find the device you want to program using a search facility. Then you can read in the data on the chip or program it as well as carry out a number of other functions such as blank check.

Here is a photo of the programmer and accessories:

The programmer was bundled with a number of adapter sockets (MSOP8/SSOP8 SMD, PLCC44-DIP40, PLCC32-DIP32, SOIC8 ZIF, SOIC8-DIP8 ZIF), USB cable, driver CD, and a PLCC extraction tool.

A screenshot of the MiniPro Programmer software (no programmer connected):

I installed the software from the CD first and then connected the programmer. When I ran the MiniPro software from the installed folder I was prompted to reflash the firmware, which I did. As a simple test I used the programmer to read a ROM which so happened to have text that was readable in the dump window and since then I have successfully read from and written to a flash chip many times.

The TL866CS doesn't support ICSP (In-circuit Serial Programming) unlike the TL866A but it is possible to add ICSP to a TL866CS which may end up saving you money.

A couple of minor complaints: some spelling mistakes (e.g. Calculater instead of Calculator) - as only Chinese and English are supported presuambly the English wording was translated from the original Chinese. More unfortunate is not being able to save a block of values as anything other than text yet the entire data can be saved as BIN or HEX file.

For the official manufacturer page:

http://www.autoelectric.cn/en/TL866_main.html

This is the best place to download new versions of the software.

You can visit a TL866 wiki which has some useful info.:

http://minipro.txt.si/

Soldering irons

853D Hot Air Solder Rework Station

As my then current temperature controlled soldering iron had broke I looked at getting another one, especially one that would help with desoldering. I ended up getting an 853D Hot Air Solder Rework Station from ebay for about £85. Supposedly it has typical features that would be needed at an electronics workbench which in this case are a hot air gun, soldering iron, DC power supply, a DC voltmeter, ammeter and an RF meter.

View the image below of the 853D:

The soldering iron is ESD safe, very light to hold and heats up in seconds; an LED display indicates the current temperature. A couple of buttons allow adjusting of the temperature but unfortunately there is no quick way to drop the temperature for when you will not be using the soldering iron, although you can just switch it off. It was stated that the station came with 11 soldering iron tips but there were actually 12 as there was already one in the soldering iron.

As for the hot air gun, like the soldering iron it heats up in very little time and the current temperature is shown on a display and can be changed with a couple of buttons. Unlike the soldering iron, however, the heat is automatically cut whenever the hot air gun is returned to its holder. This is most likely a safety feature as the hot air gun is more dangerous than the soldering iron. The hot air gun must be left to cool down enough before the station can be switched off to ensure long life of the hot air gun.

For use with the hot air gun four nozzles were provided of different sizes to cater for the different uses of the hot air gun. I did a test with some thin heat shrink and found that it was far more effective than using a soldering iron for heating heat shrink.

An LED display keeps track of the current voltage of the DC power supply when set to output and a dial adjusts the output voltage from 0V to about 15V. By checking with my multimeter I found the display appeared to be quite accurate but when turned down to OV the unit was still outputting a small voltage. You could use the station's DC output power supply not only to power a circuit that is being tested or fixed but you could also connect a light to illuminate your work area.

When set to input the display functions as a voltmeter but is almost a pointless feature because the display is slow to react (it updates much slower than even my cheap multimeter). 

While I'm happy with the station my first impressions of it weren't that good. The soldering iron stand arrived damaged (but fixable), the switches on the station were upside-down (that is, 0 for on) and the hot air gun gave out a lot of white smoke upon first use and smelled of burning (but appears to be working fine). The soldering iron gave out a little white smoke when I first used it which is expected.

Duratool D00672 desoldering station

I've been managing with a soldering iron and hand operated desoldering pump for some while and they are still useful for quick fixes or for removing components from boards that only have a few connections. However, for more difficult components, such as large pin count connectors I decided to invest in a desoldering station featuring an electric pump. After some research I settled on the Duratool D00672, which appears to be synonymous with the ZD-915 which is possibly a clone of another desoldering station.

I bought the D00672 from Amazon for around £104, which can be viewed below:

It came with the following items:

The desoldering station main unit, instructions (including a separate sheet with cleaning information), UK & EU 'Kettle' lead, desoldering gun stand (has metal tray with sponge and clips to the side of the main unit), desoldering gun, 3 cleaning rods, air pump cap, spare filters, and 2 desoldering gun tips.

At the back of the main unit is the 3-pin mains input (rated at 230V, 140W) with integrated fuse holder. At the front is a large LCD with blue backlight which shows both the current and target temperature, as well as the message 'Wait' when the desoldering gun is heating up or error if the desoldering gun isn't connected. Also on the front are the up and down buttons for setting the target temperature, deg C/deg F buttons for switching between the two units of measurement, the power button (locking type), the desoldering gun connector and the air pump connector.

The D00672 is designed for lead free solder which means it should (and does) work well with lead solder, which has a lower melting temperature than lead free solder. The desoldering gun is rated at 24V, 80W, it uses N5 series soldering tips, and is not too heavy, but the button to activate the pump is very sensitive and I found myself knocking it accidentally. Because the desoldering gun barrel is translucent you can see where solder has accumulated so that should give an indication when to clean the gun. The manual does warn that if the suction drops then cleaning should be carried out. All of the soldering tips, including the one already fitted in the desoldering gun have solder in them, so when first using a tip you must heat up the desoldering gun and then operate the pump to remove the solder.

The instruction manual is the usual badly translated document but has the model number stuck on the front and inside in the operating instruction section the line about removing the 4 screws from the control system has been crossed out (this was necessary in older versions), which is very unprofessional. Looking online, the 'removing 4 screws' has confused a lot of people, concerned that they will break the machine if they don't remove the screws (which aren't even there).

To use the desoldering station you have to connect the desoldering gun plug to the main unit, which the instruction manual mentions, but it doesn't say anything about fitting the suction pipe, although very straightforward to do. The unit is very loud when operated as it has a fan that is always on and the pump itself is noisy although reassuring that the suction is operating. People have done mods to quieten the noise which involves fitting a better fan.

When I first tested the desoldering station the pump would not work but not wanting to have to contact the seller and ship the item back I opened up the unit. I found that one of the pump motor wires had come off (despite the heat shrink) so it was a simple fix to get the pump going again. While inside I noticed that the mains switch seemed to be the double throw type but only the live had been connected; this has been mentioned by someone online and it would make the unit safer if the switch operated both live and neutral.

To desolder using the D00672, after power on let the temperature reach the set value (I used 320 deg C)-it is quite slow to heat up. Then place the soldering iron part of the gun over the solder joint and against the circuit board. Wait a few seconds for the solder to heat up and operate the pump-the solder should be removed. I was able to easily remove connectors with 6 pins as well as more common components such as capacitors.

The Duratool works well with single sided boards, making it a breeze to remove components with many connections but it does struggle with even basic components on some double sided boards, which is because of there being solder on both sides of the board and the solder being very little which is harder to heat up. I found that the tip that was already fitted on the desoldering gun has a bit too small hole for my liking but 1 of the 2 extra tips had a slightly bigger hole that was better for most joints I work with. It would have been helpful to have a tip with an even bigger hole for large joints but then the suction may not be strong enough.

Fixing faulty tools

Whether you buy tools new or second-hand they will eventually become faulty or fail to work as good as they first did. With electrical tools that are portable it is always best to check the batteries first and replace them if need be. With mains powered equipment that completely stop working check the fuse and the mains cable; some tools also have a fuse inside them that will need inspection. Another thing to check are input leads as they can break over time; this is were it's good to have spare equipment to test tools that become faulty. For example, if your multimeter has stopped working you could test its fuse and leads using another multimeter (or just a battery and bulb) but of course with the fuse/lead removed from the multimeter.

Tips for working with test equipment

There are many different types of tools and test equipment and these should be handled with care so that you get long life out of them as well as protecting yourself and others from injury. Get to know your equipment, what its limitations are, and how to get the best use of them so that designing and testing circuits is as straightforward as can be.

Keep your fingers away from test probes and circuit components as not only can your body influence readings you could also get a nasty shock!

Components are usually tested when a circuit is without power and you should discharge capacitors before testing them. If there are other components in parallel with the component you are trying to test then you should disconnect it from the circuit before testing.

Do not wrap test probes around a multimeter as this can weaken or break the wires.

All content of this and related pages is copyright (c) James S. 2007-2023