I learned vacuum tube (valve) technology over half a century ago at a technical school, and now put my knowledge to use repairing WWII SCR-625 mine detectors. Working as a technician in the disk drive industry for thirty-five years has taught me a lot about troubleshooting and electronics repair. A big challenge is finding a source of modern-day parts to substitute for the old parts that fail, and knowing what the critical issues are. I learn more with every repair I do, and willing to help others, but I don’t have any magic pixie dust to make your detector work.

The notes here are really disorganized, as I add information periodically as I learn it, and there is a lot of repetition. With the little interest there is in SCR-625 repair, it’s not worth my time to make a finished document.

Schematic with parts values and tube pinouts at bottom of page. The repair technical manual is TM 11-4016.

December 2023 - One of my detectors needs fixing again. The symptom is that it is insensitive to detecting metal, and a constant tone output, which can not be adjusted out.  V21-2 control grid voltage (cap, or pin 8) is +.25 volts, where it should be slightly negative. Caused by a (electrically) leaky capacitor 14-1. After replacing that capacitor (with 470 pf), the voltage is now -2 v, and the detector works fine.

July 2022 - I learned that a great cleaner for batteries that leaked is vinegar. A very light coating of dielectric grease may be helpful, to prevent corrosion. Just enough to wet the connector. That has made a world of difference in my trailer electrical connector. Just enough to wet the contact, not coat it. Wipe off all excess. Find dielectric grease at an auto parts store. If you are not handy with tools or soldering, look for a local Repair Cafe or Fix-It Clinic. Try contacting them ahead of time, with information.

June 2022 - See bottom of the page for typical voltage measurements.

June 2022 – Repair / circuit modification. One mine detector I had was functional, but very insensitive. Measuring the DC voltage at tube 21-1 cap/pin 8/control grid, the voltage was more than -4 v DC. It should be much less? I worked on it for quite a while, but found no cause. I verified the tube is good, also. When probing with the DVM, the operation improved. By adding a 4.7M ohm resistor from 21-1 pin 8, to ground, the operation was excellent. Update - I found the root cause - When one capacitor at 18 shorted previously, it drew excessive current through the compensator, causing the resistance to increase. I replaced the compensator, and all is well, with the 4.7 M ohm resistor removed.

July 2021 - The mine detector amplifier that I just bought on ebay did not work. The inductor/coil  labelled part "2" was open. These commonly fail, and when there was no sound out at all, I check the DC voltage on both sides of variable capacitor "15" (by tube 21-1). One side was 80 volts (my battery voltage), the other side was 0 volts.  With batteries removed, I verified with an ohmmeter that inductor "2" was open (measure across variable capacitor "15"). That part can be replaced by a 22K ohm resistor across variable capacitor "15", without disconnecting part "2" or "15". There was a slight hum after replacing that part. The AC peak to peak voltage (with an oscilloscope) at pin 6 of tube 20 was just 10 volts, so I replaced capacitors 17 and 14-2, as I do on every amplifier. The amplifier worked great after that.

June 2021 - A local collector's mine detector quit at a show. The main battery was at 75 volts, and dropped down below 50 volts when power was switched on. Later he replaced the battery, but still no indication of functioning at all. He said the battery compensator was getting hot. In this case, the battery dropped from 105 volts down to 90 volts when power was turned on. Tube plate voltages were near zero volts. Removing the batteries and checking resistance, there was zero resistance across the capacitors labelled "18", which are the main power capacitors. Disconnecting the power wire from the capacitor, one of the capacitors had a dead short. Even with those capacitors removed entirely, the detector now worked. I replaced those capacitors with a new 1 ufd, 250 volt capacitor.

Important! If there is a short in the main power circuit (like the main power capacitor(s) 18), the whole main battery voltage is across the battery compensator, causing it to overheat, possible burn up. I think it might be a good idea to replace capacitor(s) 18 if you are working on a mine detector.  The metal can style should be OK, but replace it if it is four black capacitors. See picture further down the page.

Use a 1 ufd 250 volt capacitor, but a lower value might work, too. Maybe some of the bad compensators I replaced before were caused by temporarily bad capacitor "18".

Dec 2020 - I checked out all three of my detectors, and found one was not working. There was a constant tone out, which control adjustments did not change, and putting a large metal object very close to the search coil was detected. Swapping parts with other detectors, I determined it to be the amplifier. The oscillator is functioning. No obvious voltage problems when measured with a voltmeter. I connected my oscilloscope to tube 21-1 input, and adjusted the control box for 0 p-p volts into tube 21-1. With a metal object near the search coil, the tube 21-1 input was as high as 400mV p-p. There was no output signal at V1. There were no bad parts, but after poking around, and resoldering some connections, it began working again. Possibly a bad tube connection or a bad solder connection.

Nov 2018 - One of my detectors went bad. Symptom - Low tone out all the time, unless battery compensator turned too far, which produces a loud tone, which is normal. I found the control grid (pin 8 and top ) DC bias voltage of the first stage audio amplifier tube 21-1 was too high to work. AC signal at pin 8, read with a voltmeter, and verified with an oscilloscope, was as high as several volts AC when a metal pan was close to the search coil. That is normal. There was no signal out of tube 21-1. The compensator potentiometer had been replaced with a 1K pot by me. Disconnecting one end of the pot and measuring the resistance, the pot measured 1.8K, which is far too high. I have had to replace several battery compensator pots in units I have fixed, but this is the first replacement pot I have had go bad.

Nov 2017 - A local collector brought over a newly purchased SCR-625 for repair. This was the 15th SCR-625 I have repaired. There was no sound out at all initially. While poking around, I found that shifting the D cell batteries slightly produced a loud sound, or nothing, or a slightly working condition. I have seen that several times when using my detectors. The D cell connectors looked clean, but were still causing a problem. Always clean the connectors with 200 grit or finer sandpaper, and bring sandpaper along if you are doing a demo.

April 2014 repair: A local collector brought over a previously running SCR-625 for repair. It was not functional at all. The battery was good, but the case ground to tube plate voltage was only 20 volts. The problem was the battery compensator potentiometer was high resistance, causing most of the battery voltage to drop there. A quick hook up of a 470 ohm resistor across the pot confirmed that, making the detector functional (temporarily). I replaced it with a 1000 ohm pot I had on hand.

Always clean connectors as a first step before trying to repair a detector. Wiggle the tubes (valves) a bit to make a better connection.

 It can be difficult to make voltage measurements on the detector, because the circuits are sprayed with an anti-fungus coating, preventing a good reading. Solder a ground wire somewhere to a case ground, and do voltage probing with a sharp tip. Be careful when probing tube pins, that you don't slip and short one pin to the next.

The most critical repair is the replacement of capacitors that have changed in value over the years. Wires broken internal to a cable can be an issue, but not insurmountable, if you are able to locate the problem. Broken wires are usually found where the cable comes out of the amplifier or control box. Helpful hint: Old wires typically corrode over the years, and if you need to remove the corrosion before soldering. Crimping on a connector helps make a better connection, too.

If you have replaced capacitors, and the mine detector is not working right, try flexing the cables. Be firm but not extreme. If the mine detector performance changes, that is probably the issue. Good luck.

On several units, I found the choke, labeled as part "2" on the schematic, to be open (bad). A 22k ohm resistor can be substituted in it's place. See technical not further down.

As noted at top of page, the battery compensator fails sometimes, changing to a higher resistance. With no batteries installed in the amplifier, you should be able to check the resistance while installed. The range minimum to maximum should be 0 to 750 (or 1000) ohms. If there is a permanent or temporary short in the high voltage circuit, it can burn out the compensator.

The "Test" button never seems to work on any of them. Leave it alone, lest it should stick in the test position.

In all the units I have repaired, it was absolutely necessary to replace capacitors 17 (.012 ufd, 200+ volts) and 14-2 (470 pf, 200+ volts) in all of them. If you replace these parts and clean the connectors, there is a 50% chance the detector will work somewhat.

Some of the bypass capacitors have been known to be bad, but there are too many to replace them all. As a matter of practice, I replace or add to many of the bypass capacitors, and replace many tuning and coupling capacitors except for 16, which is inaccessible, and variable capacitor 15. Bad switches or corrosion might be a problem, but the tubes are usually good. 

A key part of having a sensitive detector is cleaning both connectors of the amplifier to control box cable. For the female end of the connector, use a small, loosely fit drill bit inserted into the connector and twisted by hand.

April 2011: A quick, easy way to clean out the female contacts, if you have the right tools, and are careful: For the smaller female connector, use a #43  .089" drill bit, or #44 .086" drill bit, whichever is not tight in the connector, and a Dremel tool or electric drill using slight contact to burnish the connector. For the larger female contact, use a #31 .116" drill bit. Be sure the drill bit is not tight, if using a power tool.

On the male end try fine grit sandpaper, or a .45 caliber bore brush twisted by hand. Clean connections are critical, for both the amplifier to control box connector, and the batteries. If a previously functional mine detector does not work, be sure that the battery connections are clean, and try shifting the battery a little bit.

Before attempting any repairs or making a battery, you should first clean connectors, then try step three below, to test for short circuits or open circuits.

If you are having trouble getting any vintage electronics working, I want to emphasize trying and retrying all connections: cables, tubes, switches and batteries. If it is not working right, disconnect and reconnect all physical connections, and jiggle connections with power on when possible, to see if it changes. With no power applied, work switches back and forth numerous times to wear off any corrosion inside.

Be sure you know what you are doing before attempting a repair, and have the right equipment. A 30 to 35 watt soldering iron is about the right size for this job. A soldering iron that is too hot, if used carefully, is better than one that is not hot enough. A voltmeter is a necessity, and a oscilloscope is helpful.

My Youtube video of using a SCR-625:  https://www.youtube.com/watch?v=YtY_fqSVPFI

My Youtube video of a 1920's radio speaker hooked up tp a SCR-625, to test the speaker:  https://www.youtube.com/watch?v=LtYr098snMA

  Adding caps instead of replacing. You can use the case ground for the bypass caps.

  Circuit Description:

Tube 20 (1G6-GT) is a dual triode oscillator that generates a 1kHz signal that goes to the two transmitter coils. The nominal plate voltages should be up to 150 volts differential peak to peak with an oscilloscope, or about 100 volts AC when measured with a DVM.  Over time, as capacitors 17 and 14-2 age and change value, and the amplitude will drop significantly. This is the most common failure mode with the mine detectors. With no metallic objects near the search coil, the coarse and fine adjustment knobs should be adjusted to null out the two transmitter signals in the receiver coil. When a metallic object near the search coil disrupts the magnetic field, the difference is amplified in a two stage tuned audio amplifier consisting of two 1N5-GT pentode tubes, 21-1 and 21-2. The audio output can be heard with the M-356C speaker or HS-30 headphones. Be aware that the HS-30 headsets can be a problem of their own. Re crimp the ring terminals if the headset doesn't seem to work. 

Tube pinouts: Since the schematic and parts list do not list pinouts, here it is:

If you do not know what I'm talking about here, leave the mine detector alone.

1N5-GT: plate (top) is pin 3, filament and cathode pins 2 and 7, control grid is the cap and pin 8, screen grid is pin 4.  pin 1 is no connection, but may be used as a wire junction.

1G6-GT; plates are pin 3 and 6, control grid is pins 4 and 5, filaments are 2 and 7. Control grid 4 goes with plate 3, and control grid 5 goes with plate 6.

Tube pin numbers are read clockwise from the keyway, when looking at the bottom.

Technical Notes:

1a. If you have replaced  capacitors 17 and 14-2, and the mine detector is not working right, but there is sound, try flexing the cables. Be firm but not extreme. If the mine detector performance changes, that is probably the issue. Good luck with that. It can take many attempts at setting the controls to get that right, especially if you are new at it. See the "SCR-625 adjustments" page for setup.

2. I have had four amplifiers with the choke (component labeled 2) open (bad). One I replaced with a choke scavenged from another amplifier, on the others I have substituted a 22K ohm 1/2 watt resistor in place of the choke and variable capacitor (component labeled 15) combination. It seems to work to just solder the 22K resistor across the variable capacitor, leaving the open choke and the variable capacitor in the circuit. Using an ohm meter, if the resistance across variable capacitor 15 is open, the choke is bad.

3. Some resistance readings:

M-356 speaker/resonator: 1K ohms,  Choke (component 2), as measured across the variable capacitor 15 (the white component): 2.5K ohms. Resistances at the battery connections of the amplifier, with everything hooked up, and switches in the on position, and NO BATTERIES INSTALLED: at BA-33 connectors: 15M ohm, at A-1 connectors: 4.5 ohms, at A-2 connectors: 5.0 ohms. Your readings may be off 25%, and probably not a concern.

4. Test for excess current draw in 103 volts - items needed: two jumpers with alligator clips, a 470 ohm (or 390 to 510 ohms) resistor, DC voltmeter, a scrap of paper and a piece of aluminum foil. Also, add a piece of electrical tape near the amplifier terminal to prevent the jumper from shorting to the case.

Install the 103 volt battery, positive end first, with the piece of aluminum foil at the + battery terminal, and a piece of paper between the aluminum foil and the amplifier + terminal. b. Run a jumper from the aluminum foil to one end of the resistor, and run the second jumper from the amplifier + terminal to the other end of the 470 ohm resistor. Power up the mine detector as normal, and measure the voltage across the 470 ohm resistor. The voltage should read about 4 volts, and the mine detector should function as normal, if there are not any problems. If the voltage is higher than 8 volts, you have a problem, if it gets hot and burns up, you have a short in the amplifier 103 volts.

Voltage / Resistance = current. Typical current draw on the 103 volt supply is .010 amps (10 milliamps). If you are using a different value resistor, voltage divided by resistance equal current.

Though most multimeters have a current measurement mode, hooking up something wrong can cause problems.

Test the oscillator output (tube 20, 1G6-GT): The DC voltage from ground to both sides of capacitor 17 should read very close to main battery voltage (100 volts). If the oscillator is working, the AC voltage ACROSS capacitor 17 should read about 100 volts AC. If it reads below 75 volts AC, a possible cause might be the value of caps 14-2 and 17 are off. This assumes that you have a multimeter that reads AC frequency up to 1000 hz. Most modern ones do.

Test the amplifier output (tubes 21-1 and 21-2):  If you have a spare 0.1 ufd capacitor or something close to that, holding one end of the capacitor, touch the other end to the top cap of tube 21-1. You should hear a faint scratchy noise in the speaker. Doing that more than once in a short time will have no effect. Try the same on tube 21-2.

If the detector is adjusted and working, and a large piece of metal near the pickup coil, the AC voltage across capacitor 16 may go over 50 volts AC. With no metal nearby, the voltage should be a couple volts or less, unless the battery compensator is set too high. Capacitor 16 is way down inside the amplifier, so be very careful. However it is easier to probe that than across transformer 3 inputs. If you can get up to 50 volts AC at that point, and down to just a few volts, that would imply that the oscillator and the amplifier (up to transformer 3) is working.

To test just the first stage amplifier, tube 21-1, if the detector is adjusted and working, and no metal near the head, the AC voltage across variable capacitor 14 should be no more than a couple millivolts AC (0.002 volts AC). If a large piece of metal is placed near the head, it may read up to 2 volts AC. If the detector is adjusted, and no metal near the head, adjusting the coarse adjust screw one half turn gives a reading of 0.05 volts AC, and one turn off gives a voltage of 0.1 volts AC.

Some DC voltage readings: DC voltage measured from ground to capacitor 19-4 top should read about 5 volts lower than battery voltage, to 19-5 should read 10 volts lower than battery voltage, and 19-2 should read 15 volts lower than battery voltage. The voltage on both sides of capacitor 17 to ground should read about the same as the battery voltage.

Parts:

The parts normally required for a repair generally do not cost over $10 (excluding batteries and related parts), but minimum orders and shipping charges will add significantly to that. Order lots of spare parts to avoid a second or third purchase in case of screw ups or unexpected needs. Switches and the potentiometer are expensive, but may not need replacing. I can give you part numbers if you need to replace one. Ordering spares may not be advisable unless it looks bad or does not move freely.

Most of my parts I purchased from Mouser Electronics. As an alternate source I would consider Digi-Key, which is the preferred electronics supplier where I work. Suggestions on capacitor type selection:

1.      Select capacitors rated at 200 to 250 volts if possible. A lower voltage is risky, and much higher voltage could physically be too large, and the actual capacitance value may be different at a lower voltage.

2.      Do not use electrolytic, tantalum or mica capacitors. They have a higher failure rate, and the electrolytic capacitors are prone to catastrophic failure (they literally blow up).

3.      On bypass capacitors (19-1 through 19-6), these are rated at .1 uFd, but going .15 uFd  may be better.  A tolerance greater than 10% is acceptable here. The type is usually a metalized polyester.

4.      On the tuning and coupling caps, 10% is acceptable. The type is usually ceramic or metalized polyester.

5.      Be sure the lead length is one inch (25mm) or so.

If you need tubes, search online for sellers, by tube type. Look for NOS (new old stock) tubes rather than used. Try vacuumtubes.net . I have some for sale, too.

3/32 heatshrink is recommended for battery construction.

Battery compensator: P/N Rv4NAYSD102A 

Amplifier power switch: try Carling 2GK50-73, Mouser P/N 691-GK50-73. This switch has two sections instead of one, in case you should want to switch filament power with the same switch.

 If you can find someone qualified locally to do the repair, that would probably be best. Look up antique tube radios for sale on your local craigslist, and see if any of the sellers repair the radios. The circuits in the mine detector amplifier are similar to what you would find in an old radio. Show them a copy of the schematic and refer them to this webpage.

June 2022 voltage measurements:

Operating voltages

Test conditions: Battery voltage 105 volts. Detune control box coarse control ½ turn clockwise (always a tone output if the detector is working). Compensator in the middle of its range. Measurements with a digital DVM. Add a chassis ground wire to attach to, and possibly wires to pins 3 and 6 of tube 20 to probe at. All voltages are approximate, and may vary with battery voltages.

Tube 20 (1G6-GT)

DC voltage (chassis) from ground to pin 3, and ground to pin 6 = 94 volts

AC voltage between pins 3 and 6 (or across capacitor 17) = 121 volts AC

Tube 21-1 (1N5-GT)

AC voltage ground to cap (pin 8) = 0 to 0.4 v as metal is placed by search coil.

DC voltage ground to cap (pin 8) = slightly negative v DC. (-1 or -2 volts)

DC voltage ground to pin 2 = +1.5v DC

DC voltage ground to pin 4 = about the same DC voltage as pin 3

DC voltage ground to variable capacitor (15) right side = 70 v DC

DC voltage ground to variable capacitor (15) left side = 79 v DC

AC voltage ground to variable capacitor (15) right side = unsteady.

Tube 21-2 (1N5-GT)

AC voltage ground to cap( pin 8) = 0.3 to 12 v as metal is placed by search coil.

DC voltage ground to cap (pin 8) = slightly negative v DC. (-1 or -2 volts)

DC voltage ground to pin 2 = +1.5v DC

DC voltage ground to pin 4 = about the same DC voltage as pin 3

21-1 Output as measured at transformer (3)

AC voltage from ground to transformer pin 5 (21-2 pin 3)= 58 volts AC.

DC voltage ground to transformer pin 5 (21-2 pin 3) = 97 volts DC.

I do free email advice, if you want to do some testing before attempting a repair. deweyhassig@gmail.com I have been an electronics technician for 35 years, most of that in an engineering area. I was an infantryman for 4 years in the U.S.  First Infantry Division.

Parts list (some values may be listed differently in other sources):

Resistors

11                    .5W     100 ohm

5                      .5W     22K ohm

10-1 , 10-2       .5W     1M ohm

6-1, 6-2            .5W     10K ohm

9                      .5W     4.7M ohm

8                      .5W     3.3K ohm

7                      .5W     2.2K ohm

12                    1W      750 ohm pot (1K ohm would work)

 Capacitors, as listed in one source:

18                    1uFd total Capacitance at 400V

19-1,                .02 uFd            200V

19-2, 19-3        .05 uFd            200V

19-4                 .25 uFd            200V

19-5                 .05 uFd            200V

19-6                 .1 uFd              200V

19-7                 .04 uFd            200V

19-1 through 19-7 in another source are all .1 uFd. I use .1 ufd or greater for all.

13                    .001 uFd (1000pf)       200V

14-1,14-2         500 pF (470pf is OK)   200V

15                    500pF adj

16                    800 pF                      200V

17                    .01 uFd (10,000pF)     200V

17A                 .002 uFd (2,000pF)     200V  

or 17+17A      .012 uFd (12,000pF)  200V 

I can send you a .pdf or .jpg copy of the schematic below, with pin numbers and values. deweyhassig@gmail.com