Reverb recovery stage with AC125 transistor in Philicorda GM751 organ - problems

[bram9206]

Hello all,
It's been a while I posted on this forum, but I still remember all the great help and info here when I was building my two Pultec EQP1A (fully point-to-point) clones, so I was hoping you guys could help me again. Here I go:

I have here an old Philicorda GM751 Philips organ of which the reverb is broken (it's not fully broken, it's just very very quiet), schematic of the amp section and reverb recovery is attatched. Now, when I touch the springs I can here them rattling, but that rattling sound is much too quiet compared to when the reverb was still working. This makes me think the problem lies within the reverb recovery stage.
I've already replaced all electrolytics in the organ, replaced C679, C680 and C681 (see schematic), measured all resistors in the reverb recovery stage (they measure out just fine) and replaced the AC125 germanium transistor as well. Also checked for cold solder joints (chop stick test) and broken wires, but no results.

Still no reverb, BUT... here's the strage thing and I discovered this by pure coincidence: when I connect a clip wire to ground (chassis) and touch the point where R678, R679 and C676 meet (so I actually short that point to ground) I hear a pop in the speakers and the reverb is back! Again, I didn't connect that point to ground, I simply touched it with the clip wire that is connected to ground. So when I do that, the reverb is back, but only for a short moment. Sometimes it's back for 20 secondes, sometimes 1 or 2 minutes and than suddenly it's gone again (or at least it's very very quiet). I replaced R678 and R679 just to sure, eventhough they measured out fine, but that didn't make any difference.

I'm totally puzzled by this and I was hoping someone could shine a light on this for me.

Thanks a lot!

 

[JohnRoberts]

Inspect for bad solder connection, perhaps a leaky cap.

JR

 

[bram9206]

Already went through a bad-solder-joint-test a couple of times without results. All solder joints seem to be perfect. Leaky cap seems out of the question too since I replaced all caps with new ones (expensive TAD and Spragues ones).

 

[JohnRoberts]

was problem happening before caps were changed...  if not check polarity of electrolytic.

If you don't have a scope, perhaps use a VOM to compare DC operating voltages before and after messing with the ground.

JR

 

[RuudNL]

Did you check C676?

 

[bram9206]

@JohnRoberts: Good idea. By the way, I do have a scoop! Still pretty new ground to me, so you'll have to guide me through making measurements with it...

@ RuudNL: C676 is a brand new cap, so I don't expect any problems there...

 

[mjrippe]

@ RuudNL: C676 is a brand new cap, so I don't expect any problems there...

Ah, but that's not what he asked ;-)  Brand new caps can be bad, have the polarity marked wrong, be of the wrong value or voltage, etc.  Rare these days, but it does happen and I've seen it first hand.  Take a minute to double check just in case.

 

[JohnRoberts]

@JohnRoberts: Good idea. By the way, I do have a scoop! Still pretty new ground to me, so you'll have to guide me through making measurements with it...

@ RuudNL: C676 is a brand new cap, so I don't expect any problems there...

+1 brand new parts can still be bad or installed wrong.
=========
You don't want to pay what I want an hour for personal hand holding,  8) but we are both in luck. The schematic has typical operating voltages shown nearby important circuit nodes.

These will not be exact votages but check them with your VOM or scoop and any serious outliers will point you toward where to dig deeper.

I'll try to help from a distance. Good Luck... It's always something.

JR

PS: Some guy PM'd me on another forum today asking my personal advice about what modern op amp to use............ seriously? 

 

[dmp]

It sounds like the bias of the transistor might be off. Check the voltages at the pins. They should be close to the -10 and -4.3 as noted on the schematic.
If you post a good pic of the circuit around the transistor someone here might be able to spot something.

 

[bram9206]

I'll check that brand new 100uF cap tomorrow, just to be sure... I sure is installed with correct polarity

As for voltages at the AC125, I have -4,45V at the base and -10,55V at the collector. Seems pretty, if not very close to what it should be (see schematic). I also measure -4,36V at the emitter, is this normal? Schematic doesn't say anything about a voltage at the emitter.
All other voltages mentioned in the schematic are very close or even right on to what I measure with my VOM.

I installed a NOS AC125 and reverb still doesn't work. But NOS doesn't always mean it works, right... maybe the AC125 can be suspected? Or do the voltage readings imply the AC125 works fine?

 

[CJ]

that AC125 might be biased too cold, when you short the cap, you bleed off the charge, the transistor turns on but the voltage on the cap builds back up and shuts off the transistor,

hook your probes from E to B with pwr off, turn on power and short out the cap like you did before, watch the voltmeter, i bet it drifts down to 0.09

you only have a diff of about 0.09 for Veb, so something else is going on, check the pc board for cracks in a trace with a mag glass, or a possible solder bridge when you replaced parts,


Ge transistors need about 0.3 volts to turn on,  measure all points around that transistor and post that,

 

[PRR]

> Ge transistors need about 0.3 volts to turn on,

Sometimes less.

Who has a handy  1960s transistor manual..... oh, right, me. 2N407 says 0.15V at 2mA. 2N109 says 0.2 to 0.4V but at 50mA, which hints at 0.1V-0.2V at a half mA. 2N270 shows 0.19V at 19mA, which might be very close to 0.1V at half mA.

Agree that 0.09V looks low. But not for-sure wrong.

 

[bram9206]

Voltage across base and emitter is 0.087 volts.

Here's the frustrating part, the more I get into this organ, the more I get puzzled:
I changed out the new 100uF cap and guess what: the reverb worked just perfect after the replacement! Turned the organ off and on, still worked. Switched it off again, went out for dinner, switched it back on: reverb still worked!
Then went to bed, turned on the organ the next morning and guess what... reverb was gone...
Then I did this short-thing again and the reverb kicked in again. Then switched the reverb off on the organ, waited for a minute and switched the reverb back on (switch is named SK10 in the schematic), result: reverb was gone again.
Then did the same short-thing over again and the reverb was back, BUT... now it stays, even when I switch it on/off for a minute, reverb stays. Switch the organ off for a minute, switch it back on and reverb is present. So now it seems to work again, but I bet if I go to bed and switch the damn thing on again in the morning it won't work.

As you guys can see, this is a very weird situation and it seems there is no sense behind it...

I also made that measurement across emitter and base, which is 0.087 volts. Does that mean the AC125 is good? I also measure -4,35V at the emitter (in respect to the ground). Is this normal?

Also tapped all the solder joints and traces on the PCB board, but no results there... I've once red that there are bad solder joints that work well when cold, but after a warm up they don't work anymore... maybe something similar here?

Almost going mad...

 

[RuudNL]

I am sure that the transistor is not the problem.
If it works sometimes, that means the transistor is doing its job.
Transistors simply don't stop working one moment and start working again some time later...
I suspect one of the traces on the PCB or a cold solder joint.
It should not be too difficult to trace the problem: 1 transistor in a common emitter configuration, 4 (5) resistors and a capacitor...

 

[bram9206]

I've once red that these germanium transistors can sometimes act strange and that they can work differently when subjected to for instance cold weather. Or is this complete nonsense? I'm not really familiar with transistors...

 

[JohnRoberts]

+1 check for marginal solder connection.

JR

 

[bram9206]

Are there other ways to check for bad solder joints or bad traces in the PCB? Or is tapping them with a non-conductive stick the only way to do so? Because I've already went through this test a couple of times without any result...

 

[JohnRoberts]

Pencil eraser is the official tool... another troubleshooting aid is spray cold... The fact that your symptoms take time to come and go suggests maybe flexing the PCB, but don't break anything..

It might be helpful to repeat your DC measurements when passing signal and when not.

It's always something...

JR

 

[bram9206]

I've gently twisted the PCB in the hope to find a possible crack in the traces, but no results, reverb still works. So now we are at the point the reverb works (eventhough we can't say the problem is fixed) and we don't know what is wrong... Makes it only harder now to try and find the problem... I can only hope that it will break down again and I can continue my search.

 

[Nele]

Did you check all the fuses? I had this exact same problem with a GM754. After changing the driver transistors, caps etc. the reverb still didn't work.  It turned out to be a hidden fuse at the bottom of the case whose connections were oxidized..