Decouple this preamp?

[CJ]

Got a Fender Frontman 212R that has a strange problem, once in a while it will chop off the bottom half of the output wave,

common problem, about 300 complaints on the internet but nobody has come up with the answer,

the amp will latch up in this failure mode and then start working again.

touching a voltmeter lead to the cathode of D4 makes the amp work right again.

messes with a new TL072, messed with the feedback circuit, took some good and bad voltage measurements,

bad Veb volts shown on pic,

the diff pair works off the +/- 42 volt rails which have a bit of dirt.

i was thinking of putting in a resistor / cap filter section on the rails to the diff pair,

maybe bad filter caps?

do you think this will help?

Thanks!

full schematic attached as pdf

 

[Philbo King]

You might try a reversed Schotty diode (Q11-C to anode, Q11-B to cathode).

 

[EthanSpaulding]

Might be worth checking to see if the 0.22R resistors on the output are still soldered in solidly. Their extra weight may be able to cause them to break their own solder connection when tossed around. Then the intermittent connection is disrupting the feedback loop.

 

[CJ]

I reflowed all the power resistors and big transistors also.

I snipped the 10 if f/b cap too, that made it a bit worse

Weird how the take f/b off the plus and minus spk wires,

Also bridged a .01 directly across opamp, no change

Strange that I can make the amp switch back to clean sine wave simply by touching that diode.

 

[heatwalk]

I had this problem with an old SS Ampeg a few months ago that ended up being an intermittently failing output transistor. Wouldn't fail till it was passing signal for a few hours.

 

[CJ]

Thanks for the suggestion!

I have some filter caps ordered. I notice one was getting slightly warm but not much.

This has to be a design problem as there are so many threads on the net with the same problem, the same clipped bottom half, and even the same "touch the screwdriver to the diode" fix.

Once you unlatch it with the screwdriver trick it runs fine for a while.

These direct coupled amps can be tough as the problem in one area will affect 5 other places.

There is about 0.65 volts offset on the plus spk wire when latched and just a few millivolts when working fine.

Very interesting problem.
Wonder if it could be a layout thing?

 

[cyrano]

Weird how the take f/b off the plus and minus spk wires,

Wouldn't that make the circuit microphonic?

 

[CJ]

Actually the minus speaker lead sits on a parallel pair of 30.22 ohm resistors to provide a small f/b voltage, then it goes through a network where it gets combined with the f/b coming of the plus spk lead. Why the extra path I do not know.

 

[Gus]

CJ it looks like mix of voltage(from +) and current(-)feedback

The opamps are coupled to the input pair sometimes you will see series resistors in the base leads for stability(IIRC)
Think of the amp like a big opamp
Did fender change the orignal speaker to a different one? I ask because of the mixed feedback

look at page three source impedance effects
https://www.jensen-transformers.com/wp-content/uploads/2014/08/an001.pdf
https://www.ti.com/lit/an/slyt087/s...68990&ref_url=https%3A%2F%2Fwww.google.com%2F
https://sound-au.com/articles/current-drive.htm
https://sound-au.com/articles/guitar-amps.htm

 

[Matador]

Strange that I can make the amp switch back to clean sine wave simply by touching that diode.

Maybe solder in a 1M resistor permanently to simulate the voltmeter probe?

 

[CJ]

i think that is what Philbo King was hinting at, i got a Bat41 going in there right now,

thanks Gus! i think your right about the current f/b,
i was thinking that phase shift from the f/b network might be having some effects on this circuit,
so i swept the amp up to 50 k hz but thae amp did fine,

 

[JohnRoberts]

Symptom sounds like maybe an open DC path at op amps + input.

JR

 

[CJ]

Thanks John, i will take a look,

i think i might have had a bad solder joint on q11 because after soldering just one end of the diode to the collector, the amp works, going on 3 hours now.

diode does not harm the sine wave in any way, so i stuck the other end on there as well and stuffed the chassis back in the box, show time in about 5 hours...

​

 

[CJ]

after 5 hours of working good it just crapped out again, so got it back on the bench,

 

[CJ]

tried putting an rf generator antenna near the circuit on both Am and FM while sweeping, can not get the amp into failure mode

tried heating up all the transistors with the soldering iron, no change,

tried running the amp at higher wattages to generate more heat, no change,

then, i accidentlly touched the speaker leads together while running at about 5 watts and the amp when into failure mode (bottom of wave chopped)

so there is something unstable about the feedback loop from the speaker leads,

 

[Gus]

CJ

That is the reason I posted
Did fender change the original speaker to a different one? I ask because of the mixed feedback.
Maybe the amp was designed with a different speaker and then it was changed during production?
The speaker is part of the feedback network and inductive etc.

 

[CJ]

Speaks are stock ,

I am going to replace the transistors that come after the diff pair, right before the amp craps out it makes some fizz that sounds like a P/N junction going south.

 

[JohnRoberts]

The 1.55 Veb for q13 marked on the schematic suggests a faulty device.

Veb should only be around 0.6V.

JR

 

[moamps]

By the name Veb, CJ probably meant the voltage between the emitter of the transistor and the negative supply voltage. After all, if that transistor is faulty, the whole amplifier wouldn't be able to work at all.

And to be precise, VEB on that transistor should be around -0.6V. VBE should be 0.6V.

 

[JohnRoberts]

as usual I am not as good of a mind reader as others...
====
A Vbe of 1.55V is very unlikely (without magic smoke release) unless junction is reverse biased.

JR