Faint "fizziness" in Ampeg GT10

[yaryaryar]

Hello.

I'm a relative beginner to amp repair, but I've gotten pretty comfortable inside of tube amps.  I don't know the first thing about solid state, though.  What better way to learn than just to dive in?

I bought an old Ampeg GT10 (schematic attached) that has this really faint distortion that sort of sits on top of a note.  It's tough to describe, but it's not like the note itself is distorted, it just has this "fizziness" that follows it, and is most apparent when I let a note ring out.  It occurs at any volume.  It almost sounds like a problem with the speaker, but I've tested the amp with a known-good speaker in a different cab, and the problem persists.

Just as a blind guess, I went ahead and replaced the filter caps - C30 and C31 on the schematic - as well as C29.  No change.  Can anyone suggest a potential next step? 

[Seeker]

Howdy

I too am new at ss amp repair, so I'm afraid I can not offer assistance.  However I will be interested to read the replies.  I have the same issue' I think' with my '75 Yamaha g100.  I describe it as the way the note/sound "decays" and I too thought of speaker problems but tried what you did with no effect. Volume does no effect it either.  I do not make my living with this amp and in reality it does not bother me (well just a bit).  I do wonder if it's a "precursor" to a bigger problem down the road.
Thanks for your post,
Seeker

[Seeker]

In addition....
I had not used this amp for over a year due to another issue...Just played it for the last hour and I lied..it really bothers me.  I think the "frizzyness/decay was even more predominant after the amp "warmed" up.
Seeker

[J M Fahey]

Crossover distortion caused by poor biasing.

Now don't go randomly turning the little pot wheel hoping to fix it, you may very well burn the amp.

The amp schematic probably includes a note explaining desirable bias adjustment.

If not, post the schematic and we might be able to suggest something.

[yaryaryar]

Thanks for the replies here.

I don't see any bias pot...  although it's entirely possible that I'm missing something. 

I thought I did post the schematic, but who knows.  I'm new here.  I'm going to try to attach it to this post as well.  Here's hoping.

[Seeker]

Thanks
Seeker

[Enzo]

The Yamaha very likely has one, but none on the Ampeg.

Just for science, plug the guitar into the reverb footswitch jack, with the reverb up to hear.  That bypasses the preamp.  Is the fizzy still there?

There are DC voltages on the schematic, are any of those way off?

All those small electrolytics in the signal path are likely in need of change, but don't do that "just because", troubleshoot to find out if they are the cause.  My working suspicion  is a low level DC problem somewhere.

[Seeker]

Hey Enzo
Yamaha here,

Tried the reverb thing and yes, the frizzies are still there.  I'm too tired and frustrated tonight (not having a good health week) to find the proper schematic.  I'll look tomorrow when I'm rested and hopefully feeling better, in regards to the dc voltage etc.  Thanks again.
Till later
Seeker
PS: good to narrow it down to the power amp..thanks....will remember this "trick". My amp does not have pre-out or power in...I forgot that I could use the foot switch control circuit.  One quick question...Is the distortion circuit between the pre and power amp or "in front" of the pre-amp....hmmmm I guess I will find out by looking at the schematic.

[Enzo]

The reverb footswitch trick was for the AMpeg, it is only coincidence if it works on your Yammy.   I see that it would.

This is why we prefer to have separate threads for separate amps, it gets real confusing working on two totally different circuits at the same time.

Is this your schematic?
http://bmamps.com/Schematics/Yamaha/Yamaha%20G100%20Service%20Manual.pdf

If so, post it in a new thread, which you can start for your amp.

[yaryaryar]

Thanks for the idea Enzo.  I would have never thought about plugging into the reverb switch jack.  Plugging in from there, the problem persists.  So that means it's somewhere in the power section, right?

I checked voltages where they were notated on the schematic.  I'm attaching here the schematic with my readings in red.  In general, my voltages were a touch higher, but as far as I know, that's normal considering this schematic was written in 1971.

The only place I thought was a little too high is right around Q7 and Q8 - the schematic has it at 6.5v and I'm reading 7.5v.  Is that still within the margin of error for 1971 voltages vs 2016 voltages?

I hope the attached schematics/readings are legible enough.  Please let me know if I can post this in a more clear or understandable way.

[J M Fahey]

Confusing:

Hey Enzo
Yamaha here,

The reverb footswitch trick was for the AMpeg, it is only coincidence if it works on your Yammy.   I see that it would.

This is why we prefer to have separate threads for separate amps, it gets real confusing working on two totally different circuits at the same time.

Is this your schematic?
http://bmamps.com/Schematics/Yamaha/Yamaha%20G100%20Service%20Manual.pdf

I'm attaching here the schematic with my readings in red.
(AMPEG schematic attached)

:loco

Back to the Tech side and supposing the OP is actually asking about an Ampeg:
data shown backs my suspicion of crossover distortion due to underbiasing:
1) it's a very crude power amp, straight drive to power transistors without drivers and they are not Darlingtons either.
2) there is only one diode (presumably similar to 1N4002) trying to bias two BE junctions (Q10 and Q11).
3) measurements show 0V drop across bias diode.
Not very trusty because both are measured relative to ground, please measure and post actual voltage across the bias diode D1 .
4) also measure voltage across R44+R45 or emitter to emitter at Q10/Q11

[yaryaryar]

Sorry for the confusion.  We're back to the original question about the Ampeg GT10.

Across diode D1:  -.7v

Across R44+R45:  0v

Across E10 to E11 (which is really the same as R44+R45):  0v

(To be exact, those 0v readings did read at .1mv for a second, and finally rested at 0.0v.)

The diode seems right, compared to the schematic.  But what about emitter to emitter?  Can that be right?

[Enzo]

it is "right" because Juan is right, the thing is biased too cold, by design, so the output transistors are not idling while turned on.  That results in crossover distortion.  When the outputs are not biased to turn on, no current flows.  No current flowing means zero voltage drop across those emitter ballasts.

The two output transistors need to be biased just at the onset of conduction.  And as Juan said, that means there are two junctions to overcome, the base to emitter of each.  Even the voltages on the schematic say only 0.8v between the bases of Q10,11.

A cheap and dirty way to do this might be to add one more diode in series with the existing D1.  But chances are that would push you a little too far into conduction, and cause overheating.  Maybe.  So in parallel with that new diode, we could wire a little trim pot wired as variable resistance.  I have no idea, maybe 100 ohms?  Start with the pot at zero ohms, then slowly advance it until the crossover notch disappears.  Or until a little voltage drop starts to form across the emitter ballasts.

[yaryaryar]

Thanks so much for looking in to this.  This is definitely over my head, but it's so interesting.

I know you guys are the experts here, so I don't want to sound like I'm second guessing.  But I do want to understand this problem a little better.  So if what your'e saying is right, and this is just a design flaw, then wouldn't this problem have been present since the amp's birth?  We're talking mid-70s.  And while there isn't a lot of info about this amp on the internet, this particular model was in production for probably 8 or 9 years.  Don't you think Ampeg would have had to address this problem at some point?

Or, on the other hand, is this a problem that, given the design flaw, wouldn't necessarily be present from the beginning?  As in, the poor design caused this distortion to only appear after a certain amount of use/time?

I will try the trim pot idea - that sounds safer than the possible overheating.  I'll have to order the part, so it'll be a few days before I can report back on anything.  In the meantime, if there's any alternate theories here, let me know!

Thanks again.  You guys are amazing.

[J M Fahey]

^^^^^ What Enzo said [tm]
The answer is correct, only problem in *this*  amplifier is thatsince it's missing drivers, it passes a much higher current through D1 , so the classic 100 ohms trimpot in parallel with one diode happens to be too large "here".

IF you find a 10 ohms trimpot, I *guess* it will do, but in practice, I would add a series extra diode in series with D1 with a fixed 10 ohms resistor in parallel and check what happens.

2 precautions:
1) an error here, specially an open circuit, will fatally overbias output transistors , so build and use a lamp bulb limiter (there's a sticky thread in this Forum) , I suggest a 25W lamp, first plug amp, turn it on, and notice filament colour, it should be red or dark orange at most (it will blink for a second at turn on, not a problem) , then add the diode+resistor, turn it on again, measure emitter to emitter voltage.
You are aiming at, say, 5 to 20mA idle current which will be "seen" as 5 to 20mV across emitter resistors, since they total 1 ohm.

If less or still 0, double 10 ohms resistor to 20 ohms and so on, now you became a "human trimmer"  :lmao: and viceversa, if too much reduce it.

You will "waste" 1/2 hour doing this, still WAY less than waiting for Mail to bring a 10 ohms pot ... which might even not be right ... a lot of "seat of the pants" guessing here 

2) wait 3 to 5 minutes after soldering anything to the added diode, soldering heat changes its voltage drop big way (that's its job) .

Blowing on it might not be enough, we all expel air at 37C , so maybe a small fan (or fanning it with a piece of cardboard )  might be indicated.

Even holding the diode in your hand will change readings 

3) just leave everything "in the air" while testing, only after you find proper value you mount it in a "nice" way and solidly to last forever.
You might cut a track leading to/from D1 to some another component and solder "D2" there, then add a dab of glue so vibration does not tear it out.

4) post results for future users.
Also: horrible as it seems today, cold biasing was very common in the old days, customers "were used to it", called it "transistor sound", earliest SS power amps (VOX/Standel/Peavey/Acoustic/Carvin/SUNN/Leslie/Gibson) used transformer drive which passes very high base currents and is very hard to thermally compensate , so some didn't even try to.

EDIT: I was writing while you answered but anyway I guess I answered your still unasked questions.
So add one extra feature to SS Guitar: "Mind Reading"  :lmao: