Re: ampeg ss150 supersonic/parasitic oscillation

[ilyaa]

okay some progress, but we might need to reassess our theories.

i DONT think its an oscillation issue. looking at a larger amplitude waveform from my guitar, there is NO discernible fuzz or distortion other than what the amp naturally provides. it looks pretty clean, in other wo

[Roly]

Hi ilyaa, and welcome.

part of this weird thing with a cap right on the output. i guess its called a Zobel Network

You guess correctly.  A Zobel resistor burning up is more likely to be a symptom than a cause, but check that the Zobel cap hasn't shorted.

As an aside I'll just note that some amps fit the Zobel on the speaker side of a stability choke, which is incorrect; it should be fitted on the transistor side, and VHF-style with the shortest possible connections.

The pop/thump is also constant with the amp taking off into supersonic oscillation, and normally this isn't very kind to the output stage or speakers and should not be allowed to continue.  When it's oscillating you may also notice a reduction in output and a degree of distortion, however it would be helpful if you could confirm that it is indeed oscillating using a CRO, or a multimeter on the AC range with a small blocking cap (say 0.1uF or what have you) in series with the meter probe.

While this sort of thing is fairly common initially in home builds it is fairly rare in commercial builds.

Is it control-sensitive (volume and/or treble normally)?

Roughly how old is this amp?

Oscillation of any sort requires a feedback path, so we are looking for something like a grounding path that has changed, meaning that two stages are sharing a ground path in common and are thus now coupling.  This also happens when a power supply decoupling cap dries out or goes open, allowing signals to pass back to earlier stages via the power supply rails, however this almost always results in low frequency "motorboating" (putt, putt, putt), not supersonic oscillation.  So first I'd be looking for any grounding that might have worked loose or gone high resistance due to age and corrosion, and for any HF bypass cap that may have broken off or developed a dry joint.

If the various sockets are mounted on a PCB then have a close look for any fractured/wrenched solder joints, grounds in particular, that may be forcing signals to take the long way around.

If there is nothing physically apparent it might be a bit tricky to pin down, e.g. a small cap gone open internally, the main suspects being any HF bypasses on the supply rails, and there is often a small stability cap on the Base of the Voltage Amplifier Stage/pre-driver in the power amp.

Please PM me the circuit.

Keep us posted.

HTH

[ilyaa]

okay i made a little progress with the scope, i think.

put in a 1K sine wave and got a 1K furry sine wave. this is my first time using an o-scope to test an amp, but i compared it to the same wave into my homemade champ and, although that wave had a little dip in it (probably something i should address at some point, too), it was certainly not furry like the one coming out of the ampeg.

looks a lot like a) below



i guess b) is a parasitic oscillation, so perhaps thats not my issue, after all??

anyway, hope the scope information can guide us in the right direction. furriness is bad, right? i should be getting a nice clean sine wave, particularly since the sound out of the speaker is a totally clear tone (i have the o-scope across the speaker leads). whats a good next step? should i start going backwards across the board with the scope until i find the spot where the wave goes bad? that sounds fun, kind of, is it a good method?

[Roly]

furriness is bad, right?

In a small critter it's good, in an amp it's bad.

should i start going backwards across the board with the scope until i find the spot where the wave goes bad? that sounds fun, kind of, is it a good method?

That sounds like a very reasonable plan, but don't forget to check the power supply rails first - and keep your head screwed on about Y-sensitivity, a cranked CRO may show residual stuff that is of no significance; e.g. fur like that on a 1mV signal is to be expected, on a 1V signal is significant.

Go get 'em.   :dbtu:

[J M Fahey]

Keep it simple.
As a first step, check the input and the output of the power amp.
For now we'll be happy to know whether it receives the furry signal or it adds it to an otherwise clean one.
By the way, is this the Chorus model?
Because chorus circuits add a high frequency modulation signal to audio, which *normally* is either balanced or filtered out .... yours might be defective in that way.
I'd love to see an actual scope picture (hint hint) but the example you show points at the chorus origin and is not really dangerous (at worst , annoying).

As of fur, furry and all that jazz ..... are you making fun of *ME*?  :trouble  :trouble :

[ilyaa]

okay i think i need to rewind a bit - afraid im going to get in over my head without the tools to get out.

roly's comment on the amplitude/furriness relationship alerted me to some potential misinformation i might be providing.

im relatively new to electronics stuff and this is my first time using a scope to troubleshoot an amp, so bear with me as i get my bearings. the furriness ive been repeating indeed appears when the scope is cranked, at about 0.4 V p-p some other stuff starts to happen and the overall fur seems less apparent....help me sort it out:

here is my measurement technique -> amp plugged into an 8 ohm speaker -> scope probes across speaker leads; my computer is generating a 1K sine wave that i am putting into the amp. i was keeping the signal kind of quiet because that tone is freaking annoying, but i guess i have to crank it a bit to really see what's going on, huh? i guess ill have to get a dummy load....dont have one. i was looking at a relatively quiet signal before, so maybe the furriness was more due to noisiness inherent in magnifying something teeny on a scope.

what p-p voltage should i ideally be looking for at the output of an amp on a scope? in terms of clearly seeing what i need to to troubleshoot?

and keep in mind the amp works except for an occasional and quite loud THUMP, possibly a supersonic oscillation.

as ive messed around with the amp, changing levels and moving where im scoping, im getting all kinds of crap, from a sine wave with a weird dip in it, to something totally beyond recognition, to a curve that just looks like a cap charging and discharging. but no where am i getting a clean, normal looking 1K sine wave, except right at the signal generator output (before the amp).

in the interest of making this troubleshooting process efficient, anyone care to give me some quick guidelines as to where would be most useful to scope (right before and right after power amp, for example) AND what i should really be looking for? what range of p-p voltage? should the wave start getting all wacky if i boost any of the EQs? should gain and level be cranked as im scoping? in other words, who wants to buy me a new amp?

should this amp be giving me a nice sine wave? or am i looking for something different altogether?

thanks guys.....

[ilyaa]

oh yeah one other thing:

im supposed to be getting +/- 39v on the rails (according to the scheme), but i am getting +38 and -48...is the too large negative voltage perhaps a symptom that i should investigate further?

[Roly]

afraid im going to get in over my head without the tools to get out.

Ya jes' lookin' ol' son.  As long as you don't do anything that can't be undone you should have no worries, and a reasonable chance of winning.   :tu:

im supposed to be getting +/- 39v on the rails (according to the scheme), but i am getting +38 and -48...is the too large negative voltage perhaps a symptom that i should investigate further?

A difference of that order is certainly something I'd be following up, and it also takes us in a different direction.  It suggests that the -ve rail isn't loaded for some reason.

Disconnect the speaker and measure the DC voltage across the output, should be less than 200-300mV.

You didn't confirm that after the thump the signal level drops and is distorted.  If this is not the case then I'm starting to doubt this is an instability, but perhaps a sudden large DC shift somewhere.

I'd be inclined to hook your CRO and DMM across the output and play for a while until the thump shows up and note what happens, DC volts on the output?  Big fur on the CRO?

Now you know why most techs have dummy loads.   

what p-p voltage should i ideally be looking for at the output of an amp on a scope? in terms of clearly seeing what i need to to troubleshoot?

At full output you should be getting tens of volts, but about a volt into the speaker is normally all you can stand in the workshop.

As preamp EQ can play merry hell with your waveshape I'd suggest you try injecting your test signal at the Main In socket, then it should be an honest sinewave to the output when things are healthy.

[ilyaa]

okay some progress, but we might need to reassess our theories.

i DONT think its an oscillation issue. looking at a larger amplitude waveform from my guitar, there is NO discernible fuzz or distortion other than what the amp naturally provides. it looks pretty clean, in other words.

the amp has been behaving strangely. the previous time i plugged it in, it played PERFECTLY without thumping once for like 15 whole minutes. cranked, too! after successfully fixing my acoustic 370 via a deep clean, i thought id do the same to this one. AFTER the clean - the boards looked alright, but there was some stuff here and there, notably on some of the interboard connectors, the power supply filter caps, and on the mainline fuse - i turned it on and right away THUMP THUMP THUMP. in fact, it was doing something weird at first where it wasnt working at all as far as passing signal. it wouldnt switch channels and i could not get it to make a sound....

after i turned it on and off that went away and it was working fine, but THUMPing.

when a THUMP occurs:
1) signal goes away but comes right back afterwards
2) the LED channel indicators flicker and switch between themselves
3) DC voltage on unused speaker out - the amp has two parallel outs - goes from between 150 and 300 mV (when not THUMping) either down to near 0 or up to as high as a volt or down to like -700 mV (during a THUMP). in other words, kind of wacky. is that an okay way to measure DC volts at that point, with the amp plugged in to a speaker and the measuring tools on the other speaker out?
4) o-scope shows what you'd expect, the waveform flickers and goes nuts for a split second.

the THUMPs seem to be totally random. sometimes itll be just one in a long while, sometimes they will happen many one after another. so far, symptoms seem ambiguous

BUT

earlier i said that i measured the voltage on the rails and got something strange. where i shouldve seen +/- 39V, i get +39 and -46 to -52V. so somethings weird on the negative end of the power supply. i followed my nose and scoped the rails and saw something strange.

on the positive rails, the scope shows a small but very steady ripple - about 20 mV. on the negative rails, the scope shows a ripple that is twice as large, about 50 mV, but this ripple is NOT steady. every so often it totally wigs out, turning into some crazy zig-zaggy angular thing, sometimes for a few seconds at a time - it gets distorted and spikes and does all kinds of stuff. it returns to the normal ripple waveform, but this process is repeated and predictable.

so something is up with the negative power rails. now, unfortunately, if i have the speaker plugged in and watch the scope on the ripple, the ripple distortion/spikes do not necessarily correspond to the THUMPS. only sometimes. they happen much more consistently and frequently than the THUMPs do. what are we looking at here? bad filter caps? is this a likely cause for the amp making the THUMPs? what do people think?

a side note: at one point i was trying to measure somewhere else along the power rails and touched my ground o-scope probe to a point i thought was ground (turns out it was V+) - it arced/sparked and the amp blew a fuse!! wtf. luckily, i had a replacement handy. the fuse that blew was 250V/5A. i put in a 250V/3A slow blow. are slow blows okay for solid state mains? what the f happened, anyway, to make it blow?

ALSO, sometimes when i turn the amp off, 2 volts remain on the power rails and the Red ON indicator LED will stay lit!

similarly, sometimes after a THUMP the amp will stay on channel 2 (the channel switch wont work) and it will not pass any signal. then randomly itll start to work again.

[Roly]

what the f happened, anyway, to make it blow?

The amp supply common is connected to mains ground.
Your scope common is also connected to mains ground.
Ipso if you touch 'scope ground to a supply rail it will short it.  BOOF!

A Slo-Blo should do fine for testing (but where is your limiting lamp, eh?  Saving a small fortune in fuses is one of its functions.)

This amp has Pre Out/Main In, right?  Have you tried the "good lead between these" test yet?

Plugging a (shorted) lead into Main In will soon tell you if we are looking at the power supply/main amp (if it still thumps), or preamp (if it stops).  This is important to cut our problem in half - if you short the Main In, or feed a signal in, and still have bumps and thumps the problem is after there, but if then works fine it's coming from the preamp before there.

BTW, have you yet tried cleaning the Main In socket (in particular its bridging contacts) with contact cleaner/metho and working a plug in an out a few times?

[ilyaa]

there is a molex-style connector between the pre- and main amp. its got 5 pins, a power ground, a V+ and V+, and a signal and signal ground (i think). should i remove the connector altogether and just fashion a cable that will allow me to send signal into the main amp via the two signal pins left on the male end of the connector?

the amp THUMPs when its just sitting there everything turned all the way down nothing being played. could i just as well disconnect the preamp and just turn the amp on and see what happens? saving me the trouble of trying to inject a signal in between pre and main amp...if it THUMPs then we've achieved the same half-splitting diagnosis, right?

[g1]

 Just plug a cord into the main in, with or without signal.  This will disconnect the preamp without even having to open the amp.
 

[ilyaa]

wait, what's the 'main in'?

[ilyaa]

oh wait you mean the 'effects loop'?

sorry got confused by the terminology. if i plug something to the main in it will short out the preamp? cool ill do that.

and i did clean those contacts on the outside, but not inside - ill do that, too.

[Roly]

sorry got confused by the terminology. if i plug something to the main in it will short out the preamp?

The insertion point between the pre and main amps has several names, Fx Send/Return, Fx In/Out, Pre Out/Main In, but they are all pretty much the same thing.

Specificly, there is a switch contact on one or both of these sockets that disconnects (opens) the preamp from the main, allowing the insertion of other equipment into the signal chain.  These contacts very often give trouble, particularly in rigs where the loop is never used so they never get a scrub to keep them clean, go intermittent/open, and break the signal chain.  You will find these on the power amp circuit, top left, J1 and J2 - notice that J2 has what is called a "bridging" contact which is supposed to carry the preamp signal across when nothing is plugged into the Fx Loop.

The simple test is to plug a known good lead into both the Fx Loop sockets, bypassing this contact.  If this kills the thumping then it points directly to these contacts.

You can also inject a signal directly into the main amp at Fx Return (but make sure the volume is turned down on your MP3 player or whatever to start with - that's your only level control.)

HTH