Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Peavy PA hiss/hum

Status
Not open for further replies.

throbscottle

Well-Known Member
Has anyone experience of Peavy amps? This is a GPS 2600 that has hiss and a bubbly sort of hum on one channel. Source is before the volume control, and it dies down after a while. I'm guessing PSU fault, but can anything else cause it? Any known issues?

Thanks :)
 
Has anyone experience of Peavy amps? This is a GPS 2600 that has hiss and a bubbly sort of hum on one channel. Source is before the volume control, and it dies down after a while. I'm guessing PSU fault, but can anything else cause it? Any known issues?

Thanks :)


just as a general starting point
what have you done so far ?
checking electro caps etc ?


Dave
 
Has anyone experience of Peavy amps? This is a GPS 2600 that has hiss and a bubbly sort of hum on one channel. Source is before the volume control, and it dies down after a while. I'm guessing PSU fault, but can anything else cause it? Any known issues?

Do you have the circuit diagram for it?.

If the source is before the volume control, it's not very likely to be the PSU - although this obviously depends on the circuit.
 
Oh I so wish I did! Most I could find was a user guide.
I haven't actually done anything yet, I needed to fix my 'scope first, Cleared up most of the faults on that yesterday so I can inspect the amp with it. Also was waiting for any responses here - so thanks guys! Actually doing anything to the amp at all (more than sticking a probe in) is something of an undertaking, so I want to try to get a good idea where I'm looking first. I guess I'll start with the volume control and work my way back to the input...
 
The noise seems to be originating from an NE5532 dual op-amp, or something connected to it. I guess it's just gone noisy - are they known to do that?
But something Really Bad happened...
I fed in a 25mV square wave from my scope's calibrator so I could trace the signal path. Whilst doing that, suddenly the transformer started making getting-hotter sounds and the output stage started getting hot. It looks like the output is shorting the power supply. It's a bipolar totem pole design with lots and lots of parallel output transistors, some of which may be failing (well, one definitely has). This happened with no speakers connected, which I didn't think would cause any harm! (The handbook doesn't mention it) I suspect it was going to fail anyway :(
So, I'm thinking it's going to be the biasing or the driver that's causing the problem but since this is a over 1000W per channel PA there's going to be more to it!
Anyone got any suggestions what I should be looking for?
 
Running transistor PA's with no speakers doesn't cause failures - but I'm VERY concerned by you feeding the calibration signal from a scope in to a PA amp :nailbiting:

You mention 25mV?, is this a specific external calibrator?, and not just the scopes built-in probe calibrator? - which are usually 1V p-p or so.

Did you have the master volume turned up?, and if so why? (as you were looking for preamp faults) - it's not likely to do much good feeding massively over driven square waves through a PA amp, it's not something I would do, and it's quite likely to destroy the power amp.
 
Definitely 25mV p-p, built in scope probe calibrator. It is a very old scope though. I started with the volume turned down to zero, and turned it so I could just see where the signal was coming out after volume control, no more amplitude than the noise signal itself at 5 - 10mV after the control. Enough to identify which o/a does what.
I think I may have just been unlucky, since one of the output transistors is shorted but it's balancing resistor is open so it's effectively out of circuit. I think that's a historic fault though. So it may have been on it's way out anyway.
I was looking on that site last night. I ddn't realise you can d/l without making a donation!
 
I fed in a 25mV square wave from my scope's calibrator so I could trace the signal path. Whilst doing that, suddenly the transformer started making getting-hotter sounds and the output stage started getting hot

very bad idea .... no wonder things were heating up

but I'm VERY concerned by you feeding the calibration signal from a scope in to a PA amp :nailbiting:

me too as I said above, a really bad idea, audio amplifiers don't like square wave signals, it will push the amp into strong distortion

Always use a sine oscillator
 
One dead 2SA1302, one dead 0.33R 7W resistor, and one dead (3 diodes short) bridge rectifier for the low voltage stage output supply. OUCH! Can't find any other failed components.
What I think happened is that the transistor failed first, blew it's balancing resistor, turned all the other transistors on in that stage, which then overloaded the rectifier and killed it. I still think the transistor may have been ready to fail anyway - I just pushed it over the edge.
So, can someone explain how such a small signal caused this damage? Considering it was turned down so what came out of the volume pot was much smaller than what was going into the amp to start with. Is it purely because of it being a square wave? I would have thought the amp's response would filter out the dangerously high harmonics!
I see now that the speaker acts as the ground return to the transformer for the entire channel (which was quite a surprise!) so without it, uhmmmm, one side conducting would push the other side into conduction, but they should be being turned off by the input signal at the same time so nothing should happen. So does this come down to phase, then?
 
Last edited:
I guess it's time I built myself a sig jenny.
 
Thanks for the responses btw :) Really appreciated :)
 
Okey dokey. Amp back up and running. I replaced the suspect op-amp. Next problem was humming when an input connected. Dodgy xlr connector it seems. Switch cleaner and jiggling about sorted that.
So, it still hisses when switched on, but for a few seconds now instead of a few minutes, so I'm hoping the improvement is due to the op-amp being replaced! The only other possible culprits afaict are the 2 resistors in series with the o/a inputs. Don't know if it's worth disturbing them. Do metal film resistors ever get noisy?
 
Do metal film resistors ever get noisy?
Not generally, to my knowledge, but if they've got damp could be...

I rather disagree with the other contributors above though - I don't think driving a square wave into an unloaded amp should cause any trouble. Obviously it pays to go very gentle with an amp you suspect is faulty, but it doesn't sound like you did anything untoward.
The only danger I can see would be cross-conduction due to the fast rise times - one set of output devices could be turned on before the other set had turned off completely (because BJTs will turn on quicker than they will turn off). Even then, the slew rate would normally be limited at the voltage amplifier stage, so that seems very unlikely.

More likely is accedentally causing or a short or - more subtly - changing something like gounding, screening, stray capacatence etc, and causing an ultrasonic oscillation which could destroy the output devices by the mechanism above.

It's not unknown for op-amps to go "eggs and bacon" (hissing, popping, crackling), especialy is they've been abused in the ways that happen to the inputs of audio amps.

You have proably done anyway, but its worth checking all the emitter resistors, just in case there are more open. Keep an eye on the bias currents, too, and check that the two channels stay at roughly the same current (working on the assumption that you're unlikely to have two channels faulty in exactly the same way). Also check you've no open B-E resistors in the output stage (a potential source of less-obvious problems).

Finally (you're probably waiting for someone to tell you this) you should really replace the entire bank of transistors to get decent load sharing... however I'm well aware that's proably more than you'd be willing to spend. For practical purposes, I'm sure it will be fine as it stands.
 
Parallel transistors should be matched. Not identical, but close.

Back in the 80's, I bought a Harrison Labs (they were bought by HP) power supply, Odd beast: Selectable 0-32V, 10 A; You have to change taps and you end up with a fine adjust, Just got a manual not too long ago.

There were a few out of place transistors and when I did a load test (automotive headlight) they started to pop. I matched them and it's been good to go ever since.

Square waves might make the protection circuit (DC on speakers) upset with no load.
 
Parallel transistors should be matched. Not identical, but close.

No need, if the circuit is correctly designed - and if it's not correctly designed, even matched transistors will fail in a fairly short time.

You simply use balancing resistors in the emitters, end of problem - and this amplifier (like anything decently designed) already has them.
 
I'm glad I probably don't have to replace any more transistors, and I did think I might have to. I did worry because all the others have b-e and b-c forward voltage of about 0.46 volts, but the replacement (and one other of the originals) are closer to 0.6. But my thinking was like Nigel says - the emitter resistors should take care of it. I couldn't find any other components that looked out of spec. I guess they just shouldn't run it at full power...
Now you mention it Tomizett, there are a couple of what look like water-marks, so there may be some mineral deposit there. Don't think I can clean it without un-soldering stuff though.
I've come across op-amps that have gone noisy before. I like the term "eggs and bacon"! I'll have to remember that.
Given that the output failed when I was probing op-amp pins with the 'scope probe, I suppose it's most likely to have caused a transient it didn't like.
 
I'm glad I probably don't have to replace any more transistors, and I did think I might have to.

My concerns in such cases isn't down to the specs of the transistors - but why they actually failed.

In your case it's probably because you 'did something it didn't like' :D

However, it's commonly down to another faulty component (usually a transistor elsewhere), which may also be intermittent.

When you repair amplifiers professionally, as I've done' it's good technique to replace all transistors in the DC chain - this greatly reduces 'come backs'.
 
Thanks for the heads up! But I'm doing this for cost, otherwise it would be a different story!
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top