Ashdown MAG250 problem

[shane]

Hi Folks

I've been trying to fix an Ashdown MAG250 which a friend of mine owns. It had blown a fuse and was not producing any sound.
I replaced the output transistors and checked all the others on the power board but still it was showing an apparent short (light bulb limiter stayed lit up and the amp produced a loud hum).So checking more deeply I found that the positive supply was being pulled way down and the negative was pretty much OK.
So I did some more measurin' and figured I may have a shorted smoothing cap(one did measure low resistance) and changed the main caps.
Now plug her in (still with light bulb limiter) and hey she's OK...BUT... turn it on with the speaker connected and we're back to the positive side pullin loads of current...ahhh.
Weirdly if I turn it on without the speaker connected then plug the speaker in its fine! even if I then turn it off and on still with the speaker connected it's OK but if I leave it a while and turn it on with speaker it swings again and I have to disconnect the speaker and plug it back in to get it to run OK.

Ok so I'm stumped...(not difficult!).. Anyone got any Ideas? I know amps can tilt at turn on but dont normally stay that way!

Cheers
Shane

[Roly]

By "fine" I assume you mean that the half rail is balancing to half the available rail voltages.

On the face of it, if the half rail is latching then it implies that the negative feedback loop isn't operating (at that time).

I suspect that this may be because the two current sources, TR6&7 and TR12&14, are not getting sufficient voltage to operate correctly, that the bottom one is dropping out of conduction and allowing the other to pull the output to the +ve rail.

Can we have some supply rail voltage measurements under no-fault and fault conditions please?

[shane]

Hi Roly
Thanks for helping out it's very much appreciated!

Yes thats what I mean by fine. As for the rail voltages they are as follows:

No load attached: -66.5V and 66.5V
with load and fault occuring: -17.5V falling slowly to -9.7V (tho previously it was about-35V) and 8V
Disconnect load and reconnect load: -66V and 66V

These readings have changed, as I said above, during fault conditions the neg rail was sitting at about-35V but now its reading as I quote above.

Do these readings agree with your ideas about the current sources not having enough voltage? if so what could cause that?

I have little experience with solid state power amps having been a tube amp guy for quite a few years and only doing this and another SS amp( a Warwick xtreme 10.1 thats eating output transistors like it going out of fashion!). for some friends.

Cheers
Shane

[Enzo]

Some thoughts.

Your light bulb gets bright when the amp tries to draw excess current.   If you have no load on the output, the output can swing over to a DC voltage, but there will be no current draw without the load.  But when DC is present on the output, a load there will try to draw all the current Ohm's Law predicts.   That is why your amp makes a loud hum and draws current with the load, there is DC there.

When an amp seems to settle OK when unloaded, but acts funny with a load, often it is because the output cannot supply the current the load demands.

First this:  troubleshooting is troubleshooting, I don't care if it is tubes or solid state.   Shorted output devices, be they 6L6s or transistors, will likely blow fuses.  Problem solving is all about tracking down the source of the problem, not guessing Gee, maybe it is this part, over and over.

You replaced the output transistors.  Did you check ALL the associated resistors?   Did you check all the -.33 ohm ballast resistors for opens?  Did you check that 470 ohm R12 for open?  TR8 TR13 nay check OK with a hand meter, but if I have to replace the output transistors, I replace the drivers as well.  Which is what those two are.   They may be unable to reliably operate at current loads in the amp, even if they seem to work at the minimal current and low voltage of your meter.

R21 is a current sample off of one output emitter back to the bias, is that part healthy?

And another important question.  Is the power amp input still connected to the preamp?  It should be.   I don't know specifically about this amp, but on MANY solid state power amps, the input differential stage has its ground provided by the incoming signal cable.  If that cable is off, the inpit ground floats and cannot reference itself, and that will upset anything the feedback tries to accomplish.

[shane]

Thanks Enzo
I'll go over it all again and check those parts you mention. I have checked the emitter resistors and they seem ok as do all the other transistors, but did only check em with a meter so...

I really take on board what you say about troubleshooting too. For year I definately had a blind spot over SS amps but have realised that I should 'get over it' and get to grips with them. Thanks for the advice and the gentle kick in the pants!
Cheers
Shane

[shane]

OK.. so as per Enzo's good advice I went over all the resistors and they all measure good. And out of curiosity I checked the old output transistors again and guess what.. none of em were bad they all measured ok with the diode test on my meter!
Guess that'll teach me not to jump to conclusions!
If the drivers were bad surely they would be bad all the time?

Amp still has the problem.
Turn it on with no load and there's no DC on the output.
Plug a load in while it's on, still fine. Turn it on with load connected and bam.. light bulb lights up and theres about 6.6V DC on the output, (Enzo definitly on the money there).--disconnect load and its back to normal.
Reconnect load while on, still OK...arrrghhh.


One thing I notice is that when its pulling the excess current the Fan doesnt come on nor does the meter on the front panel light up so it must be pulling all the voltages to the pre-amp board down sinc thtey all come off there.
I'm wondering if the problem is in the mains transformer in some way. Dunno why it would be so intermittent and load dependent though, maybe I'm clutchin' at straws! Just at the mo' I'll clutch at anything to save pullin my hair out...!

Cheers
Shane

[J M Fahey]

Then you don't have a problem.

*IF* the amp turns on "right" with no load, when you add a speaker (without turning it off ) it stays normal and when you apply audio you can hear it through the speaker, (up to a couple watts) , while at most the lamp bulb limiter starts blinking along the music, then the amp is fine.

*****MANY****** amps are "stupid" when turned on with a current limiter and load applied.

But they work properly when connected straight to mains, so test it as suggested and post results.

[Roly]

The thought behind my post above was that the current source/sinks are not working correctly when a limiting lamp is in series and you try to drive load (pulling the available voltage down).  With a limiting lamp I would expect an amp to only be able to produce a fairly small output before collapsing.

If you are worried about whacking it straight on the mains (despite having the mid rail balance correctly, which is generally a sign that all is well), then simply try lamps with increasing wattage.  I expect that as you do you will find that the amp behaves more and more as it should, then you can go to unlimited mains (speaker off) check that the half rail is, and there is no DC across the speaker connection, then you can try reconnecting the speaker and see if you have normal operation.

[shane]

Well, I tried doing as JM said and it played fine with the lamp connected, not even a flicker of the lamp!
So bit the bullet and tried it straight to the mains and again it was fine.. so I guess you guys were on the money! Experience shows!!!

Roly I think your reasoning is correct, a case of the helpful device not being helpful eh...

I'll keep trying it over the next couple of days but I reckon its OK now. Keep your fingers crossed.

Thanks so much for your help and advice I learn something new every day here!

Cheers
Shane

[Enzo]

The helpful device is very helpful, within the context of what it is designed to do.

[shane]

Absolutely Enzo, I use it every day, Just learning its limits! :tu:

[J M Fahey]

Just in case, for the first time apply your NASA calibrated Finger-O-Meter Mark XXVI to the heat sink, say every 10 minutes, until you are satisfied it does not overheat.

As long as you can touch it and leave your finger there, it's fine.

[shane]

OK JM will do!