Smokin' Bass Amp!

[zebra50]

I'm having some fun and games this morning. :sad:

I'm looking at a Trace Elliot GP11 bass amp for a bass playing friend - he say it was making funny noises and not passing audio. I hooked up a junk test speaker and blew it up! (And the speaker fuse twice).

So I wired up another test speaker, this time with a chunky 33ohm resistor in series to limit current. The resistor started to smoke within seconds of turning the amp on - and red hot to the touch (ouch).

I measured the DC voltage across the output: -57V DC !

:shock:

Well that explains the symptoms, but I'm stuck looking for the cause. Has anyone had any experience trace elliott amps and seen anything like this? Anyone got a schematic?

Some more details:
1. The power supply is dual-rail ±57V.
2. I don't get continuity between the output and the -ve rail when the amp is switched off, so I don't think it's a simple short.
3. The 'bodies' of the output transistors are also at -57V.
4. The preamp section of the amp seems to work just fine.

Any ideas or suggestions would really be appreciated.
Many thanks in advance.
Stewart

 

[dale116dot7]

Any transistor, resistor, or diode that's wrecked in the power amp usually will cause a railed output. And that's what you have. There are a minimum of five but more likely 10 or more transistors in a standard power amp. Sometimes there are mirror-image input stages that go push-pull to the voltage amp stage. Check each transistor and diode in the power amp with a diode checker. If you don't have a schematic, check out the schematics of any standard "Lin" power amp and likely thing is it's a lot like that. If there are rail fuses then check those, but I assume you probably already checked. One other thing is a DC coupled input can cause this if the input is railed, though this is not that likely.

 

[Tekay]

Is the power amp a fet amp? or is that the GP7!
I'll check for the schematics when I'm in the workshop tomorrow!

In on of Traces latest dual outputs amp they use a relay that switches over the second amps output transistors to the first channel. The output transistors are not located at the same board ( mos fet's )
Have two burned out heads down at the workshop! When they starts to oscillate they burn out almost everything in the poweramp.

 

[gyraf]

If it's not a FET poweramp, start out with ohming the lower (negative side) output transistor (most likely a PNP). A short here will rail the output, errors mostly anywhere else will generate output voltages around 1V less than rail.

If it's the transistor that's shorted, also check it's associated driver transistor, otherwise you'll risk blowing the new output transistor you solder in.

If there's a pot for quesient (spell?) current setting, check this also.

Power up with 10R ½W resistors in series with the rail supplies. This can often save a set of power transistors if there's still an error.

If you have access to a vario transformer (in US: "Variac"), then use this for powering up.

Jakob E.

 

[Tekay]

A standard light bulb in serie with the main power is also a good thing to have! This was mentioned in the Peavey CS800 manual back in the late 70'th! 75-100w
I always have a cable with a serie lamp at my bench! Saved a lot of transistors that way!

 

[Gus]

Check the stuff in the above posts

Some other thing to check. First I have not been in a trace amp. This is what I would look for first.

Smell
Burn marks
check for power supply voltages
loose connections this is a bass amp thing vibrate alot.
Then
Maybe there are production diodes wired "backwards" from the output to the + and - supply and one is shorted.

If you can remove the transistors and check them for shorts from the output to input devices. Sometime the input diff pair can get a current surge and go bad causing a massive DC shift.

When you get a hard output to + or - it is often an output device: however it could be the drivers or a cap shorted causing a massive offset voltage.

Check that the feedback loop is OK this is often how the amp is DC controlled.

With higher power devices I like to test the devices out of the circuit It seem to save time and the fix is often a long term fix. Often if the SS amp Has multable output devices the ones that did not go bad work harder and are stressed and should be checked or changed.

 

[zebra50]

Thanks for all the tips. I'll check it out and let you know how I get on. I have asked Trace Elliot for a schematic.

Ther is a trim-pot on the board but without a schemo I have not yet figured out what this is doing - perhaps this is the quiescent (?) supply that Jakob mentions. What should I be checking for here?

Power supply ± voltages seem sensible and matched. The input/preamp section works fine and is pretty quient. It runs off the same power rails so I'm pretty sure they're OK.

There is no obvious sign of any burning or anything like that. I will check the diodes.

here's what it looks like:

The small transistors are marked F422/F423.

Thanks again.
Stewart

 

[Tekay]

That a fet poweramp! I think that those output transistors are obsolete now? Normaly there's one 1N4148 in serie with a Zener (10-12V) between the fet's gate and the output rail on both sides for safety.

 

[thermionic]

Tekay is right - Hitachi stopped making mosfets in T-03 cans several years ago. If you find some (I occasionally do) they are usually silly money.

If you find it is the T-03 ops then my suggestion would be to try Profusion for an Exicon / Semelab equivalent, although I?m not sure how well these will partner with the remaining Hitachi ops?

Justin

 

[Kev]

:green: :green: :green:

yes
I stock piled some when the news came through that they were to be discontinued. I have a few ZPE amps and I needed at least a few spares. I also made amps with the Hitachi/Tilbrook topology....
J49/J50/J56 and the other half 134/135/136 ... I think that's the grouping.
Some companies did stock pile many and so the source dried up very quickly.

I think there are a few equivalents out there but I don't know about their quality BUT if it gets you going again ... who cares ??

google search
good luck

 

[gyraf]

F422/423 are probably BF422/423:

http://www.hy-switch.com/zhiliao/zhi/t0-92/BF422.pdf
http://www.hy-switch.com/zhiliao/zhi/t0-92/BF421_423.pdf

Jakob E.

 

[zebra50]

Many thanks for all the tips. At least I can get started now.

I don't recognise the small blue diodes. Could they be the zeners to which tekay refers?

Stewart

PS - exicon/profusion have replacement equivs, so I should be OK -Thanks to Justin.

 

[gyraf]

If they're hooked between gate and source, they're probably 10 or 11V Zeners, protecting G-S from voltages higher than this...

 

[zebra50]

So, it's working now. But I don't really know how I fixed it, which bothers me.

Out of habit I replaced the small electro cap on the board last night, and turned the trim pot back and forth a few times in case it had gunked up and stopped bahaving correctly. I just plugged it in and measured the voltage at the output - 33 millivolts! Much better. Plugged in a speaker and connected the preamp - passes audio fine and is pretty quiet for hum and hiss.

The thing is I don't understand why either of these two things would fix it. Neither of them has more than a couple of volts on them compared with ground. The cap is connected to one of the driver transistors so maybe this was it.

It's one of those situations where a schemo would help., but trace elliot haven't been back in touch yet.

I guess I should just be happy it's working, but I'm worried that the fault is intermittent and will return.

:?

 

[Gus]

I think you might have fixed it. High Leakage in a cap can upset the the circuit in places and cause a massive DC offset that can't be corrected by the feedback. Or maybe the wiper of the pot was not making good contact.

 

[zebra50]

Thanks gus. I think this might be right. I am keeping the amp for a couple of days before I return it to the owner to make sure it stays fixed.

I certainly hope this is it as it is much cheaper than replacing the output transistors.

:thumb:

 

[gyraf]

If it was the set current pot that was open-circuit, that could have dragged one side of the push-pull set to rail..

 

[zebra50]

That makes sense - I can understand that.
Many thanks!
Stewart
:sam: :thumb: