Solid State amp help? Not passing signal and f'dup scoping

[anticpunk]

Hey all,

My buddy has a 1978 Randall Sustainer 140. Someone partially recapped it and apparently replaced all of the FETs but it still doesnt work, so now the project is mine. The stock FETs are NLA, the replacements are almost identical in spec according to online references I have found.

Poking around with my scope, no input signal, I am finding an odd wave. It looks like a Sine wave on top, but the negative sweep is a tight V. It's definitely an AC signal, but I have no clue why it's there and why it looks as it does. The supply has a pretty clean output, Bi-polar +/-25. The + pole feeds the preamp while the power section is Push-Pull transistors running on dual rails.

The power amp is fine, if you break the connection to the preamp, its got nothing on the scope and very little audible noise. The preamp makes an audible (loud) low freq HUMMMMMM. on the scope, the wave changes (just gets fuzzier and thicker) when you turn the various knobs. Presence control tightens up the trace, but nothing makes it stop. Probing at the FETs shows the wave all the way back from preamp output to the Hi-z input.

There are no schematics for this amp, closest anyone can offer is the model up, the switchmaster 150. Similar, but not the same......

Can anybody throw me a bone before I start throwing parts at it? Am I better to just build a different front end, skipping the troubled board entirely?

Thanks!

Jay

[The Scum]

It looks like a Sine wave on top, but the negative sweep is a tight V.

Where do you see this? Power pins? Signal pins?

What's it's amplitude, in Volts p-p?

And what's it's frequency/period? I'm guessing 8.3 milliseconds.

[anticpunk]

Voltage that I'm referring to is found in the signal path, not the supply path.

voltage pk to pk is about 100 to 150 MV

period around 8ms.

I could try to get a pic of the wave if that helps at all.

any ideas?

thanks,

Jay

[The Scum]

Just a little piece of pedantry: MV means MegaVolts. Do you perhaps mean mV?

8msec? Just as I thought. 1/.008 = 120 Hz. You've got rectified AC finding it's way into things.

How much ripple is there on the supply?

[anticpunk]

Definitely milli, not mega. Pretty sure something would be on fire if it were in the mega range. lol

Ripple on the supply is really minimal. I replaced the 2 main filter caps and the bridge rectifier just to be safe and I still get the same wave on the scope. Caps were leaking and were apparently skipped over in the last recap job anyway.

Seems as though whatever is happening is keeping the Fets in an off state perhaps and thats where the lack of audio comes from?

If supply ripple was making that kind of noise in the preamp, wouldn't it also be presenting the same problem in the power stages? The only thing you can see on the scope at the speaker outs with the preamp disconnected is a little bit of "buzz", and thats all you can hear.

Should I grab some new FET and try replacing them with something closer to the stock ones? or maybe I just need to add more filtering on the DC supply to clean it up?

Anything else I can check that might help narrow down the problem?


Thanks!

Jay

[The Scum]

From a troubleshooting perspective, it's pretty cut and dried. Figure out the source of the hum, then figure out how it's getting into the rest of the circuit, and stop it.

Follow the hum upstream as far as you can...either towards the input, or the power supply.

Without a schematic, any comment on the FETs is purely a guess. If it's working, but really hummy, then the FETs are probably good. You say you were testing without input - is the input jack a switching type, that shorts to ground when nothing is plugged in?

A couple other suspicions: are the first-stage filter caps in the supply all good? Are all of the ground connections solid? The wave described exists in the power supply, so that's likely the source.

[suppositron]

What do you mean by "it doesn't work"? Doesn't pass signal at all?

Also, do you have your scope and amp plugged into different circuits? You could be creating a ground loop.

[The Scum]

One more thought:

Is there continuity from the ground pin of the plug to the amp chassis? Is the outlet grounded?

[anticpunk]

Ok, I pulled this thing apart and did some more research. The original transistors were TIS58, the ones stuck in it currently are BF245C. The datasheets don't seem too similar, doesn't seem like a good cross for this application to me.

Anyway, the transistors were pinned out differently. Pulled them and lined them up right and voila, it works....... kind of.

The big hum and ugly wave form are gone, but now the signal coming out it very very low even with controls maxed.

Is it possible that the gains of the original transistors and of the replacements are that drastically different?

TIS 58s are NLA, so any advice on a proper replacement would be appreciated.

BTW- The outlet was grounded and the scope was plugged into a different circuit.

Thanks!

Jay

[anticpunk]

Actually looking further into some older spec and cross sheets, looks like a 2N5458 might be a better solution.

The sheet at hand is :

http://www.datasheetarchive.com/TIS58-datasheet.html#

Has the TIS 58 on page 6, the 2N5458 is on page 5.

-Jay

[The Scum]

BF245C is an RF FET. They may not make very good audio amps.

2n5458's are 25 V max parts - you'll be right on the edge hanging them on a 25 V rail. J201's may be a better choice, rated for 40 V. Both are cheap jellybean parts.

JFETs are really loosely spec'd. It's very possible that Randall sorted and selected the original transistors to fit the circuit.