KAT88 playing up circuit diagram needed

[gilbodavid]

hi.

I have a Kat88 power amplifier that seems to be going into oscillation with all speakers.

My repairer, Henry, needs the circuit diagram to do the repair.

Does anyone here have one? Thank you!

David

[gilbodavid]

some closeups pics of the circuit board

[Richard]

Hi David,
Looks a nice build from a while ago. There were one or two changes early on but yours looks to have included them. Has it just started playing up? 1 or both channels, HF or LF or a hum?
Send me a PM with your direct email and I'll send the info you need.

[Richard]

Thanks David, info sent

[shunt54321]

I built one of these back in 2000 I think. I have lost the schematic too so a while back I took it apart and draw out the circuit with full intention of rebuilding it.
Sadly it is now defunct as there are a number of quite serious design issues, one of which finished it.

The main theme is almost everything seems operated right on its maximum rating, like a 3W resistor has 3W dissipated in it etc. and then put in a hot box. The tubes are biased far too hot. An attempt at "pure class A"!

I rebuilt this amp many years ago after it red plated a tube, resulting in it's cathode bypass capacitor exploding, stinking out the dining room for a few weeks!

About a year later after repair it ate some more tubes with a little firework display so gave up on it for a decade or so.

I now have it stripped to find one of the output transformers has it's primary open cct. EOL!

When I draw out the cct. I realised just how many serious errors there are in this design. Like 8 Volts of ripple on the screen grids. This explains why I was never impressed favourably with its sound quality. That will have modulated the audio with 100Hz.

So put 1KHz in and out comes with 900Hz & 1100Hz sidebands.

Also as far as I can tell the preamp HT supply over volts it's supply filter caps on power up, until the filaments warm up. Couldn't quite believe that so had to triple check my cct.

It's all fixable but not in my one. So if you want any spares let me know.

If you want advice on how to make it sound better and make it safe to operate then let me know. It needs the preamp supplies sorting. It needs its screen grid supplies smoothing. It needs a re-bias to class AB to stop it lunching tubes.

But if nothing else IT NEEDS HT FUSES !!!!!!!!!!!!!!

Regards
Stu

PS. If anyone has an original cct I'd be intrigued to see how accurate my scribbled cct. is?

[bob orbell]

A freind of mine built one of these in 1995, it is still going and sounds very good with a few mods that Richard has spoken about, may be you just got a bummer or it was poorly assembled. Bob

[shunt54321]

I built it!

Exactly as instructed. That's where I went wrong.

As I said the sonic problem with this design was the 8 Volts of ripple on the screen grids. Since there is virtually no G2 current this would be very easy to smooth. The 16V of ripple on the anodes doesn't matter that much as it is the screen voltage that controls the gain of the tube in a tetrode/pentode.

This modulates all your music with 100Hz

This is not exactly rocket science. Even my old Sound City guitar amp does this. Standard practice.

This amplifier is a cross between fixed and self bias. I assume the thinking was to get the best of both worlds, i.e. the performance of fixed biased but with no need for adjustments.

It looks like the HT was around 590V from measuring the main TX, possibly more where I live. My mains is always >245V. I've calculated the bias is set to a nominal 60mA. This involves a bit of iteration between the calcs and the plate chart (JJ's), and some extrapolation, as the chart doesn't go to 600V!

Remember that 600V HT on the anodes is under no signal conditions. Under full drive Va will go beyond 800V (Limiting Value).

Anyway that puts the plate dissipation at around 35W, nominally. That is spot on the design limitation for a KT88

So how fixed bias is it and how self biased? We have about 27 volts of fixed bias and about 9 Volts of self bias generated across the cathode resistors (@ 60mA). So it's ¾ fixed and only ¼ self bias.

Since it is majority fixed bias, tube to tube variations will be considerable. Matched tubes are matched to each other within the set. They are not matched to any nominal value of bias point. So one set of matched tubes put in this mostly fixed (literally fixed), bias amp could run way hotter than the next.

Therefor you may get a set of tubes that run really hot and just fail early. Or you are lucky and get a set of dull tubes that run a bit cooler and go on forever.

Of course those dull tubes will draw less HT current, creating less ripple and thus probably even sound better as a result. But sort that G2 ripple problem and you'll wonder how you ever put up with it.


Just trying to help.

Oh yes, did I mention? Fit some frigging HT fuses!!!!!!!

That simple and absolutely necessary and standard precaution I see on every single guitar amp I've worked on would have saved my output transformer. At least it wasn't an expensive one!

All spares available if anyone can use them.
I'd love to see an original cct to compare to my scribbled one.