Hiwatt custom 20 tube amp hum problem

[Nigel Goodwin]

Although I'm not too sure what the max input voltage to the Y amp is on any of them...

If you're doubtful about the scopes input, then use a x10 probe - but in either case, switch the scope to AC coupling, as you're wanting to measure the ripple and not the voltage.

But if you're not used to valve gear, be prepared to be horrified at the amount of ripple present

 

[mel8030]

If just wanting to determine if 60 or 120 hertz, attach scope to the speaker. I don't think a dried filter cap would give 60 hertz hum, but a leaky diode could.
Just a thought, what about one open diode and reduced filter capacity?
problem sounds more like floating ground problem on the input or filament leakage.

 

[frank57]

Should have some kind of answer this week.
I gave my tech the schematic that Nigel worked so hard on last week, so I hope it's not in vain.
I'm going to call my tech Monday and see what 's going on.

 

[frank57]

Okay, here's Nigel's schematic with a few corrections and just for fun, squiggly resistor lines.
It's not final, as I would need the amp to put in the remaining capacitor numbers.
I put in the tube socket numbers and pin numbers.
There were one or two errors on my layout so I'm going to post that later. Just one or two values were off. On the phase splitter I corrected a value(220k) that was a goof on my layout. Also put some of Nigel's comments in the top left corner.
If anybody spots a goof let me know.

 

[frank57]

Good news I hope!
Looks like the amp was fixed !
I'm going to pick it up tomorrow and ask my tech exactly what was wrong with it.
From what he told me, he put in a trim pot to adjust the bias and replaced a cap with a bigger one.
With the standby on the voltage was -24 and with it off -14.
Or vice -versa? Not sure.
The buzz appears to be related to the same problem.
I think he put the voltage at 35 and the problem was solved.
Perhaps somebody can figure it out exactly, but I'll check it out tomorrow.
Couldn't have done it without Nigel's help!
Hopefully no surprises tomorrow.

 

[frank57]

Okay, so I was told the amp was biased too low.
Now it's a at 35 volts. Does that make sense?
A cap was replaced and also a resistor I believe.
I'll take a shot of the board and test out the amp today.

 

[frank57]

Amp sounds terrible!
Chords are cutting off is it over biased?

 

[Chippie]

It would be better looking at the signal path through out the amp with a scope...That should give an indication as to what is happening....

 

[frank57]

He added a pot in place of the 10k resistor at R38 for the bias.
C23 was replaced with a higher voltage cap.
He told me it's not a cathode biased amp and that the original bias of -14 was too low,
and that it's class ab. Does that make sense?
What exactly is going on here? Is it still cathode biased or not or what was it originally?
I might tell him to put it back the way it was and to hell with it. Not sure what to do.

 

[Chippie]

No its grid biased, the cathodes are tied to 0v, the method of biassing the valves is achieved by applying a negative voltage to them which is what the diode and resistors are doing...The cap voltage increase has nothing to do with the bias voltage which is set by R38.....

To set the grid bias correctly requires the anode current to be monitored with a milliameter...

If the biassing is set incorrectly you could end up overdriving the valves causing them to overheat the anodes...( look for a red glow! ) or just poorly performing...

 

[frank57]

So this amp originally was not cathode biased? Just to be sure.
What exactly is the biasing type now with the pot? The same as before?
Is it a good thing or a bad thing that was done by putting in the pot?
Originally with the resistor the bias was -14.
The tone of the amp has changed dramatically and not for the better.
No singing tone here.
He told me the cap was no good and he replaced it with a higher voltage one.
I could leave the cap and put back in the 10k resistor does that make sense?
Leaving the pot a good idea?
I can't believe Hiwatt doesn't know the proper bias for their own amp.
Not sure what to do.
Aren't amps cathode or fixed bias?

 

[Nigel Goodwin]

You appear confused about biasing?.

It's really VERY simple - valves are normally fully ON - and by biasing the grid negative, it turns the valve further and further OFF, until you reach a point where it's entirely OFF. For an audio amplifier you want it biased somewhere in between those two extremes.

The thing to remember is that the grid is negative - WITH RESPECT TO THE CATHODE.

So if you're measuring with respect to the chassis you can either bias the grid negative (with respect to that chassis), or connect the grid to chassis and make the cathode positive.

The preamp stages use cathode bias, the grid is connected to chassis, and there's a resistor between cathode and chassis - as avalve is normally fully ON, currnet passes though the valve (and the cathode resistor). Ohms law shows you that a current through a resistor causes a voltage drop, so the cathode of the valve shifts positive, by the amount of drop across the resistor. This means the grid is now negative with respect to the cathode, so current drops, and the circuit stabilises at a level set by the resistor - it's often called 'auto-bias' - because it's a negative feedback technique, that adjusts itself.

Output stages sometimes use a similar method, but it has the disadvantage that you lose power across the cathode resistor - so they often revert to the far older method of grid bias - where you apply a negative voltage to the grid, and adjust the voltage for the current you want. This has the disadvantage though that current though the valve will change as it ages, and be considerably different everytime you replace a valve. Because of this it's usual to have adjustments, either just one, or a seperate one for each valve - meaning you need to adjust it, using a meter, every time you change a valve.

Here's a circuit of a Marshall amp that has adjustable bias https://www.drtube.com/schematics/marshall/1987u.gif

 

[frank57]

Is the Hiwatt a fixed bias amp?
He put in a pot at r38. Is that a good thing?
I'm just going to tell him to put it back the way it was and get rid of it.
Good idea or not?

 

[Nigel Goodwin]

Is the Hiwatt a fixed bias amp?

Yes.

He put in a pot at r38. Is that a good thing?

It might be - as long as it's adjusted correctly.

I'm just going to tell him to put it back the way it was and get rid of it.
Good idea or not?

 

[frank57]

Should I just restore back the way it was and get a refund if possible?
Don't know if he screwed anything else up in there

 

[Nigel Goodwin]

Have you spoken to him about it?.

 

[frank57]

Yes, I left two messages and am hoping he will do as I say. I said leave the cap in.
I went there yesterday and although the amp sounded horrid, he insisted -30 was good and the amp sounded good.
He did not want to put back to -14.
The bias is not answer. Adjusting it lower won't solve it and he already tried that.So what is the point?
I think there seems to be problem with the power tubes here and that one of them isn't equal to the other. Something like that.
I think we'll split the cost of the bill and settle it that way.But if he's telling me it's fixed he's nuts.
I said what was the bias Hiwatt set it at and he told me -30. They set it at -14 so what is this?
".

 

[frank57]

Okay got the amp back.
We compromised on the bill and he admitted he can't eliminate the hum.
He went ahead and adjusted the amp anyway which is what I expected he would do.
He said he forgot to seal the pot and it moved to -34.
It is now at -24. The amp still has a hum problem, but the buzz with the distortion seems to be okay.
Amp sounds like before at least and is useable.
I have seen some guides where -24 is considered okay for el84.
So we're sort of where we were with some small improvement and with me poorer.
Can I actually adjust the bias myself?
Shouldn't it be a trim pot in there?
I will post some pictures and check the amp further.

 

[Chippie]

The bias setting ought to be fixed ( I didnt see a preset on the pcb layout you posted) ...or at least the maximum setting should be that the valves arent overrun to the point of excessive dissipation..

"I have seen some guides where -24 is considered okay for el84." It depends on the HT voltage....Like I said before, the setting is a function of anode current without over stressing the valves..

For me there are still a lot of unanswered questions....

Still if your happy that's the main thing....

 

[frank57]

You mean it's still no good?
Check the schematic Nigel did.
I posted an update of it.
I think Hiwatt had it at -14.

His answer to his critics was "They don't have the amp in front of them"
He admitted he didn't know what to do for the hum.
I agree: a lot of unanswered problems here.