Quieting an old beast

[black_labb]

I have a couple old amps that run 4 6550's. one of them has a problematic hum issue since I got it. What is curious is that the hum is coming from the power section. Pulling the PI tubes doesnt eliminate it. I've replaced the bias diode/cap as well as the main filter caps which has likely been in there since the 70's. both have been replaced with slightly higher values and voltage ratings.

Here is a schematic of an amp from the same company with the same power amp/power supply (but values are a bit different as mine is for 6550's not el34's)
http://www.ozvalveamps.org/holden/schematc.gif

Notice the voltage doubler and that the grids/preamp are fed from the mid point on the series caps. The amp does not have any resistors in parrallel with the main filter caps and the pc board does not have a place for them either.
Is this a problem? I would put some in to try it but I cant find any that would be a suitable value (but I have every other possible combination).

Otherwise the schematic is pretty much the same.

I can see that the grounding scheme wasnt very well thought through when I look into the amp. I dont think the hum is from a ground scheme issue though as it is from the power amp. I've checked the ground connections and they seem to be in good condition, though I may give them each a bit more solder incase that is the issue.

The other odd thing I've noticed is that the filter caps drain naturally within 5ish minutes after replacing the big 4 caps. (remember that those cap bypassing resistors are not in the amp). Is this a sign of something problematic?

The voltage on the transformer is 280vrms, giving the plates around 700v and the grids 350v.

Are there any other things that I'm overlooking?

[black_labb]

quick update

retouched the ground connections, same hum.

The caps draining i'd say is from the tubes still being warm enough to conduct as they didnt drain when the amp had no tubes in.

[jmaf]

Two ideas: Are the tubes well matched? And do the heaters have a virtual ground or center tap to circuit ground?

[black_labb]

Thanks jmaf

I'll check the tubes. I dont have a well matched set but I have 12 used 6550's, so I'll be able to see if it makes a difference. The grids are fed from the mid point of the voltage doubler which is only running half wave rectification so this could well be the issue.

the heaters are centretapped to ground for the pream/PI. the power tubes run from 24vac running the power heaters in series.

[jmaf]

The grids should be fed from the negative bias supply which is ok to be half wave, but it is not ok if that filter cap is leaky, it'll inject hum right into the most sensible circuit of the power amp.

I like how the suppressor grids are fed the signal and a negative bias voltage instead of hooking it up to the cathode. I think this was used by Traynor on some model(s?) to preserve EL34 tube life IIRC. It's one of the advantages or using a "true pentode".

[black_labb]

Thanks Jmaf, I typed a reply a while ago but must of not clicked send.

the hum in the power amp is gone. was partly the power tubes and partly some imperfect grounding scheme. I'm slowly working my way to the front of the amp in terms of hum reduction.

I've cleaned up the ground scheme and completely redid it to a star ground scheme. I've left the output jacks grounded at the sleeve but the rest is at the star ground point, so its not perfect as the negative feedback current is going through the chasis but its better than most amps around instead of ground scheme. it's still has a fair bit of buzz which I expected to be gone after the grounding overhaul.

The buzz I am trying to get rid of is still fairly strong with the preamp tubes pulled, but the LTP PI tubes in place. The treble control of the fender/marshall tone stack just before the PI is able to effect the buzz dispite the tube feeding this being pulled, so I've narrowed it down as something causing trouble at the input of the PI.

The odd thing I've found is when the pre tubes are in place and I probe the power supply rail to the preamp I get a bang coming through the speaker. The supply rail shouldn't make any noise. I have recently replaced the filter cap here so it shouldnt be the problem (the hum was there before as well). I've checked the resistance from ground to the cap so it's not a pcb trace broken. Is this a sign of something? It doesnt make sense to me but I cant see why it happened like this.

I also find that one of the input channels (basically 5f6a bassman preamp) actually cancels out the buzz at ~11oclock on its volume control which again suggests the buzz is coming from the PI input when you consider the phasing.

thanks

[jmaf]

Hi black_labb, if you are 100% sure the caps are fine, then the pop may have come from your ground which may be somehow deficient and you modulated the zero v rail instead of +B when you touched it. That would indicate there's a remaining ground problem.

When the negative feedback side of the output is grounded to the chassis, you've got to invert that lead on the jack and ground the other end of the output transformer instead. Keep the negative feedback tap on the same wire, just not grounded. Right now you are running without any negative feedback.

The buzz cancellation at 11 oclock may indicate another problem, it's likely you've got supersonic oscillation at that level and the oscillation is damping the buzz, not eliminating its source, you just don't hear it. Double-check for that possibility. If the plate voltage on the preamp tubes drop a lot at 11 oclock, it means they're oscillating, you're just not hearing it.

Hope this helps!

[black_labb]

Hi black_labb, if you are 100% sure the caps are fine, then the pop may have come from your ground which may be somehow deficient and you modulated the zero v rail instead of +B when you touched it. That would indicate there's a remaining ground problem.

When the negative feedback side of the output is grounded to the chassis, you've got to invert that lead on the jack and ground the other end of the output transformer instead. Keep the negative feedback tap on the same wire, just not grounded. Right now you are running without any negative feedback.

The buzz cancellation at 11 oclock may indicate another problem, it's likely you've got supersonic oscillation at that level and the oscillation is damping the buzz, not eliminating its source, you just don't hear it. Double-check for that possibility. If the plate voltage on the preamp tubes drop a lot at 11 oclock, it means they're oscillating, you're just not hearing it.

Hope this helps!

Thanks again Jmaf

I've just checked the grounding and the positive rail in the preamp and everything seems fine. I'll have to try another cap. If the cap were gone then that would explain the oscillation you are predicting.

I've explained the negative feedback thing badly, its not an issue and wasnt worth mentioning but it is working as it should.

I'll do some more poking around and see if there is infact oscillating in the preamp.

[black_labb]

The voltage on the plates doesnt change as I adjust the volume control to where the buzz goes away. I get the same bang noise from when I probe with the multimeter before the voltage dropping resistor in the B+ line as well which is very odd. could be my cheap multimeter i guess.

[black_labb]

The buzz seems to be coming from the PI area. I'll try replacing the CC resistors.

[tubeswell]

Hiya black_labb

I have found that the 'galactic' ground method advocated in Merlin's article and also in R.G. Keen's article seems to produce the quietest amps I have made so far. But with this method its very important to have each ground return path going - via its own separate wire - to the same ground return point as the relevant supply filter cap. JM2CW

[black_labb]

Thanks tubeswell. I've done that thinking that that was the issue but it made no noticeable difference.

I'm pretty stumped at the moment. I thought I had found the issue when I found that the resistor connecting the PI and -ve feedback to ground was blown so I replaced that. It cured the slight scratchyness on the presence pot but didnt do a thing for the buzz. Probably a good thing I found it though. I've also replaced the resistors in the PI section of the amp. I've been really good at narrowing down the possible places but it doesnt help me find the issue.

[black_labb]

I've just measured voltages on the PI which seem correct. When probing pin 2 or 7 (the grids) of the ltp 12a*7 the hum goes down alot, so the issue is at the input of the ltp somehow. I've replaced the resistors involving the LTP and disconnected the wire leading to it to see if that was picking up hum with no noticeable improvement.

[tubeswell]

try bypassing the presence pot (to ground) with a bit of wire. (I had a similar problem in a 5F6A once where putting my meter on the input grid of the LTP got rid of the noise. Turned out to be a dodgy presence pot - just a hunch.)

[black_labb]

wasnt the issue tubeswell, had me excited though. This amp has a resistor bypassing the presence pot which had blown and I've replaced previously, but wasnt the issue.

you know when you've narrowed the issue down to such a small spot and it still doesnt present itself and you feel like throwing the thing out a window dispite being so close?

edit: this is the schematic I've been staring at for too long with the same PI/poweramp as mine except mine has a presence control in parrallel with the 4k7 resistor at the bottom left of the schematic.
http://www.ozvalveamps.org/holden/sl...rcit200wka.gif