old Gibson Atlas top hummm

[edenexter]

Dear all,

So I rebuilt my friends Gibson atlas amp. It seemed like a fun project and I could use the practice. The amp had so many problems plus there wasn't much original to save within the amp, so gutting it rebuilding it from scratch seemed like the best idea.
Here's a schematic if you're interested: schematic

The only parts that are still original are the power transformer, the choke and the chassis. I used carbon composites for most of the resistors and orange drop 715p caps for pretty much all the coupling capacitors.

For the elco filter caps I got 22uF's instead of 20uF as stated in the schematic.

I modified the bias circuit to make it adjustable, and I set my matched 6L6's to draw around 35mA at idle. Maybe a little low but bringing it up also brings up the hum I'm trying to get rid of.

I've star grounded the circuit at a point somewhat in the middle of the amp.

I'm using a proper three stranded power cable and the chassis is grounded with it (not at the star-ground point I'm afraid).

So my problem now is the hum and noise the amp produces even without a signal going through it. I've made a little video so you can get an idea of what I'm talking about. See: YouTube - Gibson Atlas Hum

My guess the problem is coming from the power transformer, if so I might not be able to fix this amp any further which would be nice to know. With the amp on standby the transformer hums somewhat audible to about the same frequency (to my ears at least). I've tried using different, known to be good, tubes with no result. Also poked around with a stick to see if the wiring might be causing the problems. Although I know I could have done a better job at wiring I don't think its the problem. Click for full size Sorry for the crappy phone cam picture... I can get better ones if needed.

I'm interested in whether all the wires are connected right since there aren't any markings on the transformer. For example the white and the black/yellow wire are connected to each other in this amp. Now I should probably mention I live in the Netherlands (230 v mains) and my guess is the wires are connected perhaps for this reason. I'm not even sure if this amp was produced for european usage. In any case I think the power transformer is original, the number on the cover matches the one from the schematic.

The output transformer probably isn't original since this amp now has a 4, 8 and 16ohm output.


I'm pretty happy so far, the amp sounds nice and warm. A lot of low-mids, which I kind of like. It's different.
In case you have any simple adjustments to the circuit that could improve it's sound, I'd be happy to know them. I read somewhere that lifting the .047uF caps (c14, c15) from the ground opens up the sound, which I might try.

[txstrat]

Can't make it out good enough in the pic. Do you have all four filter caps grounded in the same place? (grounded on the wire alongside the board?)
If so, I'd change that. I'd ground the main filter and the screen filter, as well as the bias ground, close to the power transformer (perhaps using a PT bolt) and ground the rest of the circuit close to the input jacks.
Just my 0.02$

[edenexter]

Thanks txstrat! You're right about the grounded wire running across the board. I'll improve my grounding and we'll see what happens! I'll also get some better pictures up.

[Bruce / Mission Amps]

My $.02
I Listened to it with my headphones and there are two frequencies mixed in there... 60Hz-50Hz and 100Hz-120Hz.
That means it is regular old power supply hum (filter caps and proper more quiet ground points) and filament supply 60Hz hum.
Where is the filament supply center tap grounded, where is the center tap of the HI-V secondary grounded and where is the first input stage grounded with respect to the cathodes and the input jacks?

[edenexter]

thanks for the help Bruce!

I don't have much experience with power transformers plus there wasn't much I could find about this particular one. The amp was somewhat working before I got to work on it, so the wiring at that point is pretty much all I had to go on.

I'm puzzeled by the white and yellow/black-dashed wires that are connected with each other. I expect this has something to do with getting/using the right voltage for this country, but this is a wild guess.
Click for full size
There is a yellow/green-dashed wire and a yellow/red-dashed wire, which I believe to be the center taps of the filament supply and the HI-V secondary. I'm afraid I don't know which is which.

I've now slightly adjusted my grounding points according to txstrat's suggestions.

Both the yellow/green and yellow/red wires are grounded at a star ground point on one of the output-tran's screws.
Click for full size
In addition I have grounded my inputs here and everything else in the circuit that needs to be grounded except for the four 22uF filter caps and the bias ground(R34, c16, c17). These are now grounded, as seen in the first picture, to a screw near the power-tran (to which the ground wire on my power cable is also connected).

So I believe the first input stage is grounded using the copper rod running along the the circuit board. Which is then grounded at the output-tran screw, as are the input jacks and the cathodes.

Thanks again!

[Bruce / Mission Amps]

Find out if that one wire that looks to be red/yellow is the center tap for the high voltage supply. The three wires in question are the red ones that go to your two full wave rectifier diodes and the third will be the hi voltage center tap.
With the amp off and the power supply fully drained... your ohm meter will tell you if the three are from the same winding.

If it is that red/yellow on it is probably grounded at the wrong place if not connected directly to the negative lead of the first maim filter cap.
Actually it looks like there are too many grounds, star ground straps and grounds on tranny bolts, (etc.) on this build and most likely some built in ground loops now.
Also, you can induce buzz and hum into a pair of power tubes by having your fixed bias supply (usually a half wave rectifier) built with too little filtering or the actual -vdc node right at the power tube grid load feed point, not having any or little filtering.
At that point it only takes a 22uF-47uF at 100v cap. Don't forget, it is reverse grounded though with the positive end at ground.

[edenexter]

The red/yellow wire is indeed the center tap for the High-v supply.

Seems I've made a mess out of my grounding! I've got my four main filter caps grounded to the same bus wire. You suggest I ground them otherwise and have the high-v supply ground tap connected directly with one of them?

I'll sort out my star grounding. I might have over-done it a little with the lugs (I probably don't need my pots individually wired to the star...).

About the half-wave rectifier. Click for full size
Seen here in the schematic if I'm not mistaken. These caps were actually of a much higher value before I rebuilt the amp so this might help. Which of the two do you suggest I swap out?

Sadly I live in a different time zone so I have to go to bed. Thanks so much for lending a hand.

[txstrat]

Just for the grounding issue you could have a look here: Grounds
Only the first two filter caps are grounded on the PT bolt, as well as the power tubes cathodes and the bias ground.
I don't use a special point for the ground of the OT secondary. I just use the speaker jacks ground and my amps are dead quiet. Maybe some room for improvement here, though.

The rest of the circuit (preamp, preamp filter caps, etc.) is grounded at the input jacks.
I usually solder a solid wire over the back of the pots for this purpose, but I think your wire alongside the board is fine too.
The wire over the pots backs is sort of a substitute for a brass plate in older Fenders

[edenexter]

Thanks for the link! I've now grounded the amp according to those instructions with quite a bit of improvement!

Here's another video of the amp so far: (WARNING: turn down your volume around the 1:20 mark because of a high pitch shriek) (I also mean modification not moderation).
YouTube - Gibson Atlas Hum part 2 (turn down speakers before 1:24: Shriek)
I've still got the the following grounds connected to a bus running along the board, that is in turn connected to the bus running along my pot's and input jacks.
Click for full size

Maybe I'll try increasing the value's of my bias supply caps.

thanks

[txstrat]

The shriek sounds like oscillation that might come from induced hum/buzz being amplified by the signal chain.
Take good care of Bruce's advices. He's one of the Gurus here and really knows his stuff.

[edenexter]

I've read a few things about proper lead dressing and I see I should definitely improve mine! I'll post after doing so with results.