Bad hum - multi-section caps?

[casey73]

I just finished a homebrew and am still tweaking it a bit. Preamp is a 6SL7 into a EL34 set up as a triode and a solid state rectifier. One volume control between the triodes. Real simple set up.

I have a real bad power supply hum, even with volume turned down. I carefully star grounded the amp and isolated the input and output jacks. My only PS cap is a JJ 40/20/20/20 multisection cap top mounted to the chassis. I chose this because space was at a premium. I used one of the smaller hammond chassis for this project. I've used LCR dual caps a few times in the past with little problems, but I've read that multisection filter caps are prone to hum issues. Could this be the source of my hum? I've built a number of similar small amps in a similar fashion and haven't encountered this type of hum issue, but it's the first time I've used a four section cap.

I've thought of a possible fix, assuming the multisection cap is the problem. I thought I'd parallel all the sections together to give me 100uf on the plate supply. I have just enough room on the preamp side of the amp to add up to two Sprague Atom 22uf caps that I could used for the preamp supply. Before I start shoe horning these in, I'd like to hear other opinions.

[stokes]

Did you form the caps before firing up the amp?If not that could be your problem.Adding capacitance as you said is a good idea,with the SS rectifier you can safely use 100uf or more,but if you didnt bring those caps up slow,I would do that first.

[casey73]

Yes, I formed the caps. Brought power up slowly over a period of around 6 hours. The cap was brand new.

[dai h.]

the multi-section cap might be the problem, since the noisy filter cap negative grounds (towards the input of the B+ supply where there is still a lot of ripple since it hasn't gone through the filtering stages yet) are sharing grounds directly with the preamp filter cap/de-coupling cap grounds which are a signal path. It isn't a good idea to just use one multi-section cap for the whole amp for this reason. If that is the only source of the problem, separating the grounds by using an individual cap(or caps) should solve the problem.

[bob p]

I've used that same model of can with good results in a Deluxe, so the possibility exists that you may just have a bad cap.

At the risk of stating the obvious, I'm sure that you know that determining what frequency of hum you're dealing with would narrow down the possible source(s) of the problem.

One thing that you might want to try is mapping out the supply voltages in the amp to try to determine which stage the hum is coming from. If you don't have a scope, you could just take both AC and DC voltage measurments at each cap terminal to see how well the PS filtration is working. You might find one stage that has more AC ripple riding on top of the DC than you would expect. You might get lucky, and find that only one section of the can is causing your problem.

Just a thought...

[Enzo]

Is this a single ended amp? They naturally hum more than push pull amps. RAther than larger caps, how many stages of filtration are there between retifiers and the powere tube? Try adding a choke and cap after the first filter cap.

The frequency of the hum is a great question to ask too, since power supply ripple is mostly 120Hz, while grounding and shielding issues usually present 60Hz. Make sure which, since they are both the same note an octave apart.

Bob's list hits the target. Rather than ASSUME it is filter caps, FIND OUT. You could have a hummy preamp tube or a bad EL34.

[Rattler66]

how can i tell the difference between 60 and 120HZ, what do they sound like?

[Chuck H]

60hz sounds exactly like the hum in your head when you stick your finger in a light socket:D

Seriously, I would like to know of a way to tell them apart without a scope.

But I didn't post just to be a joker. If you've built similar amps before then you know how much hum it should naturally have. But there is a way to make it even quieter once you've solved your root problem. I've used this trick on two single ended amps and it works great.

Instead of the standard parallel wiring in most single ended amps, go ahead and wire up the preamp tube the humbucking way using a false CT on the filament secondary made from two 100 ohm resistors. Then instead of grounding the false CT, send it to the top of the cathode resistor for some positive bias.

I redid the filaments in a friends VibroChamp like this and the results were amazing.

I don't think this trick would work as well with an EL84 amp though because there just isn't that much voltage on top of the cathode resistor.

Chuck

[Alex/Tubewonder]

how can i tell the difference between 60 and 120HZ, what do they sound like?

Connect your loudspeaker directly to heater supply. This is 60Hz. 120Hz is 1 octave higher which should be easy to distinguish.
You may want to connect a resistor around 33 Ohm in series with the speaker.

[Enzo]

Also, the 60Hz stuff tends to be more or less sine wave shaped, even if distorted. The ripple TENDS to be more sawtoothy. That gives it more of an edge to the sound. Experience goes a long way to recognizing it. On the bench I often lay a finger on my scope probe tip to see how many cycles are on my screen. If the hum signal has twice as many - it's 120Hz.

[Alex/Tubewonder]

I'm not a musician but a humble amp designer and I can easily hear which is 50/60Hz hum and which is 100/120Hz hum/buzz.:p :D

PS. Enzo, why are up at this early hour?:)

[Rattler66]

Thanks Chuck, I did elevate the heaters and it didn't make a difference. I am getting the hum in my power amp. i've been tied up this week that's why i haven't looked into it further, but I'll recheck and repost.

[Rattler66]

Connect your loudspeaker directly to heater supply. This is 60Hz. 120Hz is 1 octave higher which should be easy to distinguish.
You may want to connect a resistor around 33 Ohm in series with the speaker.


oh cool, thanks this'll help a lot

[dai h.]

you might try listening to the 60 and 120Hz tones here also:

http://www.realmofexcursion.com/downloads.htm

and plugging in a cable into the input with no gtr. connected, then putting your finger on the plug will give you hum (not sure which).

[Enzo]

Touching the guitar cord provides 60Hz, since that is the ambient electrical field. 120Hz comes from ripple, and there is no ripple in your finger - unless you've been working out a lot.

ALex, I work musician hours. My shop opens at 6:30PM and we close about dawn.

[Rattler66]

it's looks to me to be 60Hz. but i need to verify, I am going by memory