Reducing noise in homebrew amp

[bobshbob1]

Hey all, I just finished building a semi-clone of a Dumble ODS, based mostly on this schematic, but just the overdrive channel, and with 6L6s instead of EL34s. I'm very happy with the sound and responsiveness of it, and also surprised with the amount of gain on tap, much more than I'd expected. Probably my new favorite amp to play.

It's a little noisy...not as much as, say, a 5150 at similar gain/volume settings, and definitely not enough to bother me, especially in a band setting. But obviously I want to make it as dead quiet as possible.

I've done a fair amount of moving wires around, knowing lead dress is pretty critical in high-gain point-to-point builds, but there seems to be a lower limit to what I can achieve by just moving wires. One thing I've noticed in poking around inside is that when I get my hand near the 220k/390p resistor/cap just after the first volume knob, the noise level increases dramatically.

In this build, it goes volume pot wiper -> wire to turret strip -> resistor/cap across turret strips -> wire to V1A grid. I'm thinking using shielded wire and soldering the resistor/cap either directly to the wiper lug or directly to the grid pin lug could alleviate this issue and maybe quiet things down a lot.

I know I could just try whatever configurations and see if it makes a difference, but I'd really like to know if anyone has a scientific answer as to which would make the most sense and why. I've seen amps where these interstage resistors/caps are connected directly to the volume pot lugs, and ones where they're not, and I'm wondering what difference it makes, if any, and what causes that difference.

Thanks!

[Enzo]

Noise is not generic, it comes from a million places and has many forms. You need to describe what exactly you mean by noise here. There is hum, which is mains frequency related. There is 60Hz hum which is what you hear when you touch the tip of a guitar cord, it comes from all manner of things. Each source of it has its own cures. You could have five things contributing to it, and five solutions each would reduce part of it. There is also 120Hz hum, which comes from power supply filter currents, or just bad filtering.

But when you mention 5150, then I start to think of other noises, like hiss or "white noise". The sound of a radio or TV station off the air. Hiss comes with gain.


But we need to know what you are chasing. lead dress often involves hum, such as a grid wire too close to a heater wire.

I don't doubt you can stick your finger by that grid input to the tube and hear noise. The cure there is simple - don't stick your finger by the parts. I assume at the gig, no one will have fingers in it?

Commercial amps shield the open side of the amp chassis. Look inside the cab where the chassis sits. We see things like aluminum foil glued to the wood, or thin sheets of metal or steel window screen stapled to the wood. Peavey uses a conductive black paint. ALL of which, when the chassis is installed becomes a lid for the chassis. It completes the shielding around the circuits. Is the open side of your chassis covered by a grounded piece of metal of some kind? I mean in the cab. Of course you could put an actual metal lid on the chassis itself too.

[Mick Bailey]

A picture of the guts would be helpful. Component type has a bearing on hiss - particularly plate resistors. I now use metal film in high gain amps and it does help. Bringing a finger close to certain components will couple hum from your body into the circuit. We live in a sea of increasing electrical noise and this is picked up nicely by your 'body aerial'. Sometimes you might even get Mexican radio.

Your grounding scheme is important as well.

Peavey uses a primitive noise gate in some amps - a pair of back to back diodes in series to gate the noise floor below the doide's forward voltage.

[Chuck H]

+1 to all above, and add...

I seem to have trouble with noisy preamp tubes every time I built an amp anymore.!. It's usually one tube in the batch I bought for a project. I've tried Shuguang and most of the Russian brands. The results have been similar. About one in four or five is unacceptably noisy. Microphonics not withstanding. I'm talking about hum and/or buzz and/or hiss. So if the OP has access to other preamp tubes, it can't hurt to try swapping some in and out. Use the first preamp socket as a test bed for unknown tubes because it will most readily show noise differences. But remember that microphonics, if not excessive, probably aren't a problem for any socket other than the first preamp tube. So your listening only for noise level.

[big_teee]

A picture of the guts would be helpful. Component type has a bearing on hiss - particularly plate resistors. I now use metal film in high gain amps and it does help. Bringing a finger close to certain components will couple hum from your body into the circuit. We live in a sea of increasing electrical noise and this is picked up nicely by your 'body aerial'. Sometimes you might even get Mexican radio.

Your grounding scheme is important as well.

Peavey uses a primitive noise gate in some amps - a pair of back to back diodes in series to gate the noise floor below the doide's forward voltage.

Are Metal Oxide as quiet as metal film?
Too many choices and too many part types to sort through.
I was thinking of mainly using 1 and 2 watt Metal oxide resistors in my next eyelet board build?
The 1 watt metal oxide are as cheap as the 1/2 watt metal film.
So what is the big diff, in all the modern resistors, as far as noise go?

[Mick Bailey]

Metal film has a lower noise then metal oxide. From best to worst; wirewound, metal film, metal oxide, carbon film, carbon comp.

I've built hi-fi amps with metal oxide throughout and they've worked out fine, but in a high gain guitar amp there may be a slight noise increase compared to metal film in certain positions (V1 plate, for instance).

[bobshbob1]

Thanks for the tips. I'd say I'm mostly getting hiss, with some (I think) 120hz buzz which is what is affected when I stick my hand near a certain area of the chassis. I tried swapping out the EH I had in V1 for a JJ, which was an improvement. I feel like I've gotten a couple less-than-perfect EH preamp tubes in the past, too.

The resistor I have there now is a carbon film, I got mostly metal oxide/film for the rest of the circuit but I had a few carbon film 220k's on hand so I used one there. When I get a chance to work on it this week I'll probably swap it out for a metal film, and also use some shielded wire.

[Mick Bailey]

You may already know this, but when you connect the braid on your shielded wire only connect one end.

[bobshbob1]

Finally got a chance to open it up last night; using shielded wire and connecting the components directly to the pot definitely reduced the noise (buzz and hiss) caused by turning the first volume knob up. I'll probably do the same thing with the second volume knob ("Drive" on the schem). One step at a time...eventually there won't be any unshielded signal wires in this whole thing...

[Chuck H]

Keep in mind that shielded cable has capacitance. If you use it all over it can add up to a change in tone. Loss of high end and associated harmonics. Just something to keep in mind as you change out leads. Sometimes a little noise is better than a loss in fidelity. Just use your ears and decide as you go.