Reducing Hiss

[Daver]

Does anyone have some tips to help reduce hiss in a high gain amp? I know it comes with the territory. I've heard this on a lot of high gain amps. I'm currently building an amp with 4 gain stages and a 5th that can be switched in with its own gain control and master. I've used metal film plate resistors, have things star grounded and am using selected low noise tubes. I've also tried many other preamp tubes both new and NOS. I just want to do as much as possible to keep noise to a minimum. Any tips? Thanks.
Dave

[defaced]

The Peavey JSX has a switchable and adjustable noise gate that in one of my builds, noticeably reduced hiss. It's a 1 meg pot and two diodes wired in parallel with opposite polarities. It's at the top center of the first page between the two triodes of V2.

[Daver]

Mike,
Cool. I will try that. I was mostly interested in preventing/reducing hiss in the first place, but that noise gate might be the ticket. Did you notice any change in the tone of the amp when you added the noise gate? Thanks!
Dave

[6267]

If there are carbon comps in your amp as plate resistors, replace them. Possibly elsewhere too, although carbon comp is an ideal grid resistor (low inductance). Most of the truly objectionable hiss I've experienced results from old, moisturized carbon comps + high sensitivity speakers.

[Daver]

6267,
All the plate resistors are metal film. All the others are 1 watt carbon film. No carbon comps anywhere. Thanks.
Dave

[defaced]

I probably could convince myself that there might be a tiny reduction in high end, but I wouldn't swear to it. Since the parts for the design cost less than 5 bucks, I'd say try it.

[mooreamps]

Mike,
Cool. I will try that. I was mostly interested in preventing/reducing hiss in the first place,

Oh right...... The major of the thermo noise content generated in a vacuum tube amplifier, {or any receiver amplifier}, is in the very first gain stage. In the Peavey schematic, as posted, this would be R33. If you pull this resistor out, and replace it with fixed bias, it should null about 90 percent of the generated hiss content.

I would still consider the noise gate, as it may also help midigate the 60 cycle hum induced from the pick-up from the instrument.

-g

[defaced]

Fixed bias as in a diode to set the bias?

[Jag]

If you pull this resistor out, and replace it with fixed bias, it should null about 90 percent of the generated hiss content.

How so Gary?

[mooreamps]

Fixed bias as in a diode to set the bias?

It can be done in two ways.

1. a -1.5 negative bias voltage, as I have seen in a couple other prints.

2. replace the bias resistor, and cap, with an led. I use an ultra-bright.

Now, if you choose either one, lower the plate coupling cap to 4.7 nF, or so, because using this method will really open up the bass responce of the stage. If you still want to do pre-amp voiceing, that can still be done on the later pre-amp gain stages.


Just giving an answer to the question.




-g

[Daver]

defaced,
I tried out the noise gate and really didn't like it. It changed the tone and had a minimal effect on the hiss. Using a Decimator pedal in the effects loop worked better.

Gary,
I may try out the fixed bias configuration. I'm leaning towards reconfiguring the preamp to a 3 and 4 stage setup rather than the folly of the current 4 and 5 stage setup. I've built plenty of 4 stagers without excessive noise problems. More experimentation......

Thanks.

Dave

[mooreamps]

defaced,
I tried out the noise gate and really didn't like it. It changed the tone and had a minimal effect on the hiss.
Dave

Perhaps because Peavey puts their noise gate too far inside their pre-amp. I'd put it just past the plate off the first gain stage. I"ll have to try it and see. Not for the hiss. My equipment does not generate but a very little touch of hiss content. It would be more for the induced 60 hz hum induced by the outside environment.


-g

[anson]

It seems that if one were to replace the cathode resistor (R33 in the schematic) with a fixed 1.5 v bias supply, then the bias supply must be built robust enough to supply all the cathode current that that stage will draw. I forget that the cathode current for a 12ax7 is, but a quick look at a spec sheet will tell how much currrent needs to be supplied.
Anson

[mooreamps]

It seems that if one were to replace the cathode resistor (R33 in the schematic) with a fixed 1.5 v bias supply, then the bias supply must be built robust enough to supply all the cathode current that that stage will draw.

Yes, indeed. It must be robust enough to supply up to
two milliamperes of dc current.

Also, as a side thought, one could also tie the cathode straight to ground, and apply the bias to the grid.


-g

[anson]

Heh Heh, good point. I don't know what I was thinking when I wrote that last thread. Long day at work...