Resistors and Hiss

[Mike Sulzer]

From your description, you have a lot more hiss than you should. Are you sure you do not have an oscillation?

To judge hiss level, plug in a guitar with 500K pots. Switch to one pickup, and turn the volume down just a bit to get 250K for the series and shunt legs of the divider (its an audio pot). This gives you effectively about 125K in series with the input (more at the pickup-cable resonant frequency) and should swamp everything including the input grid stopper resistor. You should be able to hear this hiss by turning the amp volume up if everything in your amp is as quiet as it should be using a proper design and good components.

(You can turn the guitar volume up and down to try and find the point with maximum hiss.)

[bob p]

Since when did we want the lowest possible distortion in a guitar amp?

LOL. I like the idea of a having a clean option, just for the sake of versatility.

[Dave H]

I'm trying to dump some signal in a build, using the voltage division trick like all those uber-gain amps do, using some high-value resistors. I've read that series resistance can add hiss, and that's what I'm getting: LOTS of hiss, not affected by any of the controls and that wasn't there before I added the voltage divider.

There's not much in the way of places to add the voltage divider, so right now I have it after the Volume control and before the 2nd gain stage.

You could do this - remove the voltage divider so the amp is as it was before (without hiss) then change the volume pot to a lower value and add a resistor in series with its cw terminal to make the total resistance up to the original value.
e.g. If the vol pot was 1M use a 250k vol pot and 750k resistor for a gain reduction of 12dB, that way it shouldn't hiss any more than the original circuit did with the vol pot set to 250k:750k

[Justin Thomas]

Hey everyone,

Thanks for the options, I have lots of exploring and learning to do.
I can't rule out oscillation, but as it started with the addition of the voltage divider, I'll keep at it. I don't have a scope, and again, being a true PTP, hard to tell what the heck goes on in there! I'll draw up a schematic later and put it up, withOUT the added resistors, so we all can know the starting point.

Justin

[Justin Thomas]

Well, seems I made a rather gross error in wiring my tone stack that I never caught before. Guess I'll fix that first!

Edit: wow, I REALLY messed up that! How I didn't watch that, wow... But I refuse to quit.

Justin

[Chuck H]

It happens to the best of us... And me.

[Justin Thomas]

Okay, after fixing some awful wiring mistakes (wow...) And swapping out a noisy Volume pot, I may have oscillations... the pot is scratchy, and there's no DC on any of the pots (though, on one of the ends of the volume pot, I go from 0.000VDC to -0.009VDC; yes, negative. It moves with the control.). I also found a 220k plate resistor, which I swapped out. I may be running the PI fairly hot, too... So, it'll continue to be a work in progress. And I promise, I'll get a schem up!

Right now, it's chill-out time. Oh, the adventures of building PTP one-offs without a scope... But worth it.

Justin

[Gingertube]

Justin,
My thoughts FWIW.
Resistor noise is the major contributor really only on the input stage.
Keep the grid stops on the input stage lowish. Merlin recommends 10K (in lieu of the Fender standard 68K).
Also don't use noisy carbon composition resistors for the anode load or cathode bias resistors. The only value in Carbon Comps is their low self inductance which makes them usefull for grid stops. Anywhere else their noise makes them a bad choice.
For subsequent stages the tube noise is more likely to dominate.
One of the higher tube noise sources is grid current noise. It is "statistical" in nature, that is to say it has random noise spikes (hiss), in addition to its DC value.
The way to keep that low is to keep total resistance to 0V from the grid low (grid stop + grid leak [looking away from the grid]). Often the grid leak will be the lower arm of your interstage divider, so keep the divider resistances lowish. Think of this as shunting the grid noise to 0V, the lower the resistance of the shunt the more effective. This (lower divider resistances) also has the effect of lowering the effective anode load of the driving stage which is the load resistor in parallel with the interstage divider (to 0V) resitance. That lowers the effective gain of the driving stage (often described as "gain dumping"). Note that "gain dumping" means more signal current is needed to swing in the reduced effective anode load (to generate the same signal voltage) and that will be accompanied by an increase in distortion. More current swing means more change in the tubes rp, and the change in rp vs the effective anode load IS the distortion level.
Cheers,
Ian

[Justin Thomas]

As Requested, Schematic!

Okay, here's what I'm currently working with... Pretty much a straight-up Fender preamp. Plans include changing to fixed bias if I can cram the parts in. I'm still trying to wrap my head around stuff like input impedance, because I just plain suck at math, while theory and concept I'm a lot better at. I'm wondering about changing to SF PI values along with a 12AT7, for one. I'm also beginning to think I'm facing oscillation, as the volume control is scratchy, despite being replaced with a brand new one and no voltage on any of the caps, which should rule out a leaky cap? I'll go ahead and check all of the others while I'm at it.

So, it's not AS hissy as it is now (in this schematic), but it does seem to go up very quickly in volume... Oh - rectifier tube is a 5R4WGB, because I wanted to have the voltage drop of a 5Y3 but with the voltage handling of a 5AR4.

Justin

[Chuck H]

Alright Way to step up.

Below I've shown the split plate load. I didn't indicate resistor values but the total resistance for both resistors would be 100k/120k ish. Together they form the voltage division for dropping some gain. The signal is shunted to ground via the filter cap.

[Justin Thomas]

Dang, looks like I drew it myself... :P
Anything concerning you think I should know about?

Justin

[Chuck H]

Anything concerning you think I should know about?

Huh?..

[Dave H]

Anything concerning you think I should know about?

Is that middle cap in the tone stack really 0.01? It's usually 0.1
According to TSC 0.01 will reduce the bass quite a bit, not necessarily a bad thing as you have 0.047u going into the PI. It's usually 0.001u or even 500p as the PI input impedance is very high.

On this schematic you don't have the 220k/220k potential divider between the second stage and PI.
Adding negative feedback will reduce gain (and hiss)