Another thread on preamp 'gain'

[codhead]

Hold on guys, I think some of you may have this backwards.

If you bypass the pot with a piece of straight wire, what's gonna happen? You're gonna get a whole lotta output. Nice if you want to turn your preamp into a "tube buffer", but kind of useless if you want to use your preamp for the function that God intended.

So what's the impedance of a piece of straight wire? Pretty close to a short - or zero ohms. Remember, as the impedance numbers get bigger (like 100K) less signal gets through. So if your preamp comes with a 100K pot, and that's too much signal for you, you'll have to come up with something with a bigger number (like a 200K pot). If you install a 50K pot, you'll just add to your problems.

Remember, when you turn that knob clockwise (the loud direction) the ohms are heading for the zero end of the scale - i.e. more signal is getting through.

[mikebse2a3]

Many volume control circuits in tube preamps are wired like this simplified schematic. (Example One)

Just changing(Lowering) the Volume Control's Value would only increase the load for the Source Componet. If the source has a beefy output stage that can handle the increased current demand then the output voltage will stay the same across the Volume Control. If the output stage can't deliever the increased current demand then the voltage will drop and distortion of the source's output stage will increase.

Example Two: Shows an Attenuator installed(R1 & R2).

R1 & R2 Values can be chosen to maintain the desired input impedance with the amount of attenuation desired.

For Example: If your (Volume Control is say 100K) and R2 is 100K then make R1 50K and this would maintain the orginal input impedance of 100K.

I have found that it usually sounds best if you use the minnimum amount of attenuation possible to acheive your needs.

mike[]

[Erik Mandaville]

Who said bypass the control with a piece of wire?

Erik

[Erik Mandaville]

Codhead:

"So if your preamp comes with a 100K pot, and that's too much signal for you, you'll have to come up with something with a bigger number (like a 200K pot). If you install a 50K pot, you'll just add to your problems."

Actually, you have it backwards -- sorry, but it's the truth. A potentiometer doesn't work that way, Codhead. The greater the value of the potentiometer, the LOUDER the volume will be at a given level. If Kudret is using a volume control with a value of 100K, and finds it gets loud far too quickly, it would compound the problem if one of a higher value were to be used. The reason he asked about using a 10K pot, which is more in line with what would be needed, is because the sensitivity goes DOWN, not up. The only way one can help calm down a hyper sensitive volume control is 1) replace it with one of LOWER (not higher) value, or to use series resistance on either a) the input into the preamp circuit, or on the output between the preamp and amplifier. As I also said, he could also use two linear taper trimmer pots on the inside of the preamp in series with the volume control. These could be set at a level that was balanced out with the main control, and thereby provide a greater amount of useful attenuation with his current potentiometer, which I remember now is 25K ohms, not 100K.

Erik

[mikebse2a3]

Eric said:

The only way one can help calm down a hyper sensitive volume control is 1) replace it with one of LOWER (not higher) value

-----------------------------------------------------------

I don't believe simply changing to a lower value volume control would work.

Eric lets take for example a CD player with a standard 2 volt output level capability then look at my example circuit in my previous post you will see that the 2 volt potential will be available with even a lower value control as long as the load we present to the CD players output stage isn't driven beyound its capability(current supply) thus causing a voltage drop due to current starvation.

Looking at it another way: if both controls have the same taper then turning the control half way down would still have the same voltage potential applied to the grid of the tube in my example from either control. The lower resistance control would just demand more current from the CD players output stage.

mike[]

[Erik Mandaville]

Mike:

I understand what you're saying, however I have used this technique many times to lower the sensitivity a bit. However, leaving the output of the CDP as a constant fixed output of say two volts, an easier and more effective way to do this on the preamp side would be to use fixed resistance, using the same value of the potentiometer (which I can't remember exactly, but believe will bring about a 6dB or so reduction), from the input into the preamp, which is often off the inner two poles of the selector switch into the volume control.

More effective and flexible still would be to use the trimmer potentiometers in series before of after the main volume control. I've built preamps for people with this feature, exactly for the reason described.

Erik

edit: thanks for that little schematic, by the way! That helps illustrate what you're saying. Kudret could also certainly try something along those lines.

edit: Mike I've done something similar to the schematic show. What you have is a resistor in series with two resistors (one variable, one fixed) in parallel. In terms of how that relates to the total resistance with the volume control turned all the way up, you're right, you'd maintain the original value. This works too.

edit 3[] And Mike, you do in fact indicate that resistor in series with the other one to ground combined with the input of the volume control (also referenced to ground).

[codhead]

Actually, you have it

backwards -- sorry, but it's the truth. A potentiometer doesn't work

that way, Codhead. The greater the value of the potentiometer, the

LOUDER the volume will be at a given level...

Erik

Well, I've certainly been wrong before. But I respectfully disagree...

The top meter reading in the attached photo is a 50K audio taper pot at

full volume. The meter reads .6 ohms, which means you'll be getting

most of the signal out that you put in.

The bottom meter reading is with the volume turned all the way down. At 43.7 K ohms, the signal is significantly smaller.

An easy way to verify what you're claiming would be to make up a couple

of interconnects. Put a 1 ohm resistor in one, and a 50K ohm resistor

in the other. If what you're claiming is true, the music is going to be

louder when played through a 50K ohm resistor than it will be through a

1 ohm resistor!

[mikebse2a3]

Mike:

I understand what you're saying, however I have used this technique many times to lower the sensitivity a bit. However, leaving the output of the CDP as a constant fixed output of say two volts, an easier and more effective way to do this on the preamp side would be to use fixed resistance, using the same value of the potentiometer (which I can't remember exactly, but believe will bring about a 6dB or so reduction), from the input into the preamp, which is often off the inner two poles of the selector switch into the volume control.

More effective and flexible still would be to use the trimmer potentiometers in series before of after the main volume control. I've built preamps for people with this feature, exactly for the reason described.

Erik

edit: thanks for that little schematic, by the way! That helps illustrate what you're saying. Kudret could also certainly try something along those lines.

Eric if you are talking about a resistor or control used as R1 in my example without R2 in the circuit then yes I agree with you. You will be creating a voltage divider(using R1 and the Volume Control) that as you point out will give you a reduction in voltage to the volume control. The thing for someone to look out for doing it like that would be a chance that the preamp could be a little more sensitive to noise/hum due to the input stages higher impedance with that method.

mike[]

[Erik Mandaville]

Codhead:

That's fine, I've been wrong before millions of times, too! However, higher values of resitance on input stages -- either fixed in the form of grid leak resistors in tube amps, or variable resitors in preamps, are more sensitive than lower values.

And! I still respect that you may disagree with me on that.

Erik

[Erik Mandaville]

Mike:

"The thing for someone to look out for doing it like that would be a chance that the preamp could be a little more sensitive to noise/hum due to the input stages higher impedance with that method."

This is absolutely correct, thanks for pointing this out. Higher impedances are more sensitive and are thus less immune to problems with RF and other pollution that would not be such a problem for systems of lower impedance.

BTW: The circuit you mentioned above works just as described. I just a minute ago grabbed 100K pot, which I paralled with a fixed 100 K R, and then combined two 100Ks for the 50K ohm in the schematic (didn't have a single 50K on hand), and it's perfect. As you said, I think this would be better than a single pot of lower value, since it maintains a suitably high input impedance. Cool!

Erik

[Erik Mandaville]

"An easy way to verify what you're claiming would be to make up a couple of interconnects. Put a 1 ohm resistor in one, and a 50K ohm resistor in the other. If what you're claiming is true, the music is going to be louder when played through a 50K ohm resistor than it will be through a 1 ohm resistor!"

I'm sorry, but this not what I'm claiming. I can direct you to more reading than you might want concerning the behavior of volume controls, various types of L and/or T-pad attenuation (as Marvel mentioned), voltage dividers, etc. The analogy you are using concerns simple series resistance, only. A volume control is a little different.[]

Erik

[NOSValves]

Sorry but Erik is correct on this one the lower the value of the pot the less over all volume or signal will get to the preamp from the source. What isn't being clearly stated or understood is that the VC is referenced to ground when the control is at zero ohms there is NO signal getting through. When the VC is at say 100K it has moved the control(or signal)100K above ground letting signal through. Therefore if you have a 250K pot you can and will allow more signal through because you can adjust it even further above ground. The resistance of the control is not what adjusted the volume or signal its the reference to ground. At least in the hand drawing done above which is the normal way to wire a VC.

Craig

[codhead]

Okay, I see what you're getting at. Duh!

I just ordered a little Luminous Axiom passive, and it shunts to ground in a similar manner to what you're discussing.

I was going from past experience with a preamp that actually used the

potientiometer to control the signal. I swapped the pot for one of

higher value, but found that even though it helped with low volume

settings, it got loud in a big hurry past about 10 o' clock.

Finally ended up putting the original value pot back in, and adding a

resistor in addition to the pot, which worked a lot better (essentially

the same result as an inline attenuator).

[mikebse2a3]

Sorry but Erik is correct on this one the lower the value of the pot the less over all volume or signal will get to the preamp from the source. What isn't being clearly stated or understood is that the VC is referenced to ground when the control is at zero ohms there is NO signal getting through. When the VC is at say 100K it has moved the control(or signal)100K above ground letting signal through. Therefore if you have a 250K pot you can and will allow more signal through because you can adjust it even further above ground. The resistance of the control is not what adjusted the volume or signal its the reference to ground. At least in the hand drawing done above which is the normal way to wire a VC.

Craig

Maybe this will make it easy to see why I say that changing the volume control's value will not lower the volume settings(Assuming same taper rates and both turned half way) as long as the source componet can handle the lower impedance load of a 50K versus 100k control(Or Any Value We Choose). The voltage presented to the grid is the same in both examples only the current demanded from the Source has changed. As can be seen the voltage potential from ground will be the same for either value control.

mike[]

[Erik Mandaville]

Craig:

You're right, the reference to ground is certainly part of this equation. While I have used lower value pots to....let's say 'influence' the input signal, one of the reasons I mentioned to Kudret that I didn't think a 10K ohm potentiometer would make much of a difference compared to the 25K he was using had to with what Mike is mentioning above -- given equal taper rates of both controls.

The schematic above looks like a 'shunt' type of input control, where we have greater resitance in series with the input, but where the original over-all value of the potentiometer is maintained across the circuit. This works. I used this type of setup with the Foreplay years ago, but just started using simple resistance in series off the selector switch. The above is similar to a variable L-pad in a crossover, where the impedance of the driver being fed remains the same regardless of the position of the wiper in the pot.

Mike: A great addition to this thread! If Kudret is still following this, I think it would be good to try the above input circuit using the 25K ohm control you're now using. Also, if the CD player has a variable output, that could also be used to attenuate the signal prior to the preamp, although this has never sounded as good to me -- just my own preference, I guess.

Codhead: This stuff always leaves me scratching my head, and I remember when I was having this same discussion with my dad about 15 years ago. He just smiled, and kept exlplaining it until it became clear. What you were thinking, and it seems logical (at first) that the use of a higher value pot would present MORE resitance to an incoming signal than would a lower value one. And in most cases, you would be right, but Craig has included some information above that might help with this. IOWs, I completely understood what you were thinking and why you were thinking it! Thanks for the inlusion of the meter shot, too![]

Erik

[codhead]

Erik,

Yeah, that was a good screw-up on my part. Reading back through this

thread, you actually did say something about the circuit - right before

I convinced myself that this was backwards. Had I been paying

attention, I would have caught myself right there.

When you went on to claim that "a potentiometer doesn't work that way", that really

convinced me this was backwards. I went right for the trusty Fluke. Had

I continued to read the thread, I would have seen what you were getting

at.

I once built an AES kit, where the grid was being fed from the

potientiometer. I believe the AE-3 kit is still being sold this way

today, while the AE-3 DJH (or at least the one I have) uses shunt

volume.

I also have a passive which runs the signal straight through an ELMA

stepped attenuator. I ordered the little Luminous passive, as I want to

see if I can detect the claimed sonic difference between these circuits.

[NOSValves]

I've never actually tested these scenario's but I for one do not see how its possible with both controls in the same position in the schematic above to present the same 1V to the input tube. The signal has a 50% closer path to ground. I just don't see it as electrically possible.

[NOSValves]

I stand corrected. I just mocked it up quickly with my function generator and used a 10K and 25K pot and step for step the signal passed out the wiper of the controls is identical no difference. Hmmmm still makes no sense to me LOL !!

Craig

[Erik Mandaville]

Mark:

You also use input level controls on the Peach, which is what I have done with my own preamps. This is in series with the volume control, and can be set at level/s best suited to the combination of components on either end (source/amplifier). I think this is what Kudret would find useful, and I didn't think a 10K pot would be an improvement for him over the 25K he was presently using -- but he could certainly try to see what happens.

There are other ways of getting some input attenutation that I didn't mention. If the Seth uses a bypass capacitor across the cathode bias resistor to ground on the input grid of the first stage, he could also experiment with the removal of it. I have tried that whole issue with and without bypassing many times, and find I prefer using a bypass cap. It can effect the gain of the amp, and he might just try it. Actually, I've even used bypass capacitors of values high enough to be frowned upon by other 'who know,' such as in the realm of 220uf and even twice that without ill effects that I could notice.

The shunt control is a neat approach I had forgotten about, but had found useful. I think the load concern is important. I've used 10Kohm potentiometers before without any obvious distortion in terms of what I heard, but each case is different. As I mentioned, I think a series resistive element on the input into the volume control would be helpful, and has worked for me without the inclusion of an extra shunt element to ground. Mike's reminder of the the shunt control would be a perfect application here, and is absolutely effective!

Erik

[mikebse2a3]

A few pages from Radiotron Designer's Handbook about Attenuators.

See Fig 18:12 for the L Type that we have been talking about.

mike[]