1K Audio Pot Used As Cathode Ground Reference?

[plexi50]

My wheels are spining again. I have a circuit in a build i am doing with a combination 2204/AA864 board

The faceplate of my amp has two Master pots but i dont want to use a master on the bassman circuit

Is it possible to make use of a 1K Audio pot in conjunction with lets say 820 ohm resistors on V1A &B allowing for varying cathode values?

I could use a rotary switch for this but was thinking it would be cool to use an audio pot as a sweep if realistic

Example=820+1K=1.8K

This would raise or lower (SMOOTH OUT) the preamp tubes gain at will / Not a gain pot

I dont want this faceplate to have a non functioning pot in place and i dont know what else to do with it other than to remove the pot and put a snap button in place

That may just be the best idea. Like i said my wheel are spinning (There always spining)  

[RicharD]

This won't have much effect on gain.  Really all you're doing is moving the operating point up and down the load line.  Yes, as you increase current, B+ will drop a little, but not enough to have much effect.  To decrease gain, you want to steepen the load line (make it more vertical).  You can do this by lowering B+ and lowering the value of the plate resistor.  The load line is plotted by selecting your B+.  That is the right most point of the load line (red).  The second coordinate is B+/Ra  (where Ra is the plate resistor.... roughly).  That's the left most point of the load line.  The 2 blue horizontal lines show the approximate range you're describing.  Neg bias/current=cathode resistor.  The blue vertical lines show Vout=55V for 1V of swing Vin.  It doesn't matter where you are on the load line, gain will be the same.  What does change is maximum output.  I didn't show this but at -1, you can get roughly -75V to +100V out.  For -1.5V you get approximately - 100V to +75V. It might add some nice quality, but it's gonna be very slight.

If I were gonna fill an extra hole, I'd use a switch and select between several different cathode bypass caps, .68, 2.2, 5, 15, 22 uFs.  That will make for a nice bass contour control.

-Richard

[plexi50]

You are right. The ole rotary switch again. LOL! In the past i have used it for like you said several different value res or cap combos

I am trying to keep it simple but your idea seems to be the only way out other than to plug the hole
 
A bass contour sounds like a good idea. Or maybe a slope frequency switch

[VMS]

I have also been thinking about this type of control.

What do you guys think of my approach? Is it bad to only have a pot as a cathode resistor?

Would this be effective? Is there a risk that this would be a scratchy control?

[plexi50]

There would be a slight wisp as you turned the pot from the DC on the pot

If the carbon on the wiper ever wore out you would loose your cathode voltage all together and no sound

After thinking some more it would be better to have multiple bypass caps on a rotary switch to get the desired tone changes

Maybe some good old Frankenstien experimenting would be good to do and see first hand

[Merlin]

Is it possible to make use of a 1K Audio pot in conjunction with lets say 820 ohm resistors on V1A &B allowing for varying cathode values?

Yes you can do that, in fact you don't even need the 820R resistor. You could just use a 2k pot or similar, allowing you to vary the bias from zero to 2k. This would vary the gain from around 60 to 30 if the resistor is unbypassed. If it is bypassed then the gain would remain the same, but the clipping characteristics (headroom) would change.

[plexi50]

Ok here a wild question: How low could you safely go in lowering a preamp plate resistor value?

Does the plate voltage affect the cathode voltage? I say of course but to what tonal degree? Is it all relative?

Time to put on the tonal boots?

[HotBluePlates]

I have also been thinking about this type of control.

...

Would this be effective? Is there a risk that this would be a scratchy control?

You could use a regular resistor for the cathode resistor, and a pot wired as a variable resistor between the cathode and the ungrounded side of the bypass cap. Tubenit has done that in a build, with this control as one of 2 tone controls. The design came from O'Connor's Soma 84 amp.

[jjasilli]

I have also been thinking about this type of control.  What do you guys think of my approach? Is it bad to only have a pot as a cathode resistor?  Would this be effective? Is there a risk that this would be a scratchy control?

Is it possible to make use of a 1K Audio pot in conjunction with lets say 820 ohm resistors on V1A &B allowing for varying cathode values?

Yes you can do that, in fact you don't even need the 820R resistor. You could just use a 2k pot or similar, allowing you to vary the bias from zero to 2k. This would vary the gain from around 60 to 30 if the resistor is unbypassed. If it is bypassed then the gain would remain the same, but the clipping characteristics (headroom) would change.

I'm confused.  Given cathode bias, if the cathode "resistor" is reduced to a dead short, won't that cause the tube to burn out?

[DummyLoad]

what i perceive will happen with grounded K amp. as you decrease Rk, Ik increases, you'll clip more of - half before + clips, as you'll be running closer to saturation. as you increase Rk, Ik decreases, then you are running closer to cut-off and will clip + half before - half clips.

stated simply - a user controlled variance from warmer to colder bias points.

[VMS]

You could use a regular resistor for the cathode resistor, and a pot wired as a variable resistor between the cathode and the ungrounded side of the bypass cap. Tubenit has done that in a build, with this control as one of 2 tone controls. The design came from O'Connor's Soma 84 amp.

Yes, I know the Soma84 control and that is probably a better way to do this.

I got this idea from radiotron handbook. There is a picture of partially bypassed cathode with two cathode resistors in series and capacitor connected parallel to the bottom resistor. 

[Merlin]

I'm confused.  Given cathode bias, if the cathode "resistor" is reduced to a dead short, won't that cause the tube to burn out?

Nope, provided the anode resistor is greater than about 40k, the current in the valve can never become large enough to cause any harm even with zero bias.

[jjasilli]

OK, thanks Merlin.  (Maybe someday I'll learn to read those pesky tube charts   :book1:   )

[plexi50]

There wont be a huge difference in tone but there will be some difference in tone. Of course that depends on the values of the  resistor and pot value combo

I would want to go with an 820 ohm as a bottom base and then raise it from there accordingly to taste

It fill's that master hole and add's some spice i suppose. I'll let you know when i get the PS caps to finish this amp up next week

[RicharD]

>Ok here a wild question: How low could you safely go in lowering a preamp plate resistor value?

That depends on a few thangs.  The general rule of thumb is for the plate load resistor to be no less than 2x the internal plate impedance.  A 12AX7 has an internal plate impedance of 62500 so using a 100k is already bending this rule.  With geetar amps, rules are meant to be broken.... sorta.  Another rule of thumb is for the load resistor (volume pot) to be 5x the plate resistor.  This way when drawing a load line, you can forget about Rl.  This is why you commonly see a 100k plate resistor and a 500k volume pot.  Counting on thumbs, 500k in parallel with 100k is still 100k, actually 83,333 ohms, but close enough. 

OK, so you really really want to decrease the plate resistor.  Looking at the curve chart, you'll see the current scale ends at 3.5mA.  We'll consider this the max current.  Your 2nd coordinate is B+/Rp=I.  You can flip that around to be B+/I=Rp.  300/.0035=85714  200/.0035=57142  So this means if you lower B+, you can lower the plate resistor.  (See attachment 2)

>Maybe someday I'll learn to read those pesky tube charts

Let's look at some more traditional schemes.  (see attachment3)  The red lines show 100k/1k5, and the blue lines show 220k/3k3.  Here's how the load line thang works:
Pick a B+.  Pick a plate resistor.  Divide B+/Rp.  Now you can plot the load line.
Draw a line straight down from where your load line crosses 0 bias volts.  Call that Vcutoff.  Now ideally you want to set Vtube 1/2 way between Vcutoff & B+.  Draw a line straight up From Vtube to your load line.  Draw a line straight across to see the bias current.  Now look at the point where these 2 intersect on your load line.  Call this your operating point.  The red one intersects at roughly 1.5Vbias and 1mA.  1.5V/.001=1500 ohms.  This is your cathode resistor.  Now go up your load line .5Vbias from your operating point and draw a line straight down.  Go down your load line from your operating point .5V and draw another line straight down.  You've graphed 1V swing along your load line.  Measure the distance in volts between these 2 point.  222V-160V=62V  AV=Vout/Vin.  AV=62/1 = 62
 

[JayB]

I have also been thinking about this type of control.

What do you guys think of my approach? Is it bad to only have a pot as a cathode resistor?

Would this be effective? Is there a risk that this would be a scratchy control?

That would be more effective if you ask me. If you're worried about the pot as a cathode resistor, just use the resistor and use a reverse audio pot in series with the cap.

[VMS]

What about the frequency response of these two control methods.

I'm thinking that in 'A' the boost gets thicker as you dial it in, and in 'B' it gets thinner.

Could someone maybe plot these circuits? It would be interesting to see how they behave.

[jjasilli]

Thanks Buttery, that's great.  You just saved me YEARS of therapy!!!

This also sparked (no pun intended) me to google the topic, and I found this also helpful:  http://diyparadise.com/tubeloadline/tubeloadlines.html

[jjasilli]

VMS:  A bypass cap increases the gain of the tube; and the value of the cap affects frequency response.  A 22uF cap boosts the whole audio spectrum.  Smaller caps boost higher frequencies only.  So the better option is to Switch different bypass cap values.  That way you get to keep the gain provided by the bypass cap, AND vary frequency response.  Geezer has a circuit posted in the Schematics section.

Sometimes the additon of a bypass cap causes too much gain and the circuit goes into oscillation -- squealing.  If so, a resistor up to say 10K under the cap is a fix.

[VMS]

VMS:  A bypass cap increases the gain of the tube; and the value of the cap affects frequency response.  A 22uF cap boosts the whole audio spectrum.  Smaller caps boost higher frequencies only.  So the better option is to Switch different bypass cap values.  That way you get to keep the gain provided by the bypass cap, AND vary frequency response.  Geezer has a circuit posted in the Schematics section.

Yes, I know that. But doesn't the resistor parallel to bypass cap also affect the frequency response?

Those are just arbitrary values that I put on the example circuits.

[PRR]

> general rule of thumb is

No, that's not the question.

"How low could you safely go..."

What is "safely"? No human fatalities? 99.99% reliability over a lifetime of tough touring?

Take the 12AX7's 1 Watt plate rating. Plot it on the 12AX7 curves.

Follow the zero grid bias line until it hits the 1 Watt line. You may have to tape some paper past the edge of the chart.... well, don't bother. Ignoring a mild wiggle at the bottom, the zero-bias line is very nearly 40K resistance from less than 0.1W to over 0.5W, and probably follows that slope to 1W.

What is the maximum voltage on 40K to stay within a 1 Watt limit?

Up to that voltage, a 12AX7 is "safe" (<1W) with NO plate resistor.

If you have a higher suply voltage, simple voltage-divider figuring will give a minimum plate resistor to stay "safe".

If this resistor is higher than your total goals, split it so some is your audio load, then B+ cap, then some more to hold the power down.

But is this really what you want? Even with non-zero resistor a small resistor gives small output. OTOH it eats a LOT of supply current, which has to be buzz-filtered even a bit better than a normal stage. Meanwhile the grid distorts bad, unless the source happens to be a very low impedance.

But won't cost much to jury-rig so your ears can see.