Cathode Biased Power Tube Cap Value?

[Structo]

After spending about an hour googling about the value of a cathode biased power tubes and the value of the cap used, I really haven't come up with much.

I understand that it is a RC circuit and that the cap value has a reactance effect, I'm still not sure I understand what I want to know.

Specifically, what audible effect does a small cap (25uf) verses a larger cap value (100uf)?

For instance, a Fender 5E3 Deluxe uses a 25uf cap on the cathode bias.

But some 18 watt Marshall type amps use a 100uf cap.

Eventually I will simply tack some in and try different values but I was wondering if somebody can tell me something like, a 100uf bias cap will have more bass or that a 25uf cap will make it tighter.

[Alexo]

Like you said, it's an RC circuit, so if you have a 5E3 with a 25uf bypass cap and a 250 ohm Rk, you get a low roll-off starting at 25 hertz.

If you have a 125 ohm Rk, as in an 18 watt, in order to maintain a roll-off of 25 hertz, you have to increase the cap value in proportion to the decrease in resistor value, and you'd need a 50 uf cap.

Why they went to 100 is just a design choice, I suppose.

Really big caps (hundreds to thousands of UF) start acting in a way that's similar to the filter caps in the power supply - that is they start to stabilize the cathode voltage and make the bias setting less susceptible to fluctuations as the current through Rk increases and decreases, making the amp act a little more like a fixed bias amp.

[Andy Le Blanc]

I favor the 100uf, the useful gain of bass is extended but you'll get better
reproduction of the upper bass and low mid.

[collinsamps]

Edited - Chart was preamp tube bypass cap values & bass cutoff freqs

[Alexo]

Very interesting.

...why is this not working for me? 250 ohms plus 25uf gives a roll-off of 230 hertz? That's just not right. Am I missing something? What tube type was this designed for - if it had a very low internal impedance, that would explain the discrepancy.

[collinsamps]

See above.

[Alexo]

Yep. (As in the title of this thread )

[collinsamps]

This is another way to get there, quoted from another site.

""A bypass capacitor will have a noticeable effect on the sound. It can be calculated using C=1 / (2 * pi * R *f) where f is the frequency that you'd like to start the bypass. You can play with the frequency you select to change the sound of the amp""

[Alexo]

C=1 / (2 * pi * R *f)

2 x 3.14 x 250 (ohms) x 25 (hertz) = 39,250

1/39,250 = .0000254 farads = 25.4 microfarads.

But you need to put Rk in parallel with the internal resistance of the tube, cathode to plate and grid, etc., Aiken has a great article on how to calulate that figure.

[Structo]

Thanks.

But certainly we don't need a guitar amp to go down to 25Hz.

Guitar is roughly 80 to 1000 Hz so perhaps a 100uf cap would better suit that range.

I have some caps on the way, I'm going to try a 50uf and a 100uf.

As a side note I was also looking into using a conjunctive filter on the OT primaries.

I found an article or forum dialog about this, using a 10K 10 watt and .022uf across the primaries.

http://www.google.com/search?sourceid=n ... ive+filter

[collinsamps]

Frequency talk. Just some of those things that make you go hmmm

One of the most popular guitar speakers in use starts at under 50hz and is resonant just above that (55 I think)which the cab is never designed to allow, yet it's in use everywhere.

Speaking of cabs - take three popular 1x12 speaker cabs, say a tweed deluxe, deluxe reverb, and 1974 18 watt. All open back, all completely different sizes, small, medium and large. Which one is tuned for the speaker it shipped with? The short answer is none.

Most guitar speakers roll off about 4 - 4.5k but harmonics beyond that seems to be a topic of discussion now & then especially in class A groups.

Whats all this got to do with selecting a bypass cap value for a power stage? Nothing, just saying that sometimes a perfect design isn't even available yet it's widely lauded.

Back to power stage caps.........(sorry to much coffee)

I tend to utilize coupling caps and bypass caps to get where I want to go before the power stage. Meaning that if you are hitting it from the PI with a certain spectrum, it's not going to pick up any extra bass or treble from an anonymous source!

Back to your regularly scheduled thread

[Guitarman18]

Structo wrote:

Specifically, what audible effect does a small cap (25uf) verses a larger cap value (100uf)?

I've just tried this with my champ clone and found that 25uf seems softer and looser (nice for cleans) and the 100uf is tighter, particularly on the bass, which for me works well on overdrive. I suspect it's a similar effect to larger stage 1 filtering.

Is there a safe way to set this up on a switch, by putting another cap in series so that the two values are selectable? Or would it need voltage balancing resistors, changing the bias resistor setting?

Cheers,

Paul.

[Andy Le Blanc]

The cap blocks DC volts, so put the two values on the ends of a pot. with the wiper to ground.
At the extremes you'll get either value and in the center you'll get an
un-bypassed tone. Youll have to fiddle with the value of the pot. to find the
most effective to your ear.

[Structo]

The pot would have to be a large wirewound right?

I'm talking two 6V6's on the power amp.

Right now I have a 10 watt resistor on the cathode and it gets pretty warm.

Craig, I understand what you are saying, I was just trying to wrap my head around why different amps have different cathode bias caps.
Maybe that is what gives a Fender it's character and Marshall theirs.

I did change most couplers to .002uf from .01uf. so it's not as woofy.

Like I said, I have a couple different caps coming from Hoffman so I can see which one I like.

Mathematically, it looks like 50uf would work for me.
Right now I have a 560 ohm cathode resistor but I may lower that value and change the bias.

I can't figure out why this amp is pretty much full volume by the time it is at 12 o'clock, then from there on it just turns to mushy distortion with no definition.

It only has two gain stages (one 12ax7) then the PI and power tubes.

Maybe it has too much gain on V1.
I have 100K plates with 1K5 Rk, I have the bypass caps on a switch.
When the resistors are not bypassed the amp is pretty clean up to about noon, then some good distortion after that.
But when I kick in the bypass caps it starts sounding like crap above noon.

I also still get a squeal when I dime the treble and volume or if the presence is dimed.

I had tried switching the phase before, at the PI output but that was worse.

[Structo]

OK, I have the output power switch in this amp now and while it seems to work it is not a real dramatic difference.....


I am having trouble finding an answer about the cathode bias resistor on push pull amps.

This amp has two 6V6's.

One article I read said that if you are using a single cathode resistor and want to go to separate resistors, one for each tube that you double the value.
I was running a 560R 10 watt, so I went to two 1000R 10 watts.

Seems to work fine and I'm getting about the same cathode voltage at around 22-24v at idle.
But in looking at other schematics that have separate cathode resistors, it seems that they are going smaller with the resistors if there are individual cathode resistors.

So which is it? Double the value for the resistors on a 2 x 6V6 or half the value?