Cathode bypass capacitors: Electrolytic vs. film?

[Rhodesplyr]

In playing around with V1 cathode bypass capacitors in a simple guitar amp, I decided to try making up the nominal 25uF value with a 3.3uF Dayton film cap I had for some crossovers in parallel with a standard 22uF electrolytic.

There is an audible difference to me in terms of slightly more pronounced note attacks with a film cap in the mix compared to electrolytic alone.

So here's the question: I mainly play piano and organ and don't really consider myself a guitar player. What do guitar players want? Would you prefer a slightly smoother, more compressed attack or a sharper note attack where what you're doing with the pick comes through more clearly? This is clearly a matter of taste thing to me, and I'm highly aware that I risk accusations of cork-sniffery just by posting it.

[nsubulysses]

Did you measure the capacitance of the capacitors?

I would guess it is most likely a change in tone because the paralleled pair has slightly lower capacitance compared to the single 25uF

As a guitar player, I'd guess a player that plays more clean tones or lighly overdriven would probably want the woofier, looser sound where as a player that has a an amp with a substantial amount of gain would want pick attack to cut through. So you'd most likely see cathode bypass caps in the .68-10uF range or so for a gainy amp, whereas in the clean amp you'd probably end up seein the 10uF and up

Not sure if the material is really the main thing driving different tone here but I am not sure becasue I have not messed around with different materials for tone. I'd guess capacitance first

[tubeswell]

Besides upping the uF slightly, the only other difference I can think of, is that bypassing an e-cap with a film cap lowers the ESR for the frequencies that the film cap can pass.

If you take a 1k5 cathode resistor on a 121AX7, 3.3uF on its own can bypass all of the standard tuned guitar notes (bottom E = 82Hz), the half-boost point is:

f=1/(2Pii x Rk x Ck) = 1/(44/7 x 1500R x 0.0000033F) = 32Hz

41Hz is the 1st even-order harmonic of the fundamental bottom E note, and this would also be boosted in a V1 stage, so would be likely to be picked up in a PP amp (esp where you use global NFB to get more fidelity). So I'd say what you are noticing is probably just the effect of the film cap improving the ESR ever so slightly (and thereby also improving the S:N ratio).

[Chuck H]

I would guess it is most likely a change in tone because the paralleled pair has slightly lower capacitance compared to the single 25uF

Paralleling caps gives a sum capacitance. Paralleled cap value will always be greater than any single cap in the circuit.

[Chuck H]

It's like tubes said ^^^^^^^^^^

At that sort of uF you probably wouldn't hear a difference between 22uF, 25uF or 28.3uF. Most likely what you're hearing is the improvement in bypass efficiency because of the lower ESR. Although this is also going to be quite small (tiny really) I don't doubt you can hear it.

Something Mesa does (and I have done also) is to use dry tantalum capacitors for cathode bypass circuits. I know tants have a bad rep. Ignore this. These are dry tants in the bullet shape made by Kemet, not the little epoxied blobs in 70's gear that are notorious for burning up and stinking. Also, preamp tube cathode bypass circuits rarely see more than 3 volts so the probability of a reverse or over voltage failure are minimized if their use is limited to these circuits. The ESR of the Kemet dry tants is markedly lower than electrolytics and markedly more compact than paralleling film caps for higher uF values. Most of the old timers here will be quick to voice their distaste for ANY tantalum capacitor, so the choice is up to you.

I used them because dry tants last much longer than electrolytics and that minimizes work when a cap job is necessary. The lower ESR and any tonal advantage therein is something I'm not even sure I can hear. When you consider that most preamp cathode bypass circuits allow a gain advantage of maybe 20% @ frequency and the ESR difference between modern cap types (film OR electrolytic) is a difference of, at most, a couple of percent, you're looking at a performance advantage of 2% of 20% (roughly)? So about .4% advantage Like I said, I'm not saying you can't hear it. I'm just saying it's a vanishingly small difference.

Of much more value to me is the idea that the dry tants aren't going to crystalize and cause problems in ten years like some modern electrolytics. If you actually hear a tonal improvent as well, so much the better

[Enzo]

I wonder about the longer life of the tant being a thing. I expect a good 40 years out of a common electrolytic as used in the cathode bypass circuit. SO if the dry tant lasts longer than that, I won't be around to worry about it.

[Chuck H]

I wonder about the longer life of the tant being a thing. I expect a good 40 years out of a common electrolytic as used in the cathode bypass circuit. SO if the dry tant lasts longer than that, I won't be around to worry about it.

True enough.

[nsubulysses]

Paralleling caps gives a sum capacitance. Paralleled cap value will always be greater than any single cap in the circuit.

what I meant was unless measured you don't know what it is. What if the 22uF cap is 19uF and the 3.3uF is 2.9uF. That's 21.9uF. What if the 25uF he is comparing it against is 27uF when measured?

I was guessing his different tone he heard was becuase of actual values of the caps in circuit when measured, not just the values printed on the label.

[Chuck H]

Ok. I don't think I would be able to hear a difference at those values in a typical preamp triode cathode bypass circuit. The difference at 82Hz between, say, 20uF and 28.5uf is only .03dB and there's not much below that to think about regarding guitar or guitar amp speakers.

[bob p]

In playing around with V1 cathode bypass capacitors in a simple guitar amp, I decided to try making up the nominal 25uF value with a 3.3uF Dayton film cap I had for some crossovers in parallel with a standard 22uF electrolytic.

There is an audible difference to me in terms of slightly more pronounced note attacks with a film cap in the mix compared to electrolytic alone.

Equivalent circuit diagram for a capacitor:


Frequency response characteristics:


Comparing capacitor types:

[Enzo]

nsu -

22.9uf versus 21.7uf, or whatever. usually when trying different bypass caps, unless you use at least a factor of two you rarely hear a difference. SO if you have 22uf, if you want to actually hear a difference, you might select something at the 10uf level or lower or the 47uf level or higher to compare. Comparing 25uf and 27uf is a laboratory experiment, but not usually an audible factor.

I have to point out the notes on the old Fender schematics that everything on there was a +/-20% deal. The caps were not precision anyway, so 25 and 27 were within 10% so the stock cap could measure either one.

[nickb]

Another factor to consider is the time constant was changed. The effect is more significant on stages with big signal levels as the signal is not symmetrical (even order distortion) and this leads to the cathode capacitor charging and so will affect the signal.

Taking bobp's charts the ESR goes from about 4 ohms to 0.01 ohms, both much less than the typical 1.5K ohm cathode resistor. The ESR difference is very roughly about 0.1dB at the corner frequency assuming a single stage is modified. That's very small so it might be hard to hear amongst all the other small distortions.

Since we don't have the schematic or even how many stages are being modified only general statements can be made.

Lastly, the only way to be sure there is any audible effect is to do an ABX test. Unless the differences are huge it's all but impossible for a human to accurately do it due to cognition bias effects. It would be interesting to record the output for the two cases using identical source material, then post the recordings but don't tell as which is which

Edit: Here are some plots for the two cases, one with frequency response and the other showing the DC shifts by looking at the voltage on the cathode with a sine wave burst.

Freq Cath Caps.pdf

DC Shift Cath Caps.pdf

[Enzo]

Darn hard to hear 0.1db. 1 decibel is very roughly intended to be about the minimum difference the ear can hear. And it is logarithmic.

[Mick Bailey]

I see some Hifi designs (particularly those described in 'Audio Valve Amplifiers by Rainer zer Linde) where every electrolytic is bypassed with a low-value film cap, even in the PSU - usually 220nf or 470nf. I can't detect any A-B difference with/without no matter how hard I listen. I'm not convinced of the value in doing this in a guitar amp. If there's anything it's outside of my hearing range. Either that, or I'm not a True Believer.

[Steve A.]

If we are talking about a BF or SF preamp I've always liked to use 2.2uF film caps instead of the 25uF electrolytics for the first gain stage. Or use a center-off SPDT switch to toggle between 2.2uF, 0uF and 0.68uF (Dan Torres sold such a switch as his Triple Killer kit. Only I think he included a 22uF e-cap...)

Steve