• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Cathode Bypass Capacitor

Hi folks,

I need some advice on choosing a make/model of cathode bypass cap. I need a 100uf 10-25v, preferably axial.

Is there much reason to use anything other than a low impedance, high temperature, general purpose? I've done some reading on what spec to look for, and on recommended caps, but a lot of what I've read has been contradictory.

Thank!
 
Radial electrolytics will have slightly lower inductance, and higher voltage ones will have slightly higher ESR (to over-generalize) but neither is too significant for audio use. 105 C rating is sure nice to have inside a hot amplifier chassis. If you have some steady DC voltage across one, and it's very large electrically, so that no signal appears across it, an electrolytic really isn't terrible, despite their rep.

All good fortune,
Chris
 
Having to replace some electrolytic caps 3 time in 42 years on my dynakit amp, I am partial to long life caps. Most of the ones that have an 8000 to 10000 hour spec have decent ESR also. Unfortunately they only come in radial lead, not tall cans or axial. I'm partial to buying from farnell.com as they have the hours life in the selection table. Other vendors you have to download the datasheet, and then read it carefully to see if the size you are buying has the best hours life. Sometimes when I need something from mouser that newark (farnell usa) doesn't stock, I go to the trouble of downloading the datasheets. I've ended up lately with various panasonic, nichicon, and rubicon radial lead caps, nothing under 3000 hours service life. Every cap I've installed has made the organ or amp sound better or produce more power, even the 1000 hour caps I bought in 2009 before I figured out how to install radial lead caps on solder terminal strips below the deck. This includes some tiny 22 uf 25 vdc caps I used to replace 25 uf 3 vdc cathode bypass caps the Hammond amp came with.
 
Last edited:
Thanks for the info.

Let me clarify my question a little further, I'm new to vacuum tubes so bear with me. I guess what I'm more asking is how much this cap will contribute to the sounds/performance of the amp, and in what ways. Will a general purpose cap like a Nichicon VX, or some sort of Sprague, or Mallory yield any different results in this circuit than say an Elna Cerafine/Silmic or Nichi Muze of some flavour?

Using an axial will make things a little easier for me, but "better" caps are available as radials.
 

M Gregg

Disabled Account
2010-06-28 11:04 pm
UK
Well,

Different caps will give different results...it depends where in the circuit it is used. In the bypass of a cathode resistor in a preamp you will here it..and it depends what type of sound your looking for. There is always the LED trick (so people that say the caps don't make a difference and then say but you can hear an LED??)..but if your looking for the old tube sound a standard 105Deg cap will give a warmer sound warm bass. The silmic have two "Flavours" I like the Silmic II its a sort of cross between old warmer sound with detail..similar effect using Hex diodes in a bridge ...The Cerafine are sharper slightly more edgy..<<HIFI talk.

Elna Cerafine and Elna Silmic Capacitors homepage

I have no connection with the above..Muse are also good..

From a mix point of view think one type for the pre sections and another on the power section to balance the sound..you might like Silmic II in the OP section and Cerafine in the pre..again it depends on type of coupling cap etc...You wont get wax paper sound with polypropylene etc. High temp tend to give longer life as does using higher voltage values.. Again watch mounting position don't mount close to hot/warm components it reduces the life of the cap<<dry out.

Regards
M. Gregg
 
Last edited:

M Gregg

Disabled Account
2010-06-28 11:04 pm
UK
I should also add,

The type of cathode resistor will also make a difference..(Off topic)
In the pre section this will show..how much bass, how clean, how warm etc...Low wattage mills..carbon...tant.etc or LED..
I have also heard some marked changes with adding feeder resistors to LED's in the cathode of a preamp tube. (increasing current)
The effect is a warmer sound with the feeder..you won't know this until you try it..

Regards
M. Gregg
 
Last edited:
The use of an electrolytic cathode bypass capacitor makes no discernible difference to the sound other than the value affects the frequency response. The choice of resistor also makes no change to the sound other than carbon types might add a little noise.

I agree with other posters that a good long life electrolytic is the way to go. Just a not about quoted life hours. Be aware these are usually quoted at the rated operating temperature so that if you operate it considerably below that temperature the life will be extended.

Cheers

Ian
 
The use of an electrolytic cathode bypass capacitor makes no discernible difference to the sound other than the value affects the frequency response.

To you, it doesn't. To many people, it does! Please accept the fact that we are not all the same - some of us *can* hear the sound of a capacitor.

If you must use a bypass cap, the Elna Silmic II.would be a good choice. A CCS anode load+LED cathode bias would be a good way, too...
 
To you, it doesn't. To many people, it does! Please accept the fact that we are not all the same - some of us *can* hear the sound of a capacitor.

We are not talking about the sound of a capacitor but the sound of a cathode bypass capacitor. It is well documented that in equalisation circuits and poorly designed coupling and bypass circuits that different capacitors do have different sounds. In properly designed bypass and coupling circuits there is no evidence of any discernible difference. If you have such evidence I would be pleased to see it.

Cheers

Ian
 
I assure you I can hear the "sound" of any coupling or bypass cap. It is especially obvious when I use good headphones and good tubes, in a minimalist, transparent topology. PSU caps have an influence, too.
We both know this is very subjective and difficult to measure with the instruments we have.
 

M Gregg

Disabled Account
2010-06-28 11:04 pm
UK
time to exorsize those bypass caps and replace them with diodes or led's where you can...

:D why if you can't hear a difference..if you can and you prefer the sound of a cap..then what? Then why is there a difference?

Oups the circuit design is not correct ..or you should not hear a difference..:joker:

OK there have been lots of measurements so we know there is a difference SY has done a lot of work on this...if the diode can be heard perhaps I should say not heard then suddenly anything else put in circuit in place of it should make no difference not heard...because cathode components cannot be heard? If they can then try putting a low watt mills in and have a listen..

Regards
M. Gregg
 
Ok, the disagreement is pretty much in line with what I've read. I'm in the camp that caps in certain applications can change sound from my own experiences experimenting with input coupling caps on a solid state amp doing blind A/B tests (luckily easy with stereo!). Some will change the signal very little/not at all compared to a jumper, some will change it in a more noticeable way- of those that change the sound some do it in a nice sounding way, and some do it in a way that sounds not so nice. If all caps were ideal (C only) capacitors there would be no sound difference between construction type and brand, but caps aren't ideal, and have different electrical properties, what sounds best is really a discussion of type and magnitude of flaws (from an electrical standpoint) in the capacitors performance. What sounds best will depend on a whole host of variables, most important of all being the listeners preference (and expectation). I'm far from convinced that exotic/expensive caps will always sound better/best, but I can accept that caps will be designed with "flaws" that generally sound more pleasant, hence audio specific caps like the Muse.


Anyway, without going further down the rabbit hole of whether caps have sound, lets assume that different properties of capacitors are audible in some applications. Is the cathode bypass on of these applications. I understand whether you have a bypass cap, and what its value is will effect the output in easily quantifiable ways, but beyond that there seems to be little agreement on whether the ancillary properties (flaws) of capacitors will impact the output signal in a meaningful way here. I guess I've got another point to experiment with. I think an excuse for buying an osciloscope is forming.

As no one seems to have any strong preferences for particular caps in this location, and based on what I can get cheap or already have on hand I think I'll try

Nichicon VX
Vishay AML 138
Sprague Atom (given to me by a friend, based on things I've read I wouldn't bother otherwise)
Nichicon Muse
Maybe Elna Silmics if I still have some of the right value somewhere.

Any other recomendations?

Can you tell me more about using a diode in this location? What is that supposed to accomplish?
 
To follow up on a point made earlier, classic era electronics have woefully small bypass caps almost without exception. Today, the incremental cost of adequately large cap values is insignificant for DIYers. The bigger the (ideal) capacitance, the less signal appears across it, so the less it can affect the sound.

All good fortune,
Chris
 
And of course, should anyone have forgotten by now WHAT the purpose is of the cathode-resistor bypass capacitor... It can be thought of two different ways: either in conjunction with the resistor and the operating point of the tube, it acts as a small self-contained voltage source, raising cathode ... synthetically making more negative the grid ... or if you prefer ... turning the cathode resistor into low-impedance A/C constant voltage source, again for the same purpose of making the grid more relatively negative.

The "problem" with using LED or DIODE (normal, or zener) is that the voltages are fixed not by the operating point of the tube but by the said-to-be-normal operating point of the theoretical tube that was designed into the circuit. With brand-to-brand variation, and even era-to-era variation, the bias-point can vary by a surprisingly significant amount. The resistor (A/C bypassed by a capacitor) approach adapts to different tubes more gracefully. And as they age... The adaptation isn't perfect, of course, but it tends to accommodate all but the most egregiously-out-of-spec valves.

The idea of adding a constant-current source as the anode-load is helpful in that it serves as its own kind of cathode-resistor-bypass. Indeed, if the constant-current source is "near perfect", one can remove the cathode-resistor bypass capacitor altogether!

So, if the goal based on some real golden-ear listening preferences is to remove the cathode resistor cap, then ... CCS is the way to go. If the goal is to just implement an inexpensive grid bias (by way of our cathode resistor), at fixed voltage, then choose a cap of high enough value to have a -3 dB point say down near 5Hz or 10Hz or even lower. They're pretty cheap these days. And, supposing you want to try the glowing red/green LED light show ... then that too is OK - but first, with your tubes in place and glowing, remember to measure the exact cathode voltage, anode voltage, B+ supply! You'll need these numbers to engineer the LED substitute.

GoatGuy

EDIT/PS: The cap value can be found with this formula: 1000000/(6.283 F R) where F is the -3 dB corner frequency you're trying to engineer, and R is the nominal value of the cathode resistor itself. e.g. 1,000,000/(6.28 x 5 Hz x 560 ohm) = 56 uF
 
Last edited:
I assure you I can hear the "sound" of any coupling or bypass cap. It is especially obvious when I use good headphones and good tubes, in a minimalist, transparent topology. PSU caps have an influence, too.
We both know this is very subjective and difficult to measure with the instruments we have.


The only coupling or bypass capacitor you can hear is one is a badly designed circuit. Please give full circuit details of this minimalist headphones amplifier so we can judge for ourselves.

Cheers

Ian
 
And of course, should anyone have forgotten by now WHAT the purpose is of the cathode-resistor bypass capacitor... It can be thought of two different ways: either in conjunction with the resistor and the operating point of the tube, it acts as a small self-contained voltage source, raising cathode ... synthetically making more negative the grid ... or if you prefer ... turning the cathode resistor into low-impedance A/C constant voltage source, again for the same purpose of making the grid more relatively negative.

It has always been my understanding of the function of the cathode bypass capacitor is that it provides a low impedance AC path to ground. This is significant, since the cathode bypass capacitor is in the signal path, being in the both the input and output (AC) current loop. Since it is in the signal path, capacitor quality is just as important as any coupling capacitor.