18Watt.com

Will 35V rating capacitor do in the power section cathode capacitor?
I see they are in various rating in various schematic.

Thanks

35V should be fine. You get about 11V at idle across that cap, although it will tend to move up and down as the amp gets played. Very large cap values will get less transient fluctuations across them.

how large is very large? i.e. 470uF? 1000uF?

25uF and up would be "large"


i've seen no amp with cathode bigger than 320uF

Are you talking about in the preamp? Typical 18W EL84 cathode cap values range from 100uF to 500uF. But I'm talking about 1000uF to 2200uF at the power tube. This helps get rid of some graininess from the distortion you can sometimes get with cathode-biased amps.. I've done this with all the cathode-bias amps I'm currently using.

so in the preamp the biffer the cap the woofier it could be?

but in the power cathode the bigger... the better the bass?







PS: in the preamp with 820R - would i hear any differences between using 25uf or 4,7uF , even when the CUT is 60hz???

I believe that a bigger Ck in the power section helps to tighten the bass - Phil, correct me if I'm way off base here.

Did you say "way off bass?"

OK, so with a huge cathode cap, a cathode-biased power amp behaves more like a fixed-biased on. So it's punchier and with firmer bass, since the bias voltage is less affected by bass notes.

richon wrote:PS: in the preamp with 820R - would i hear any differences between using 25uf or 4,7uF , even when the CUT is 60hz???

I think you should be able to hear a difference, between a preamp stage with a low-end break point of 60Hz vs one of 13Hz, even though the guitar low E is at 83Hz. Personally, I would prefer to roll off at around 100Hz or even 150Hz, since the low notes from an electric guitar pickup seem to always be much louder than the higher notes. I would generally expect 3.3uF or 2.2uF to be a better choice.

Z did you keep the resistor value same or raised that from 125 Ohm?

I never use 125 ohms, as I usually use NOS tubes that I want to last a few years. At the moment I'm favoring 180 ohms to 220 ohms. With the real big cathode caps you can run the power tubes colder, and therefore more safely, without detriment to tone. This is because the cathode voltage doesn't keep zooming up so high that the power tubes go into blocking or starved mode. Instead the cathode voltage just follows the average amplitude envelope and adjusts the bias voltage to suit.

FYI, I just read a post on AX84 today, that recommends 2200uF caps.

Zoe N Iain wrote:Have you tried the 'stupid big cap on the PA cathode' trick yet, Jamie? Credit where it's due to James Marchant, cos that one really does turn a good cathode biased amp into a great one. Doesn't lose anything of it's cathode-bias-ey-ness, either. Just sounds tighter and nicer.

FWIW, I think 2200uF is the sweet value. 1000 is good and 4700u and 10000u sounded a bit wooden to my ear, but 2200 is just right.

wow,
I just ordered 1000uf, maybe should have added a 2200uf, I think I have some 150 Ohm and some 250 Ohm 5 watt resistor. I will try with a 150 Ohm, and a 1000uF to start with.
Thanks

zaphod_phil wrote:FYI, I just read a post on AX84 today, that recommends 2200uF caps.

Zoe N Iain wrote:Have you tried the 'stupid big cap on the PA cathode' trick yet, Jamie? Credit where it's due to James Marchant, cos that one really does turn a good cathode biased amp into a great one. Doesn't lose anything of it's cathode-bias-ey-ness, either. Just sounds tighter and nicer.

FWIW, I think 2200uF is the sweet value. 1000 is good and 4700u and 10000u sounded a bit wooden to my ear, but 2200 is just right.

I don't know if I can agree with that. Part of the tone and feel of a Cathode biased amp is the looseness and compression. That comes from the looser power amp. If you use a big cap and that tightens up the low end like a fixed bias amp, you might as well go fixed bias and get the extra output from the power amp (Yes, I know fixed biased is more complex).

I'm not saying it isn't valid to use a big cap, because it certainly is. But the thing about most cathode biased amps that is cool is the compression caused by the power amp trying and failing to keep up with each note. After the initial attack the power amp 'catches up', which is what causes the bloom of the held notes. I also wonder if the stiffer power is harder on the tubes - that they are held to higher voltage for too long.

Maybe a good compromise solution would be to use On-Off-On switch for standby and switch in two different Cathode cap values. One for the stock spongy feel and another for the tighter feel.

True, but the voltage across that big fat cap still fluctuates with the average signal envelope, which compresses the output signal. So it still preserves a lot of the feel of a cathode-biased amp. It just doesn't zoom up and down on every bass note or pick click.

Ok, but a fixed biased amp isn't exactly fixed either. A lot depends on how severe the power supply filtering. For ex - a JTM-45 vs a Marshall 100 watt.

The more I think about it, I think having a 50uf and a 2200uf on a switch would be a great option.

rjgtr wrote:Ok, but a fixed biased amp isn't exactly fixed either. A lot depends on how severe the power supply filtering. For ex - a JTM-45 vs a Marshall 100 watt.

You really made a good point there....

Here's my question - where do y'all find 2200 uf caps?

2200uF caps at around 25V or 35V can be found just about anywhere that sells electronics components. They get used a lot in SS stuff.

I know I should try to build it and try for myself, but I'd like to ask anyway. How much of a difference in sound is there between big cap cathode bias, normal cathode bias and fixed bias? For example, is it easy to tell which is which if you do an A/B comparison between big cap cathode bias and fixed bias?

And I was also wondering how much the bias voltage swings with normal cathode bias? Zaphod mentioned zooming up and down, so I assume it's reasonably large

The "large" is a relative thing. It's in comparison with an amp's idle cathode bias voltage level and affects different kinds of tubes to different extents, depending on their gain. Paul Ruby measured a 15V bias shift in one of his 18W amps when operating flat out, with a steady 200Hz sine wave, which is an awful lot for an EL84. His amp, which was running at somewhat lower voltages than a typical 18W, had an idle bias voltage of 7.5V. You would need a good storage scope to see some of the faster transients when using lower value cathode caps.

I'm now coming around to the view that, given that the typical bias voltage for many 18W amps is around 11 or 12V, this suggests that 25V may be on the margin for a cathode cap, and 35V or above may be a safer bet.

hmmm,
rjgtr and Z has two different opinion here.

rjgtr what value caps do you use for your amp?

Maybe I stay in between and use something like 470uF with 150 Ohm Resistor.

Ultimately I most likely will end up testing it myself.

Thanks