SEX 2.1 Cathode Bypass Cap?

[Daikini]

There is a cap that is labelled "Cbias" on the SEX 2.1 schematic.  After doing some reading of old posts, it is clear that there was a cap with the same label in earlier versions of the SEX amp that was NOT a cathode resistor bypass.  However, my interpretation of the schematic for the 2.1 is that this capacitor is in fact bypassing the cathode resistor.  Am I interpreting this correctly?

If this is actually a cathode bypass cap, and I wanted to experiment with using a film cap in that position, how would I calculate the relationship between cap value and low frequency roll-off point?

[mcandmar]

Watching this thread with interest

I actually bought a crap ton of 1000uf 50v caps for testing, my plan is to test them as dc blocking output caps in a small tube amp to see can i pin down any sonic differences between the different brands and models.  The end game is to use the best ones in my portable tube amp, and also as the cathode bypass caps in the S.E.X. amp.   I never thought about the actual capacity, i just assumed 1000uf was required which ruled out any poly caps.

My test selection will include Panasonic FR, FM, FC, Nichicon KA, KT, FG, KZ, KW, vs the ELNA Silmic II i am currently using.

[Paul Birkeland]

There is a cap that is labelled "Cbias" on the SEX 2.1 schematic.  After doing some reading of old posts, it is clear that there was a cap with the same label in earlier versions of the SEX amp that was NOT a cathode resistor bypass.  However, my interpretation of the schematic for the 2.1 is that this capacitor is in fact bypassing the cathode resistor.  Am I interpreting this correctly?

The old SEX 2.0 had some unusual design characteristics to enable the output transformer to be wired as an autoformer.  Conceptually, Cbias and Rbias are fulfilling the same role in both amplifiers, with the exception that all the high voltage current in the SEX 2.0 travels through Rbias to develop the negative fixed bias voltage, while Rbias in the SEX 2.0 is moved to just the cathode of the output triode in each channel, which runs a little less current.

If this is actually a cathode bypass cap, and I wanted to experiment with using a film cap in that position, how would I calculate the relationship between cap value and low frequency roll-off point?

I think you'd have to start by finding 2,000uF worth of film caps that you could fit inside the chassis.  The amp is designed to avoid the contribution of that particular capacitor as much as possible.  To put the coloration of that cap back into the circuit, you could move the wires going from A3/B3 to the output transformer to the other side of the 680R resistors and 1,000uF cap (the black one in the pair) to ground the primary of the output transformer. 

-PB

[Paul Joppa]

The 1000uF is overkill, though some think overkill in that position is desireable.

To roughly estimate where it starts to affect the frequency response, I use the inverse of the tube's transconductance. Transconductance is about 5500 micromhos at the operating point, so figure about 180 ohms. That would be about 0.9Hz with the 1000uF cap, 9Hz with 100uF, and 90Hz with 10uF. In actual practice the turnover will be a bit lower but it gets complicated to calculate, and the effect is not simple to begin with, so I use this conservative approximation which is close enough.

As PB points out, the effect of this cap is much reduced by connecting the OPT to the cathode as we do in SEX. But I'd never say it has no effect at all. Try the comparison suggested by PB and see what you hear!

[Daikini]

Thanks for the calculations.

I see the value of connecting the output transformer to the cathode to minimize the effect of the cathode bypass cap. However, my experience tells me that even in this configuration, tinkering with the cap has a noticeable effect on the amp's sound.  I bypassed the 1000 uF electrolytics with some 0.1 uF film caps that I happened to have handy, and the sound thickened/warmed noticeably, which was more of a change than I had expected.

I think that I can live with a 9 hertz rolloff point.   I ordered some 100 uF film caps which should come in today.  I'll put them in tonight and report back in a few days once they've had time to break in a bit. 

[mcandmar]

Interesting the bypass cap had such a pronounced effect.  Keep us posted on any developments...

P.S. You are aware that 100uf film caps are usually huge?

[Daikini]

Yes, they are pretty big. I got Dayton 250v 100uF caps, which were marginally smaller than Solens, and I'm glad I did. They barely fit in the base. I was imagining what it would look like on stilts there for a bit, but I got them to fit.

I'm about 20 hours into burn in and reluctant to make a judgment just yet on the sound. They should be pretty well settled sometime this weekend, I'll let you know what I think then.

[Daikini]

Ok, it's had plenty of break-in time.

The changes I hear are mostly pretty subtle.  It sounded really good before, and it sounds really good now.  But if I had to put my finger on what's different, I would say that there is more detail and clarity.  I can hear small details, especially little quiet things as notes fade away, that I did not notice before. 

Going all the way to 100 uF film caps is probably excessive, but I knew I would want to try it eventually, so why not go all the way.  But for the money I spent on those two caps, I could have bought a selection of smaller value caps with which to bypass the original electrolytic.  These would certainly be smaller and easier to fit on the chassis, and there are lots more cap flavors to choose from in lower value caps. 

[Daikini]

I'm liking the sound of the amp more as I listen more.  Maybe the break-in on caps in this position is longer than I thought, or my ears are just getting used to it.  Hard to say.  But I definitely have a subjective experience of more clarity and detail, in a way that really helps my fullrange speakers do their magic realism with vocals and acoustic instruments.  And there is a smoothness in the midrange that I don't remember with the electrolytic bypass caps.

It's a very addictive amp.  When I don't want to get loud, I actually prefer it over my Stereomour.

[mcandmar]

I recently picked up two pairs of Mundorfs for a song on eBay, one set is 150uf and another 220uf so i really have to excuses not to try this

[Daikini]

Are they enormous?  I could barely fit 250v 100uFs in the box.

[mcandmar]

They are, the 150uf are 2.5" x 2.25", and the 220uf are 4.5" x 2".  No chance they will fit inside the enclosure so i would have to bodge something together temporarily to test them.

[Grainger49]

Testing is easy.  All you have to do is use jumpers.  Fitting them in afterward is the problem.

[JamieMcC]

I recently picked up two pairs of Mundorfs for a song on eBay, one set is 150uf and another 220uf so i really have to excuses not to try this

AKA the ebay sniper you out bid me in the last second But I do have a pair of 100uf Mundorfs and a couple of much slimmer 89uf Epcos that actually sound rather good so watching with interest.

[mcandmar]

Ohh really, somebody got a 2nd username i dont know about