Parafeed cap range for S.E.X. 2.0 w/ Iron Upgrade

[Dr. Toobz]

I completely rebuilt my 2009 S.E.X. 2.0 from the ground up last summer, which included the new plate chokes and transformers that now come with the 2.1, as well as the newer-style C4S boards that mount on their own standoffs. It dawned on me the other day, however, that when I had the Magnequest iron upgrade in tow, the recommendation was to increase the parafeed coupling capacitors to 2.0uF, up from 1.0 with the Specos, presumably because of the greater inductance of the upgraded plate chokes (e.g., 50H vs. 30H stock). No such instructions were provided for the Bottlehead iron upgrade, however, so I have the same old 1uF Solens in there now. I'm aware that there is no exact science to calculating the optimal value (or range, as it may be), but I am not mistaken in that the new irons would do better with a somewhat higher value than 1uF? Just want to make sure that the -3dB roll-off point in the bass is low enough where it will not affect the quality or quantity of low bass frequencies on my speakers and HD650 headphones.

[Paul Joppa]

The value we ship is 1.5uF,  for 40 henries. My approximate calculation gives 1.25uF, so 1.0 is pretty good as well. Experimentation is in order if you feel the need to explore changes!

[Dr. Toobz]

Thanks Paul - that's a smaller difference than I had expected. Can't say I hear anything wrong with the 1uF ones in there now, though for some odd reason I had always preferred a larger value in the amplifier's first iteration. Sounds great either way - and the amp is super clean and presentable, with nary a change since I built it! Hence my not wanting to mess with anything under the plate unless it really matters.

The upside is that 1.5uF capacitors are a bit cheaper than 2.0, so that gives a little more budget for something exotic!

[denti alligator]

I have the SEX 2.0 with MagneQuest iron and never knew about this. Which caps should I switch out and what recommendations would you make?

[mcandmar]

Paul, is there any downsides to running a larger capacity, i.e. 2.2uf

[Paul Joppa]

I have always said that my approximate calculation is the center of a good range, from 0.5 to 2 times the nominal. That's based on simulations assuming a resistive load, which speakers are not in the deep bass. That's why I suggest experimenting, and doing so with the speakers you intend to use, and preferably in the room where you intend to use them - deep bass is also strongly influenced by room modes.

In the early days of parafeed revival in Japan, it was common to use a different calculation which generally gave a capacitance value something like 10 times larger than mine. It's a different sound, with some advantages and some disadvantages. The technical part is complex, trading off small-signal extension, power bandwidth, impedance flexibility, and distortion, and requiring a sophisticated model of core losses in the choke and output transformer.

[mcandmar]

Sounds like i'm still in the ball park with that value.  I found an old post of yours with the formula C = 2*L / (R*R) , are 40H and 8K the right values as that gives me 1.25?    Also any hints on how to calculate the interstage when you have a resistor instead of an inductor. ...i'm just curious.

Thanks,

Mark

[Paul Joppa]

You got it.

The interstage coupling capacitor/power stage grid-to-ground resistor have a time constant which is R*C, its inverse is 2*pi*f - f is the high pass cutoff frequency. Around 5Hz seems to work well in most cases. A lower frequency (longer time constant) slows recovery from grid current in transient clipping, while a higher frequency (shorter time constant) limits the bass. In theory, even a 20Hz cutoff should be basically inaudible, but I would not count in that without doing the experiment myself...

[kgoss]

I have the SEX 2.0 with MagneQuest iron and never knew about this. Which caps should I switch out and what recommendations would you make?

The upgrade instructions state that you should replace the stock 1uF parafeed capacitor with your favorite 2.0uF cap.  Just be sure to use a cap with a voltage rating at least as high as the stock.  I believe it was a 630V 1.0uF Solen.  I put Jupiter HT caps in mine.

Here is Paul's post about cap values for the S.E.X. amp.  He was kind enough to do one for each of the Bottlehead amps.  You can find them with a search but I think these should be made sticky posts inside each amp's forum page.

http://bottlehead.com/smf/index.php?topic=3148.msg26814#msg26814

[denti alligator]

Thanks. What about a Mundorf 2.2µF / 600V? That's approximately what's in my price range.

[kgoss]

Thanks. What about a Mundorf 2.2µF / 600V? That's approximately what's in my price range.

Yes, in Paul's post he states 600V minimum so that cap will be fine.

[denti alligator]

Any other recommendations for this price range?

Deciding between:

Mundorf F_Cap 2.2uf / 600V, Supreme

Clarity Film Capacitor 2.2uf / 630VDC, SA Series, Capacitor (cheap)

Auricap XO Film Cap 2.2uF / 600VDC

[JamieMcC]

The Jantzen Superior Z-Cap and Ampohm both sounded very good to me

[denti alligator]

I realize that I've already got Obbligatos in there, which I'm pretty sure are 2.0uf...

[Dr. Toobz]

As an update, I started off my capacitor journey with Mundorf Supremes in each coupling position. I was astonished at how big these things are! The 0.1uF variant itself dwarfed the larger valued Solens previously charged with parafeed duties. In any case, the 0.1uF and 1.5uF capacitors are now installed and sound very good - even prior to break in, with the best description I can come up with is that a layer of "haze" has been lifted from the sound. Bass is a bit better, at least for very low pulses or "booms," but there was not much difference between the 1uF and 1.5uF values, as expected. I've used Mundorf oil capacitors previously in my Seduction, and these might find their way into my yet-unbuilt Eros.

I am using the 4 ohm taps right now, but I've dithered about rewiring for 8 ohms instead. The amp is dead silent over headphones as a result, but probably would be noisier on the 8 ohm taps. On the other hand, the ongoing Klipsch speaker paradox continues. The Heresies seem to like lots of current, so 4 ohms is great, but the lower impedance (and resultantly lower THD) also has the opposite effect of making the sound somewhat austere and bass-shy. A subwoofer is probably next on the list....