SEX 2.1 initial measurements

[christos]

Hi everyone

I finally ended up building my SEX 2.1 with the C4S upgrade. I measured everything fine but the following two:

W/O C4S               with C4S
A5=65,5V               A5=  59V, OA=58V (should be 75V)
B5=90V                  B5=  88V, OB=86V (should be 75V)
Are these OK (given that they should be around 75V)? I notice that, although I have more extension in highs, presence, naturalness, emotion in voices, I miss that bit of resolution and dynamics. Is this connected with the above measurements?

I consider to replace the stock pot with a Gold point or Dact and the o,1μF cap with some upgrade of Jensen or Duelund. Will this improve things in terms of resolution, dynamics? Can anyone please help me as to which matches most from the above?
I would appreciate any advice.

Christos

[Paul Birkeland]

I bet the A5/B5 voltages will switch channels if you swap tubes.

75V is 75V +/-15%, which is 63-86V.

[christos]

Yes Paul

I swaped the tubes, now they measure

OA=87,5    A5=87,5
OB= 59      B5=59,1

I suppose this is ok? BTW why is that deviation from 75?

Could also anyone give me some clue about the upgrades' merits? Is it worth to go ahead? Which one is recommended?

Christos

[Paul Birkeland]

Yes Paul

I swaped the tubes, now they measure

OA=87,5    A5=87,5
OB= 59      B5=59,1

I suppose this is ok? BTW why is that deviation from 75?

Yes, this indicates that the circuit is working as expected.  75V is what the 6DN7 datasheet says the voltage should be.  They create these datasheets from a "bogey" tube, so that's a bit arbitrary.  The 15% deviance is just from tube-to-tube variation.  If you see something like 100V at A5/B5, that would be more like a tube on its way out.

[Paul Joppa]

I should point out that the C4S upgrade results in a greater variation in this voltage. Not a problem, just characteristic of the way the circuit works.

[christos]

Thanks for the instructions.
I bought (not delivered yet) 4 russian K42Y-2 (.1, 630V) PIO caps at a low cost. Has anyone tried them with the SEX? Are these an upgrade to the stock ones? Could you give me some impressions?
I also found some Vitamin Q but of a 0.047μF value each. Could I use two in parallel to get the manufacturer's specification?

Thanks
Christos

[mcandmar]

Thanks for the instructions.
I bought (not delivered yet) 4 russian K42Y-2 (.1, 630V) PIO caps at a low cost. Has anyone tried them with the SEX? Are these an upgrade to the stock ones? Could you give me some impressions?
I also found some Vitamin Q but of a 0.047μF value each. Could I use two in parallel to get the manufacturer's specification?

Thanks
Christos

I've not tried those ones myself but i've only ever had good experiences with Rusky caps so let us know when you get them installed.

[physicsmajor]

Thanks for the instructions.
I bought (not delivered yet) 4 russian K42Y-2 (.1, 630V) PIO caps at a low cost. Has anyone tried them with the SEX? Are these an upgrade to the stock ones? Could you give me some impressions?
I also found some Vitamin Q but of a 0.047μF value each. Could I use two in parallel to get the manufacturer's specification?

Thanks
Christos

Warning: I hadn't seen the K42Y-2 designation before. I looked it up and while the capacitance and voltage ratings are correct, these are PIO (paper-in-oil) capacitors, NOT teflon/film caps! They are Russian, but I don't know if these are going to be an upgrade at all over the stock Solens.

When people reference Russian teflons, or KK teflons, etc. they mean military cold-war production teflon film caps. The FT-3 series is what I used, and they usually run about $10 per (and are available in matched sets, for example here: http://www.ebay.com/itm/0-1-uF-600-V-STRONG-MATCHED-RUSSIAN-TEFLON-AUDIO-HI-END-CAPACITORS-FT-3-/281305311703 )

[Paul Joppa]

...
I also found some Vitamin Q but of a 0.047μF value each. Could I use two in parallel to get the manufacturer's specification?

Thanks
Christos

You can use them by themselves. The -3dB point is 6Hz stock, so 12Hz if you use half the capacitance. A lot of listeners (but not everybody!) find something not right with paralleled caps.

[christos]

You can use them by themselves. The -3dB point is 6Hz stock, so 12Hz if you use half the capacitance. A lot of listeners (but not everybody!) find something not right with paralleled caps.

Paul
Using 0.047 instead of the design 0.1 will not affect anything? Is it a sole matter of shifting the -3db point? And to which extent can one go changing the design capacity in the coupling cap and the output cap? Is it acceptable lowering  or increasing it?

Likewise can one accept a 400V voltage rating instead of the design 630V?

There must be some tech theory behind it which I am interested to cherish.

Christos

[Paul Joppa]

It does depend on the particular point in the circuit.

The interstage cap (0.1uF) in conjunction with the grid to ground resistance (249K ohms) gives a low frequency cutoff of -3dB at 6.39Hz. This is fairly standard practice, allowing for many such cutoffs in the chain from recording microphone to loudspeaker. If one of those filter frequencies is above 20Hz you might possibly hear it, and above 40Hz you will hear it for sure in a typical audiophile sound system. Of course if you did that everywhere it would be quite bad, but in a single place it's not a problem.

The voltage in the SEX amp is around 400v; a bit higher at startup for a few seconds. Most caps can handle a greater voltage for a short time, but also most caps will have a longer life at a lower voltage. You will likely be fine with a quality cap (such as the Vitamin Q) rated 400v but I won't recommend less that 600v because I don't know all the different caps that people might use.

For the output (parafeed) cap, the value is chosen carefully to match with the plate choke and output transformer. Usually you can go between half and twice the design value with relatively little change. But because of the way parafeed works, in the deep bass it is possible to get higher voltages than the power supply, so I definitely recommend at least a 600 volt rating.

Hope that helps a bit!

[christos]

Yes Paul it helps but also raises more questions as a consequence that, alas, I miss some basic theory of amplification electronics.
 I wonder if there is a thread or link to deploy the theory (and practice) behind the coupling capacitors in signal length in relation to the frequency bandwidth.

Christos

[Paul Birkeland]

The "Analog Circuits" paragraph here gives a pretty concise explanation with lots of links to other topics.

-PB