Output Cap (Cpf) value

[Ivan303]

The more I look at this the more confused I become.

Normally the value of  Cpf (Capacitor Parafeed?) would be calculated using both the load of the 300B and the input impedance of the amp the BeePree is connected to?  Somehow the load is in parallel with the input impedance of the amp it's driving (B+ is signal ground and old PhD EE used to tell me)? I'm guessing here.

But then we add a pot (also in parallel) which can be any value from 0 Ohms to 10K but...

The Cap always sees full 10K value of the pot to ground which is in parallel with the input impedance of the amp(and the load of the tube)?

So, rather than try to figure all of this out we just make the Cap large enough so that it falls out of the equation?

 

[Paul Birkeland]

The cap and pot form a high pass filter.  The presumption here is that the input impedance of the gear following the BeePre is sufficiently high, but even if you have a 10K load on the output of a stock base BeePre, 10uF is still big enough. 

 

[Ivan303]

C=10uF R=10kΩ

1.59Hz

If this is correct (and it may not be) do I really need to go that low, even if the BeePre is driving a sub?

 

[Paul Birkeland]

Yes, it's very low.  It's designed that way because the first upgrade drops the attenuator impedance a little bit (and at the same time dramatically increases the impedance of the 300B plate load).

 

[Ivan303]

Yes, it's very low.  It's designed that way because the first upgrade drops the attenuator impedance a little bit (and at the same time dramatically increases the impedance of the 300B plate load).

How much is 'a little bit'?  Plan to upgrade but might just upgrade the active load part and leave the pots as is.  Maybe.

Love having a balance control.

 

[Paul Birkeland]

The stock circuit has 3K plate loads and 10K level controls, the first upgrade increases the plate load to constant current sources and the level control is no longer a constant loading impedance but wanders a bit between about 3K and 7K.

 

[Ivan303]

The stock circuit has 3K plate loads and 10K level controls, the first upgrade increases the plate load to constant current sources and the level control is no longer a constant loading impedance but wanders a bit between about 3K and 7K.

Out of curiosity, how is a constant current source as a plate load calculated a load impedance?  Infinite load to AC? Fixed load to DC? For calculating the frequency cut-off of various capacitor sizes?

Is it kinda like a choke?

Would like to choose a smaller value Cap of a Teflon type but not to go so small as to not so low as to not be able to run a sub (3dB at 30Hz would be fine though).


 

 

[Paul Birkeland]

A choke is a reactive load that will vary in impedance rather substantially with frequency.  A constant current source doesn't typically wiggle around so much.  For the C4S, it's safe to assume a 500K+ impedance for the CCS.

-3dB at 30Hz means the preamp will start rolling off a 300Hz, which wouldn't be acceptable to us.  The 10uF ODAM caps are a popular choice.

 

[Thermioniclife]

I installed 10uf ODAMS in my Mainline and They transformed the amp drastically for the better imho.
all you have to do is use smelling salts after you faint when you pay for them.

 

[Ivan303]

  The 10uF ODAM caps are a popular choice.

Great suggestion!  Will do, thanks. Have used his small 'Teflon' (Fluoropolymer Film) caps in past builds with good results.

Ok, 8.2uf 250V as physical size(diameter) is just about the same as the Dayton and with an active load it's seeing a higher impedance even with the stepped pot and two load (amp and sub amp) so should be fine.

Either that or the 3.3 uf Fluoropolymer Film he has 'on sale' at $395 each.  :-\

 

[Ivan303]

Either that or the 3.3 uf Fluoropolymer Film he has 'on sale' at $395 each.  :-\

OK, math was never my big thing but... 

I would appear that, if the stepped pot/active load upgrade increases the load on the 300B for 3K by an order of magnitude, or thereabouts, and the stepped pots drops the volume control's load on the 300B by just half, then maybe I don't need a 10uF output cap?  And if I really want to keep the volume control/balance control as-is for now and only do the active load part of the upgrade, then a 3.3uF output cap should work as well as the 10Uf does in the stock circuit?

However...

I plan to leave things as-is, with no cap upgrades or other nonsense until AFTER I do the active load part of the 'sweet whispers' upgrade and see if it does anything for the 'hard' upper range I'm hearing (which might be more 'detail' or 'dynamic range'  than I am used to with the prior 6SN7 pre?).

Currently unning some NOS AVVT AV-5'Bs (low power 300B's with max 16 watt plate dispensation designed as driver tubes for BIG triodes but also used in  'Lapazator' DACs and such).  They sound very good but still, leading edge of piano notes (piano is a 'percussion' instrument) about knock one out of their listening chair.

http://www.elektronenroehren.de/avvt/datashts/av5b.htm

That said, if dynamic range in a tube pre-amp is something one needs, the BeePre2 certainly has it.  And the MonAmour monblocks are SO good that whatever the pre-amp sends is exactly what you get.  Plus EdgarHorns are not going to dampen dynamics much either.

But some if not all of the best custom built pre-amps I've heard use DHTs as output tube, albeit with BIG chokes loading them.  Here's hoping the active load part of the 'sweet whispers' acts as said BIG choke; one with a flat frequency response.  Gotta be better than a 3K sand cast resistor, I'm thinking.

 

[Paul Birkeland]

I would appear that, if the stepped pot/active load upgrade increases the load on the 300B for 3K by an order of magnitude, or thereabouts, and the stepped pots drops the volume control's load on the 300B by just half, then maybe I don't need a 10uF output cap?

You need the 10uF cap. 

And if I really want to keep the volume control/balance control as-is for now and only do the active load part of the upgrade, then a 3.3uF output cap should work as well as the 10Uf does in the stock circuit?

This is correct.  There is a way to retain the balance control with the stepped attenuators as well if you want to drill a hole into the wood to relocate the balance pot.

But some if not all of the best custom built pre-amps I've heard use DHTs as output tube, albeit with BIG chokes loading them.  Here's hoping the active load part of the 'sweet whispers' acts as said BIG choke; one with a flat frequency response.  Gotta be better than a 3K sand cast resistor, I'm thinking.

The circuit is not really suitable for a plate choke, and keep in mind that we stock several so it would have been minimal effort on our part to use one. The C4S load will behave approximately like a 6000H plate choke.

How does the preamp sound running an actual 300B?

 

[Ivan303]

How does the preamp sound running an actual 300B?

We'll leave the math for another day.

KR 300B is just too microphonic (was unusable in a 300B amp years ago for the same reason), very ZINGY.

EH shipped with the kit is thankfully not microphonic (many 300b;s are) but really needs more break-in.  Ran for about 6 hr. a day for a week is all.

WE 300B (Charlie Whitner's late 90's WE re-issue version with some miles on them).  Probably best sounding of all (as it should be).

Excepting the micropohic KR 300bs, not all that much differences between tubes.

No noise now as the DAC I am using has a ton more output than the DENAFRIPS Terminator for some reason, so the volume control is VERY low and no hum is heard at all with no music playing unless you stick your head into the horns and even then, every little.

As regards plate chokes vs. active loads, most custom builders aren't quite up to designing their own active loading circuits so for those folks a choke just has to do.   

The 'Sweet Whispers' are on order and look to be a pretty straight-forward upgrade so will know soon.