Understanding how various capacitor types influence different SET circuits

Downhome Upstate · 7468

0 Members and 1 Guest are viewing this topic.

Offline Downhome Upstate

  • Sr. Member
  • ****
    • Posts: 289
Hi all,

Can someone point me to one or more well-written sources designed for the lay reader and describing why certain capacitor types are better used in certain audio (SET) circuit positions than others? 

By 'better' I mean producing a result (whether in regard to power supply filtering, signal coupling or otherwise) that aids the designer's goal(s) for the circuit and for the sound of the device (or its contribution to the sound of the system).

I am that lay person facing the usual steep learning curve but eager to learn. Thanks in advance for any help you can offer!

Mike

"Too soon old;
   Too late schmart"

    The late Mr. Fox, Fox's Deli, Rochester, NY

Mike P.


Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19751
There are some very long, technical white papers here and there that attempt to take a look at what you're referring to, but they are very dense reading and will make your eyes bulge by the second paragraph.

The essential influence boils down to the image below:

(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Frichardsears1.files.wordpress.com%2F2011%2F06%2Ffigure-4-single-ended-signal-loops-with-ccr.jpg&hash=277a03c22f8d33b74ca02b2d95de2c9224674779)

The image shows a conventional single ended voltage amplifier.  This is similar to what's in most of the commercial amps out there, and also is somewhat applicable to a basic driver stage as well.  The red dashed lines and green dashed lines represent your AC signal current.  The green line shows the input signal current, and the red line the resulting output.  In each case, there is a capacitor (or two) intersecting with those loops (the capacitors are denoted C1/C2).

Where the signal current passes through a capacitor, the quality of the capacitor will become very important.  In Bottlehead kits, Parallel Feed topology is employed, and one positive outcome is that the size of the final capacitor for a power amp shrinks considerably, allowing for a higher quality cap to be used (see left side image below).
(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Frichardsears1.files.wordpress.com%2F2011%2F06%2Ffigure-5-se-para-and-direct-feed-with-ccr.jpg&hash=5ef815b3ec5bbbfe5a70b3671e1e965beb1faa86)

Additionally, there are often two amplification stages stuck together, and often times you will see a capacitor between them to block DC voltage, this is referred to as a coupling capacitor, and occupies a similar position as the capacitor in a parallel feed arrangement.

As far as why certain capacitors sound better than others, that becomes a hard question to answer.  The preference for film type capacitors over electrolytic capacitors can be boiled down to chemistry and measurement, but what makes one metalized polypropylene capacitor sound better than another metalized polypropylene capacitor can be quite elusive.  For more reading on this, you can check out this capacitor shootout: http://www.laventure.net/tourist/caps.htm

From my personal perspective, I shoot for film and foil capacitors whenever possible (imagine long stacks of alternating metal foil and plastic film rolled up and stuffed into a tube).  Next down the list are the metalized capacitors (imagine depositing a very thin layer of metal onto a layer of plastic, then rolling that up), with polypropylene as the best film material.  Last up would be electrolytic capacitors, though there are times where they must be used, and certain electrolytic capacitors exist that perform quite well.

Feel free to request more specific info.  Hopefully this thread doesn't become "my favorite capacitor is XYZ".

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man


Offline Downhome Upstate

  • Sr. Member
  • ****
    • Posts: 289
Thanks, Paul.  I'll have a look at that capacitor shoot-out tonight.

Best,

Mike

"Too soon old;
   Too late schmart"

    The late Mr. Fox, Fox's Deli, Rochester, NY

Mike P.


Offline Bill Chase

  • Jr. Member
  • **
    • Posts: 25
PB,
   Thanks for stripping away the eye-bulge material!  It's really helpful for the way I learn to understand the key difference, then learn the more subtle pieces.  This one "clicks" for me-- much appreciated!
Cheers,
Bill

Bill Chase
Digital music collection enjoyed thru:
   Mac Mini to Jolida Glass DSD DAC
   BeePre & Stereomour I
   Seductor EL84 Amp
   DecWare Ultra Preamp w 5x Paramount 300Bs with Magnequest Iron
   Crack with Speedball
   SEX v1
Guitars and Ukulele enjoyed by Tode


Offline Downhome Upstate

  • Sr. Member
  • ****
    • Posts: 289
Paul,

I don't understand how parallel feed topology differs from the direct circuit arrangement, or how it affects capacitor performance. Can you explain?

Thanks

Mike

"Too soon old;
   Too late schmart"

    The late Mr. Fox, Fox's Deli, Rochester, NY

Mike P.


Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19751

I don't understand how parallel feed topology differs from the direct circuit arrangement, or how it affects capacitor performance. Can you explain?

Sure, have a look at the top image to see the direct feed AC current loop in red.  Notice that it passes through C1, and that C1 is the final capacitor in the power supply.  Based on where this capacitor is, it generally ends up needing to be quite large in value (47uf to 100uf generally in our products) and a 47uF 630V film capacitor is one pricey part, at least $20 per cap.  You don't have to look too hard on other forums to see folks spending $50+ per capacitor on coupling capacitors, but not touching the final capacitor in the power supply (sad).

In the Parallel Feed image, the red current loop passes through the parallel feed capacitor, which is generally at least an order of magnitude (if not closer to two) smaller than the last capacitor in the power supply.  Take the SEX amp, for example, which has a 100uF/450V capacitor in the power supply before the output stage, but a 1.5uF/630V film parallel feed capcitor. 

What isn't shown on in the image I posted is the plate choke, which assists us in keeping that signal current away from the last cap in the power supply. 

There are also some conditions where the Parallel Feed capacitor will begin approaching the value of the last capacitor in the power supply, and at that point the design has to be reconsidered (I ran into this when building my Cold War amplifier).

-PB

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man


Offline Grainger49

  • Hero Member
  • *****
    • Posts: 7175
Mike,

FYI, the Cold War Amp is the one in Paul's avatar.  It uses 6C33C Russian Military surplus output tubes.  They are large and Paul should post a link to the pictures of it.  It is a thing of beauty.
« Last Edit: October 24, 2012, 11:18:53 AM by Grainger49 »



Offline RPMac

  • Full Member
  • ***
    • Posts: 220
Paul,

In the parafeed image, you don't show a cathode bypass cap. Is that because returning the parafeed cap to the cathode takes it out of the output signal loop or is the cathode bypass cap not needed? Does returning the parafeed cap to the cathode change the function or value of the cathode bypass cap?

Originally, my parafeed amps returned the parafeed cap to ground. I now have them wired as you show.



Offline Grainger49

  • Hero Member
  • *****
    • Posts: 7175
Robert,

Good eye, good question.  I have the above cathode resistor return on my MQ transformers on my Paramours.  It would be nice if the cathode bypass cap were not necessary.  It is the ONLY cap in them I have not upgraded.



Offline Paul Joppa

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 5833
The bypass cap is needed. At very low frequencies, the plate choke impedance is not high relative to the cathode resistor, so the current loop is no longer confined to the tube/parafeed cap/OPT. In practice, removing the bypass cap causes the bass to roll off around 50Hz (depending on the specific component values).

Paul Joppa


Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19751
Yeah, I borrowed that drawing from an open website.  The plate choke and the bypass cap are missing. 

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man


Offline 2wo

  • Hero Member
  • *****
    • Posts: 1261
  • Test
Here is something interesting from Tomas Mayer's Vinylsavor. The output transformer is a conventional SE unit. He calls it "ultrapath" but except for the connection of the 60H choke, it looks a lot like parafeed with the cathode connected cap. I have been trying to redraw it in my head but it doesn't look to me like a true parallel connection?   

John S.


Offline Paul Birkeland

  • Global Moderator
  • Hero Member
  • *****
    • Posts: 19751
It's not typical parallel feed since current flows through the primary of the output transformer.  Ultrapath generally would refer to taking the cathode bypass capacitor and connecting it between the cathode and B+ instead of the cathode and ground.  That drawing from Mayer (who I certainly respect) looks a lot like an LC stage with the C grounded at the cathode feeding a conventional SE amp.

There's some good reading on this topic here: http://www.tubecad.com/2008/08/blog0147.htm

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man


Offline 2wo

  • Hero Member
  • *****
    • Posts: 1261
  • Test
"It's not typical parallel feed since current flows through the primary of the output transformer."

Ah, click.

Thanks...John

John S.


Offline RPMac

  • Full Member
  • ***
    • Posts: 220
PJ,

If I understand, even though the cathode bypass cap is out of the output signal loop, it still has an effect on the performance of the output signal loop, namely the output tube?
Back to the OP's original question, with it out of the signal loop, does it matter if the bypass cap is electrolytic or film? And will it still need to be of large value for the low hz?