Coupling cap voltage rating

[Klimon]

Hi all,

I'm building a C3G cathode follower headphone amp for my father and am a little uncertain about the required min. voltage rating for the output caps. The designer writes that it's not a luxury to use caps with 2* the required AC voltage rating, he uses 660V AC caps; I've got some good sounding MKV's "420Veff, 600Vn", both DC. Would these be adequate for this particular amp (http://www.jogis-roehrenbude.de/Leserbriefe/Siemens-C3g/c3g.htm)?

What's the formula to calculate voltage requirements for coupling caps? Is it the same game for anode and cathode followers; I never thought twice about the voltage rating of anode follower coupling caps and had no fatal accidents yet...

Thanks a lot,

Simon

 

[Tube_Dude]

Hi all,

I'm building a C3G cathode follower headphone amp for my father and am a little uncertain about the required min. voltage rating for the output caps. The designer writes that it's not a luxury to use caps with 2* the required AC voltage rating, he uses 660V AC caps; I've got some good sounding MKV's "420Veff, 600Vn", both DC.

Hi Klimon

First of all , it's not a cathode follower but a common cathode amplifier , with a pentode in triode mode.
If it was a cathode follower , the output signal must come from the cathode and not from the anode, as in this case...

About the voltage rating , your 600 Volt capacitor is perfectly adequate , because C8 and C9 that decouple the high voltage rail are only 350 Volts rated .

With 600 volts DC your capacitor is adequate by a large margin...

Cheers

 

[fdegrove]

Re: Re: Coupling cap voltage rating

Hi Klimon

First of all , it's not a cathode follower but a common cathode amplifier , with a pentode in triode mode.
If it was a cathode follower , the output signal must come from the cathode and not from the anode, as in this case...

About the voltage rating , your 600 Volt capacitor is perfectly adequate , because C8 and C9 that decouple the high voltage rail are only 350 Volts rated .

With 600 volts DC your capacitor is adequate by a large margin...

Cheers

Hi,

Probably....
I didn't take a look at the design but I'd like to add that it's NOT because the B+ rail is only 350VDC that AC rating of the coupling caps would only have to be that voltage.
While going from 350VDC to 660VAC seems like an enormous voltage swing it's still not impossible. Hence my remark.

Cheers,

 

[Tube_Dude]

Re: Re: Re: Coupling cap voltage rating

While going from 350VDC to 660VAC seems like an enormous voltage swing it's still not impossible.

Hi

Not in this case,a voltage swing greater than the rail voltage is only possible in case of a choke anode load or a bootstraped resistor load ,neither exist in the present case...

 

[Svein_B]

I try to select coupling caps that are dimensioned for the max DC voltage they may encounter, which is normally at startup, until the tubes start drawing any current.

(but I have also on occasion ignored my own advice, since I recon there is usually some safety margin)


SveinB.

 

[Klimon]

Thanks for your replies, I'll take my chances with the 600v caps. PIO's aren't the most fragile caps so I'll probably have more than enough margin.

Simon

 

[kevinkr]

I have built a number of White CF based and other headphone amplifiers and the one thing no one has mentioned so far is that in a cap coupled design like this the other side of the cap is sitting at 1/2VS - if that capacitor ever fails for any reason some level of high voltage will be present on the headphone drivers (possibly whilst on your head) which in some circles could be considered hazardous.

Presumably a cap failure in use could result in some exciting noises in the headphones, followed by the transducer vc possibly opening up due to fault current, at which point a shock hazard might exist depending on the headphones insulation and its ability to stand off as much as 300V.

IMO The 6.8K plate load resistor will probably prevent sufficient current from flowing to open the vc.

Caveat Emptor! I'd recommend transformer coupling as a much safer alternative, parafeed or normal SE would be suitable.

As always YMMV, I just wanted to mention the potential safety issue in any design relying on a capacitor to block dc to a pair of headphones.

 

[Brian Beck]

Of course, in that case the headphones act as a handy fuse to protect the output tubes.

There can also be pretty dramatic turn-on and turn-off thumps through the headphones with cap coupling.

Seriously, as tempting as direct coupling is for headphone amps, this is also a place where an OPT can be used to especially good effect. I use oversized OPTs with a push-pull pair of class-A 7119s for my headphones. The low plate Z insures very deep bass response with little bass distortion.

 

[Tube_Dude]

For double security,we can use a capacitor feeding a transformer... just in case!

 

[Klimon]

Well.... I'll take the leap and hope my father doesn't get a hell of a surprise one day amidst airy pastoral soundscapes.......... An extra pair of trafo's is simply above my budget and considering cap-coupling is quite commonly used with headphone-amps + pio's aren't too fragile I'm not too scared for the big bang. Thanks again.

Simon