Pye Black Box valve amp circuit - UK Vintage Radio Repair and Restoration Discussion Forum

martin.m · 21st Aug 2018, 1:58 am

I have one of these amps, the version with a pair of EL42s running in Ultra Linear mode. It is working fine after changing the Hunts capacitors and the ECC83's 220k anode load resistor, but I am slightly puzzled by the output circuit. The EL42 cathodes are connected together and returned to HT negative via a 470 ohm resistor that has no bypass capacitor. Is this for negative feedback? Also wouldn't there be mixing of the two out of phase audio signals as well? It sounds fine so I will leave it alone but I am just curious.

GrimJosef · 21st Aug 2018, 9:10 am

In a perfectly balanced push-pull circuit, as long as the amp stays in Class A the total current flowing through the cathode resistor will be pretty much DC since the current in one valve rises at just the same rate as the current in the other valve is falling. With only DC through the resistor there would be no current flowing through any capacitor across it. So the capacitor might as well not be there. Once you get out of Class A things get more complex and then the absence of a capacitor effectively provides some local feedback. In most amps, including the Black Box one, there is also a global feedback loop and the interaction between the local and global contributions can be a bit difficult to unpick when it comes to an overall analysis.

If the amp happens to be unbalanced, say because one EL42 has a higher gain than the other one, an out-of-balance signal will appear at the common cathode and the resulting feedback will tend to suppress it.

Cheers,

GJ

martin.m · 21st Aug 2018, 12:39 pm

Thank you GJ. I understand now. After 45 years my knowledge of valve theory is a little rusty.

Regards
Martin

kalee20 · 21st Aug 2018, 12:56 pm

It's really elegant, the common cathode resistor - unbypassed. Saves a capacitor. In fact, biasing the valves separately would use 2 resistors and 2 capacitors, so we're 3 components to the good!

As GJ says, if one valve has higher gain, then it will give an unbalance voltage across the resistor which gives a bit of cathode drive to the other valve, restoring things.

Unfortunately it does not help DC conditions. If one valve takes a bit more standing current, it gives a higher drop across the cathode resistor which turns the other valve OFF a bit more. So valves have to be well-matched.

For this reason, I personally prefer separate resistors. They can be connected together from an AC point of view (a few hundred microfarads, non-polarised) such as a pair of back-to-back electrolytics, negative terminals together and taken to chassis via 10kilohm to polarise them) or just bypassed in the normal way.

21st Aug 2018, 1:58 am	#1
martin.m Hexode Join Date: Mar 2018 Location: Bishop Auckland, County Durham, UK. Posts: 373	Pye Black Box valve amp circuit I have one of these amps, the version with a pair of EL42s running in Ultra Linear mode. It is working fine after changing the Hunts capacitors and the ECC83's 220k anode load resistor, but I am slightly puzzled by the output circuit. The EL42 cathodes are connected together and returned to HT negative via a 470 ohm resistor that has no bypass capacitor. Is this for negative feedback? Also wouldn't there be mixing of the two out of phase audio signals as well? It sounds fine so I will leave it alone but I am just curious. __________________ Regards Martin

21st Aug 2018, 9:10 am	#2
GrimJosef Dekatron Join Date: Sep 2007 Location: Oxfordshire, UK. Posts: 4,311	Re: Pye Black Box valve amp circuit In a perfectly balanced push-pull circuit, as long as the amp stays in Class A the total current flowing through the cathode resistor will be pretty much DC since the current in one valve rises at just the same rate as the current in the other valve is falling. With only DC through the resistor there would be no current flowing through any capacitor across it. So the capacitor might as well not be there. Once you get out of Class A things get more complex and then the absence of a capacitor effectively provides some local feedback. In most amps, including the Black Box one, there is also a global feedback loop and the interaction between the local and global contributions can be a bit difficult to unpick when it comes to an overall analysis. If the amp happens to be unbalanced, say because one EL42 has a higher gain than the other one, an out-of-balance signal will appear at the common cathode and the resulting feedback will tend to suppress it. Cheers, GJ __________________ http://www.ampregen.com

21st Aug 2018, 12:39 pm	#3
martin.m Hexode Join Date: Mar 2018 Location: Bishop Auckland, County Durham, UK. Posts: 373	Re: Pye Black Box valve amp circuit Thank you GJ. I understand now. After 45 years my knowledge of valve theory is a little rusty. Regards Martin

21st Aug 2018, 12:56 pm	#4
kalee20 Dekatron Join Date: Feb 2007 Location: Lynton, N. Devon, UK. Posts: 7,088	Re: Pye Black Box valve amp circuit It's really elegant, the common cathode resistor - unbypassed. Saves a capacitor. In fact, biasing the valves separately would use 2 resistors and 2 capacitors, so we're 3 components to the good! As GJ says, if one valve has higher gain, then it will give an unbalance voltage across the resistor which gives a bit of cathode drive to the other valve, restoring things. Unfortunately it does not help DC conditions. If one valve takes a bit more standing current, it gives a higher drop across the cathode resistor which turns the other valve OFF a bit more. So valves have to be well-matched. For this reason, I personally prefer separate resistors. They can be connected together from an AC point of view (a few hundred microfarads, non-polarised) such as a pair of back-to-back electrolytics, negative terminals together and taken to chassis via 10kilohm to polarise them) or just bypassed in the normal way.