If you had a set of tubes in push-pull with a load impedance of 4KP-P, if you kept all voltages the same and paralleled another set of tubes, would the load impedance then drop to 2KP-P?
Thanks,
Ray
Thanks,
Ray
Short answer is yes for class A, class AB1 and AB2.
I'm not sure about class B but Isuspect it follows the same math.
I'm not sure about class B but Isuspect it follows the same math.
Load impedance will be the same (it depends on what is connected to the secondary), but load of tubes decreases (resistance that an anode sees will increase twice), so distortions will go down significantly. My first Pyramid was built such a way:

You may mean optimum load impedance i.e. the best load impedance to use, the one which best balances output and distortion. This halves when you put two valves in parallel.
The actual load impedance is set by the output load and the transformer turns ratio, so stays the same until you change the transformer.
So it all depends on what you mean by load impedance. I assume you meant the first one above.
The actual load impedance is set by the output load and the transformer turns ratio, so stays the same until you change the transformer.
So it all depends on what you mean by load impedance. I assume you meant the first one above.
Load impedance will be the same (it depends on what is connected to the secondary), but load of tubes decreases (resistance that an anode sees will increase twice), so distortions will go down significantly. My first Pyramid was built such a way:
Seems to me a tube has a sweet spot for lowest distortion. If the output transformer reflected impedance takes the tube load out of this sweet spot (halving the impedance) the distortion can go up.
As an example the 6v6 likes to see about 6000 ohms. drop the load (increase the impedance to the tube by 2X) and the distortion almost doubles.
See last page.
http://tdsl.duncanamps.com/link.php?target=0125C507
Seems to me a tube has a sweet spot for lowest distortion.
It seems to me a sweet spot in terms of load impedance is the optimal ratio between distortions and electrical efficiency. I agree, I paid for increased sound quality by decreased efficiency. More parts, more heat. So what?
After that I've found better tubes to go with than ancient 6L6, and now I use only a pair of GU-50 for the same sweet pristine clean sound at twice higher output power.
By the way, the current development, Pyramid-9 monoblock, uses 4 of GU-50, but for a different reason: twice higher power (160W compressor limited), while 6L6 were paralleled for 4 times lower distortions on 40W only output power. Also, screen grids in all cases are powered from significantly lower voltage regulated sources than anodes.
Last edited:
Wavebourn, what schematic are you using for the two GU-50?
I posted all versions several times here. It is called Pyramid. The latest was Pyramid-VII-M, I posted both phase splitter and output stage schematics. However I did not post complete schematics because 800V B+ is dangerous for novices.
However I did not post complete schematics because 800V B+ is dangerous for novices.
Same situation I have with my 500W Williamson; the 1A 600V programmable SS powersupply is not a bench lash-up; out of the question too !
richy
- Status
- Not open for further replies.
- Home
- Amplifiers
- Tubes / Valves
- Push-Pull vs Push-Pull-Parallel