New Build 5/2025

[martin manning]

Using the formula Martin suggested from Merlin’s website, I got this:
with 5U4GB & 6F6s
Plate voltage 311 Plate to cathode 296
2 x (296 - 50)^2 = 60, 516 % 9200 = 13 watts

If this amp sounds good as it is, then this is of academic interest only.

Fooling around with the 6F6 data sheet and scaling things (data sheet is at 250V Vg2, where you are at 232) I believe your 9k2 load line is crossing the Vg1=0 grid curve at around 100V, and you are only getting around 6.5W output power. So much of the load line is Class A (both tubes conducting) that the simplified equation above isn't accurate. A scope could confirm this.

In theory, at the same voltages, a 12k load line would pass through the Vg1=0 knee instead of above it, and the output power would go up to 7.7W.

[johnnyreece]

Look for an old Akaia/Roberts R2R carcase on eBay. I found this one for cheap. It is just what you need...

Funny you mention this...I just fixed an old Akai R2R for my wife's aunt and uncle! Just waiting for the new pinch roller to arrive before calling it done.

[WRC34]

Martin, you said:

“Fooling around with the 6F6 data sheet and scaling things (data sheet is at 250V Vg2, where you are at 232) I believe your 9k2 load line is crossing the Vg1=0 grid curve at around 100V, and you are only getting around 6.5W output power. In theory, at the same voltages, a 12k load line would pass through the Vg1=0 knee instead of above it, and the output power would go up to 7.7W.”


This is interesting. I have always been under the impression the higher the impedance the lower the output power, all else being the same. But you are saying if somehow the reflected impedance were to be increases from 9.2K to 12K the output would increase!

[martin manning]

I have always been under the impression the higher the impedance the lower the output power, all else being the same. But you are saying if somehow the reflected impedance were to be increases from 9.2K to 12K the output would increase!

Power output is about the voltage and current swing you can get when you hit Vg1 = 0. If the load line is above the knee, you are getting less than if it went through it. Below is where I see your load line, scaling your 248 - 16 = 232V Vg2 to match the data sheet's 250V.

[pdf64]

... I have always been under the impression the higher the impedance the lower the output power, all else being the same.

That's pretty much the deal for solid state amps. They have very low output impedance, very low loss in the output devices such that the output voltage signal can swing almost rail to rail (of the power supply).
Typically, the main limitation is how much current can the output devices cope with. So they are ok with load impedances above a certain level. As the load impedance is reduced below that, the power output will carry on increasing, but the output devices are at risk of overheating.

But you are saying if somehow the reflected impedance were to be increases from 9.2K to 12K the output would increase!

Valves are very different, the output pentodes we use have a high output impedance, to devise the best loading for them, a loadline can be drawn on their anode curve.
The maximum power output, at the minimum power wasted in the valve, will occur with a 'through the knee' loadline.