• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Push-pull load lines for GU-50-Triode - did I do this right?

After a recent thread about load lines for push-pull tube amps, I decided to try matching triode-wired GU-50s to a couple of OPTs I have lying around. One has 8k ohm (plate-to-plate) primary, the other about 5k ohms. Check my work? Did I do this right?

GU-50_Triode PP 8k and 5k OPTs compared.jpg

I chose Vp-k of 380V and Ip of about 80mA (Vg1 of -60V) because that's a reachable set of operating points with power transformers I have available.

What would be the disadvantages to using an OPT with 5k pri? Are there significant advantages to using an OPT with 8k pri?
What you did is not correct. There is indeed a change from 4k to 2k loadline (and from 2k5 to 1k25), but that happens when the other tube shuts off, so not around the idle point as you did now. Do you know this site?
There you can see where the loadline changes.

As an example for a 5k load line with 380V and 80mA as the idle point (left part of picutre). Note that the two light blue lines have the same length: while the voltage on the plate of valve 1 goes down, voltage at plate 2 goes up, and both are conducting (class A area). At circa 570V the valve 2 stops conducting, the amp goes into class B, reducing the load on valve from 2k5 to 1k25, shown by the steeper angle.

Using 8k with the same operation points is shown on the right side. Valve 2 can now swing up to 680V while still conducting, so this ensures a much bigger voltage swing in class A. Only at circa 50V on the plate of valve 1 one sees the change to class B. The calculated ouput also shows this: 5k gives zou 16W class A, while 8k gives you 25.6W. 5k gives you higher AB power though (51W vs 34W).

Hope this helps.


  • loadlineGU50.png
    48.6 KB · Views: 37
Last edited:
I see that I was sliding both the Class A and the Class B lines across the graph, which is incorrect.
If I'm reading the Valve Wizard instructions correctly, the Class B line is drawn as Va divided by one-quarter of total primary impedance of the OPT, then the Class A line is calculated from 1/2 the OPT primary impedance and slid horizontally until that line intersects the bias point. Then you draw a composite line from the origin of the Class B line on the y axis (plate current) to its junction with the Class A line, then continuing on the Class A line through the bias point down to the x axis (plate voltage). That resulted in these:


Still incorrect?

The bias point doesn't agree with the results from vtadiy.com, but I figure they modeled a GU-50 that doesn't match the bogey device in the datasheet curves shown above. Otherwise they seem pretty close.

8k primary, Vp = 380V, Ip = 75mA


5k primary, Vp = 380V, Ip = 75mA


Both 8k pri and 5k pri look like they'd work well for push-pull GU-50-Triode. No?
Last edited:
You can also use the VTADIY simulator to predict the level of harmonic distortion, 2nd, 3rd and 4th. Use the Universal vs. the Power Stage calculator. When you use the output headroom value it will add the distortion prediction. It can be surprising how it moves around with the operating point. Quite often the pentode prediction will be much lower that the same tube in triode although you would have to use some feedback, local or global, to get the output impedance lowered for pentode or UL although I wouldn't believe it without actually measuring. I have plans for using some GU50's but will use pentode with local feedback. Better power efficiency and very likely lower distortion.

Good luck with project,