• 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.

GU-46 push-pull & single-ended amplifiers

Jump to Latest
#83
Bandersnatch said:
On that load line, for what ever its g2 voltage is( 5 or 600V IIRC ), it should be reduced so the g1=0V line happens at that current delivery. I don't quite see this GU46 running U-L. Local FB pentode will deliver the required output impedance...which is a fine quantity to reduce...🙂
cheers,
Douglas
Click to expand...
Eg2 was 600V in the Ep-Ip chart (post 21). Below is the Ep-Ip characteristic with Eg2=450V. The operating conditions are Ea0=1000V, Eg1=-34V, and Ra-a=5k. The Spice model isn't that accurate, so please ignore the saturation region, instead use the diode line shown in green to get the estimates for Ia_max and Ea_min, in this case, they are 720mA and 200V respectively, so Po is still around 280W as calculated before.

GU-46 Class A_450V.jpeg
 
#92
The Zungunder is not what I would considered hi-fi, it's just clobbered together with whatever parts he had on hand, I think you can do better. Why not just start with the ClassZ SE design, which is quite straight forward.
 
#93
jazbo8 said:
Yup, this is one expensive space heater. :snowman:

On a more serious note, below is the Ep-Ip characteristic with an even lower Eg2 of 300V. The operating conditions are: Ep0=1000V, Eg1=-20V, Ra-a=10k, with a Po of "just" 170W.

View attachment 651264
Click to expand...

If you bump the load 2x, you notice it can swing a lot higher delta towards cut off than towards saturation. If you must run a 10k a-a, might as well bump B+ to 1k5V...🙂 Then the 300mA idle current will net higher anode dissipation.

A good 10k a-a is difficult around 60W...not much chance of success going bigger.
cheers,
Douglas
 
#95
Interesting, the SE project does not seem to draw as much attention as a high power output push pull approach does.

This would be my first SE project, so I would stick to ClassZ's approach, as proposed by Jazbo.


Such a SE OPT will also be a challenge, will it? Any important points to consider / take in account when going for such a big SE OPT?
Thomas
 
#96
The OPT is going to be a problem since the output power is quite high (by SE standard), ClassZ's design used a Hammond 5k but it was just for 30W, not sure you would want to repeat that given all the effort and costs put into the project. Although his driver circuit could be adapted for your needs.
 
#97
Well, I am currently trying to find some transformer builders to find out where they see their limitations for OPTs.
I think it could be an alternative to use two paralleled OPTs instead of one, wouldn't it?

This would half the required wattage and ampere they need to handle, probably optimize mids and highs, and be 10k in instead of 5k?
Does this work for pp and se?

Lets forget about weight at the moment...
 
Last edited:
#98
Yes, paralleling OPT’s is a viable option, but you do not need to double the primary impedance, since putting two OPT’s in parallel does not change the turns ratio. For off-the-shelf OPT, the Hammond 1642SE may work, it can handle 300mA, and has 4-8-16 taps, so you can get the load matching required. In the above example, you can put the 8R speaker on the 16R tap to get a primary impedance of 2.5k, which is close enough.
 
#100
I got the first offer for SE OPTs

- 2.5k primary, 4 + 8R secondary, 170W, 1200V at 600mA, Ei 150/92 core, 14kg each, approx. 360Euros pp
- 5k primary, 2 + 4R secondary, 80W, 1200V at 300mA, Ei 120/73 core, 7kg each, approx. 180Euros, to be used in parallel

Custom wound, an additional winder did not yet call back.
 
You must log in or register to reply here.
Top Bottom