Power supply issues with Borbely style 12AU7/MOSFET hybrid

[jjbunn]

I'm building a tube/MOSFET hybrid power amplifier based on a circuit in AudioXpress by Cozza (5/01) which in turn is based on a design by Borbely using low voltage tube operation.

The schematic of the built circuit is shown below:



(The original circuit uses 6DJ8 and IRF9540/IRF540, but I'm using 12AU7 and IRF9140/IRFP140 because they are what I have to hand.)

My problem is very simple, and related to the 45/0/45 power supply. With R2 out of circuit the supply provides 45/0/45 as measured. I can adjust the plate currents in the 12AU7 over the range of a few milliAmps using R14, as expected.

However, with R2 connected, the supply drops to about 10/0/10 :-( My (limited) understanding of the circuit is that the power MOSFETs should run a quiescent current of an amp or so through R2 and R3, so it looks like my power supply can't handle the current - but why not?

The transformer TR1 is a very beefy and heavy unit that came out of vintage HP test gear. B1 is a heavy duty bridge rectifier rated at up to 25 Amps.

I'd welcome some advice on what may be wrong.

Thanks!

[M Gregg]

Put your multimeter,

In place of R2 BUT fit a fuse of 5A in series DC current 10A range.. On second thoughts just try the fuse and R2..current might be to high for the meter.

I bet its off the scale or blows the 5A fuse..

What voltage have you got across D1 zener?

What is the voltage from R6 to gnd with the tubes cold and hot?

Does the fault happen if you let the tubes run (get warm) without r2 in place and then connect R2 with fuse?
(might be a good idea to fit a 5A breaker in series with R2 when testing )

The connection by R16 is connected to PE power supply? (not missed out?)

Regards
M. Gregg

[jjbunn]

Originally Posted by M Gregg

Put your multimeter,

In place of R2 BUT fit a fuse of 5A in series DC current 10A range..

I bet its off the scale or blows the 5A fuse..

What voltage have you got across D1 zener?

Regards
M. Gregg

Thanks - I'll measure the current through the MOSFETs, as you suggest. I measured the voltage across D1 at 5V, - I'm pretty sure with R2 in circuit, but will check.

Do you think Q1 may be faulty?

[jjbunn]

Originally Posted by M Gregg

Does the fault happen if you let the tubes run (get warm) without r2 in place and then connect R2?
(might be a good idea to fit a 5A breaker in series with R2 when testing )

Regards
M. Gregg

Yes, I tried letting the tube warm up, measuring the power supply voltage, then connecting R2 - the voltage immediately drops to 10/0/10 from 45/0/45.

[kevinkr]

JJ, when you figure this out I think you might want to check out the 6DJ8 as it's performance on low plate voltages IMLE is generally much better than the 12AU7.

[jjbunn]

Originally Posted by kevinkr

JJ, when you figure this out I think you might want to check out the 6DJ8 as it's performance on low plate voltages IMLE is generally much better than the 12AU7.

I started with a 6DJ8 but found the heater current a bit too much for the low voltage supply I have, so switched to the 12AU7, which has the advantage of lower current, and I have a whole bunch of them! (I sold several 6DJ8/ECC88 about a year ago, in a moment of stupidity, so only have one left). My intention, once the circuit is operating correctly, is to try also 12AT7 and 12AX7 to see what has the best linearity at these mA currents and low plate voltages.

[M Gregg]

Originally Posted by jjbunn

I started with a 6DJ8 but found the heater current a bit too much for the low voltage supply I have, so switched to the 12AU7, which has the advantage of lower current, and I have a whole bunch of them! (I sold several 6DJ8/ECC88 about a year ago, in a moment of stupidity, so only have one left). My intention, once the circuit is operating correctly, is to try also 12AT7 and 12AX7 to see what has the best linearity at these mA currents and low plate voltages.

You can run the ECC88 heater with your supply its the cold heter current that is the problem..use a soft start and they will work,,(in series)

[M Gregg]

Q2 must be ok,

Its ony the bottom half thats the problem..

[M Gregg]

Measure R3 current.. without R2

It could be a short mosfet in the lower half just make sure you have correct voltage on the gate resistor..

[M Gregg]

Here is a thought..

turn Q1 off IE disconnect the gate and connect to gnd via resistor..if you can't turn it off the its stuffed..In this case its to the - rail..not gnd

Q2 would be output..