A guide to building the Pass F4 amplifier

Some data and a question. With amp thoroughly warmed, DC offset at 0mV,and bias set to ~270mV, heatsink temps on the 4u chassis are 131F (55C)... This is about right ( ie max)... Correct?

Btw... For any considering a build, would suggest you swap 10K resistor at R9 for 4.7K as 6L6 suggests in build guide. With 10K, I could only get bias to 65mV cold. Reworking that part after everything is assembled is a touch tricky... Or so I found anyway. The 4.7K provided plenty of adjustment.

Bryan
 
My F4 project with pics


Hi Guys,

Just posting my F4 project with pics from the early stages to the final F4.


I decided to build a dual mono F4 - this was my first DIY project ever. Had so much fun doing it!! :)



My dual PSU boards mounted to the 5U chassis front

IMG-2816.jpg

— —

My dual transformers. (these were 400VA 230V to 2 x 24V)
I later discovered that I was going over board with the 24V. I ended up returning the transformers and going for 200VA 230V to 2 x 18V - as I’ll be using two transformers, 200VA is fine. Thanks to 6L6 for the tips :)





— —

My mounted F4 right board

IMG-2807.jpg

— —

Almost wired up

IMG-2831.jpg


— —

Close up of the right PSU wiring


Final wiring - this is where I got stuck. I’m from the UK so we use 230V the guide only had the 115V wiring with picture.

6L6 helped me out here, with a hand drawn diagram (it was a great help!) :)

IMG-2844.jpg


— —


Testing

So I have replaced the transformers and successfully completed the final PSU wiring.

I tested the PSU boards first and all was working, so I connected the F4 boards.


IMG-2846.jpg


— —


I had to replace my JFETs and MOSFETs for matched items, this is something I didn’t do at first and it was giving me problems.

Ordered my matched JFETs and MOSFETs - and managed to complete this F4. Its sounds amazing!!

It runs at 300mV bias and heatsinks get to around 45C average. It only went higher on the hottest day of the year but it still only got to 50C.

Just need a good preamp now! (I’m currently building the Pumpkin Shunty)

See final finished pics below.


IMG-2869.jpg



IMG-2878.jpg



IMG-0375.jpg



IMG-0382.jpg



IMG-0386.jpg


IMG-0393.jpg


IMG-0393.jpg
 
Last edited by a moderator:
hi guys, finished power supply and now working properly. My 22 volt transformer is giving 31.3v dc, (AC is quite high in this part of Switzerland, around 226 volts AC). So with this output DC, bias should be set at less than .13 volt cold and .2 volt hot for standard 18v transformer, what would you recommend as a good starting point for bias with this voltage? .11 volt cold and not let it get above .17 volt at source resistors when hot would be an appropriate target or less?

Chassis is similar to the 4 model in the diy store, and weighs about 12 kg and each heatsink is 38 cm x 16cm x 3.2cm.

Everything is ready, just have to solder in transistors and put on heatsinks probably on Wednesday.

Thanks, John.
 

6L6

Moderator
Joined 2010
Paid Member
I think your target bias is a great starting place. See how that works, and increase a little at a time until whichever you hit first -

1) 65C devices and 55C heatsinks, and/or
2) 150W total bias (Half your transformer VA)

Bias is Vdrop (on source resistor, I.E., the .17V bias voltage) divided by the source resistor value, times rail voltage. Times number of devices.

So - .17/.47 = .36 (Amps) *29 (Rail voltage under load) = 10.4W bias per device
10.4*12 = 125W

Doing the math backwards shows that a drop of .2 across the source resistor will be approximately the max bias you want due to transformer limits.