MOSFET source follower for output stage of AB763 blackface

[martin manning]

No, this new 1k would be in the lead going to the input lugs of the bias pots, but yes(!) they can share it. I was thinking the current in each branch would vary (as it does in some bias circuits), but in this case it doesn't.

So, it would go like this:

The anodes of the new FW rectifier diodes go to CP10.

CR1 (the original bias rectifier) is removed and replaced with a jumper. This connects the rectifier output to the 100uF reservoir cap C36 through 22 ohm resistor R69.

The trace connecting 220k resistors R56 and R57 is cut such that one of these resistors is still connected to 22uF cap C29's negative lead and the existing bias pot wiper connection point, P18.

A new 22uF cap is placed with it's negative lead going to the now free end of the resistor that was separated from C29 and its positive lead connected to to the grounded end of 10k resistor R59. This cap can be placed on the top of the PCB and connected to the exposed leads of the separated 220k resistor and R59, or holes could be drilled for its leads so they can be soldered to the appropriate pads.

The new bias pot's wiper is connected to the junction of the separated 220k resistor and the negative lead of the new 22uF cap.

The wire going to the existing bias pot's input lug from connection point P18 is now used to feed the input of both bias pots through a 1k resistor.

The new bias pot's third lug is grounded to its case or the chassis through a 10k resistor.

[SixStringBender]

Hi martin, the wire that goes to the input lug of R58, the original bias pot, is the wire I was talking about cutting and putting two 1K resistors in.

So I don't need two 1K resistors there. Okay, then I'll just put one in the line to the original bias pot's input lug and connect the second bias pot's input lug to the first bias pot's input lug.

This was my original plan using a 1K or 2K trim pot.

Can I use heat shrink over the 1K resistor and put it in the wire?

The rest you say is how I plan to try to do it if I can find a trace to ground to drill through. What I meant by sharing the same ground is I know the best grounding is for the new bias pot to have it's cap's positive lead to the same place C29 and R59 is going to ground. Same for the new 10K resistor from the new bias pot.

If we can't find a ground trace to drill through we'll just solder to ground for them at R59. If that is impossible to reach or I don't like how they will be laid over other components I'll just put them on the pot from the lugs to the shell. I don't want any shorts.

[martin manning]

Can I use heat shrink over the 1K resistor and put it in the wire?

A better way would be to attach a terminal strip (one lug insulated from ground is all you need) to the chassis next to the bias pots and solder the resistor and the lead to that, or maybe find a place for it with your bipolar power supply. The last thing you want to lose is the bias.

[SixStringBender]

Update: I received an email from Heyboer today. They said the new power tranny shipped yesterday.

I paid on April 27th and today is July, 17th. So be aware of the wait time if anyone else is following and planning the same mod.

It'll be a while longer before I follow through with the mod. I have things that must come first. I need the amp up and running for now.

Martin, what is your opinion on reducing the .1uF post PI coupling caps to shorten bias excursion recovery time? Based on calculations .020uF still extends well below E flat.

According to the calculator the low impedance drive to the output tubes doesn't reduce recovery time much. However, the excursion is reduced. Reducing the coupling caps to .020uF reduces the recovery time greatly and it reduces the excursion as well.

Calculate with a low impedance drive and the higher 35-40K impedance drive with .1uF vs .020uF.

Bias excursion calculator
http://www.ampbooks.com/mobile/amplifie ... excursion/

[martin manning]

Easy enough to try reducing the coupling cap values and see if you like it. 0.047uF or 0.022uF are often used.

[Mark]

What became of this project, was the mosfet follower implemented?

What did it look like in the amp?

Did you think it was worth doing to the amp?

[martin manning]

Breif mention of it here (last item in my post), and in the answer to my question: https://ampgarage.com/forum/viewtopic.p ... 17#p397817