Converting to adj. -bias on an Ampeg B-25-B

[Willabe]

The bass player from church asked me to fix his Ampeg B-25-B. Needs new tubes/filter caps.

I'm thinking making R-38 about 1/2 it's value, ~22K to 26K and standing it on the 50K trim pot Doug sells. With the pots wiper centered that should set the -bias around where it was set from the factory and give a little range up/down for new tubes.

Yes?

 http://el34world.com/charts/Schematics/files/ampeg/Ampeg_b25b.pdf


            Thanks,   Brad   

[HotBluePlates]

Yep, that sounds reasonable.

[Willabe]

Ok, thanks HBP.

If I don't get enough -dcv I guess I'll try making the 100K/2w series R (R 40) smaller.


       Thanks,       Brad        

[stratele52]

Ok, thanks HBP.

If I don't get enough -dcv I guess I'll try making the 100K/2w series R (R 40) smaller.


       Thanks,       Brad        

Start with higher R38 and R39 , if not enough - DC ,  try R40 smaller

[HotBluePlates]

Or R39 bigger. It creates a voltage divider for the raw a.c. before it hits the rectifier diode.

Yes, it uses more parts than Fender's setup; I think the intent is to be certain to have the lowest a.c. output impedance possible from the bias supply (the bias feed resistors see only a bias cap to ground) but the highest d.c. impedance (the bias feed resistors see only a bias cap to ground).

Start with higher R38 and R39 , if not enough - DC ,  try R40 smaller

Yeah, what he said!

[Willabe]

OK guys, I get it, thank you.

Yes, it uses more parts than Fender's setup; I think the intent is to be certain to have the lowest a.c. output impedance possible from the bias supply (the bias feed resistors see only a bias cap to ground) but the highest d.c. impedance (the bias feed resistors see only a bias cap to ground).

Yeah, I figured that they wanted to make sure the -bais was stable because of the 500dcv plate and 490dcv screen.



             Brad     

[Willabe]

Yes, it uses more parts than Fender's setup; I think the intent is to be certain to have the lowest a.c. output impedance possible from the bias supply (the bias feed resistors see only a bias cap to ground) but the highest d.c. impedance (the bias feed resistors see only a bias cap to ground).

Would you go deeper into this please? I think your saying the 'bias feed R's are the series R's?

I see they put a voltage divider before the diode to knock down the B+ fed to the diode, then they have a pie filter.

So having the 2nd filter cap as the last thing the -bias feeds the 220K grid leaks is what makes it low Z?


                  Brad       

[HotBluePlates]

Yes, it uses more parts than Fender's setup; I think the intent is to be certain to have the lowest a.c. output impedance possible from the bias supply (the bias feed resistors see only a bias cap to ground) but the highest d.c. impedance (the bias feed resistors see only a bias cap to ground).

Would you go deeper into this please?

After looking at the schematics for the Princeton Reverb and for Marshall amps that tap bias from the high voltage winding, I may have to retract my statement.

When I said "bias feed" I mean the grid leaks for the output tubes, which feed bias voltage to the grids. Looking at the Fender and Marshall schematics, they have a resistor to ground from the junction of those grid leaks, but there is a cap to ground as well. The cap would still be effective at holding that junction at a.c. ground even if there is a resistor present.

I think I was looking for an explanation on why Ampeg used so many more parts than they needed to do the job. I guess now I don't have a good answer for that.

So it looks like the Ampeg circuit could have been simplified; either way, I think you know how to tinker it to get the needed range of bias voltage.

[stratele52]

Or R39 bigger. It creates a voltage divider for the raw a.c. before it hits the rectifier diode.

Yes, it uses more parts than Fender's setup; I think the intent is to be certain to have the lowest a.c. output impedance possible from the bias supply (the bias feed resistors see only a bias cap to ground) but the highest d.c. impedance (the bias feed resistors see only a bias cap to ground).

Start with higher R38 and R39 , if not enough - DC ,  try R40 smaller

Yeah, what he said!

I don't see where he is talking about R 39

[HotBluePlates]

Yeah, what he said!

I don't see where he is talking about R 39

My post was speakeing to Willabe; so when I said, "What he said," I meant what you (Stratele) said. You posted while I was still typing my reply.

[stratele52]

Ok I understand ,Thank's


We should begin post by 
 


Hi HotBluePlate ,


( answer here )

[ernie_jr]

Another option that I have used on a scratch built b 25 is after R40 (the 100K off the p/t) build a bias circuit that SUNN uses in their 200s, etc. Sunns also used about 500 volts on the plates and this circuit gives a fairly wide range of adjustment.
Good luck,
Ernie

[Willabe]

Thanks Ernie, I'll look at that.


            Brad