AB100 Class AB Power Amplifier

[Nelson Pass]

This question is for Nelson. Is it possible to change the front end to accept balanced signal?

Yes, of course. Just about any diff pair input stage can be made balanced. Easier for high input impedance if they are fets. Instead of feedback loop going to ground, it goes to minus input node.

 

[osscar]

with R3 100R and r4 22K gain will be 36db or x63.
with 1K and 22K there will be some 26db or x20 (I have this configuration).
yes 6.8k affects the bias current regulation.

 

[jk556]

thanks oscar,


will try swapping out r14 for a bigger one.


I just re read eveything as well, and realized that with the schematic it has 36db gain, my bad for some reason I thought that was supposed to give 20x gain, dunno where I got that from.


I will also up the power supply rails soon, was using about + & - 25DC, and gonna up to around 35
when the new transformer gets here,, juts had that one lying arounnd.


Thanks for the tips and help will report back soon

 

[osscar]

Bingo, I found your problem. you definitely don't have an R35 220R (on original NP pcb it was removed as far as i can remember). the simulation shows it. if there is a jumper instead of 220R - gain is like you have 200x.

 

[osscar]

with 1K +22K 220R resistor effect is minimal but with 100R + 22K - huge!

 

[jk556]

hi oscar,


again thanks for your help and tips,


ur right my r35 was not 220R but only 10R, as shown on the prasi board/schematic, I changed that for a 2.2k and unfortunately it did not change anything


However, I took ur earlier advice and swapped r3 from 100R to 1k and that reduced the gain nicely, almost perfect.




What exaclty is the purpose of r35? In the original NP schematic it was 2.2k but you oscar said it as 200R, and on the prasi schematic and boards its refered to as 10R optional?


Thanks for the help

 

[prasi]

I used R35 as an optional 10R ground lift, but forgot to show the value of R3 to 1k from 100R.

 

[osscar]

In Prasi version 10R is like a ground divider. But in the original scheme the signal ground is taken from r35 / r8 - then gain = 36db.
if there is a combination 1K + 22K (gain 26) the effect of r35 is small on the gain factor. But in the Prasi version, nothing changes if you change 10R to 220R.

see post 111 ground point:
AB100 Class AB Power Amplifier


In the original scheme R35 = 220R (2210 = 221 and multiplier 0)

 

[osscar]

Prasi version with 100R have gain x200 instead x63 due different input connection.

 

[osscar]

to provide 36db or x63 gain in the Prasi version 100R should be replaced by 330R

 

[pattison]

Hi,

I recommend you DOGC-H amplifier by Dr. Borivoje Jagodic. It uses current-dumping topology. It is class AC and sounds very good! (better than class AB amps). If you want PCB, PM me and I will send you PCB in Eagle format.

I like your circuit very much, can I get a PCB? My Email: zy2trading@hotmail.com. Thank you!

 

[thimios]

to provide 36db or x63 gain in the Prasi version 100R should be replaced by 330R

Please post an update schematic compatible with Prasi's pcb

 

[osscar]

You have to ask him ... but in my opinion it doesn't matter if you use r3 1K, then there is no change needed. Becausein his schematic R3 = 1K from the beginning. and 26db gain respectively.

 

[thimios]

You have to ask him ... but in my opinion it doesn't matter if you use r3 1K, then there is no change needed. Becausein his schematic R3 = 1K from the beginning. and 26db gain respectively.

Ok. thanks!

 

[osscar]

I add both input variants - in NP variant - it turns out that there is another resistor in the negative feedback - R35 (in my circuit r6)

 

[Vrystaat]

Thimios,

Here are the PDF files from Prasi.
https://www.diyaudio.com/forums/pass-labs/227503-ab100-class-ab-power-amplifier-82.html#post6167110


Great sounding amp !

Regards
Jan

 

[osscar]

I agree - this NP amp sounds very good. One of my quietest amplifiers also - noise around 30uV at output...

 

[thimios]

Thanks but R3 in this schematic is 100R.
In the first post R3= 1000R

 

[osscar]

no - in first post 100 R 100 with zero multiplier. because how else will 36db gain ?

this is the most mystical circuit with multipliers - the hidden treasure of NP

 

[prasi]

Please post an update schematic compatible with Prasi's pcb

Here is revised sch for both 2 rail caps version and 4 rail caps version, the only change in both sch is the value of R3 which is now updated to 1k from earlier 100R.
Shown in bold for convenience.

This refers to the gerbers of following post no. 812
AB100 Class AB Power Amplifier

Thanks to Osscar for eagle eyes and analytical work!