AB100 Class AB Power Amplifier

Hi,

Only have plug n play, but here you go

Regards
 

Attachments

  • AB100 20Khz.png
    AB100 20Khz.png
    7.2 KB · Views: 1,477
  • AB100 10Khz.png
    AB100 10Khz.png
    7.1 KB · Views: 1,475
I'd greatly appreciate a direct comparison AJ to AB100 !! :wave:
Honestly I do miss bass precision on my Aleph. :crying: :Ouch:


Rails: +/-50V or so
Nelson, any idea what an appropriate voltage range would be? How low
could I go? :scratch1:

Unfortunately I have just some transformers on spare with either +/-18 V
(giving ~24 V rails) or +/-42 V (58 VDC). Power is not any concern to me.

Thank you again for all the effort !! :up:
 
Love your AB100! What problems has mine?

Hi guys, I'm pretty new in the audio diy but I'd like to get better, and I could really use a help from you all, please. I've been trying to build a class AB amp myself with TIP 142/147 Darlingtons, so I saw the AB100 and I found it great and not so different from mine.
Here is the schematic, I tried some simulations on Spice and it seems to work fine: but when I built it, I had (and still having) tons of problems. For instance, when I try it with no load it actually works pretty fine, but then I try a 15 Ohm resistive load or a 4 Ohm speaker and it starts ringing like there's no tomorrow (around 4-6 MHz). Ringing kinda decreases a little when I bias more the output stage (slightly higher values for R7), but nothing good anyway. Do you have any advice for me? Some changes to be made? Or would it be better to start over with a new schematic?

Thank you very much!
Giovanni

An externally hosted image should be here but it was not working when we last tested it.
 
Administrator
Joined 2007
Paid Member
Layout, grounding and signal returns can be crucial on designs like this.

I don't see any Zobel network on the output.

It might be worth looking at adding a small (10pf or less) across R18 and see if that changes things.

Also add a small cap across the vbe multiplier (say 4.7uf or 10uf)
 
Using a resistor load (R5=1k5) is unusual.
Have you considered and rejected the usual options of Bootstrapped and Active (current sink)?

You have matched the collector load currents of the LTP.
But ignored the current injected by the VAS.
The LTP currents end up not being matched.

C4 (Miller compensation) is unusually large. With only 1.1mA available to charge it, slew rate will be severely curtailed.

Consider where local decoupling can be added to maintain performance at HF.
 
Thanks for your answers, first of all. Well, I didn't thought of a Zobel network actually, but I'll try. I imagine that I'll have to measure the series inductance of my speaker: given that I haven't got an impedance-meter, should I measure it manually with a sine wave at different frequencies or is there a simpler method?

Then I'll add a few caps where you told me and I'll tell you...just a few questions for AndrewT: can you provide an examples of what you call "Bootstrapped and Active (current sink)"? A resistive load was the simplest solution that came to my mind and no, I did not think of the usual solution (maybe I didn't even know...)! One more thing: what do you mean by VAS? Because if you mean the gain stage (Q8 with its R19) I thought I could neglect that base current...what should I do to match the currents correctly?
 
Member
Joined 2000
Paid Member
AB100 Boards

Hi Guys,

I want to build a Class AB power amp for a friend that's turning 70 this year.
Maybe should build a 70w amp to be symbolic. ;)
A class A amp might be too much for her in terms of weight and heat.
Was thinking about the ACA, but then would need high efficiency speakers.
Might be hard for me to build/find high efficiency speakers in a small package.

I did read that the AB100 runs cool, which is a step in the right direction.
Then I could get her whatever speakers fit her style and sound.

She loves classical, operas, orchestras, etc. So, I think some power is needed.
I consider her an audiophile, because I don't know too many women who record
classical broadcasts to a digital tape recorder, although I'm sure they're out there.
I want to make something for her that sounds really good.

I'm also curious as how this amp would sound for me, so...

Does anyone have PDF or Gerber board files for the AB100?

Thanks,

Vince
 
Member
Joined 2000
Paid Member
ZM, that's what I thought. I think the DIYAB is 4 stages, maybe three if omitting the cascode stage. I could be wrong either way. Usually am. :)

I've look on the net for a simpler design, but keep coming back to the AB100.
That said, the DIYAB look fun to build, if that's what you're after. It's nice to have all that support on a project.

I'll just have to learn Eagle PCB software. :D