Is this any good, new DIY design?

Any comments on this design .. it's a mixure of designs found on this forum, the www and some patent :)....

any ideas how to give it some global feedback ?
 

Attachments

  • newdiyamp6.png
    newdiyamp6.png
    3.1 KB · Views: 816

Nelson Pass

The one and only
Paid Member
2001-03-29 12:38 am
I think it would be worth exploring. I don't see where
it is patented, and I would be surprised if it was.

I think, though, you will want to alter R11 and R12 to
200 ohm (or so) potentiometers, as you can't assume
that the MOSFETs Vgs figures will be the same as the
ones used by the designer, the result being a bias
figure which might be catastrophically different.

Start with R11 and 12 at minimum resistance and bias it
up to taste.....
 
Like, I'm surprised that Jam wants current sources?
Actually, in this case I happen to agree with him. Whether as an amp or a preamp, I'd think current sources would be a decent possibility.
The values for R5 & R6 look a bit scant to me. I'd go for 10k as an absolute bare minimum so as not to load down the preceding stage or source.
Jam,
You're wanting push-pull outputs on a preamp?

Grey
 
Thanks for the reply guys!

I'll consider all comments

Dear Pass: funny you say that about the patent.. this design is also based on patent 5,376,899 but with the stuff left out that I don't understand :)

My only wish is some way of including the power stage into a feddback loop...this should improve liearity above class A bias point in 8 ohm loads..

I realy wanted a headphone amp and I hope the whole thing works

Thanks again..

greetings,
thijs
 
Grey,

The comp. output stage would work if it was biased high enough. Just in case you think I that I have lost it, my choice of output stage would be a single ended, current sourced (had to get this in somewhere) mosfet.
Cascodeing the comp.(push-pull) output stage might be a worthwhile experiment though (pre-amp only).

Jam
 
Thijs,
A headphone amp? Sounds like an interesting idea. Go for it.
Jam,
I looked at the schematic at work last night but don't have it at hand now, so I'm quoting from memory--hence liable to be faulty. (Don't have time to download & look it over again...maybe later.) Seems to me that you're going to end up with excess gain. Yes? No? What will you do with the excess? NFB? I'll try to look at the schematic again later today.

Grey
 

Nelson Pass

The one and only
Paid Member
2001-03-29 12:38 am
thijs,

The part left out would be the patented part.

The circuit could be turned into a SuperSymmetric
amplifier, but I don't want to pull that circuit of the
hat until later this year.

Until that happy day, I suggest you guys go over the
schematic on the front page of the patent at the
passlabs.com website. The concept of Super Symmetry is
really very simple, and the implementation is even
simpler. I was simply lucky that nobody noticed it before
me.
 
Hi,

About the patent.. I can't wait ... I've been trying to design a proper implementation of it but haven't succeeded satisfyingly.

That's why I wanted to try a more simple efficient design, just as an headphone amp and maybe more.. But as I read the other threads I'm not the onlyone and in fact others have done it better..I guess I'll just build a BOSOZ.. seems it willl drive my headphone just fine...... thanks for al the support!

greetings,
thijs