Hello,
I allways loved class A amps, but it's allways problem to get all heatsinks needed (mostly because their high price), so I designed my amp like alephx which can work in class AB. Both output halves are biased to 1A, which can drive my 8 ohm speakers 20Wrms in class A, but circuit is able to run more power in class AB. I know I made some sins like complementary differential input stage etc. But the question is - what do you think about it and what would you make other way?
The amp is running for 4 months and I'm very satisfied with results (sorry no measurements). I'm going to integrate 4 chanels to one chassis and use active crossover for my speakers.
Attached schematic is very simplified so here are some details : for thermal stability I change current biasing diff pairs (this sulution took me some time to develop, but works and it's simple). Source resistors on output pairs (IRFP240 & IRFP9240) are 0R47 and currently I have 16 output transistors per chanel. In my future project (4 chanels) I'd like to run only 4 devices (minimum for this circuit) per chanel.
So, what do you think?
I allways loved class A amps, but it's allways problem to get all heatsinks needed (mostly because their high price), so I designed my amp like alephx which can work in class AB. Both output halves are biased to 1A, which can drive my 8 ohm speakers 20Wrms in class A, but circuit is able to run more power in class AB. I know I made some sins like complementary differential input stage etc. But the question is - what do you think about it and what would you make other way?
The amp is running for 4 months and I'm very satisfied with results (sorry no measurements). I'm going to integrate 4 chanels to one chassis and use active crossover for my speakers.
Attached schematic is very simplified so here are some details : for thermal stability I change current biasing diff pairs (this sulution took me some time to develop, but works and it's simple). Source resistors on output pairs (IRFP240 & IRFP9240) are 0R47 and currently I have 16 output transistors per chanel. In my future project (4 chanels) I'd like to run only 4 devices (minimum for this circuit) per chanel.
So, what do you think?
Attachments
It's hard for me to compare, because I haven't had opportunity to compare to other amps on the same gear. Currently I have Canton LE103 speakers, which I heard playing on Krell amp, but there was different source, than I have, so hard to tell. Within few months I'll compare amp on Dynaudio BM6 with Meridian (there will be a pair of golden ears with me, so we'll see).
According to my ears the amp has very smooth sound, L/R localization is perfect, F/R localisation can't check, speakers are not so good and currently I don't have good source. Bass is strong and well controled.
Rail voltages are +-25V with 180mF capacitance
Send me an email for component values (I have them only in Eagle pcb format)
toudy@chello.sk
According to my ears the amp has very smooth sound, L/R localization is perfect, F/R localisation can't check, speakers are not so good and currently I don't have good source. Bass is strong and well controled.
Rail voltages are +-25V with 180mF capacitance
Send me an email for component values (I have them only in Eagle pcb format)
toudy@chello.sk
That’s very interesting, I’ve never heard of a class A\B push pull amp describe as “very smooth”. I wonder if it’s distortion from the output stage. Is there a high end roll off?
How is the noise floor, is there any hum? This is the Achilles heal of common source operation as you know. However the symmetry must help out some what.
I also wonder if that is why Mr. Pass didn’t use this topology in his amps.
How is the noise floor, is there any hum? This is the Achilles heal of common source operation as you know. However the symmetry must help out some what.
I also wonder if that is why Mr. Pass didn’t use this topology in his amps.
LBHajdu said:
Nice amp (I got similar idea). One thing I wanted to avoid in my desigh was use of separate supply for driving circuit and output stage.
I would use 4 resistors to ground around Vgs multiplier to decrease open loop gain and increase linearity. Also I would adjust current from both current sources instead of changing values of drain resistors of diff pair.
Would it be possible to make the diff pairs out of j-fets. And if so would there be any benefits to doing that? The input impedance would be really high.
Do any of the mosfets making up the diff pairs/ccs have to be mounted for thermal coupling with any other part? This is probably better answered when we have part numbers.
I thought this amp would get a much bigger response, it is the only AB x-amp that that will take a single-ended input. Plus the symmetry is very pleasing to the eye.
Leve
Do any of the mosfets making up the diff pairs/ccs have to be mounted for thermal coupling with any other part? This is probably better answered when we have part numbers.
I thought this amp would get a much bigger response, it is the only AB x-amp that that will take a single-ended input. Plus the symmetry is very pleasing to the eye.
Leve
from N.P. --"looks great, congratulations" and that's it! cool!-- - Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.