Nico Ras said:
I am not sure why you want to bias the output in class A if the driver is class B, you already have cross-over distortion present from the op-amp, class A output stage is not going to make this go away.
Rather run in class AB with lower quescient current, why heat up the environment if this does not have any positive effect.
I will give it a try tonight. The idea is to bias both opamp and output stage in class A when operating in class A (I don't know if this will make any improvement in audio quality since my heatsinks really don't encourage class A bias).
Anyways distortion from the opamp is much lower than distortion from output stage and does not happen at exactly the same point since voltage and current are not in phase.
I thank any advise since i'm rather new to discrete output stages. Since i'm planning to buy some audio mosfets, has anyone listened to various of them and can give some recommendations (some are really expensive!)?
Anyways distortion from the opamp is much lower than distortion from output stage and does not happen at exactly the same point since voltage and current are not in phase.
I thank any advise since i'm rather new to discrete output stages. Since i'm planning to buy some audio mosfets, has anyone listened to various of them and can give some recommendations (some are really expensive!)?
This sounds well? I remember doing one with zeners and resistors and the results were anything but nice.
Are you using the same mosfets than me? If this has 100x the distortion of the driver its a killer amp.
Are you using the same mosfets than me? If this has 100x the distortion of the driver its a killer amp.
ionomolo said:
You are basically increasing the current drive of the op-amp. Mosfets work nice because they do not need high current to drive it and the op-amp driving them should be fine. Simulation indicates that THD improves although this is unlikely.
I never made one, only quickly simulated this a moment ago to see, just try it, you have all the components at hand anyway and it should work without any problem.
I will try the next week since i havn' any power mosfets at the moment. The only IRFP's i had left are now soldered (without bridging) in a board to make some nice computer loudspeakers. Yes, the design is not optimal but sounds way better than the average gainclone.
Theoretically my design does not increase current drive of the opamp if both current sources are balanced and current-source loading should give great power supply rejection, but yours is so incredibly simple that it seems it's not worth the improovement.
Any reason why opamp-driven output stages are not more popular among the diy community? I know that 12W it's not much but opamp rolling is really cool and 50W bridged matches the more common gainclone design.
Theoretically my design does not increase current drive of the opamp if both current sources are balanced and current-source loading should give great power supply rejection, but yours is so incredibly simple that it seems it's not worth the improovement.
Any reason why opamp-driven output stages are not more popular among the diy community? I know that 12W it's not much but opamp rolling is really cool and 50W bridged matches the more common gainclone design.
Hard to tell why. Maybe it is not sophisticated enough. 12W is actually plenty, the average listening level for most mortals are between 4 - 6 watts. 100 watt amps are the reason neighbours become enemies.
I've realized i was using 47K resistor instead of the 20 stated on the schematic. Probably higher nfb will also lower the distortion and change sonics.
I will buy some varied 2SK/J's to try how they sound and listen also to your design. I can't wait to have this thing boxed and with proper power supply and heatsink.
I was planning to add some sort of EC once finished biasing. Is this a good idea or would add too much propagation delay?
I will buy some varied 2SK/J's to try how they sound and listen also to your design. I can't wait to have this thing boxed and with proper power supply and heatsink.
I was planning to add some sort of EC once finished biasing. Is this a good idea or would add too much propagation delay?
Nico Ras said:
Hi Nico,
I guess I'm not mortal, I need MUCH power! 4-6 watts will not even drown out the voices I hear in my head...
🙂
BTW, all of my neighbors hate me.😀
ionomolo said:
"EC" sorry I am unfamiliar with some of the DIY jargon, I only found out today what IMO meant (in my opinion) could not figure this out try inter, modulation ... took me a while, I think I am a little stupid.
MJL21193 said:
No John, I guess your not mortal. I must admit, I use a monster quad-amped system also, but since I built a 32 w/ch class A, I have not turned on the monster in the lounge even once in four months.
I now even receive Chistmas wishes from the neighbours, in the past they threw rocks on my roof, and I did not even know this.😱
Wow, i could not sleep thinking i was the only one who thought that IMHO had something to do with intermodulation the first time i read it! EC i meant error correction (or cancellation).
ionomolo said:
Thank you Ionomolo for enlightening me, I feel so much better, one more word I learned today.
ionomolo said:
Thank you kindly, ionomol. That was stupid of me too, I did not know the Spanish flag, but forgive me, I learned about history some 45 years ago and like as my ears, my memory is fast also diminishing.
- Status
- Not open for further replies.