That may be true, but this thread is not about high end audio market; it's about a project to attempt to get the best performance without resorting to global feedback!
Do You considering to use tube VAS and short circuit output with triac for DC protect?
Regards
Regards
Last edited:
I tried being serious, with similar results.
I would be thrilled if you, like me, were disappointed by that state of hi-end.
jd
If you defined 'best performance' we could have a more meaningfull discussion.
Let me help you:
If your definition is 'more accurate reproduction of music', you'r on a loosing streak.
If your definition is 'I like it', anything goes. Neg feedback, pos feedback, no feedback, you name it. There's no objective yardstick for 'I like it'. Ask 3 people and you get 4 opinions 😉
jd
Or you are measuring the wrong thing. I can prove umbrella usage causes car accidents by repeated tests.
Everybody, could we please keep discussions for and against feedback now turning toward car accidents out of this thread ?
I would like to have this thread being technical about this specific design, where is decision about no feedback have been taken.
Soren
I would like to have this thread being technical about this specific design, where is decision about no feedback have been taken.
Soren
Tube in input, yes, that would be pretty easy, just replace the LSK389 and two BSP88 with an ECC88, actually I have a schematic in my system where you can select between them....
Triac in output, I did consider it, but decided against it as I didn't like the consequences if there is a short in t.ex. a output transistor, as I don't want fuses in the power rail.
Soren
The Panasonic APV1121S (so8) and APV1122 (dip) are normally stocking parts at both Mouser, Digikey and Farnell....
Soren
Soren
Oh, I have finally found your thread (was waiting pointlessly for a pm)
That includes input/VAS?
Agreed.
Very nice choice. It was about time that someone made an electronic relay. I was going do make an electronic fuse in the PSU, à la JLH (and nothing else), but this works even safer.
300khz at -3dB?
And which is the highpass?
I'd like to know 1W and 10W THD20 into 8 Ohm. Can you do these measurements?
Good. The bigger version IMO can be adapted to run in class A at roughly 20W lowering the voltage rails.
Separate PSU, frontend and OPS would be nice, but probably too costful. At least keeping separate the PSU pcb?
That includes input/VAS?
Agreed.
Very nice choice. It was about time that someone made an electronic relay. I was going do make an electronic fuse in the PSU, à la JLH (and nothing else), but this works even safer.
300khz at -3dB?
And which is the highpass?
I'd like to know 1W and 10W THD20 into 8 Ohm. Can you do these measurements?
Good. The bigger version IMO can be adapted to run in class A at roughly 20W lowering the voltage rails.
Separate PSU, frontend and OPS would be nice, but probably too costful. At least keeping separate the PSU pcb?
Telstar,
Yes, the output of the complete amplifier have been measured to swing to about 4.5V below the power rails, with load.
I have chosen the parts for -3 dB at 1.4 hz and 300 Khz.
I will post complete measurements once I have the real hardware, which should be in about 10 days, assuming planes with my parts can still land here....
My plans are to run them with high bias, but I also believe you can easily do more than 20W class A, I would say maybe one fourth of the class B specs.
I will personally do a three way active crossover system, planning to bias the amplifiers differently for each frequency range.
I want to make a really low cost and simple to use module, with everything on one PCB. You can always not mount the power supply parts if you want to have a separate power supply.
Soren
Yes, the output of the complete amplifier have been measured to swing to about 4.5V below the power rails, with load.
I have chosen the parts for -3 dB at 1.4 hz and 300 Khz.
I will post complete measurements once I have the real hardware, which should be in about 10 days, assuming planes with my parts can still land here....
My plans are to run them with high bias, but I also believe you can easily do more than 20W class A, I would say maybe one fourth of the class B specs.
I will personally do a three way active crossover system, planning to bias the amplifiers differently for each frequency range.
I want to make a really low cost and simple to use module, with everything on one PCB. You can always not mount the power supply parts if you want to have a separate power supply.
Soren
Exactly what I have found in many test open listening, blind, and double blind. About 85% can hear this amount of error.
Beg to differ- I can make that thing oscillate in a heartbeat, no problem at all. Been there done that.
Real circuit measurements- very cool. Nothing personal but I really tire of specs quoted from simulations. I suppose if the music is simulated then the sim sounds good also, haha!
I like no global feedback for circuits driven by and driving know impedances especially at RF frequencies or for extremely low noise. Speakers are not even loads but in stereo pairs one speaker makes the other speaker produce a signal so a speaker is another source. This means the amp needs to be a little more robust than the one shown.
Lots of luck with it though. Will be curious to see the finals. Congratulations on the efforts!
planning to bias the amplifiers differently for each frequency range.
This is something I was thinking of for some time, great project.
All the best
This is something I was thinking of for some time, great project.
All the best
I'm doing the same but with completely different amplifiers for the various ways.
The hardest thing to do is the tweeter amp.
- Status
- Not open for further replies.