Did you try a bessel low pas filter, a 24 dB octave, because I did sim with it quite good, use the middle of filter for feedback.
I have in sims seen that the use of very high speed mosfets and feedback opamps did very well, like the opa 2365 is a very good candidate, but needs low voltages, with this you get lowest hd.
I do go for high end, I try to make a audiophile version, but not now, first the circlotron and the making of the transformer, both allfet and hybrid.
regards
Sim doesn't tell how good the sound is
Why?
Because your simulation can only simulate a single frequency. No mix. Not represent a music signal which is very mix and dinamic.
No body can make instantly a good amp directly from a very good simulation. You should make it real and hear it. And frequently you will need to check between trial and simulation until you get final result
Why?
Because your simulation can only simulate a single frequency. No mix. Not represent a music signal which is very mix and dinamic.
No body can make instantly a good amp directly from a very good simulation. You should make it real and hear it. And frequently you will need to check between trial and simulation until you get final result
Last edited:
Yes you are right, because some amps with terrible HD has sometimes a very good sound, and a amp with 0.0001 HD sound often clinical.
I did try a hybrid with only plate followers and no feedback, was the best sounding amp ever made by me. The digital versions I did not yet try, so coming winter I do, now I am busy with making a coil winder for the circlotron, programming it in a pic who I do not myself but a other guy.
Th bessel 24 dB filter is a nice approach to get good sound because it it's faseliniar, but needs 24 dB octave, with it I can a digital amp without feedback to try and with it, what I have done is feedforward to get it UCD.
regards
Last edited:
Fan7293
Hello friend, dead time, I applied it in r gate, and in fact, this driver handles, two mosfets per side, fdp2614 200v 62a 27 miliohms Qg 76 nc vcc 85v, enough power, at 450 khz, very high fidelity, greetings . This Fan7392 is exactly the same as ir2110
Hello friend, dead time, I applied it in r gate, and in fact, this driver handles, two mosfets per side, fdp2614 200v 62a 27 miliohms Qg 76 nc vcc 85v, enough power, at 450 khz, very high fidelity, greetings . This Fan7392 is exactly the same as ir2110
In this topology, more switching frequency, more fidelity, in amplifiers of any power.
Wrong diodes
Is correct, the diodes that you mention, but in your diagram, you put 1n4148, they do not work there, you need diodes, anti saturation.
Is correct, the diodes that you mention, but in your diagram, you put 1n4148, they do not work there, you need diodes, anti saturation.
There are manufactured better faster mosfets, so in time we can go higher in switch frequency.
PS how high is your R for dead time?.
regards
An externally hosted image should be here but it was not working when we last tested it.
..
I want design PCB for this amp
I have Professional Software for Design PCB to produce sample byself for small money in Chinese PCB Factory
Somebody interested for finished PCB Layout with Gerber Files to produce sample in PCB Factory ?
Thanks for the link, but PCB Factory
Cheap PCB Prototype | Pandapcb | Quick turn PCB
not accept like this
You need Gerber files
Cheap PCB Prototype | Pandapcb | Quick turn PCB
not accept like this
You need Gerber files
I did some more things, one is a fase shift version, who is very low hd and a open loop multilevel version.
I do not now however what it does in real time, but sim looks good.
I have IR2110 I think 20 pieces, but sim model is not really a good one, so did behavior models.
Attachments
Hi. Can you give Gerber the files for Your amplifier? I would appreciate your help