@andyr:
It's the loop gain, not FR if that's what you were thinking? Makes output impedance lower at LF. I saw this in the output impedance plot too, and saw it mentioned somewhere in the thread too. With the feedback loop connected, the FR of the amp is basically flat.
I find this interesting, since I usually like the mids/treble of amps with low OLG (like the JLH), and the bass sounds better with high OLG, could be a nice compromise like this.
@AKSA:
Thank you Hugh for the plots! Do you know what is causing the hump around 30MHz? Have you done the same analysis on the 'current control loop'? (That's what I wanted to try too, and it looked funny too, since I can't trust my sim as it is). From my understanding, humps like that can sometimes be due to unstable local loops that might not look significant in the overall loop? Not saying it's a problem, just curious
It's the loop gain, not FR if that's what you were thinking? Makes output impedance lower at LF. I saw this in the output impedance plot too, and saw it mentioned somewhere in the thread too. With the feedback loop connected, the FR of the amp is basically flat.
I find this interesting, since I usually like the mids/treble of amps with low OLG (like the JLH), and the bass sounds better with high OLG, could be a nice compromise like this.
@AKSA:
Thank you Hugh for the plots! Do you know what is causing the hump around 30MHz? Have you done the same analysis on the 'current control loop'? (That's what I wanted to try too, and it looked funny too, since I can't trust my sim as it is). From my understanding, humps like that can sometimes be due to unstable local loops that might not look significant in the overall loop? Not saying it's a problem, just curious
Last edited:
Thanks, Rf.
(Yes, I thought it was the FR graph.)Not sure what having "output impedance lower at LF" causes.
As I would hope it would be!
Andy
Andy: Well, basically you could describe output impedance as a series resistor on the output of the amp, but it can vary with frequency, like in this case. Maybe damping factor is more familiar to you? Damping factor is calculated as speaker impedance/output impedance (correct me if I'm wrong), so lower output impedance equals higher damping factor. Usually low output impedance/high damping factor makes bass sound 'tighter', I guess because it has some effect on the Q of the bass alignment.
Hi RF,
I have not done any current plots; I've only just mastered the Tian probe in fact.
I'm not convinced that beyond about 10MHz the Tian Probe is particularly accurate, and I'm suspicious that models and LT Spice engine are doing odd things beyond that frequency. I do know that Tian is far more accurate than Middlebrook, and I've been amazed at how useful it is in design analysis.
The OLG of the Alpha Nirvana reveals about 28dB from 1Khz to about 18KHz, but below 100Hz it rises to almost 60dB. This is quite deliberate to increase damping factor at bass frequencies. Most SS amps drop precipitously beyond 20KHz, and this amp does about 2dB from 15KHz to 20KHz but by having a flat OLG during most of the audio band you see some good sonic characteristics which are worth having.
HD
I have not done any current plots; I've only just mastered the Tian probe in fact.
I'm not convinced that beyond about 10MHz the Tian Probe is particularly accurate, and I'm suspicious that models and LT Spice engine are doing odd things beyond that frequency. I do know that Tian is far more accurate than Middlebrook, and I've been amazed at how useful it is in design analysis.
The OLG of the Alpha Nirvana reveals about 28dB from 1Khz to about 18KHz, but below 100Hz it rises to almost 60dB. This is quite deliberate to increase damping factor at bass frequencies. Most SS amps drop precipitously beyond 20KHz, and this amp does about 2dB from 15KHz to 20KHz but by having a flat OLG during most of the audio band you see some good sonic characteristics which are worth having.
HD
Thank you Hugh! I'm no expert on stability simulations myself, I just followed Sandro's tutorial when he made it, and have done a few simulations on some amps since, and they have been pretty close to reality I must say.
https://www.diyaudio.com/community/threads/sw-vfa-01-audio-power-amplifier-video-series.347673/https://www.youtube.com/channel/UCMcyPNHgpgdQ353hN4Zz5Qg)
I have played around with compensations (methods and cap values) on some amps, and most of the times the simulated values are good.
Sometimes there has been oscillation problems even if the overall loop plot seemed ok, in both cases there was a local loop oscillating that was only visible as a small hump high up in frequency and far below unity gain in the simulated overall plot (one with CFP output, and VAS on Bonsai's KX). However, the source of the problem was found by putting my finger on legs of components, rather than any local loop simulation..
I never really got the local loops fully, but I figured the current control here would be a good example of such a local loop. I also read Kokoriantz comments in the beginning of the thread, and that made me want to simulate it. I think he's quite a clever guy, had some help from him when I was playing around to improve my JLH amp.
Sandro put the probe on the negative feedback, so that's how I've been doing it so far, but in this case it's not making sense to me. I also tried the probe on the base of the 'helper transistor' for the current control, but could not really grasp what I was seeing.
Anyway, I think square wave signals and probing around would reveal if there was any stability issues. I'm just sitting around waiting for the boards, so I waste my time on this stuff, don't let me waste your time too
https://www.diyaudio.com/community/threads/sw-vfa-01-audio-power-amplifier-video-series.347673/https://www.youtube.com/channel/UCMcyPNHgpgdQ353hN4Zz5Qg)
I have played around with compensations (methods and cap values) on some amps, and most of the times the simulated values are good.
Sometimes there has been oscillation problems even if the overall loop plot seemed ok, in both cases there was a local loop oscillating that was only visible as a small hump high up in frequency and far below unity gain in the simulated overall plot (one with CFP output, and VAS on Bonsai's KX). However, the source of the problem was found by putting my finger on legs of components, rather than any local loop simulation..
I never really got the local loops fully, but I figured the current control here would be a good example of such a local loop. I also read Kokoriantz comments in the beginning of the thread, and that made me want to simulate it. I think he's quite a clever guy, had some help from him when I was playing around to improve my JLH amp.
Sandro put the probe on the negative feedback, so that's how I've been doing it so far, but in this case it's not making sense to me. I also tried the probe on the base of the 'helper transistor' for the current control, but could not really grasp what I was seeing.
Anyway, I think square wave signals and probing around would reveal if there was any stability issues. I'm just sitting around waiting for the boards, so I waste my time on this stuff, don't let me waste your time too
Hi audiostar,
We can have this conversation in PM, I wouldn't want to pollute the thread.
Thanks
Do
Yeah! Once you have this thing in front of you, it is quite impressive transformer! LOL
Here's some pictures with it installed. Enfin! Le son du silence! (At last, the sound of silence). This thing is totally quiet!!!
I need to cleanup the cable mess, some were just temporary. It will be done shortly.
Do
Here's some pictures with it installed. Enfin! Le son du silence! (At last, the sound of silence). This thing is totally quiet!!!
I need to cleanup the cable mess, some were just temporary. It will be done shortly.
Do
Thanks all for the nice comment!
Both songs are on Tidal (Bridge over troubled waters & Disturbed’s version of The Sound of Silence). I'll make sure I give them a good listen!
I'm about to take the old transformer apart to try and find out what was causing all the humming sound... Just out of curiosity...
Things to finish :
Do
Both songs are on Tidal (Bridge over troubled waters & Disturbed’s version of The Sound of Silence). I'll make sure I give them a good listen!
I'm about to take the old transformer apart to try and find out what was causing all the humming sound... Just out of curiosity...
Things to finish :
- Install 75 Celsius cutoff thermostats
- Connect both PSU grounds to chassis
- Connect Softstart "touch" power botton
- Plug and enjoy music!
Do