Happy New year to you L.C. I hope this year will be full of splendid AUDIO projects as your SSA was.
I wanted to make a sim with your version of SSA, and asked your sim file for that. But my message had no answer. Anyway, this exploration is not a surprise for you, if you had read your private message box recently 🙂
I wanted to make a sim with your version of SSA, and asked your sim file for that. But my message had no answer. Anyway, this exploration is not a surprise for you, if you had read your private message box recently 🙂
Hi Esperado
OP amp kind of error correction is actually working as OLG of the corrected amp would be much higher, it certainly can lower distortions if stability maintain unaffected. Most probably OP amp will drive the feedback signals into peaking and even into oscillations if original amp could not follow.
What VSSA in other thread surprisingly showed me is its enormous damping (extremely low Zout) gained from high OLG/bandwidth and exact feedback/compensation scheme. uV feedback precision coming out of six transistors only makes amplifier's output differentiating from unloaded conditions down to few mV.
Square waves misuse showed no instability and for me that is of more importance than mostly accepted low THD/TIMD.
Regards, Andrej
OP amp kind of error correction is actually working as OLG of the corrected amp would be much higher, it certainly can lower distortions if stability maintain unaffected. Most probably OP amp will drive the feedback signals into peaking and even into oscillations if original amp could not follow.
What VSSA in other thread surprisingly showed me is its enormous damping (extremely low Zout) gained from high OLG/bandwidth and exact feedback/compensation scheme. uV feedback precision coming out of six transistors only makes amplifier's output differentiating from unloaded conditions down to few mV.
Square waves misuse showed no instability and for me that is of more importance than mostly accepted low THD/TIMD.
Regards, Andrej
Yes it increase the ratio of feedback as much as you can afford keeping the stability. (Phase margin can be unchanged, and even, we can be increased with some compensation caps in the error feedback loop, without affecting slew-rate of the power amp itself.).
About speed, the problem with fast power amp is to find an OPA fast enough, as SSA is damn fast itself. . No risk, in my example, to saturate the OPAs, the output of the second OPA stay at <+/-700mv peak at 100Khz full power with distortion at -60dB.
I do not know if it improve the sound or not: Need a try in real world. May-be the benefit will be side of the damping factor ?
As it is simple, easy and not expensive nor destructive, it deserve a try.
Attachments
Last edited:
I think reducing distortion of the open loop of an amp is not to be opposed to the use of feedback, they are complementary. if you can reduce closed loop distortion while keeping the slewrate with a driver (i doubt) the error correction add will reduce the new distortion with the same 10 ratio.
Last edited:
Very nice Esperado 🙂
I am working on a new amp (100W CFA) and after this will do another amp with AFEC (switchable!). I think the technique has merit, but of course you have to take care with the loop compensation. The free DC offset control is also a great side benefit!
I am working on a new amp (100W CFA) and after this will do another amp with AFEC (switchable!). I think the technique has merit, but of course you have to take care with the loop compensation. The free DC offset control is also a great side benefit!
Well, something is interesting, below your question.
Because we increase the error feedback level, in a way, we reduce the slew-rate/correction-signal ratio.
But, because we are dealing with a yet low distortion, and because we can filter extreme HF from this added part of feedback; i believe we do not have to fear any added TIM.
So, the main benefit is to improve distortion on an EXISTING amp without modification of the amp Itself.
As noticed by L.C. an other approach is to increase OLG of the amp, not so easy while keeping-it stable.
This error correction present, of course, some negative aspects: it is a little acrobatic and need precise fine tuning.
Even if i had this idea long time before, while i was working to my protection system, i forgot-it with time and it was YOUR publication which refreshed-it in my mind. So all the credit is for you.
Last edited:
But you are not filtering extreme signals from the opamp. Any input/feedback signal regardless of BW will enter the opamp's output, no? Does the simulator model the effects of opamp output impedance?
I believe yes. I don't understand "enter op amp output", Keantoken. (This opa is able to drive video lines so, low impedance even at very HF and is charged by 3K. What do you fear ?)
The level of error signal is so low at audio frequencies that slew rate of the error will be negligible comparing the the ones of the OPA and Power amp itself. (both 1000V/µs)
Last edited:
ugh.... technospeak, technospeak and more technospeak. could someone please translate all of this in to sound quality of this design? i'm not getting much out of this thread or the one about fetzilla, i think... this foreign language is making my head hurt and me forget stuff now.
i was getting a little but of something when someone talked about speed, detail and bandwidth, but alas it was only theoretical talk and not a review of this design SQ. just looking for a non technical lowdown to see if maybe i try buidling one of these instead of maybe a gainclone or get a trends or even kingrex class D.
i'd like a fast and detailed amp that isn't lifeless or syrupy thick either.
i was getting a little but of something when someone talked about speed, detail and bandwidth, but alas it was only theoretical talk and not a review of this design SQ. just looking for a non technical lowdown to see if maybe i try buidling one of these instead of maybe a gainclone or get a trends or even kingrex class D.
i'd like a fast and detailed amp that isn't lifeless or syrupy thick either.
Last edited:
I think both one of the SSA options or the fetzilla would be very much the kind of sound you are looking for. Hopefully and SSA kit will be coming up soon ? ? ?
Any news on that Lazy Cat ?
Any news on that Lazy Cat ?
what's up with all the talk i keep seeing about tweaking bias etc. after the amp is built? i'd rather just get a kit that you build and be done with it with adequate instructions for someone whose never done anything more ambitious than solder speaker cables that thinks he could confidently build a speaker crossover as long as no resistors (have to be installed directionally, right?) are needed.
i was originally interested in DIY amps because of simple gainclones with just 9 parts. when i start seeing pics by people i'm guessing are more experienced than me asking if they mounted everything right, memories of my NAD blowing up when my subwoofer wires shorted come back. *pop pop pop*! flash and oh all that smoke
all the other talk about making changes to power rails and filtering too make me think this might be too big a project to attempt for me. my mind is allergic to abstractions. i understand mechanics much better as you can see what's going on.
i was originally interested in DIY amps because of simple gainclones with just 9 parts. when i start seeing pics by people i'm guessing are more experienced than me asking if they mounted everything right, memories of my NAD blowing up when my subwoofer wires shorted come back. *pop pop pop*! flash and oh all that smoke
all the other talk about making changes to power rails and filtering too make me think this might be too big a project to attempt for me. my mind is allergic to abstractions. i understand mechanics much better as you can see what's going on.
Last edited:
Shaan tried SSA with servo, PCB is at post #3428, so no adjustments, I think it's easy for beginners to build. I'm not sure anybody made it yet.
Try DX!
Search for DestroyerX's threads!
You will feel as comfortable as being at home.
Search for DestroyerX's threads!
You will feel as comfortable as being at home.
I think from what he says budget minded would prefer assembled & tested boards to which he can add a PSU & box etc - not sure what to recommend for this.
- Status
- Not open for further replies.