Depends on the reason you have for using R-D strings.
I wanted to know one good reason.
You need to test it's usefulness for that goal.
That's the question. Except that my goal is the best overall performance.
That goal could be supply filtering, supply impedance, supply sag reduction... So your question is very vague.
For supply filtering and impedance I don't need the diode... There is another goal... but anyway, the question is... to make it simpler, will you use it in Shaan schematic?
I was afraid that rough simulation will not be precise enough to predict it's behavior...
Simulation could tell you performance specs, but those specs on't tell you if you ill like the result or not if your goal is to enjoy the sound. Simulation ill help you understand hy something changes the sound, provided you already have empirical observations to begin ith. So the process in this sense is backards: test first simulate after to understand.
As for filtering, I think it is orth a try on the current sources and/or the VAS, since there are still unprotected Early junction effects, but I think the difference ill be small. I think there may be a bigger improvement if you filter the rails ith my poer K-multipliers, so the FETs get a regulated rail. Hoever my Kmultipliers seem to ork much better on BJT output amps. I don't have a lot of experience here but the small signal area of the amp seems ell protected from the rails.
Another factor is rectifier spikes on the rails, if your supply isn't ell filtered.
As for filtering, I think it is orth a try on the current sources and/or the VAS, since there are still unprotected Early junction effects, but I think the difference ill be small. I think there may be a bigger improvement if you filter the rails ith my poer K-multipliers, so the FETs get a regulated rail. Hoever my Kmultipliers seem to ork much better on BJT output amps. I don't have a lot of experience here but the small signal area of the amp seems ell protected from the rails.
Another factor is rectifier spikes on the rails, if your supply isn't ell filtered.
Attachments
Thank you, Esperado.Congratulation.
If your amp does not heat too much, and once burned out a little, you can try to reduce the bias of each power unit to 150ma. Not sure for yours, but most of the power Fet/Mostfets i know seems to sound the best around this value in class Ab amps.
Now, if somebody ask-you if you can build an amp with no servo, you can just answer:
"Yes, I can".
![]()
Yes, I can build this SSA without servo
My heatsink only warm... i can touch with my hand as long as i want.
Without support and help from you, shaan and esperado a newbie like me maybe wouldn't hear the beatiful sound of SSA.Super, so also newbie can do it hehehe
I'm using Bookshelf Speaker build by an expereinced DIY'er from my country.What's the loudspeakers you have to connect to SSA?
Simulation could tell you performance specs, but those specs won't tell you if you will like the result or not if your goal is to enjoy the sound. Simulation will help you understand why something changes the sound, provided you already have empirical observations to begin with. So the process in this sense is backwards: test first simulate after to understand.
So far I believe in simulations (coupled with good engineering judgements). I can build circuits and compare by ears, and my ears is very good enough to rely on, but I want a quick and precise approach by knowing what to look for in simulation. Then I will know how to get the best.
However my Kmultipliers seem to work much better on BJT output amps.
I don't use BJT regulators in FET amps. Sometimes I use FET regulators in BJT amps (class-A). But I prefer BJT regulators for BJT amps. I use the simplest BJT multiplier for my SSA_BJT. Might try your K-multiplier, but I'm skeptical.
Power supply should be part of amplifier design. I started with transformers I have on hand. Lowering and (often) raising rail voltage of a finished amp will reduce performance
maybe Resistor -Diode string sir, just guess.Sorry to ask a stupid question , due to my poor English: What is R-D strings ?
I always understood that its main goal was sag reduction, when outputs are fed from the same source.
Yes. In this case, why is front end sag reduction or backward current must be prevented? What is the quantitative goal?
I think it has a mix of dynamic effects, including sag reduction depending on hat the R value and here it is placed.
A silicon diode has effective R of about .033/Id. R decreases ith Id. This is nonlinear, so if you give a diode a sine current, the average resistance ill actually be loer, and the voltage drop of the diode. Thus you have a dynamically changing RC filter. In fault conditions, it prevents frontend rail sag, but not entirely because it pulls some current to turn the diode back off.
I think this behavior must be helpful for the amp, because I've heard from many people that DRC is the ay to go.
Another effect may have to do ith ringing on the rails. If there is ringing through the diode, the sine current ill raise the diode's resistance, and then the resistance ill recover after the ringing is gone. It is like a dynamic auto-damping effect. Furthermore, the nonlinear action orks like an RF mixer to convert the resonance to other frequencies here their energy is lost from the resonance.
Series and parallel resistors have been used to tailor this effect. A small series resistor may be in order to damp any resonances through the diode's parasitic capacitance depending on ho large a diode it is.
So it seems the diode may be having the effect of breaking don rail resonances. This seems useful, but it may also be a sign that the rails are resonating and decoupling needs to be reevaluated. After all the diode can only help resonance after it has started (unless of course Id is chosen beforehand to apply the correct damping resistance).
A silicon diode has effective R of about .033/Id. R decreases ith Id. This is nonlinear, so if you give a diode a sine current, the average resistance ill actually be loer, and the voltage drop of the diode. Thus you have a dynamically changing RC filter. In fault conditions, it prevents frontend rail sag, but not entirely because it pulls some current to turn the diode back off.
I think this behavior must be helpful for the amp, because I've heard from many people that DRC is the ay to go.
Another effect may have to do ith ringing on the rails. If there is ringing through the diode, the sine current ill raise the diode's resistance, and then the resistance ill recover after the ringing is gone. It is like a dynamic auto-damping effect. Furthermore, the nonlinear action orks like an RF mixer to convert the resonance to other frequencies here their energy is lost from the resonance.
Series and parallel resistors have been used to tailor this effect. A small series resistor may be in order to damp any resonances through the diode's parasitic capacitance depending on ho large a diode it is.
So it seems the diode may be having the effect of breaking don rail resonances. This seems useful, but it may also be a sign that the rails are resonating and decoupling needs to be reevaluated. After all the diode can only help resonance after it has started (unless of course Id is chosen beforehand to apply the correct damping resistance).
Today, i connect my SSA to my speaker,
and for the first time i hear a beautiful sound of this Amp. Natural Vocal, nice bass and crisp treble and more important it stable..
You've got it! Well done.
So it looks like the SSA, although a symmetrical CFB amplifier with minimal compensation and LARGE bandwidth, can be built successfully by relatively fresh diy-ers without an explosion and can provide its elegant sound to anyone who wants to build it. IMHO, Naf's experience/experiments with the SSA is a useful and important addition to this thread and will encourage others to throw their prejudices and fears against CFB/symmetrical and embrace an idea that will bring wonders to them.
I am not exaggerating, I and one of my best friends here had an exam yesterday. On the way back I brought him to my house, he has been using AltecLansing's FX5051 5.1 PC speakers for two years now, and is very satisfied with them. I have auditioned those speakers and they indeed are one of the best PC speakers available here, and looks nice with the isobaric subwoofer.
I turned on the SSA and played some random music.
It was funny looking at his expressions.
The comment that came, "it feels like the speaker is way bigger, all over the wall, when I close my ears. And a feeling of life is there, damn I'm having goosebumps".
He spent straight three hours listening to this and that and before leaving his comment was, "This thing is not the average stuff man, it's got a heart. It's only a single small box and is hotter than my monster Altec. What did you put in it?"
Heh, what else could happen. I'll update more experiments with more people. With pictures in future.
you express it very well Shaan...
"it feels like the speaker is way bigger, all over the wall......
![wizard :wiz: :wiz:](https://files.diyaudio.com/forums/images/smilies/wizard.gif)
so who's next newbie diy'er want to build this SSA... you wouldn't regret!
.. six hour past playing music through this amp
![eguitar :eguitar: :eguitar:](https://files.diyaudio.com/forums/images/smilies/eguitar.gif)
![piano :Piano: :Piano:](https://files.diyaudio.com/forums/images/smilies/piano.gif)
![violin :violin: :violin:](https://files.diyaudio.com/forums/images/smilies/violin.gif)
Last edited:
you express it very well Shaan...
so who's next newbie diy'er want to build this SSA... you wouldn't regret!
.. six hour past playing music through this amp![]()
![]()
There should be a text written on SSA sch:
Warning, use carefully, SSA could easily consume a big portion of your precious time.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Solid State
- Simple Symetrical Amplifier