peranders said:The snubber for normal people is for removing or reducing the impedance peak, but I'll guess I'll continue wandering in the darkness, Carlos.
So the real purpose of this snubber is... and the contest starts right now.
The snubber reduces the impedance over a broad frequency range. Point.
No peak in the game. The peak can be attenuated with a snubber, but it's not very effective (doesn't provide much attenuation there).
If I wanted to reduce the peak I would point to an exact (and narrower) frequency, but that would just be effective there.
And for that, one additional passive component is missing.
If you are smart you will get there. Maybe one day.
Just don't ask me too much questions because I don't feel like sponsoring your commercial ventures.
Jung regs, anyone? 😀
Mick_F said:
Yeah, I fail to understand that too.
It's even on his footer. Wrong, must I say.
This guy spreads this over at forums around the world, in several different languages.
I was going to say something related to computer viruses, but that would send me directly to Texas. 😀 😀 😀
... and I fail to see that this is true. It's only reduces a peak in a rather limited range, very limited I would say. Theory says so and also measrements in particular those made by Joseph_K. Nevertheless I have made a small test but then I have used SMD resistors and groundplane.carlosfm said:
carlosfm said:
Not trying to be a wise-*** but what do you call a 'broad frequency range' ? 10Hz, 1kHz, 10kHz ? I'm just trying to figure out why, if snubbers indeed have a broad freq. range impedance reducing effect, component values are so important in designing snubbers for different psu's.
In my experience, trial and error for component values gives me better results than calculating optimum component values and sticking with those. But that may be a difference between theory and real life.
-- in theory, there's no difference between theory and real life, but in real life, there is... -- 😉
Thanks,
Matt
matjans said:
Indeed they are, as you can see on my regulated snubberized PSU.
If you discount the peak(s), there's a lower impedance from the audio band up to the Mhz.
matjans said:
That's why I went changing values over the time.
The first values I proposed were already an improvement, but still not enough.
And yes, I was a little surprized by these 'final' values I arrived to, specially the small cap value, as I always had the feeling that lowering the resistor to below 1R would improve things.
You need to know what you're after, and then make the calcs, but the final test is always listening.
I can't imagine the development of a really good amp without listening tests.
And the listening tests are part of the development process.
I can't conceive things in any other way.
peranders said:
there's a problem applying "small signal stability analysis" to something which isn't a small signal --
Somebody will have to try really hard to convince me that a single bridge with a single 10,000uf cap per rail (bypassed or not) is a good PSU.
Not even talking about low(er) capacitance.
This has been lots of fun.
😀
Not even talking about low(er) capacitance.
This has been lots of fun.
😀
carlosfm said:
Your response is meant to deflect the impact of the real issue.
Also I have seen no evidence that the snubber itself causes lower impedance at audio frequencies. JosephK's graph showed that for his rig that was the case, but remember there are differences beyond just the snubber to consider.
The only way to get a clear picture would be to test two rigs where the only difference is the lack of the snubbing resistors in the non-snubbed PS, and work from there.
I just want to see more evidence, and not anecdotal or subective listening evidence either, as I have learned that healthy suspicion of subjective analysis and opinion is always wise.
Cheers!
Russ
Do you have any measurements of this? I thought the whole point of snubbers was to dampen out (resonance) peaks in the output impedance of a psu.
I guess "they" are you and I who still have questions to be answered. 🙂
So I think the "they" or the "we" if you will are those to whome the answer "build it and listen" is not the end of the discussion.
I just want more research and facts, purely because it entertains and interests me.
Cheers!
Russ
So I think the "they" or the "we" if you will are those to whome the answer "build it and listen" is not the end of the discussion.
I just want more research and facts, purely because it entertains and interests me.
Cheers!
Russ
Russ White said:
This is a never ending story because you guys are all lazy.
I have said 1000000 times that I have NO INTEREST in saying more than what I have already said.
The context of these particular PSUs are for use on a power amp, so BUILD ONE, LISTEN, MEASURE, wharever.
There are other applications, which I have used with success, but I'm too tired and some of you guys don't deserve to know more.
I'm not your employee and I don't take orders.
I have much more things to do.
I'm sick of the 'commercial' brigade, those are the ones that want to know more and will follow me wherever I go.
I don't give a ****.
PS: this is not for you, matjans.
Lazy.... Wow Carlos. You seem to throw accusations around like that a lot.
I have built/listened/tested and so have others, and my conclusions (and those of others) simply seem to be different than yours.
Did someone offend you by actually testing this idea of yours?
Odd that.
I have built/listened/tested and so have others, and my conclusions (and those of others) simply seem to be different than yours.
Did someone offend you by actually testing this idea of yours?
Odd that.
Hi all,
What would actually be lazy would be simply take it for granted that a snubberized supply is always better than any other alternative simply because a few people say so. 🙂 How much more lazy could you get? Blind faith?
I would never do that, I build, I listen, I measure, I audition for others who are even more critical than I. Then I try discover why things happen so I can repeat the results in a different context. I also share my knowlege with my friends which is the right thing to do, but I try to do so in a way that does not make anyone feel stupid, or that they just don't get it.
Carlos, I take exception to your assertions that those who question your approach to this PS simply don't "get it". That is hogwash, but even if it were true, the best thing to do would not be to call people names or say they are blind. It would be to lead them to the knowlege you think they lack, even if it is just a point in the right direction.
Carlos, I respect you very much and nobody could value your contributions more than I do. So, lets just talk about the technical topic and the effects of the snubber and leave out the petty personal jabs out OK?
If you have shared all you will and will share no more then I have learned all you have to teach (which is a pitty) so far, and that is fine, but I certainly thank you. I shall seek more knowlege elsewhere and think kindly of you for your contribution to my education. After all nobody simply manufacturers knowlege on their own, we all learn it from somewhere even if it is from nature itself, and that is not lazy. 😎
Cheers!
Russ
What would actually be lazy would be simply take it for granted that a snubberized supply is always better than any other alternative simply because a few people say so. 🙂 How much more lazy could you get? Blind faith?
I would never do that, I build, I listen, I measure, I audition for others who are even more critical than I. Then I try discover why things happen so I can repeat the results in a different context. I also share my knowlege with my friends which is the right thing to do, but I try to do so in a way that does not make anyone feel stupid, or that they just don't get it.
Carlos, I take exception to your assertions that those who question your approach to this PS simply don't "get it". That is hogwash, but even if it were true, the best thing to do would not be to call people names or say they are blind. It would be to lead them to the knowlege you think they lack, even if it is just a point in the right direction.
Carlos, I respect you very much and nobody could value your contributions more than I do. So, lets just talk about the technical topic and the effects of the snubber and leave out the petty personal jabs out OK?
If you have shared all you will and will share no more then I have learned all you have to teach (which is a pitty) so far, and that is fine, but I certainly thank you. I shall seek more knowlege elsewhere and think kindly of you for your contribution to my education. After all nobody simply manufacturers knowlege on their own, we all learn it from somewhere even if it is from nature itself, and that is not lazy. 😎
Cheers!
Russ
Russ White said:
You didn't test THIS PSU, and this PSU is also the recommendations for the local bypass on the chip.
When someone tested ON your/Mauro's amp and reported good results you guys said he was listening to things and had a strong imagination.
Btw, something's beginning to happen on the class D forum, and it's not my fault.
It's against the designers' recommendations, but what can you do to avoid people testing and trying to improve?
I don't care about your oppinion, Russ.
You know why?
Because you sell your amps, of course they are the best.
Listening to your oppinion about your amps (the best invention since the weel) is like reading a commercial on the back cover of an audio mag.
Biased?
Noooooo... 😀
Don't worry, you are not different from others.
You are quite right, Carlos can't present anything which proves his claims. The resistor _only_ reduces an impedance peak in the MHz region, nothing more. As said before here, small signal analysis, both in measurements and simulations.Russ White said:
I have made some measurement and notice a small peak at a few MHz and this peak can be eliminated.matjans said:
Russ White said:
This member asked my oppinion:
http://www.diyaudio.com/forums/showthread.php?postid=736534#post736534
Look at the next post, after me, who came in immediately?
For the 1000th time, Mr. Per-Anders.
Now look at the reaction of that member. He's right.
There is no patience.
peranders said:
It can be attenuated, and you must calculate the values for that.
But that's not my main objective, and that's not what gives me the best results.
If it's yours, go ahead.
But hey, you are sick about the deam snubber .
Since the beginning.
That means something.
- Status
- Not open for further replies.