Hi Gary
just a finding from my side. between the 30v and and 36V that i have tried (on the Rev C)
the 30 V accentuated the top mids and highs to a certain extend. when I tried the 36V it was more like my class A amps. with a better spread and less "bite" the same as a snubber will do on the GC amps.
just my 2 cents worth.
Cheers
Rudi
just a finding from my side. between the 30v and and 36V that i have tried (on the Rev C)
the 30 V accentuated the top mids and highs to a certain extend. when I tried the 36V it was more like my class A amps. with a better spread and less "bite" the same as a snubber will do on the GC amps.
just my 2 cents worth.
Cheers
Rudi
Russ White said:
No, I did not say that.
What I meant is that Gary picked on an existing design and tried to improve it.
'Thinking by it's own head' is not just strictly following what the designer made or recommends, it is thinking in ways to improve it.
I knew I would be badly interpreted.
Although I did not try this amp, I've been there, done that (related to PSU) and I know how to improve it.
There's no miracle bypassing arrangement that solves the problem of the big cap's ESL and the unavoidable trace inductance (in this amp the traces from the PSU caps to the bypasses are long).
There are ways to solve this, but as the climate got heavy and I don't have the patience and time for this anymore, I'm out.
PS: I'm not interested in trying the amp.
I have other projects in hands right now, and I'm very happy with my amp.
But I seriously recommend those who have built it to try 0.1R+3.3nf snubber across the last caps, on the chip.
And increase those caps to at least 1,000uf (or even better 2,200uf), if it fits.
Enjoy.
Carlos - take it easy. Absolute statements like this is what can cause other members to object to the tone of your posts. Being absolute in the audio world is not possible as what works in one system may not work in another and as per your own reasoning everyone should "use their own head".
I think it is great that these sorts of suggestions are made, however in this particular case they need to leave room for the fact that you have no experience with this particular amplifier. In general comments should leave room for others to express their opinions. It should also be remembered that all you present here is your own opinion and it should be presented in that light
Cheers & thanks for your suggestions.
Byrd said:
What absolute statement?
You and Russ must be dreaming.
Byrd said:
I have LOTS of experience with the LM3886 chip, and I know it is very demanding of a good PSU arrangement.
Whatever you do with it, the PSU is the most important thing here.
Hey, even Russ has taken my advice about op-amp bypassing on the preamp boards. Go and see what he is selling. Exactly as I told.
Wide bandwidth amplifiers? You bet. Some treat them as 30 year-old opamps.
Seems like snubberizing has become a taboo because many don't understand it.
I will not open my book, and I have clear reasons not to do so.
Just try what I suggested, that's it.
I leave you with what Phred told me, after watching intense and absurd fights here bacause of the snubber:
Fred Dieckmann:
"You have just scatched the surface of a most interesting and very important aspect of decoupling power supplies for high bandwidth amplifier circuits and most of these idiots just don't get it. I do get it."
This is absurd Carlos, are you saying you are the only one here who hold the knowlege of snubbers in power supplies?
Do you really claim that we are all so stupid as to not have thought about the technique? If so you are even more arrogant then any of us thought.
You prove your contempt for our design process in your posts. While it is true I have taken advice about bypassing, it is not true it originated solely from you. In fact the first reference I have is from the OPA627 datasheet.
It is not that we have overlooked or somehow missed or forgotten the snubber for this application. Lets be clear, it is not in the circuit because it does not need to be there. Period.
Now let me get back to my "dreaming", and leave the discussion who really care about the core circuit as you promised you would.
I think you do not open your book because you do not wish your ideas to bear the scrutiny of the group as those of us who are more open with our ideas do. Such a tactic is not the sign one who really wishes to enlighten, but more of one who wishes to demogogue.
That is most definitely not "enlightening". Snubbers are old news, while I do now undersntand them in theory and in practice, I also know when they are not practical. I do not claim to be an expert, but I am no rank beginner.
I am sure I could learn much from you if you would just participate in a way that people could respect. But as it is its more constructive to tune you out.
Now enjoy the rest of the weekend.
Cheers!
Russ
Do you really claim that we are all so stupid as to not have thought about the technique? If so you are even more arrogant then any of us thought.
You prove your contempt for our design process in your posts. While it is true I have taken advice about bypassing, it is not true it originated solely from you. In fact the first reference I have is from the OPA627 datasheet.
It is not that we have overlooked or somehow missed or forgotten the snubber for this application. Lets be clear, it is not in the circuit because it does not need to be there. Period.
Now let me get back to my "dreaming", and leave the discussion who really care about the core circuit as you promised you would.
I think you do not open your book because you do not wish your ideas to bear the scrutiny of the group as those of us who are more open with our ideas do. Such a tactic is not the sign one who really wishes to enlighten, but more of one who wishes to demogogue.
That is most definitely not "enlightening". Snubbers are old news, while I do now undersntand them in theory and in practice, I also know when they are not practical. I do not claim to be an expert, but I am no rank beginner.
I am sure I could learn much from you if you would just participate in a way that people could respect. But as it is its more constructive to tune you out.
Now enjoy the rest of the weekend.
Cheers!
Russ
No, you do not understand it.
If you took care to see my original thread, you would see a link to the tnt-audio article right on my first post.
So, this observation is silly.
That article doesn't say much, I investigated on my own and introduced this here.
It's not a technique for chipamps, not even for power amps. It's for general PSU design, unregulated or regulated.
It is all very strange for most here.
Otherwise there will be no fight.
You think what you want.
Again, I was just trying to help ONE member. Those who are afraid to try, stay out. fine with me.
I find it very unconvenient that you kick off everyone who tries it.
That's not true, and the fact is that AFTER I pointed out the flaws, you presented a new PCB layout.
It would be honest for you to recognize this, but forget it.
I DON'T WANT you to put the snubber on your PCB.
I'm glad you didn't.
It would not be nice from your part to do it.
But let the guys try, if they want. Period.
It seems that you guys found perfection and don't want any oppinion. You expulse anyone that makes a simple question or reports an improvement.
This is very arrogant.
I never claimed perfection or the 'ultimate' amplifier.
P-A's LM3886 amp was also very 'professionally' designed and then I started receiving messages on my PM reporting big improvements with the snubbers. As I predicted and was very badly interpreted.
I have been quiet on this thread. But someone would have go and try it. I don't need to make anything.
I promised?
Where?
The core is the PSU.
The problem is: I have my reasons.
Seems like you don't follow this from the beginning.
Now you became another reason.
Well, I don't know what are you talking about.
Again, anyone who points out a possible improvement to the design gets kicked out.
Even more absurd, a guy that reports improvements with the snubber is accused of hearing things and having strong imagination, changing the trafo (which he didn't)...
And how do you do that?
I'm not 'tunable'.
Cheers
Russ White said:
If you took care to see my original thread, you would see a link to the tnt-audio article right on my first post.
So, this observation is silly.
That article doesn't say much, I investigated on my own and introduced this here.
It's not a technique for chipamps, not even for power amps. It's for general PSU design, unregulated or regulated.
Russ White said:
It is all very strange for most here.
Otherwise there will be no fight.
You think what you want.
Again, I was just trying to help ONE member. Those who are afraid to try, stay out. fine with me.
I find it very unconvenient that you kick off everyone who tries it.
Russ White said:
That's not true, and the fact is that AFTER I pointed out the flaws, you presented a new PCB layout.
It would be honest for you to recognize this, but forget it.
Russ White said:
I DON'T WANT you to put the snubber on your PCB.
I'm glad you didn't.
It would not be nice from your part to do it.
But let the guys try, if they want. Period.
It seems that you guys found perfection and don't want any oppinion. You expulse anyone that makes a simple question or reports an improvement.
This is very arrogant.
I never claimed perfection or the 'ultimate' amplifier.
P-A's LM3886 amp was also very 'professionally' designed and then I started receiving messages on my PM reporting big improvements with the snubbers. As I predicted and was very badly interpreted.
I have been quiet on this thread. But someone would have go and try it. I don't need to make anything.
Russ White said:
I promised?
Where?
The core is the PSU.
Russ White said:
The problem is: I have my reasons.
Seems like you don't follow this from the beginning.
Now you became another reason.
Russ White said:
Well, I don't know what are you talking about.
Again, anyone who points out a possible improvement to the design gets kicked out.
Even more absurd, a guy that reports improvements with the snubber is accused of hearing things and having strong imagination, changing the trafo (which he didn't)...
Russ White said:
And how do you do that?
I'm not 'tunable'.
Cheers
Gary, I hope you know I do not dispute your opinion, nor your experience. I am sure you know what you are doing, and I welcome the experimentation. I also welcomed your post, I was adressing you respectfully and I believe you adressed me in the same way. So please don't be put off by the useless noise.
Peace.
I should know better than to feed the troll.
Peace.
I should know better than to feed the troll.
Re: Re: No, you do not understand it.
You learned a lot on this thread.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=51803&perpage=10&pagenumber=1
In fact, it was not you. I'm sorry.
You appeared in the middle and posted a very similar board to what others were doing, following my advices.
See as you have no clue what I was talking about?
I didn't mean the film bypasses, everyone does that and that's on the datasheets.
I mean the electrolythics on each chip's PSU pins.
I'm not caustic.
And I don't like that things turn out to this path.
But please, let people try if they want.
Scrutinity is every day.
What's the matter, do you really think the amp can't be improved?
I'm sure it can.
But I can't try every amp on the planet.
I'm also being pushed to try class D amps, but right now I have no time...
One thing you can be sure: everything is improvable and I would not just assemble the amp. I would not sit quiet.
Cool down, man.
Russ White said:
You learned a lot on this thread.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=51803&perpage=10&pagenumber=1
In fact, it was not you. I'm sorry.
You appeared in the middle and posted a very similar board to what others were doing, following my advices.
Russ White said:
See as you have no clue what I was talking about?
I didn't mean the film bypasses, everyone does that and that's on the datasheets.
I mean the electrolythics on each chip's PSU pins.
I'm not caustic.
And I don't like that things turn out to this path.
But please, let people try if they want.
Scrutinity is every day.
What's the matter, do you really think the amp can't be improved?
I'm sure it can.
But I can't try every amp on the planet.
I'm also being pushed to try class D amps, but right now I have no time...
One thing you can be sure: everything is improvable and I would not just assemble the amp. I would not sit quiet.
Cool down, man.
Russ White said:
Russ,
Yes, I can say that you have always been courteous to me and I've tried to return the courtesy. And I am grateful for all the work you've put into this project. You've done a great service to a lot of folks by providing these boards and your support of the project.
With that said, I have to admit that I find the entire dynamic between you and Carlos incomprehensible. There must be a lot of history here - you seem to read a lot more into Carlos' messages than I see and seem to take offense at things that seem pretty harmless to me.
If I can offer a word of advice, I'd suggest that we drop the topic entirely or at least we have a 24 hour cool down period where Russ and others agree not to address any mesages to Carlos.
With best intentions,
---Gary
Carlos, please tell me when I discouraged someone from trying anything?
I encourage people to try anything they like.
I never said anything to Gary that could possibly be construed as discouraging him from his efforts.
Also, electrolytics arranged as I have are no special prior art, but if you would like to take credit by all means... take it. You seem to crave that.
I invite all discussion about all things, but especially those things which can be illustrated by proofs and documentaion and which can be repeated and demonstrated in fact to be beneficial for a given application, and not just some persons opinion or suggestion.
Keep it real, and show your work. Otherwise people have no reason to trust you actually know what you are doing. I certainly don't. you have done nothing but show values for components with no logic given for the choices, or proofs for the benefit.
The choice to omit the snubber was not one of ignorance or inexperience, it has been tried, and rejected. Not because it is "bad" (which would be subjective), but because it adds nothing (nothing tangible or proven) electrically beneficial to this circuit.
This is not just my opinion, but that of those who have designed, built and tested and listened to the amp with snubbers and without. Yet you seem to insist that you know better... Really, do you? Prove it.Honestly I want to believe you, but I don't. Not yet.
Of course, there is no accounting for personal taste both in regard to audio and in regard to the satifaction one may bet by soldering extra doo-dads on their PCBs (which I am sure I am guilty of doing myself to be honest).
The benefit is not "obvious" as you pretend. And there is nothing of original thinking in considering the snubber, it has been covered in detail in the past.
You hurt your case for the snubber by insisting that any LM3886 high cap will be better with a snubber.
It is quite certainly a ridiculous assertion and only harms your credibility. That is because "better" is a subjective term. It may be better to many, but it also may be worse to many based on any number of criteria including practical considerations.
Of course anything can be made better, but making something more complex does not necessarily make it better. In this case it is my educated and experience opinion, that is all a snubber in this application accomplishes.
Now please back your assertions up with facts, objective facts, and relieve us all.
Cheers,
Russ
I encourage people to try anything they like.
I never said anything to Gary that could possibly be construed as discouraging him from his efforts.
Also, electrolytics arranged as I have are no special prior art, but if you would like to take credit by all means... take it. You seem to crave that.
I invite all discussion about all things, but especially those things which can be illustrated by proofs and documentaion and which can be repeated and demonstrated in fact to be beneficial for a given application, and not just some persons opinion or suggestion.
Keep it real, and show your work. Otherwise people have no reason to trust you actually know what you are doing. I certainly don't. you have done nothing but show values for components with no logic given for the choices, or proofs for the benefit.
The choice to omit the snubber was not one of ignorance or inexperience, it has been tried, and rejected. Not because it is "bad" (which would be subjective), but because it adds nothing (nothing tangible or proven) electrically beneficial to this circuit.
This is not just my opinion, but that of those who have designed, built and tested and listened to the amp with snubbers and without. Yet you seem to insist that you know better... Really, do you? Prove it.Honestly I want to believe you, but I don't. Not yet.
Of course, there is no accounting for personal taste both in regard to audio and in regard to the satifaction one may bet by soldering extra doo-dads on their PCBs (which I am sure I am guilty of doing myself to be honest).
The benefit is not "obvious" as you pretend. And there is nothing of original thinking in considering the snubber, it has been covered in detail in the past.
You hurt your case for the snubber by insisting that any LM3886 high cap will be better with a snubber.
It is quite certainly a ridiculous assertion and only harms your credibility. That is because "better" is a subjective term. It may be better to many, but it also may be worse to many based on any number of criteria including practical considerations.Of course anything can be made better, but making something more complex does not necessarily make it better. In this case it is my educated and experience opinion, that is all a snubber in this application accomplishes.
Now please back your assertions up with facts, objective facts, and relieve us all.
Cheers,
Russ
Hi Gary,
I agree.
I am sorry I dragged this on.
Once again, I encourage you tests, I would just recommend you try to do direct A/B tests if at all possible to be sure you percieved benefits persist.
I have found that I here things in changes I make to circuits which other percieve as better, but I may percieve as worse, or the other way around.
Our ears forget quickly.
Cheers,
Russ
I agree.
I am sorry I dragged this on.
Once again, I encourage you tests, I would just recommend you try to do direct A/B tests if at all possible to be sure you percieved benefits persist.
I have found that I here things in changes I make to circuits which other percieve as better, but I may percieve as worse, or the other way around.
Our ears forget quickly.
Cheers,
Russ
Russ,
I agree that having two versions of an amplifier is ideal for understanding how changes affect the sound.
I don't have that, but I do have several amplifiers which remain unchanged and that I use as references for comparison. These are a pretty standard LM3875 gainclone with (forgive me) a power supply along the lines suggested by Carlos in his hi cap snubberized thread. And I also have a nice set of Class D monoblocs out of japan from a company called Flying Mole. Swapping between these and the My Ref amp gives me a good means of triangulating any changes as I tweak things. And I continue to think that the addition of snubbers did help this amp, at least in conjunction with the increase in capacitor size at the C1, C2, C17, C18 positions.
As always, your mileage may vary.
Regards,
---Gary
I agree that having two versions of an amplifier is ideal for understanding how changes affect the sound.
I don't have that, but I do have several amplifiers which remain unchanged and that I use as references for comparison. These are a pretty standard LM3875 gainclone with (forgive me) a power supply along the lines suggested by Carlos in his hi cap snubberized thread. And I also have a nice set of Class D monoblocs out of japan from a company called Flying Mole. Swapping between these and the My Ref amp gives me a good means of triangulating any changes as I tweak things. And I continue to think that the addition of snubbers did help this amp, at least in conjunction with the increase in capacitor size at the C1, C2, C17, C18 positions.
As always, your mileage may vary.
Regards,
---Gary
Russ White said:
I have shown what I wanted to show, and there is logic on the selected values.
It is obvious for me that you didn't try my last recommended PSU arrangement, but you are free not to.
The original 1R+100nF values can be much improved, those were the initial values based on my initial tests.
I'm also certain that you didn't follow my threads, because I explained (up to what I want to reveal) what this is about.
There was even someone that did understand and started calculating. He was on the right path.
So, it's not as vague or as subjective as you say.
One of my threads reveals it quite well, but some are really blind.
The thing is: if you don't want to believe, you don't.
If you don't want to investigate, to think by yourself, you don't.
I have posted several schematics (amps, PSUs, etc.), with values that do work and lots of guys are already happy that way.
When you (and others) start pushing it really gets me into thinking that there's other interests behind.
I prefer not to give everything, and I don't care if you or others believe it or not. I have nothing to loose, nothing to gain.
Black magic? I don't believe it either.
Now tell me, have you offered something that YOU designed? Alone?
I can't see...
Doesn't matter, anyway.
I'll try to stay out of this thread and let you guys play.
PS: don't make me search your threads to point you out where you were rude to me and to others.
You believe in this design blindly because you were told all the arguments to defend it.
Are they all right? Some are, some don't.
I go a way for the weekend to return to this bickering and fighting?
Please GENTLEMEN... Shake hands and walk away.
You don't have to like each others work, and you don't have to like each other, but you both are representatives of this hobby and you DO have a responsibility to maintain it in a good light.
Notice no finger pointing, just the request to end the negativity.
Please GENTLEMEN... Shake hands and walk away.
You don't have to like each others work, and you don't have to like each other, but you both are representatives of this hobby and you DO have a responsibility to maintain it in a good light.
Notice no finger pointing, just the request to end the negativity.
Time for amother thread
Maybe we could start another thread about picking values and trying these. The values are related to the power transformer, diodes, and circuit.
I suspect the elcheapo trans used [n mine has a lot more inductance and leakage capacitance than a good toroid. This may be a good thing. Several very knowledgable people prefer EI over toroids because the lossy nature acts as a filter.
Long traces running parallel across the board have a fair resistance and capacitance. They may be a built in snubber of sort also.
I like reading discussions where ideas get bounced around. Who knows I might pick up something usefull.
After reading the Hagerman article I feel better using AB carbon comp resistors in my projects. They were picked because they are best as grid stoppers. This is the same effect, tube input capacitance causing ringing.
George
bg40403 said:
Maybe we could start another thread about picking values and trying these. The values are related to the power transformer, diodes, and circuit.
I suspect the elcheapo trans used [n mine has a lot more inductance and leakage capacitance than a good toroid. This may be a good thing. Several very knowledgable people prefer EI over toroids because the lossy nature acts as a filter.
Long traces running parallel across the board have a fair resistance and capacitance. They may be a built in snubber of sort also.
I like reading discussions where ideas get bounced around. Who knows I might pick up something usefull.
After reading the Hagerman article I feel better using AB carbon comp resistors in my projects. They were picked because they are best as grid stoppers. This is the same effect, tube input capacitance causing ringing.
George