Joseph K said:
My friend Joseph
Any 100 nF capacitor at 1 Mhz , will have a impedance of ~ 1,5 Ohms...
This impedance ,is function of the capacitance...ESR is another issue..
percy said:Carlosfm:
Carlos, why do you think what Upupa said does not apply to Chip amps ? Here I go again...(i love this icon)
As I said a zillion times, I also tested big caps on the chip's pins, bypassed or not.
If you and Upupa still ask the same question, then you miss the whole point, the whole discussion, the point of making such a PSU, the point of going to such a trouble and work.
These are op-amps (yes! 😀 ) and more often then not it's very easy to muck up the sound of an op-amp just because of the PSU layout, the caps used and where you place them.
It is very easy to put an OPA2132 producing horrible sounds and believe me, few can use it correctly.
Why doesn't a signal op-amp need a snubber?
Because you normally use it with small caps, after a regulator.
Put big caps (like 10,000uf) on their pins and try to listen.
I treat the LM chips as op-amps, and I have good results.
They are quite high bandwidth amplifiers and they need attention.
Konnichiwa,
Well, I suspect they could help, but 10,000uF non-inductive Film capacitors are unlikely to fit anything sensible. Wherever sensible I suggest to use Sealed Lead Acid batteries for DIY Soild State Gear.
I just have at work humming to itself an APC 1500VA UPS that I pulled out of the junk (need to get that heavy sucka home yet), which comes +/-24V worth of batteries with plenty of capacity (IIRC 24 AH) for a Gainclone and all the charging logic, plus a nice looking (if large) case.
I only need to find a way to to tap into this whole thing and include an external trigger signal that shuts down the UPS Charger for playing music.
Once you do that sort of thing simply placing a nice low inductance 4.7uF Epcos stacked film cap across the Chip's +V & -V pin and a second from -V to nominal ground and using high capacitance cables to carry your battery current and all our problems simply go away....
Not practical for a commercial Amp, but this here is DIY Audio, ain't it?
My Wife loves them and I am rather foind of her, so i cannot avoid the occasional one. Generally my taste in movies runs more along such ones as Gattaca, Silkwood, Matrix (original one only), eXistenZ, Seven Years in Tibet and anything music based (Beatstreet, Flashdance, Fame to Naim a few).
I am a great anime' fan (you must watch Wind of Amnesia, Ghost in the Shell, Megazone 29, Akira is optinal, among the serials 3X3 Eyes, Armitage III and doomed Megapolis are recommended, Keishin Haidan is also pretty cool) and my wifes stance on these is about as mine on Indian Movies. She's awfully fond of me....
Sayonara
percy said:
Well, I suspect they could help, but 10,000uF non-inductive Film capacitors are unlikely to fit anything sensible. Wherever sensible I suggest to use Sealed Lead Acid batteries for DIY Soild State Gear.
I just have at work humming to itself an APC 1500VA UPS that I pulled out of the junk (need to get that heavy sucka home yet), which comes +/-24V worth of batteries with plenty of capacity (IIRC 24 AH) for a Gainclone and all the charging logic, plus a nice looking (if large) case.
I only need to find a way to to tap into this whole thing and include an external trigger signal that shuts down the UPS Charger for playing music.
Once you do that sort of thing simply placing a nice low inductance 4.7uF Epcos stacked film cap across the Chip's +V & -V pin and a second from -V to nominal ground and using high capacitance cables to carry your battery current and all our problems simply go away....
Not practical for a commercial Amp, but this here is DIY Audio, ain't it?
percy said:
My Wife loves them and I am rather foind of her, so i cannot avoid the occasional one. Generally my taste in movies runs more along such ones as Gattaca, Silkwood, Matrix (original one only), eXistenZ, Seven Years in Tibet and anything music based (Beatstreet, Flashdance, Fame to Naim a few).
I am a great anime' fan (you must watch Wind of Amnesia, Ghost in the Shell, Megazone 29, Akira is optinal, among the serials 3X3 Eyes, Armitage III and doomed Megapolis are recommended, Keishin Haidan is also pretty cool) and my wifes stance on these is about as mine on Indian Movies. She's awfully fond of me....
Sayonara
Kuei Yang Wang said:
On a UPS I can't remember an easier way to disconnect the charger than disconnect the electrical plug from the wall.😀
And disconnect that internal buzzer.
carlosfm said:
You should not forget the wine from Alentejo
An externally hosted image should be here but it was not working when we last tested it.
Everyone should try this wine, it's like the "rolls royce" of the HIFI 😎
But Dão is very, very good! All Portuguese wine is good: clown:
Indian movies???
How the $$$$ did this get so far off track?
Oh, well........might as well make it worse......
Any of our Indian members have a chhole recipe that they care to share? Probably more productive than where this is heading. Thanks to Mr. Creosote for starting yet another pointless thread.
Jocko
How the $$$$ did this get so far off track?
Oh, well........might as well make it worse......
Any of our Indian members have a chhole recipe that they care to share? Probably more productive than where this is heading. Thanks to Mr. Creosote for starting yet another pointless thread.
Jocko
Imagine that I want to build an amp that will sound as I like.
Should I trust the numbers or build it trusting in my instint(ears) ? (to achive the sound that I want)
Can any one difine the "sound" you like in numbers ?
I think the best way is to do it using both ways, so, why not try the snubber ?
I will try the snubber in my amp.... if I like the sound I will keep it no matter what the numbers tells me: dead:
If I don't like it, I will not use it! Simple 😀
Should I trust the numbers or build it trusting in my instint(ears) ? (to achive the sound that I want)
Can any one difine the "sound" you like in numbers ?
I think the best way is to do it using both ways, so, why not try the snubber ?
I will try the snubber in my amp.... if I like the sound I will keep it no matter what the numbers tells me: dead:
If I don't like it, I will not use it! Simple 😀
I'm start to getting convinced of a very compact PS _with_ snubbers but I don't buy things easily as you may have noticed. I maybe even test a snubber! Maybe even something for Pavel!
Copper prism
Hi, What do you mean by copper prism? Very wide copper tracks?
😕
Upupa Epops said:
Hi, What do you mean by copper prism? Very wide copper tracks?
😕
Hi Elso, copper prism is on below side of PCB and by screws is connected to rail voltage. On this prism sit power transistor. Look at mail, I will send you pic.
Come on guys just try the thing, you know you want to.
It takes a lot less time to try this mod than sit there debating about it😀
I've tried it and glad I did, cheers Carlos
It takes a lot less time to try this mod than sit there debating about it😀
I've tried it and glad I did, cheers Carlos
Whoa guys, guess what, you should try snubbing the REGULATED PSU !!! I think it's the best I have gotten. It beats the unregulated snub by a considerable margin. The grass is greener, the dark is darker, the highs cleaner...
regards,
will
regards,
will
My experience with snubbers is positive in every way.
The psu consists, in placement order from the 2 bridges, of: 2*4700uf BC, 2*10000uf nichicon bypassed with polyp 100nf, resistor 1r with 100nf in series to ground, wire-------------------------------(going to the small amp case)inside the amp case: 2*1800uf kme, 1000uf*4 chemicons, snubber 1r 100nf. At the chip pins there is 100uf and 100nf.
In my suboptimal longwire psu setup I found that adding a snubber on both sides of the long wire between amp case and psu helped a great deal. I almost dare to say that my trafo emits less mechanical hum (But I dont dare to say it, seems too odd).. + that I guess that having two snubbers like that must act kind of like 1 snubber with 0,5r and 200nf? I did try to snubb at one point only with those values after all caps, but then the amp sounded bad with the 10000uf caps, and without those the lows suffered a bit.
I was at the soldering iron experimenting with placement orders of caps and snubber values the whole of last week, and this is without doubt the best setup I have hit so far.
I have now decided, after reading this thread, to rebuild the amp in a new larger case so it will be possible to make the psu compact and all of it closer to the chips. Will be fun to see if this renders snubber 1 useless.
This tank thing bothers me, are there any obvious signs of ill behaviour if one creates a tank circuit? I am thinking perhaps my dual snubber is dealing with that considering the cap spree I use?
If not, then I can´t understand how anyone would not note a difference when going snubb. With bussy music like Primal scream's "Kill all the hippies" I found that I could without effort hear high pitch tones embedded in the soundscape and follow them more easily. In a very obvious way.
The psu consists, in placement order from the 2 bridges, of: 2*4700uf BC, 2*10000uf nichicon bypassed with polyp 100nf, resistor 1r with 100nf in series to ground, wire-------------------------------(going to the small amp case)inside the amp case: 2*1800uf kme, 1000uf*4 chemicons, snubber 1r 100nf. At the chip pins there is 100uf and 100nf.
In my suboptimal longwire psu setup I found that adding a snubber on both sides of the long wire between amp case and psu helped a great deal. I almost dare to say that my trafo emits less mechanical hum (But I dont dare to say it, seems too odd).. + that I guess that having two snubbers like that must act kind of like 1 snubber with 0,5r and 200nf? I did try to snubb at one point only with those values after all caps, but then the amp sounded bad with the 10000uf caps, and without those the lows suffered a bit.
I was at the soldering iron experimenting with placement orders of caps and snubber values the whole of last week, and this is without doubt the best setup I have hit so far.
I have now decided, after reading this thread, to rebuild the amp in a new larger case so it will be possible to make the psu compact and all of it closer to the chips. Will be fun to see if this renders snubber 1 useless.
This tank thing bothers me, are there any obvious signs of ill behaviour if one creates a tank circuit? I am thinking perhaps my dual snubber is dealing with that considering the cap spree I use?
If not, then I can´t understand how anyone would not note a difference when going snubb. With bussy music like Primal scream's "Kill all the hippies" I found that I could without effort hear high pitch tones embedded in the soundscape and follow them more easily. In a very obvious way.
Joseph K, I may have missed some posts but I can't see any measurements of yours where I see without this one-ohmer. If you have got the time I would appreciate if you did a measrement with and without the resistor. How big is the difference?
EDIT: Ooops, I have found it. Around 10 dB peak, right.
I have measured today and I can detect a 6 dB peak with 220 uF// 150nF/63V (with DC on) and no peak with 1 ohms is series. I can't detect any visible dip.
EDIT: Ooops, I have found it. Around 10 dB peak, right.
I have measured today and I can detect a 6 dB peak with 220 uF// 150nF/63V (with DC on) and no peak with 1 ohms is series. I can't detect any visible dip.
Hi
Did anyone try to use a regulated power supply, originally used for Son of Zen, to power gainclones?
More precisely:
http://www.passdiy.com/projects/zenv3-5.htm
Have a look at figure 6.
I was thinking about combining this power supply with gainclones, and just add a snubber. Capacitance at the output could be low, maybe just 2x4,7 uF....(with a 100nF and 1 ohm of course)
Any idea how could this sound?
Vix
Did anyone try to use a regulated power supply, originally used for Son of Zen, to power gainclones?
More precisely:
http://www.passdiy.com/projects/zenv3-5.htm
Have a look at figure 6.
I was thinking about combining this power supply with gainclones, and just add a snubber. Capacitance at the output could be low, maybe just 2x4,7 uF....(with a 100nF and 1 ohm of course)
Any idea how could this sound?
Vix
Kuei Yang Wang said:
Thorsten,
You have valid points there, but if you read the paper, they don't claim anything other than that sighted tests introduce other factors that determine you preference other than just sound reproduction. The conclusion then is that if you want to evaluate preferences based ONLY on sound quality, you have to go the blind way. I don't think there is anything wrong with this reasoning, and their tests and documentation of it are pretty good, they don't claim anything they can't back up.
Now, this DOES bring up another interesting question, at least for me. And Mr Moderator, Sir, I know it it OT, but please indulge me.
Mr Toole has to design speakers for Harmann that sell, that's what he is hired for. Looking at the test results, what should he do? Spend more money on stronger magnets for smoother response? Or spend more money on a fancy enclosure finish? Since he has himself concluded that maybe the last option leads to more preferences and thus to more sales.
What would you do, and why?
Jan Didden
Isnt it funne how all efforts to prove the leffect of the snubber are stomped on or simply disregarded?
PEOPLE NOT KNOWING THE SNUBBER, ITS NATURE, PURPOSE OR VERY EXISTANCE HAVE MADE NOTIONS OF BETTER SOUND GC's WHEN APPLIED!!!
there you have it jannemann..
btw: shoudn't you be done with that snubbed GC yet?
or does more extensive testing of the girlfriend have a higher priority?
PEOPLE NOT KNOWING THE SNUBBER, ITS NATURE, PURPOSE OR VERY EXISTANCE HAVE MADE NOTIONS OF BETTER SOUND GC's WHEN APPLIED!!!
there you have it jannemann..
btw: shoudn't you be done with that snubbed GC yet?
or does more extensive testing of the girlfriend have a higher priority?
demogorgon said:Isnt it funne how all efforts to prove the leffect of the snubber are stomped on or simply disregarded?
PEOPLE NOT KNOWING THE SNUBBER, ITS NATURE, PURPOSE OR VERY EXISTANCE HAVE MADE NOTIONS OF BETTER SOUND GC's WHEN APPLIED!!!
there you have it jannemann..
btw: shoudn't you be done with that snubbed GC yet?
or does more extensive testing of the girlfriend have a higher priority?
Marius,
Sorry? Did I comment on snubbers? I'm not aware of that, I think it is an interesting subject, and I learned quite a lot from Joseph's measurements. In fact I used snubbers on my power supplies, described in the article series I did with with WJ and Gary Galo in Audio Amateur in the mid 90'ies. I'm not sure they were actually included in the articles, (too lazy to look it up now), but I think I did at least discuss the advantage of using lossy capacitors (relatively large ESR, which comes out to the same but is cheaper) to damp the oscillations.
Jan Didden
demogorgon said:
Unfortunate choice of words. To prove the effect you need independant, repeatable results. All we have at the moment is opinion. Proof would just get us back into the objective/subjective/blindtesting/goldenears argument yet again.
BTW, I have just installed the recommended snubber on one channel of my OPA548 testbed amp, with no perceptable difference. However I do look forward to trying them on my BGT/PD LM4780 pcb project when my output resistors arrive, as opinion has it that the LM chips respond well to this mod.
Thorsten
You mentioned earlier the use of the Epcos 4.7uF stacked film part on the V+ and V- of the chip. I'm assuming that these are the PET variety but please correct me if I'm mistaken. What characteristic favors this cap over a BlackGate or other liquid/gel variety cap? I also would like your opinion of the SMT part for this application.
You mentioned earlier the use of the Epcos 4.7uF stacked film part on the V+ and V- of the chip. I'm assuming that these are the PET variety but please correct me if I'm mistaken. What characteristic favors this cap over a BlackGate or other liquid/gel variety cap? I also would like your opinion of the SMT part for this application.
- Status
- Not open for further replies.