|
|
Post by gns on Jul 29, 2007 3:31:56 GMT
Mike My understanding is that they are normally only a problem when directly at the output of a voltage regulator. If there is a bit of PCB track between them and the VR, everything should be fine ? SandyK At 100kHz, that bit of track is an inductor! Nice tuned circuit at some frequency or other. Better than the inductive track! Most voltage regulators have sod all ripple rejection at 1MHz and next to sod all at 100kHz. Try using an = or > 0.33uF tantalum across the input with a small resistor in series before that (1-10R). Directly across the output you want a small tantalum (or something like a Panasonic ECQB mylar - minimum 0.1uF, or 0.33uF on a negative VR) which limits any ripple to around 1/1000th of the output signal level of your circuit at 100kHz, then use the small resistor or length of PCB track before the big caps. How do you calculate the tant? Try C = dI x T / dV (C in Farads, I in Amps, T the inverse of frequency, V in Volts), derived from Q = CV = IT (Q = charge in Coulombs). The little d's mean "Delta" - the "dI" difference in current for the "dV" difference in voltage delivered to the load. I'll not do a worked example - don't want to make it too easy...  But for a preamp outputting 1V into 600 Ohms it's 0.016uF, which means use 0.1uF (or 0.33uF on a neg VR). Just about what the manufacturers data says... |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 29, 2007 3:51:50 GMT
Graham
My preference is usually 100nF + 100uF, but as you pointed out , I should increase to 330nF with -ve regulators. I will do so in future.
Anyway, with a typically short PCB track as mentioned, wouldn't that series inductance be so small as to take any tuned circuit to many MHZ ? I am only guessing here !
SandyK
|
|
|
|
Post by gns on Jul 29, 2007 4:29:51 GMT
...with a typically short PCB track as mentioned, wouldn't that series inductance be so small as to take any tuned circuit to many MHZ ? I am only guessing here ! SandyK Yes! Way beyond the regulator's abilities, and we all know what RF does to the sound. Try it the way I suggested and you should hear a more "substantial" sound, or it could even sound louder. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 29, 2007 4:49:48 GMT
O.K. So what about forgetting the -ve regulator , and using another +ve regulator arse about, to take advantage of the much faster +VE versions, as some learned person suggested elsewhere ?
|
|
|
|
Post by gns on Jul 29, 2007 4:53:27 GMT
Maybe, but wouldn't that compromise the ground?
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jul 29, 2007 5:10:00 GMT
Graham I don't know. Perhaps we shpould ask Jan Didden that question ? SandyK quote: Originally posted by Lumba Ogir -Reply to sandyK I do not understand... Low ESR is a desired feature of any capacitor. (Ideally zero at all frequencies.) It can reasonably not itself cause sonic deterioration. Hi, The reason you want some ESR in a reg output cap has to do with the reg stability. The output cap modifies the reg gain/phase characteristic and you NEED some cap to roll off the reg gain before 180deg phase shift. The reg is a feedback amp basically. The small ESR damps any residual oscillatory tendencies at high frequency load currents. This is especially important in low-drop-out regs that use PNP pass transistors (to get the low drop-out). Since PNP's generally (in IC's) have less gain/more phase shift, these regs are more prone to oscillations, and often a minumum REQUIRED ESR is specified. Same goes for 'normal' regs (no low drop-out) for neg voltages, because they also use PNP pass transistors. Neg regs generally are not as good as pos regs at higher freqs. If you have separate pos and neg secondaries on your transformer I would recommend using two pos regs, and connect the "+" output of the "neg" reg to ground, and it's "ground" to neg voltage to the load circuit. Jan Didden __________________ / "Never trust systems without some form of feedback mechanism" - Prof. A J Berkhout. -------------------------------- Meet me at www.linearaudio.nl |
|
|
|
Post by PinkFloyd on Jul 29, 2007 10:38:01 GMT
Why fit a tantalum when a tantalum electrolytic has an even lower ESR than an aluminum electrolytic? Of course you are fitting a resistor in series with it to meet the minimum ESR requirement of the regulator. Tants are good if fast load transient response is important and solid tantalums are better at extremely low temperatures where electrolytics could freeze. The X-CANS (version 1) utilises Panasonic AN7815 / AN7915 regulators.  Co in the X-CANS (v1) is 100uF electrolytic. There is no ringing at all and the amp sounds truckloads better with these caps compared to the Jamicons that were fitted.... Just as the Panasonic FM (low impedance) make the v2 and v3 really cook on gas. |
|
|
|
Post by PinkFloyd on Jul 29, 2007 12:02:59 GMT
....and maybe I should have added......   |
|
leo
Been here a while! 
Team wtf is it?
Posts: 3,638
|
Post by leo on Jul 29, 2007 12:23:19 GMT
Yes, its why we just have to experiment with these caps in any circuit.
Some circuits low ESR caps can sound great yet in some others they sound like crud.
In one of my dacs I only added low ESR caps across the IC's supply pins bypassed with a 1R+10NF , it worked best in this unit.
If regulators are used I normally just check the outputs on a scope to make sure its behaving.
|
|
|
|
Post by gns on Jul 29, 2007 12:37:16 GMT
....and maybe I should have added...... There we go, spouting the Bible again... |
|
|
|
Post by PinkFloyd on Jul 29, 2007 12:40:35 GMT
There's no doubt about it Leo, in all three of the X-Can amps (and buffers) low ESR caps bring about a very noticeable "improvement" there is no doubting it, no doubt at all, the hundreds of people I know of who have tried it all agree categorically that the improvements are not only real but they are bloody HUGE. I sometimes question my ability to hear things but when someone e-mails me and says "WOW!" I know it to be the case  |
|
|
|
Post by PinkFloyd on Jul 29, 2007 12:45:21 GMT
....and maybe I should have added...... There we go, spouting the Bible again... They're your chosen brand, in case you haven't noticed them, Radial 85°C 2000hrs, General Purpose ECA-M series (M) Nice caps in certain areas In other areas low ESR is the ONLY way to go IMO. It's also worth mentioning that the Panasonic FM is also 105C rated so better suited to the environment of the X-Can series of headamps.... you wouldn't believe the amount of bulging 85C GP Jamicons I have seen over the years  Built in obsolescence? possibly..... employed in the design because they're cheap? very probably.... the best cap for the job? Certainly not! |
|
leo
Been here a while!
Team wtf is it?
Posts: 3,638
|
Post by leo on Jul 29, 2007 12:53:27 GMT
Well Mike as long as those regs are not ringing/over heating and the other important thing is they are sounding great then that’s all that matters Several occasions I've had friends fit FM's even Oscons on the regs output and noted a etched quality to the sound, also they noted the actual regs was getting hotter. others have tried FM's and Oscons after a reg and noted they sounded great Sometimes I use 47uf BC037 caps on a regs output or a 10uf tant (never higher than 10uf if it’s a tant) Its all good fun! |
|
leo
Been here a while!
Team wtf is it?
Posts: 3,638
|
Post by leo on Jul 29, 2007 12:56:51 GMT
All Jamicons I've come across was crap! I've also known these to go electrically leaky, one unit I was messing with had Jamicons in the feedback loop, it was noticed that one channel would give an intermittent DC offset, replacing those Jamicons cured it nicely
|
|
|
|
Post by PinkFloyd on Jul 29, 2007 13:11:21 GMT
Well Mike as long as those regs are not ringing/over heating and the other important thing is they are sounding great then that’s all that matters Don't know if you saw this pic Leo: Those are not low impedance Also.... check the v3 board (no regulators per se)  And the v2 board (no regulators per se)  |
|
leo
Been here a while!
Team wtf is it?
Posts: 3,638
|
Post by leo on Jul 29, 2007 13:20:43 GMT
Nice  Ah, those metallic blue sleeved Panny ECA's, the ones classed as audiograde with the special sleeve;) You using the FM's in the heater circuit Mike? should work nicely there |
|
|
|
Post by PinkFloyd on Jul 29, 2007 14:06:48 GMT
You using the FM's in the heater circuit Mike? should work nicely there Yes! Pretty much across the board in the v2 with the exception of the input / output caps which are CapXon non polars (or Nitai dependant on availability) I've got a few Tonerex and Fine Golds (100uF) which I'll try shortly. The v3 uses a mixture of FM / FC / EVOX, again with CapXon or Nitai at the input / output. The X-10 v3 uses a mixture of FA / FM I find the 63V 1000uF FA series sounds better than the FM / FC / EVOX in the X-10 v3 The X-Cans uses a mixture of FM / ECA / EVOX and quite a lot of the caps I have replaced with non polar types (sounds better!) Believe me Leo, I've tried pretty much all there is to try in these units (and really must update my website and remove some of the stuff!) and time and time again the Panasonic caps have proved to be the best sounding overall. From the cerafine mods way back in 1997  To bypassing  To more bypassing   To cap damping  To low impedance 130C  To all out stupid  To insanity  To precarious  To (well...... haven't finished this contraption yet)  To a nice sounding mixture  There have been hundreds of combinations Leo, I can't believe how many there have been... there must also be hundreds of photos over at Head-Fi, photos I can't even remember taking..... a LOT of research and listening has gone into the X-Can amps and I didn't just pick up a bag of Panasonic FC / FM and think "I'll just bung them in for the hell of it" I wouldn't put an FC / FM anywhere near a WNA amp but in the X-Can.... most definitely! All this listening and auditioning different caps has resulted in the tried and tested:  There are a few more photos here but only a very small percentage, I'll get round to uploading them all one of these days www.rock-grotto.co.uk/audiogeddon/index.htmlMental stuff.... I really should get a life ;D |
|
leo
Been here a while!
Team wtf is it?
Posts: 3,638
|
Post by leo on Jul 29, 2007 14:18:34 GMT
That’s brilliant Mike, thanks for sharing the piccys! I'm a bit of a saddo too ;D I enjoy looking at the piccy's, its nice to see others work too. Yes I can imagine all the hours you've put into the X-can mods, the beauty with your kits is that it saves all the work with faffing about with one of these from scratch, as we know all circuits differ and it takes time to get the best out of them. By the way love the insanity pic  |
|
|
|
Post by PinkFloyd on Jul 29, 2007 14:21:12 GMT
You want to try a couple? and, btw, did I send you those caps I promised? I can't remember sending them if I did |
|
leo
Been here a while!
Team wtf is it?
Posts: 3,638
|
Post by leo on Jul 29, 2007 14:37:55 GMT
You want to try a couple? and, btw, did I send you those caps I promised? I can't remember sending them if I did Not sure what I could try those interesting looking insanity caps in, they looked totally different to the normal bypass way though:D If you'd like me to try a couple in something and post the results then sure! You didn't send the VA's yet but no worries Mike, I said no rush and whenever its convenient Nice soldering By the way! |
|
|
|
Post by gns on Jul 29, 2007 14:45:02 GMT
Screechy low Z Electrolytics may sound great with lo-fi CD, but quite a few of my Solo's go into editing suites, so I'll take a rain check on all areas of this forum but the corner where I was shoved...
|
|
|
|
Post by PinkFloyd on Jul 29, 2007 14:48:04 GMT
You want to try a couple? and, btw, did I send you those caps I promised? I can't remember sending them if I did Not sure what I could try those interesting looking insanity caps in, they looked totally different to the normal bypass way though:D If you'd like me to try a couple in something and post the results then sure! You didn't send the VA's yet but no worries Mike, I said no rush and whenever its convenient Nice soldering By the way! Think about all the lead inductance though leo, it would have been a better bet to stitch the film caps onto the bottom of the electrolytic but then you end up with one very "fat" base which would probably foul adjacent components. Remind me of your details and I'll get some packed up right now. Mike. |
|
leo
Been here a while!
Team wtf is it?
Posts: 3,638
|
Post by leo on Jul 29, 2007 15:05:01 GMT
Screechy low Z Electrolytics may sound great with lo-fi CD, but quite a few of my Solo's go into editing suites, so I'll take a rain check on all areas of this forum but the corner where I was shoved... I like those Philips caps but I don't think you can buy them anymore so I normally go for BC's. I've got some Philips axial 121's in one of my hi-fi amps and no other cap came close to those in that for the decoupling and across the VBE multiplier. Wonder why a lot of folks give Tants a bad name, one of my best sounding dacs uses nothing but tants, they sounded more neutral than any others that was tried |
|
leo
Been here a while!
Team wtf is it?
Posts: 3,638
|
Post by leo on Jul 29, 2007 15:12:14 GMT
Not sure what I could try those interesting looking insanity caps in, they looked totally different to the normal bypass way though:D If you'd like me to try a couple in something and post the results then sure! You didn't send the VA's yet but no worries Mike, I said no rush and whenever its convenient Nice soldering By the way! Think about all the lead inductance though leo, it would have been a better bet to stitch the film caps onto the bottom of the electrolytic but then you end up with one very "fat" base which would probably foul adjacent components. Remind me of your details and I'll get some packed up right now. Mike. Thanks Mike, PM on its way. TBH I've not tried a lot of bypassing, normally I use either a single 100nf MMK or just the 1R+10nf depending what its used in, could always use the Canamp as a guinee pig ;D |
|
leo
Been here a while!
Team wtf is it?
Posts: 3,638
|
Post by leo on Aug 3, 2007 11:33:48 GMT
Samwha VA's arrived this morning, taa Mike!
I have just fitted some in my Burson buffer on the decoupling rails, this is one unit I use for trying caps so I fitted some little gold sockets that allows swapping easy.
I have also fitted some in the WTF is it headphone amp, I'll let them run in a while to burn in.
First I'll use the main hi-fi and then try the headphones
|
|
Built in obsolescence? possibly..... employed in the design because they're cheap? very probably.... the best cap for the job? Certainly not!
