I'll just agree to disagree with you. 🙂 No ill feelings.
I did look at my latest batch of modules though and you are right about Hypex changing the caps. The ones on the original modules had 50V 470uF caps. The newer modules have 63V rated caps. I'm sure the higher safety margin is a good thing for general use. That doesn't preclude people using a 50V part there though when they know that they have the safety margin on the power supply rails.
Just my 0.02 I don't want to ruffle any feathers.
I did look at my latest batch of modules though and you are right about Hypex changing the caps. The ones on the original modules had 50V 470uF caps. The newer modules have 63V rated caps. I'm sure the higher safety margin is a good thing for general use. That doesn't preclude people using a 50V part there though when they know that they have the safety margin on the power supply rails.
Just my 0.02 I don't want to ruffle any feathers.
Kevin Haskins said:
No ill feelings at all, no feathers ruffled, no worries here. There shouldn't be, but I don't think you're disagreeing with me now either.
Most just want a quick recipie so they come here looking for one and if they don't know enough to limite their rail voltage for a 50 volt cap it could have some bad consequences, the majority, those who would need a thread like this to tweak their amps to begin with, will be monkey see monkey do, so if you dont' tell them about limiting rails..
You know I should thank you. I took the caps off my module in order to read their values as I was pretty sure they were 63 volts.
I stuck them on my homebrew amp and now there's next to no 60Hz hum just a faint hiss, burried the noise floor on it. Really brought out the detail, but I do hear that veil others mentioned with them.
questionable caps
Chris,
The only point I was trying to make is that caps really aren't that fragile if they don’t get hot, especially the 105 deg types. Over voltage is not what usually kills a cap but ignoring the ripple current rating will do it every time. The rule of thumb or finger (with your other hand in your pocket away from ground) feel the caps and if they run much above ambient they won’t last as long and if you can't keep your finger on it, it will probably vent in short order. What happens is the high temperature causes the electrolyte to be forced out and the cap eventually dries out causing the ESR to rise dramatically. This in turn causes the temperature to also rise dramatically destroying the part and making a mess.
The veil may go away as you get some time in, let us know.
Roger
Chris,
The only point I was trying to make is that caps really aren't that fragile if they don’t get hot, especially the 105 deg types. Over voltage is not what usually kills a cap but ignoring the ripple current rating will do it every time. The rule of thumb or finger (with your other hand in your pocket away from ground) feel the caps and if they run much above ambient they won’t last as long and if you can't keep your finger on it, it will probably vent in short order. What happens is the high temperature causes the electrolyte to be forced out and the cap eventually dries out causing the ESR to rise dramatically. This in turn causes the temperature to also rise dramatically destroying the part and making a mess.
The veil may go away as you get some time in, let us know.
Roger
Hi,
I realize that once it starts to heat and vent it's a chain reaction. I think the reason it's heating and venting is all a byproduct of reforming rapidly which breaks down the dielectric, punches holes in the foil, leakage increases until an internal short occurs which allows enough current to flow to boil the electrolytic.
You might get away with overvoltage a few times without too much damage but with too little tolerance it's a time bomb in my view.
The dried out cap occurs well after the damage has been done brought on by overvoltage. Anyway we agree it's cool if you limite your rails and keep a healthy tolerance.
Theil veil is a kicker it's pretty unique. I almost dont' mind it but there's some some mild stiffling. Not so far to be unpleasant though. It's extremely dynamic now.
I'm not happy with how I put them in there though and there's some cheap ceramics in place as well, I'm going to cut them out and beef up the decoupling, might even try a snubber if I can find the parts I think I can. Let you know that works.
Regards,
Chris
I realize that once it starts to heat and vent it's a chain reaction. I think the reason it's heating and venting is all a byproduct of reforming rapidly which breaks down the dielectric, punches holes in the foil, leakage increases until an internal short occurs which allows enough current to flow to boil the electrolytic.
You might get away with overvoltage a few times without too much damage but with too little tolerance it's a time bomb in my view.
The dried out cap occurs well after the damage has been done brought on by overvoltage. Anyway we agree it's cool if you limite your rails and keep a healthy tolerance.
Theil veil is a kicker it's pretty unique. I almost dont' mind it but there's some some mild stiffling. Not so far to be unpleasant though. It's extremely dynamic now.
I'm not happy with how I put them in there though and there's some cheap ceramics in place as well, I'm going to cut them out and beef up the decoupling, might even try a snubber if I can find the parts I think I can. Let you know that works.
Regards,
Chris
classd4sure said:
Chris,
I wanted to check in and see if there was any final conclusion on snubbers. Bruno used a high ESR cap at the output devices as a snubber. Also there is the .05 ohm in series from there to the 470uf cap to decouple the outputs further and isolate them from any following series inductive ringing. A snubber was recommended at the bulk storage cap to terminate any ringing caused by the inductance of the connecting wiring. Values were in the range of .1uf - .47uf and the resistor was in the .1ohm to 2ohm range. Tuning this would have to be a very individual thing as each setup would nave different lead inductance. What I am hopping for is a general starting point to save time. I would think that even if the snubber values were off by a lot it still would have to be better than no snubber at all.
Roger
Hi,
Yeah I posted it to this thread:
http://www.diyaudio.com/forums/showthread.php?s=&postid=735837#post735837
I didn't experiement with extra resistors but I'm sure it would help, or possibly an inductor.
All the same I found that the bigger decoupling caps have a dramatic affect on response, too much is not good, too little is not good, something in between flattens it right out.
The snubbers I used were 10nF after the 470uF local decoupling cap, parallel to 10nF + two five ohm resistors in parallel, best I could do.
There's songs like the cover of Tainted Love by marilyn manson that weren't worth playing before and are outright impressive to listen to now, the delays and finer detail are right there and seems to really open up the sound very naturally, better dynamics.
I'm sure tweaking the snubber would have some effect to further enchance this but it seems you can't go wrong with even a wide variation over the standard values given.
I'm sure doing it blindingly wouldnt' improve the snubber that's on the module already, but I see how better local decoupling caps can make a huge difference, I'd probably want to try half a dozen there at least, and expect a freakshow.
I think the results expected by snubbing the PSU caps would vary with the quality of the cap since that's what it's compensating for, there may not be much to compensate if you have really good caps.
Regards,
Chris
Yeah I posted it to this thread:
http://www.diyaudio.com/forums/showthread.php?s=&postid=735837#post735837
I didn't experiement with extra resistors but I'm sure it would help, or possibly an inductor.
All the same I found that the bigger decoupling caps have a dramatic affect on response, too much is not good, too little is not good, something in between flattens it right out.
The snubbers I used were 10nF after the 470uF local decoupling cap, parallel to 10nF + two five ohm resistors in parallel, best I could do.
There's songs like the cover of Tainted Love by marilyn manson that weren't worth playing before and are outright impressive to listen to now, the delays and finer detail are right there and seems to really open up the sound very naturally, better dynamics.
I'm sure tweaking the snubber would have some effect to further enchance this but it seems you can't go wrong with even a wide variation over the standard values given.
I'm sure doing it blindingly wouldnt' improve the snubber that's on the module already, but I see how better local decoupling caps can make a huge difference, I'd probably want to try half a dozen there at least, and expect a freakshow.
I think the results expected by snubbing the PSU caps would vary with the quality of the cap since that's what it's compensating for, there may not be much to compensate if you have really good caps.
Regards,
Chris
Cap sound
Chris,
It has been reported by many that the bulk storage caps can have quite an influence on sound, even when dealing with the same basic values. What parameters are different? ESR, ESL, type of terminals, type of construction like slit foil etc. Which of these parameters is influencing the sonics and how? It might be a self damping type thing where some caps act as though they have built in snubbers already. I have seen this effect to some degree and it might be some of the answer. I would think a proper external snubber of the right values would help make the overall sound better and more independent of the electrolytics behavior. I think this would be worthwhile. The snubbers parts must be very low ESL to be functional at HF. The resistor could be one of the new noninductive foil types like from Ohmite. The cap would probably be best if leadless like in surface mount. Properly done it could be effective to several 10’s of MHz or even higher. My guess would be to start around .47uf and .27ohm. Too bad there wasn’t a way to adjust this while observing a scope.
Roger
Chris,
It has been reported by many that the bulk storage caps can have quite an influence on sound, even when dealing with the same basic values. What parameters are different? ESR, ESL, type of terminals, type of construction like slit foil etc. Which of these parameters is influencing the sonics and how? It might be a self damping type thing where some caps act as though they have built in snubbers already. I have seen this effect to some degree and it might be some of the answer. I would think a proper external snubber of the right values would help make the overall sound better and more independent of the electrolytics behavior. I think this would be worthwhile. The snubbers parts must be very low ESL to be functional at HF. The resistor could be one of the new noninductive foil types like from Ohmite. The cap would probably be best if leadless like in surface mount. Properly done it could be effective to several 10’s of MHz or even higher. My guess would be to start around .47uf and .27ohm. Too bad there wasn’t a way to adjust this while observing a scope.
Roger
New cap?
Dean,
Yes, it has been a while, no word yet. I will check with them Tuesday and post then.
Roger
Dean,
Yes, it has been a while, no word yet. I will check with them Tuesday and post then.
Roger
Hi guys,
I finally got around to ordering a pair of the UCD modules, I decided to go for the 180ST mainly because their output is more than enough for my needs and I intended on trying some different op-amps
I've had a quick listen to them last night before fitting them inside the case and I'll admit I was rather impressed even with these units using the standard coupling cap and NE5532 op-amp, even at this stage in my opinion the sound is much superior to any of the little Tripath kits I've tried so well done to the guys at Hypex
The main thing I wanted to ask is have any of you guys compared the op-amps in these units, if so what are your honest opinions?
After looking at that nice neat board I don't fancy fitting and removing several types because the print looks a bit too delicate
I have AD8620,AD826,OPA2132 and a few others
Which ever type I use I will remove and link out the two electrolytics after the op-amp and fit 2.2uf Auricaps on the input if DC offset is not a problem with the newer op-amps
Just a note, I've often found AD8620 to be a bit bright when I've tried them in other projects, do they make the UCD modules slightly brighter than the standard NE5532's? I'm not a fan of an over bright or clinical top end
Any advice much appreciated
I finally got around to ordering a pair of the UCD modules, I decided to go for the 180ST mainly because their output is more than enough for my needs and I intended on trying some different op-amps
I've had a quick listen to them last night before fitting them inside the case and I'll admit I was rather impressed even with these units using the standard coupling cap and NE5532 op-amp, even at this stage in my opinion the sound is much superior to any of the little Tripath kits I've tried so well done to the guys at Hypex
The main thing I wanted to ask is have any of you guys compared the op-amps in these units, if so what are your honest opinions?
After looking at that nice neat board I don't fancy fitting and removing several types because the print looks a bit too delicate
I have AD8620,AD826,OPA2132 and a few others
Which ever type I use I will remove and link out the two electrolytics after the op-amp and fit 2.2uf Auricaps on the input if DC offset is not a problem with the newer op-amps
Just a note, I've often found AD8620 to be a bit bright when I've tried them in other projects, do they make the UCD modules slightly brighter than the standard NE5532's? I'm not a fan of an over bright or clinical top end
Any advice much appreciated
Leeuwarden said:
Sorry,my mistake🙄 http://www.diyaudio.com/forums/showthread.php?postid=786486#post786486
I'll stick on that thread😉
UCD output filter cap replacement
Roger,
I assume that the Audience cap prototype for the UCD output filter is not going to happen?
Regards,
Dean
Roger,
I assume that the Audience cap prototype for the UCD output filter is not going to happen?
Regards,
Dean
Thanks Roger. I think replacing the output cap with a better quality one will be well worth it.
Regards,
Dean
Regards,
Dean
If I'm correct, I believe that the latest batch UCD400 (well those of november anyway) have a new output cap - it's a panasonic now!
It's really just EMI absorbing spray paint 
For those who've changed the filter cap to that BC one, was it worth it?
For those who've changed the filter cap to that BC one, was it worth it?
There must be some other nice 680nf available thats the right size
Its a shame the 5mm Evox PHE426D only go up to 220nf
Its a shame the 5mm Evox PHE426D only go up to 220nf
my opinion...
i think it was definitely worth it. i used bc mkp379 and mkp479's. Both improved detail and imaging of the ucd400's. I think it made the bass go a little bit deeper aswell (but i'm not shure). Bass 'speed' improved a little too.
CHoice between the two is more a matter of taste and I doubt i'd be able to pick them out in a blind test.
i think it was definitely worth it. i used bc mkp379 and mkp479's. Both improved detail and imaging of the ucd400's. I think it made the bass go a little bit deeper aswell (but i'm not shure). Bass 'speed' improved a little too.
CHoice between the two is more a matter of taste and I doubt i'd be able to pick them out in a blind test.
- Status
- Not open for further replies.