|
|
Post by jelosno on Mar 23, 2007 9:26:01 GMT
Opening up my equipment that is. Only thing it does is make me want to change things in there. Next I would like to change the caps in the power supply of both my CDP and amp. let's start with the amp  Well, it sounds really nice. Most of all i found the close proximity of the transformers very disturbing. On the other hand - sound is nice... Since I do not listen too much anymore to my speaker system I think transplanting the internals of the amp into a custom build housing is not going to happen. So, why not change the diodes to ultra fast and the caps to better/higher capacity ones? Can anyone shed some light on the green drop style caps shown below? Please note the input selector hanging loose and normally sitting behind the front panel and looking like being sourced out of some vintage military equipment  Alps blue poti though.  For the CDP it will be also new caps and diodes in the PS part.  The top right part (where it says CDLP) is inactive at the moment. It is the headphone out section that was fed by the tube buffer. I will disconnect the power supply to it as well. The bottom centre with the white block thingy that has JZC 7FA printed on it(transformer?) is the first mains filter stage after the ON/OFF switch. lines go into the JZC 7FA block and to the large trafo which supplies all kinds of current (2.5V, 5.0, 25V, 220V or so) to the rest of the board. Panasonics might be the best choice for all the above. Not sure which ones for the biggies in the amp at 450V 470nF 108deg. Stefan |
|
|
|
Post by jelosno on Mar 23, 2007 11:09:35 GMT
In the amp I also miss a PS filter 'network' as such.
Seems that there isn't much filtering happening with that few caps.
|
|
|
|
Post by PinkFloyd on Mar 23, 2007 11:24:06 GMT
The green jobbies are 220nF 630V metallized polyester by the look of them, you've got plenty of room there to fit 220nF 630V polypropylenes there by the looks of things.
Replacing the 1N4007 with UF4007 will make an improvement, I've just done that with my new Chinese MHZS CD33 CD player and the UF4007 definitely brought about more clarity to the proceedings. I also spot an NE5532 on your PSU board, you could remove that fit a gold plated turned pin dil8 socket and try a few different opamps in there Stefan with OPA2134 being a good starting point (It's stable and a direct drop in replacement for the NE5532) Other chips to consider are OPA2107, OPA2132 and of course all the exotic types on brown dogs etc. if money is no object.
Panasonic FM / FC are always a good cap to use in a PSU being extremely low impedance. For the biggies (the 450V 470nF ) you are looking for snap in types and I can fully recommend the 105C Panasonic types (forget the model number) any good 105C snap in will be good here.
You could also replace those wirewound resistors with some of Rick's favourite Riedon types.... they DO make an improvement.
What amp / CDP is it anyway Stefan?
|
|
|
|
Post by jelosno on Mar 23, 2007 11:43:40 GMT
WOW! I have to decode all your suggestions now the CDP is a LUA Cantilena a expensive rebadged Euroverion of the www.tnt-audio.com/sorgenti/jolida-jd100_e.html The amp is a LUA 4040C which, like the CDP, by the looks is made in China as well. As for the amp, does it make sense to replace the caps with bigger ones - if available? Stefan |
|
|
|
Post by jelosno on Mar 23, 2007 11:54:55 GMT
That is the headphone amp out section I have disconnected from the rest of the DCP. Didn't go unnoticed! I use my EAR+HD instead I am still waiting for the Tent clock to arrive with a seperate SPDIF OUT generator for my external DAC that should come in the same shipment. In many parts of course. I am looking forward to compare the tube buffer stage of the CDP to the 60 chip DAC - once it's completed - with the EAR+ and the RS-1  |
|
|
|
Post by jelosno on Mar 23, 2007 12:08:21 GMT
|
|
|
|
Post by PinkFloyd on Mar 23, 2007 12:56:23 GMT
Just remembered, the caps you are looking for are Panasonic TS-HA series.
|
|
|
|
Post by jelosno on Mar 23, 2007 13:40:42 GMT
|
|
|
|
Post by jelosno on Mar 23, 2007 13:56:18 GMT
...found that somewhere at DIY audioasylum...
"I saw a post once which made a lot of sense in terms of what caps to use in the power supply. He contended that if you used fast recovery diodes for rectification, then film in oil caps are great for taming the digital hash and the resulting sound, while slower than other possilities, results in a good balance between speed and warmth.
If you are using tube rectification, the slower speed of tubes would become too slow with oil based caps in the power supply, so some polypropylene or high quality electrolytics would be the ticket. Your findings certainly are consistent with this recommendation."
Any opinions on that?
|
|
|
|
Post by jelosno on Mar 23, 2007 19:46:54 GMT
|
|
|
|
Post by PinkFloyd on Mar 24, 2007 22:05:36 GMT
|
|
|
|
Post by PinkFloyd on Mar 24, 2007 22:10:21 GMT
...found that somewhere at DIY audioasylum... "I saw a post once which made a lot of sense in terms of what caps to use in the power supply. He contended that if you used fast recovery diodes for rectification, then film in oil caps are great for taming the digital hash and the resulting sound, while slower than other possilities, results in a good balance between speed and warmth. If you are using tube rectification, the slower speed of tubes would become too slow with oil based caps in the power supply, so some polypropylene or high quality electrolytics would be the ticket. Your findings certainly are consistent with this recommendation." Any opinions on that? It's all pretty subjective Stefan, what some people class as good sounding others find as not too good, what some refer to as warm others may refer to as neutral...... all depends on your ears and capacitors are one area you really can experiment with.... the best thing to do is try a few different combos and see what suits your ears best..... suck it and see, if you like. |
|
|
|
Post by jelosno on Mar 27, 2007 9:13:15 GMT
I have sent Powertron in Germany the link. They are a RIEDON company. The types shown at Welbornelabs are RIEDON type UB or UT. Minimum order from Powertron direct is 25 pieces per type and value. I will place a larger order - large for me - with Digikey soon. Then I want some parts from PartsconneXion and maybe some stuff from Welbornelabs. If I think of all the shipment costs...  Is it always good to put higher value capacitors in a power supply compare to the original? I was thinking of a factor of 1.5 to 2.0 when changing the ones in the CDP and amp... Stefan |
|
|
|
Post by PinkFloyd on Mar 27, 2007 12:06:49 GMT
Really need the Rickmonster to chip in here as he's the expert on resistors (as well as a lot more) but fear he's disappeared into the bermuda triangle!  |
|
|
|
Post by jelosno on Mar 27, 2007 20:37:26 GMT
maybe I can find him. Just fired up my new 20" LCD @ 1600x1200. So much space it acutally must cover the earea where Rick is hiding |
|
|
|
Post by PinkFloyd on Mar 27, 2007 21:26:40 GMT
maybe I can find him. Just fired up my new 20" LCD @ 1600x1200. So much space it acutally must cover the area where Rick is hiding Think Barry Manilow.... think Bermuda Triangle... I doubt 1600 x 1200 will cover either (especially Manilow's nose span) |
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Mar 29, 2007 23:32:10 GMT
HELL YES I HAVE AN OPINION ON THAT ! In any power supply feeding an audio circuit you can pretty much get away with murder in the early stages but the very last stage that feeds the actual power pins of the device shold use the best possible parts you can fit in/afford to use there because IT will be the dominant sound. so for instance say you have this layout AC Mains,AC rated 0.01 cap from hot to nuetral on the primary side of the power trafos,parallel (shunt) connected R/C in series across the secondary,silicon rectifiers OR valve rectification-matters not really because you are still trying to damp the AC line feed not just protect against reverse revovery spikes,some high wattage (depends on psu itself) series resistors of say 10-25 ohms then the actual rectifier,the first filter cap-likely an electrolytic and that ideally is a low impedance/low inductance type but if the voltage is a high voltage you use what is available,resistor or filter inductor to set the second stage filter cap turnover to,next stage filter cap. that is your basic "overdone" power supply buss that get you to the ballpark voltage figure your devices will use but assuming you device has two channels (stereo) you ideally should have another round of filtering/interchannel decoupling that consists again of a series resisor (value determined by how much voltage drop you need/can allow followed by ANOTHER electrolytic for each channel fed from the initial power supply LAST capacitor. So up to this point you would have 3 electrolytics before EACH channels active circuit and these caps can be anything from the mosrt expensive designer parts known to humans or off the shelf surplus crap depending on need,availability,cost,space available,etc and it will STILL come down to the quality of the bypass capacitors after the very last cap how good the power supply will "sound" in real worl conditions (doubt you can measure this but maybe.i don't know how but maybe ) ! So if you have deep pockets and unlimited space dropping a parallel combination 10uF Motor Run/0.1-0.22 PIO or Vitamin-Q right across the tube/I.C. socket power pins will have a readily heard positive effect because it is THIS that the actual part "sees" as the power source no matter what went before assuming proper filter values. I regularly toss a 10uF Solen/0.22uF film cap across ALL final power lead ins and if you think it looks less than retarded when the bypass caps are bigger than the damn storage caps in a teeny tiny op-amp based audio circuit you would be mistaken-it DOES look goofy but it sonds just fine so is the path i choose. not enough room for 10uF ? No problemo el audiomundophilio ! just back down the scale going to the next smnallest value in line until you find one that will fit.even 1uF/0.1 is better than a straight feed of an electrolytic cap unless the circuit is non-audio where it is all about MEASURED results to achieve an end not AUDIBLE results. At least that is my story and I'm sticking to it until dragged away kicking and screaming |
|
FauDrei
Been here a while! 
Posts: 489
|
Post by FauDrei on Mar 31, 2007 20:47:03 GMT
Hi Rick, nice to have you back.
Think I understood the part about bypassing the last electrolytics in PS, just not sure whether you bypass them with TWO caps 10uF + 220nF, or you bypass them with ONE OF 10uF or 220nF?
If both of them is the way to go, what if there is no space for both (existing smal enclosure)?
Furthermore, are these any good for that: RS Stock No. 190-7817?
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Apr 1, 2007 3:46:29 GMT
one of each actually because with capacitors size matters so the biggger it is the deeper into the low frequencies the effectiveness will be but you still need to address the out of band RF frequencies which is where the 0.22uF comes in by shunting them to ground (single pole low pass filter) and where for whatever reason trying to get by with a lesser grade cap even though out of the audible band is noticeably worse than using a good film bypass of which there are many of from various manufacturers at way cheap prices. I use the Orange Drops for bypasses in high voltage circuits mainly because I am comfortable with the part having used it for well over 25 years (  ) and they are STILL dirt cheap compared to some parts that at best are marginally better for the location in the circuit while the 10uF DOES benefit with an increase in part quality simply because it goes further down into the audible freqency range, Once you get to the "Fast Cap to electrolytic" handoff you are hopefully deep enough into the audible range where any negative sonics from the electrolytic are inaudible though usually better would be 47-100uF (two 47's in parallel) until you consider the SHEER ENORMITY of a 47uF film cap this puppy is 54x73 mm  while this one only 32x45 mm us.st11.yimg.com/us.st.yimg.com/I/yhst-8476489043850_1941_29619080not always about what is the absolute BEST way (or the whole damn power supply would use film caps stacked in parallel and that would be the size of a shoe box just for the caps) but the best option to suit the conditions so you try and target areas that return the biggest bang for the buck rather than use a shotgun approach of what you think are the best parts for that position in the circuit and trust me,a great cap is only great where it is best suited and may actually be a detriment elsewhere like the Solens which are pleasant enough directly in the signal path until you listen THROUGH the cap then its limitations show up while for power supply duty damn hard to better at any rational price or size |
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Apr 1, 2007 6:28:34 GMT
couldn't say definitively having zero experience with those particular parts-one of those "U.S.A./U.K." parts readily available things where you know best what you have easy access to so anything I wite is conjecture only not that thinking outload is a new rickmonster trait ...so yeah.I would say give 'er a go and see what shakes out of that tree though offhand I would think a better and as readily available choice is the WIMA MKS 63 volt caps for solid state work. In fact for a single cap solution to replace the 10uF/220nF stew that does pretty much what the more expensive and definately larger combo does (just not as deep or high up though still noticeably good) is a single 1uF/63 VDC WIMA MKS right at the power feed pin of the device (transistor or IC). Inexpensive all else being equal,effective and will FIT into even the smallest devices that when you add that all up slides right into the "Team El Cheapo" comfort zone and that is a very good zone to be in  |
|
Alps blue poti though.
