Sandy,
I have just browsed the Jaycar catelogue. Those capacitors are not low spec! They are also low impedance type (I doubt if it is low ESR as it is claimed) with high ripple current. If you compare the specs, they are not worse than Panasonic FC. They are cheaper. But whether I can trust the specs given by Jaycar / Chinese manufacturer is a different story.
I had a lot of frustrations with the quality of many products sold by Jaycar. Jaycar imports most of its stuff from China, and the quality can vary greatly. I found Jaycar is getting a lot more expensive these days (possibly because DSE, Candy and others don't put parts on their shelves any more). However, these capacitors are still cheap comparing to RS Components and Farnell.
Have you tried the 16V or even the 25V ones? Usually the higher the voltage rating, the better quality it is. They also sell the 4,700uF/10v/25v. Have you tried it?
If the JLH does not need to worry about current limit (which is provided by the current limit of the LM317/337 at 1.5A max), does it mean the larger the capacitance the better? Have you tried 2 x 4,700uf? Note that the ripple eater's low impedance output at low frequencies is provided by the capacitors, not the ripple eater.
Regards,
Bill
I have just browsed the Jaycar catelogue. Those capacitors are not low spec! They are also low impedance type (I doubt if it is low ESR as it is claimed) with high ripple current. If you compare the specs, they are not worse than Panasonic FC. They are cheaper. But whether I can trust the specs given by Jaycar / Chinese manufacturer is a different story.
I had a lot of frustrations with the quality of many products sold by Jaycar. Jaycar imports most of its stuff from China, and the quality can vary greatly. I found Jaycar is getting a lot more expensive these days (possibly because DSE, Candy and others don't put parts on their shelves any more). However, these capacitors are still cheap comparing to RS Components and Farnell.
Have you tried the 16V or even the 25V ones? Usually the higher the voltage rating, the better quality it is. They also sell the 4,700uF/10v/25v. Have you tried it?
If the JLH does not need to worry about current limit (which is provided by the current limit of the LM317/337 at 1.5A max), does it mean the larger the capacitance the better? Have you tried 2 x 4,700uf? Note that the ripple eater's low impedance output at low frequencies is provided by the capacitors, not the ripple eater.
Regards,
Bill
JLH Ripple Eater
Hi Bill
Yes,I did notice the printed specifications for those capacitors.
I take much of that wiith a grain of salt.
I haven't tried the 16V 0r 25V versions, mainly because JLH specified even lower voltage, and our PCB layout has only been improved in the capacitor area recently, to allow for a modest increase in capacitor dimensions.
I do intend trying 1 x 2,200 uF 10V low ESR and a normal 2,200uF 16V from Jaycar in parallel, to see if the neutrality can be further improved. Going too high in capacitance in that area may stress the voltage regulators further on startup. (?)
However, I may try 1 x 2,200uF low ESR and 2 x 2,200uF 16V normal electros in parallel. Incidentally, the original JLH design used 2 x 2,200uF normal type electrolytics in parallel.
Have you seen the original article by JLH ?
Regards
Alex
Hi Bill
Yes,I did notice the printed specifications for those capacitors.
I take much of that wiith a grain of salt.
I haven't tried the 16V 0r 25V versions, mainly because JLH specified even lower voltage, and our PCB layout has only been improved in the capacitor area recently, to allow for a modest increase in capacitor dimensions.
I do intend trying 1 x 2,200 uF 10V low ESR and a normal 2,200uF 16V from Jaycar in parallel, to see if the neutrality can be further improved. Going too high in capacitance in that area may stress the voltage regulators further on startup. (?)
However, I may try 1 x 2,200uF low ESR and 2 x 2,200uF 16V normal electros in parallel. Incidentally, the original JLH design used 2 x 2,200uF normal type electrolytics in parallel.
Have you seen the original article by JLH ?
Regards
Alex
Sandy, yes I have read it. Are the Jaycar capacitors in the RA you brought to me the low ESR type or the general purpose type?
JLH Ripple Eater
Bill
The JLH used 4 x 2,200uF 10V low ESR.
Jaycar cat.no. RE6306
Also 100uF 35V at the output. I am not sure if they were low ESR from Jaycar or not. You could try both the low ESR -RE6334 (100uF 35V), and 100uF25v normal type -RE6140 and see which you prefer ?
Does anybody know how to identify low ESR types ?
Alex
Bill
The JLH used 4 x 2,200uF 10V low ESR.
Jaycar cat.no. RE6306
Also 100uF 35V at the output. I am not sure if they were low ESR from Jaycar or not. You could try both the low ESR -RE6334 (100uF 35V), and 100uF25v normal type -RE6140 and see which you prefer ?
Does anybody know how to identify low ESR types ?
Alex
Sandy,
I went to Jaycar and bought some capacitors.
I seriously doubt the overall quality of them. I bought 4 x 2,200uF/16V. They have a volume about 3 times of your 4 x 2,200uF/10V. How can that be? I have also bought a pair of 2,200uF/25V, and they are half the size of the 2,200uF/16V. All have the same look.
Anyway, your "ripple eater" sounded good while mine with the Rubycon or Panasonic did not sound that much different, so I have to give the Jaycar caps a go.
Unlike your 2,200uF/10V, these 2,200uF/16V did require run-in. When first connected, the sound was very closed (vs open), dry and inaccurate. 2 hours later they were much better, but still fell short comparing to your 2,200uF/10v from memory. The difference was that the high frequencies were still exaggerated and edgy.
I should have bought the 2,200uF/10v! I turned the thing on until this morning and will have a listen tonight and see if further run-in helps or not.
Because I am changing the passive XO at the same time so I can't simply conclude what sounds right or not. The trouble is that my new passive XO requires a 0.12mH inductor which I have to unwind from a 0.22mH. I need a good computer soundcard to use Speaker Workshop to measure the inductance. But my soundcard is not good so I have to buy a new one. All these will take some time. My project is now brought to a halt until I sort out my computer.
Regards,
bill
I went to Jaycar and bought some capacitors.
I seriously doubt the overall quality of them. I bought 4 x 2,200uF/16V. They have a volume about 3 times of your 4 x 2,200uF/10V. How can that be? I have also bought a pair of 2,200uF/25V, and they are half the size of the 2,200uF/16V. All have the same look.
Anyway, your "ripple eater" sounded good while mine with the Rubycon or Panasonic did not sound that much different, so I have to give the Jaycar caps a go.
Unlike your 2,200uF/10V, these 2,200uF/16V did require run-in. When first connected, the sound was very closed (vs open), dry and inaccurate. 2 hours later they were much better, but still fell short comparing to your 2,200uF/10v from memory. The difference was that the high frequencies were still exaggerated and edgy.
I should have bought the 2,200uF/10v! I turned the thing on until this morning and will have a listen tonight and see if further run-in helps or not.
Because I am changing the passive XO at the same time so I can't simply conclude what sounds right or not. The trouble is that my new passive XO requires a 0.12mH inductor which I have to unwind from a 0.22mH. I need a good computer soundcard to use Speaker Workshop to measure the inductance. But my soundcard is not good so I have to buy a new one. All these will take some time. My project is now brought to a halt until I sort out my computer.
Regards,
bill
JLH Ripple Eater
Hi Bill
A very knowledgeable Rock Grotto mamber has made the attached comment : "A shunt reg while having sonic impact is not in the supply path in the classic sense of "in" so will pass along what has gone before (and why dual regs are always a good idea with a wideband series reg followed by a shunt reg which by definition is wideband ) "
I find that best results when using the JLH appear to be when used after the normal implementation of the LM317T/LM337T regulators. I prefer not to use "Audio Grade" capacitors in the voltage regulator area, as I have found that they all seem to add their own individual "house sound/flavour" to the overall sound. In particular, the "adjust" capacitor, where I prefer to use standard 10uF 16V electrolytics. Like yourself, I dislike tantalum capacitors at that location.
Alex
Hi Bill
A very knowledgeable Rock Grotto mamber has made the attached comment : "A shunt reg while having sonic impact is not in the supply path in the classic sense of "in" so will pass along what has gone before (and why dual regs are always a good idea with a wideband series reg followed by a shunt reg which by definition is wideband ) "
I find that best results when using the JLH appear to be when used after the normal implementation of the LM317T/LM337T regulators. I prefer not to use "Audio Grade" capacitors in the voltage regulator area, as I have found that they all seem to add their own individual "house sound/flavour" to the overall sound. In particular, the "adjust" capacitor, where I prefer to use standard 10uF 16V electrolytics. Like yourself, I dislike tantalum capacitors at that location.
Alex
I know this is a little late to the discussion, but I saw this page that seemed to relate to the the treble issue of the Rubycon ZL caps and the 317 regulator.
http://www.tnt-audio.com/clinica/regulators_noise2_e.html
http://www.tnt-audio.com/clinica/regulators_noise2_e.html
Sandy,
Last night I added a bypass cap across the "adjust" capacitor. My "adjust" cap was a 22uF/63v Pana FC. It cleared up the top quite a bit. 0.1uF MKP seemed too large. 0.01uF too small. 0.056uF seemed to be right for the bypass. It all depends on the "adjust" capacitor.
I will do a lot more "cap tests" and I hope I will finish the thing this weekend.
Is your "adjust" cap 10uF/16v the Jaycar general purpose cap?
Last night I added a bypass cap across the "adjust" capacitor. My "adjust" cap was a 22uF/63v Pana FC. It cleared up the top quite a bit. 0.1uF MKP seemed too large. 0.01uF too small. 0.056uF seemed to be right for the bypass. It all depends on the "adjust" capacitor.
I will do a lot more "cap tests" and I hope I will finish the thing this weekend.
Is your "adjust" cap 10uF/16v the Jaycar general purpose cap?
Bill
I use the normal Jaycar 10uF 16V electrolytic capacitors at that location.I often use the Jaycar KC5418 dual regulator kit these days for convenience. It really is all that you will normally need when followed by a JLH, in the majority of applications. The kit is only AU$17.95, and also saves hunting around for individual bits and pieces. To be honest, I don't believe in fiddling around with added bypass capacitors in the adjust position initially.
IMHO, that kind of tweak is best left for after you have optimised the rest of your DIY system ,particularly the speakers, or in order to fine tune the sound from commercial components etc.
So many other things can influence the final sound as it is, including the type of interconnects used, and whether they use silver or copper in their construction etc. Low capacitance interconnects such as the reasonably priced range from the U.S.
based Blue Jeans Cable , are likely to influence the sound much more than your added bypass capacitor.
Alex
I use the normal Jaycar 10uF 16V electrolytic capacitors at that location.I often use the Jaycar KC5418 dual regulator kit these days for convenience. It really is all that you will normally need when followed by a JLH, in the majority of applications. The kit is only AU$17.95, and also saves hunting around for individual bits and pieces. To be honest, I don't believe in fiddling around with added bypass capacitors in the adjust position initially.
IMHO, that kind of tweak is best left for after you have optimised the rest of your DIY system ,particularly the speakers, or in order to fine tune the sound from commercial components etc.
So many other things can influence the final sound as it is, including the type of interconnects used, and whether they use silver or copper in their construction etc. Low capacitance interconnects such as the reasonably priced range from the U.S.
based Blue Jeans Cable , are likely to influence the sound much more than your added bypass capacitor.
Alex
For those who have built the JLH.
Last night I replaced the 4 x 2.2uF polyester capacitors in the circuit with 4 x 1uF MKP, the Digikey Vishay blue/grey box. Not that I wanted to decrease the value but 1uF was the biggest value I had for MKP without mounting a huge ICW SA cap. The corner frequency would be increased from 0.7Hz to 1.6Hz, which is no big deal here. This was because I could hear clearly that familiar polyester colouration. The MKP got rid of the colouration. The sound is now more "high-end" and less coloured. I was very pleased with the change.
The 10uF cap is the most sensitive cap defining the "FR response" and I will still need to do more experiments.
Last night I replaced the 4 x 2.2uF polyester capacitors in the circuit with 4 x 1uF MKP, the Digikey Vishay blue/grey box. Not that I wanted to decrease the value but 1uF was the biggest value I had for MKP without mounting a huge ICW SA cap. The corner frequency would be increased from 0.7Hz to 1.6Hz, which is no big deal here. This was because I could hear clearly that familiar polyester colouration. The MKP got rid of the colouration. The sound is now more "high-end" and less coloured. I was very pleased with the change.
The 10uF cap is the most sensitive cap defining the "FR response" and I will still need to do more experiments.
Hi Bill
I think you will find that it is the preceding regulator stage that has the most influence. In particular the choice of capacitors. Changing component types in the JLH is more likely to fine tune overall system sound. Approximately 50 JLHs of this type have been constructed worldwide, that I am personally aware of. They have been mainly associated with headphone amplifiers, DACs etc,where any small changes are more readily perceived. Due to availability, the original 2.2uF capacitors as listed originally by JLH, have often been replaced by values varying from 1uF to 2.2uF polyester, such as Wima, with very little performance changes noted.
Consistent sound quality results have been obtained without resorting to the use of either low ESR, or Audiophile type capacitors in the preceeding voltage regulators. However, that is not to say that the JLH can't be further tweaked as you suggested to suit the actual system.
SandyK
I think you will find that it is the preceding regulator stage that has the most influence. In particular the choice of capacitors. Changing component types in the JLH is more likely to fine tune overall system sound. Approximately 50 JLHs of this type have been constructed worldwide, that I am personally aware of. They have been mainly associated with headphone amplifiers, DACs etc,where any small changes are more readily perceived. Due to availability, the original 2.2uF capacitors as listed originally by JLH, have often been replaced by values varying from 1uF to 2.2uF polyester, such as Wima, with very little performance changes noted.
Consistent sound quality results have been obtained without resorting to the use of either low ESR, or Audiophile type capacitors in the preceeding voltage regulators. However, that is not to say that the JLH can't be further tweaked as you suggested to suit the actual system.
SandyK
Curmudgeon brought this up earlier, and I can't believe it hasn't been mentioned since really, or become a valid part of the comparison, but I find the Nichicon Muse line of caps to be outstanding, and their value far exceeds their purchase price IMO.
Whenever possible in projects, rebuilds, upgrades, etc., I try to use the UKZ and UES Muse variety as often as possible. UKZ being the best polarized cap they offer, and UES being the best non-polar they offer. I believe they are currently revamping what used to be the larger UFX and UFG caps.
I tend to have a simplistic approach (too simplistic sometimes) to electronics, and that sometimes helps a LOT, and sometimes hinders. By that I mean, each component should do its job, and not show tendencies of the others, and be "ideal". A capacitor either stores energy or filters a signal (sometimes both simultaneously). If you need to limit current or voltage, use a resistor. That said, I do know enough to work as a repair tech in one of the higher-end audio shops in town.
Of course, no component is ideal. HOWEVER, if you have to worry about how much a capacitor is acting as a resistor, or an inductor, it's time to try another cap! Especially since, to my ears anyway, resistor noise is a MUCH bigger issue than anything caused by a cap.
Naive? Not really, I'm a naturally-born skeptic, and when I hear some claim of "X" making a noticeable improvement in sound over "Y", even from multiple sources, I'll go get an "X" and a "Y" and try them in the same piece of gear. I'll put "Y" in first, listen, then "X", and if the heavens do not immediately open up, or the improvement isn't noticeable at all, or, in some cases, if it gets WORSE, then from that point on I go by what my ear tells me. Sometimes what seems logical makes no difference, sometimes something that sounds crazy does. What has helped my knowledge of electronics from that approach is not only am I a naturally-born skeptic, I HATE not knowing the answer to "WHY?"
My experience with caps so far tells me that yes, different ones do in fact sound different, but not necessarily for the reasons commonly thought. For instance, I build a custom preamp that I hope to bring to market someday soon. The only signal path cap I use is one input coupling cap on each channel, a 2.2uf metallized polypropylene cap. Two things I learned from building at least 28 prototype boards, and sometimes changing out parts within one prototype:
1. A lot of people will tell you, and at first it does seem logical, that film caps are non-polar, therefore not directional. Non-polar? yep. Non-directional? Try tapping on an input coupling cap sometime. You'll learn which lead is connected to the outer foil REAL quick! Makes me wish more manufacturers marked which lead was which. Oh yeah, before I forget, you want the outside-foil lead connected to the "source side", and the inside-foil lead connected to the base of the input transistor, or the grid of the input tube. To make it even better, add a shield of some type around the body of the cap (non-magnetic).
2. Given the same materials, the way a cap is built makes a bigger difference than what it is made of! The way the cap is built (given the same materials are used), makes a much bigger difference than the brand name or the price! In other words, brand "A" probably doesn't have a secret formula for polypropylene that is better than brand "B". Now, given the same materials, and same values, logically brand "A"'s cap shouldn't sound any different from brand "B". The EEs will tend to say, "Yes, that's correct", while what some people think of as the "lunatic fringe" will argue to the death that brand "A" and brand "B" sound different. Who's correct? Believe it or not, in most cases, it's the "lunatic fringe", but not so much for the reasons they may think. Let's go back to my 2.2uf poly coupling caps. I was happy with what I was using for brand "A" for quite some time, and planned to use them in the "real" preamps. Then, one day when I was ordering more parts, on a whim (and because for some reason the 2.2uf is cheaper than brand "A" when the bigger ones aren't) I bought a few of brand "B" to play with.
Fired the preamp up, and it was one of those instances where the heavens opened up instantly! But WHY? Both had the same value capacitance and voltage, tinned copper leads, etc., that you could think of. No logical reason for it at all, at least using "adult" logic. Thankfully for me, I taught middle school kids for 18 years (yes, music), and since in a lot of ways I never grew up myself
, I can still think very much like a kid, and use "kid" logic when I need to! If you were to present the two caps to most average-thinking kids of that age, and tell them that they are made of exactly the same stuff, but they sound different anyway, and ask them why, the first answer you would likely get would be, "Because they are different colors?" (middle schoolers almost always give answers as questions😀 ). If you tell them no, the color doesn't matter, the next thing they would most likely say (besides "I don't know!"🙄 ) would be, "because they are a different shape?". BINGO! Most of you guys would then look down and go, "Huh? they are both cylindrical, axial caps, with the leads coming out of the center of each side!". And yes, you would be correct, but to a kid (and some adults), "fat and stumpy", and "long and skinny" are different shapes, and in this case, they are also correct.
The longer, thinner, cap sounded much better than the other one. This doesn't make any sense at all, until you unroll the cap on your mind. Laid out flat, the ends of the long and skinny cap will be much closer to the leads, while the fat and stumpy one will have the ends a lot farther away from the leads. This is true no matter if the leads are attached in the center to roll the cap up, or attached to one end to roll the cap up. This has two big effects that make the difference in sound. The skinny one takes many less turns to roll up completely. Much less inductive that way. Also, the entire plate will want to charge, so if the end of the plate is farther away, that's more built-in resistance. And again, to me, resistor noise is more audible than what is attributed to the difference in capacitors, and I hope this offers at least some explanation why.
NOW, if you're still with me, a lot of the above holds true for electrolytics as well. A lot of the ones more suited to audio will be longer and thinner, compared to other types, for reasons mentioned above. However, there are a few more variables in electrolytics, mainly due to the fact they are electrolytic!😀 Brand "A" may not have a better polypropylene formula than brand "B", but they may very well have a much better formula for their electrolyte!
Shutting up now!🙂
Don Taylor
Brassteacher
Whenever possible in projects, rebuilds, upgrades, etc., I try to use the UKZ and UES Muse variety as often as possible. UKZ being the best polarized cap they offer, and UES being the best non-polar they offer. I believe they are currently revamping what used to be the larger UFX and UFG caps.
I tend to have a simplistic approach (too simplistic sometimes) to electronics, and that sometimes helps a LOT, and sometimes hinders. By that I mean, each component should do its job, and not show tendencies of the others, and be "ideal". A capacitor either stores energy or filters a signal (sometimes both simultaneously). If you need to limit current or voltage, use a resistor. That said, I do know enough to work as a repair tech in one of the higher-end audio shops in town.
Of course, no component is ideal. HOWEVER, if you have to worry about how much a capacitor is acting as a resistor, or an inductor, it's time to try another cap! Especially since, to my ears anyway, resistor noise is a MUCH bigger issue than anything caused by a cap.
Naive? Not really, I'm a naturally-born skeptic, and when I hear some claim of "X" making a noticeable improvement in sound over "Y", even from multiple sources, I'll go get an "X" and a "Y" and try them in the same piece of gear. I'll put "Y" in first, listen, then "X", and if the heavens do not immediately open up, or the improvement isn't noticeable at all, or, in some cases, if it gets WORSE, then from that point on I go by what my ear tells me. Sometimes what seems logical makes no difference, sometimes something that sounds crazy does. What has helped my knowledge of electronics from that approach is not only am I a naturally-born skeptic, I HATE not knowing the answer to "WHY?"
My experience with caps so far tells me that yes, different ones do in fact sound different, but not necessarily for the reasons commonly thought. For instance, I build a custom preamp that I hope to bring to market someday soon. The only signal path cap I use is one input coupling cap on each channel, a 2.2uf metallized polypropylene cap. Two things I learned from building at least 28 prototype boards, and sometimes changing out parts within one prototype:
1. A lot of people will tell you, and at first it does seem logical, that film caps are non-polar, therefore not directional. Non-polar? yep. Non-directional? Try tapping on an input coupling cap sometime. You'll learn which lead is connected to the outer foil REAL quick! Makes me wish more manufacturers marked which lead was which. Oh yeah, before I forget, you want the outside-foil lead connected to the "source side", and the inside-foil lead connected to the base of the input transistor, or the grid of the input tube. To make it even better, add a shield of some type around the body of the cap (non-magnetic).
2. Given the same materials, the way a cap is built makes a bigger difference than what it is made of! The way the cap is built (given the same materials are used), makes a much bigger difference than the brand name or the price! In other words, brand "A" probably doesn't have a secret formula for polypropylene that is better than brand "B". Now, given the same materials, and same values, logically brand "A"'s cap shouldn't sound any different from brand "B". The EEs will tend to say, "Yes, that's correct", while what some people think of as the "lunatic fringe" will argue to the death that brand "A" and brand "B" sound different. Who's correct? Believe it or not, in most cases, it's the "lunatic fringe", but not so much for the reasons they may think. Let's go back to my 2.2uf poly coupling caps. I was happy with what I was using for brand "A" for quite some time, and planned to use them in the "real" preamps. Then, one day when I was ordering more parts, on a whim (and because for some reason the 2.2uf is cheaper than brand "A" when the bigger ones aren't) I bought a few of brand "B" to play with.
Fired the preamp up, and it was one of those instances where the heavens opened up instantly! But WHY? Both had the same value capacitance and voltage, tinned copper leads, etc., that you could think of. No logical reason for it at all, at least using "adult" logic. Thankfully for me, I taught middle school kids for 18 years (yes, music), and since in a lot of ways I never grew up myself
, I can still think very much like a kid, and use "kid" logic when I need to! If you were to present the two caps to most average-thinking kids of that age, and tell them that they are made of exactly the same stuff, but they sound different anyway, and ask them why, the first answer you would likely get would be, "Because they are different colors?" (middle schoolers almost always give answers as questions😀 ). If you tell them no, the color doesn't matter, the next thing they would most likely say (besides "I don't know!"🙄 ) would be, "because they are a different shape?". BINGO! Most of you guys would then look down and go, "Huh? they are both cylindrical, axial caps, with the leads coming out of the center of each side!". And yes, you would be correct, but to a kid (and some adults), "fat and stumpy", and "long and skinny" are different shapes, and in this case, they are also correct.The longer, thinner, cap sounded much better than the other one. This doesn't make any sense at all, until you unroll the cap on your mind. Laid out flat, the ends of the long and skinny cap will be much closer to the leads, while the fat and stumpy one will have the ends a lot farther away from the leads. This is true no matter if the leads are attached in the center to roll the cap up, or attached to one end to roll the cap up. This has two big effects that make the difference in sound. The skinny one takes many less turns to roll up completely. Much less inductive that way. Also, the entire plate will want to charge, so if the end of the plate is farther away, that's more built-in resistance. And again, to me, resistor noise is more audible than what is attributed to the difference in capacitors, and I hope this offers at least some explanation why.
NOW, if you're still with me, a lot of the above holds true for electrolytics as well. A lot of the ones more suited to audio will be longer and thinner, compared to other types, for reasons mentioned above. However, there are a few more variables in electrolytics, mainly due to the fact they are electrolytic!😀 Brand "A" may not have a better polypropylene formula than brand "B", but they may very well have a much better formula for their electrolyte!
Shutting up now!🙂
Don Taylor
Brassteacher
Nichicon KW series?
Anyone have any experience with the Nichicon KW series?
This statement in the datasheet caught my eye: "Realization of a harmonious balance of sound quality, made possible by the development of new electrolyte."
datasheet here: http://products.nichicon.co.jp/en/pdf/XJA043/e-kw.pdf
Comments?
Anyone have any experience with the Nichicon KW series?
This statement in the datasheet caught my eye: "Realization of a harmonious balance of sound quality, made possible by the development of new electrolyte."
datasheet here: http://products.nichicon.co.jp/en/pdf/XJA043/e-kw.pdf
Comments?
I have been using the JLH ripple eater for a little while now. I have actually built two used in different components.
I have to say that I am pleased with it. It is obviously better than the plain LM317/337 regulator.
As reported in some of my previous posts, the capacitors used within the JLH can affect the sound. Coupled with the high inductive output of the front end LM317/337 regulator, one has to be very cautious when using low ESR capacitors, as Sandy indicated. Using the wrong capacitors would have as much problem as what the JLH is intended to solve.
My finally implementation was like this: All Rubycon electrolytic capacitors. There were no bypass capacitors used. For the large capacitors (2 x 2,200uF), I added 0.22R series resistence. All high frequency hiss was gone. I also swapped in and out the 4 x 1uF-2.2uF a few times, and still found that I could not live the the polyester sound or the Solen MKP. The Vishay blue box was the only one that gave the best result.
I highly recommend it.
Thanks to Sandy again. Sandy, you are welcome to come and audition my system again - the speakers are now completed.
Regards,
Bill
I have to say that I am pleased with it. It is obviously better than the plain LM317/337 regulator.
As reported in some of my previous posts, the capacitors used within the JLH can affect the sound. Coupled with the high inductive output of the front end LM317/337 regulator, one has to be very cautious when using low ESR capacitors, as Sandy indicated. Using the wrong capacitors would have as much problem as what the JLH is intended to solve.
My finally implementation was like this: All Rubycon electrolytic capacitors. There were no bypass capacitors used. For the large capacitors (2 x 2,200uF), I added 0.22R series resistence. All high frequency hiss was gone. I also swapped in and out the 4 x 1uF-2.2uF a few times, and still found that I could not live the the polyester sound or the Solen MKP. The Vishay blue box was the only one that gave the best result.
I highly recommend it.
Thanks to Sandy again. Sandy, you are welcome to come and audition my system again - the speakers are now completed.
Regards,
Bill
Hi Bill
Interesting comments.
Could you please give further information about the Vishay Blue box capacitors you refer to, and where you obtained them.
Regards
Alex (SandyK)
Interesting comments.
Could you please give further information about the Vishay Blue box capacitors you refer to, and where you obtained them.
Regards
Alex (SandyK)
SandyK,
I can ensure you that the blue box makes a substantial difference and the sound is far more cleaner and high-end. If you are not convinced, come to my home and we can do an (a bit time-consuming) experiement and swap in and out for a few times.
About a year ago, I ordered 8 of them with many other parts from Digi-key. They are not expensive, possibly less than $2 a piece. If you search on Vishay MKP 1uF you will find it.
I love to have another 1-2 dozens but it does not work out cheap if freight is included. The freight cost (about $65?) is more than the part cost. I don't have other parts to be ordered at the moment. However, if you want to order them, I am happy to share the costs with you. From what I remember visiting their website last time, there weren't too many left and I doubt if they will re-order them in and sell at small quantities, unless you have thousands to order. I am worried that they will be gone soon.
These blue boxes sound very neutral. They are designed for high frequency application and of course they sound good if they are accurate at MHz frequencies with tight tolerance. I prefer them over the ICW Clarity SA caps which sound veiled in comparison. I love to have them replacing my CD output caps. The trouble with them is that they are low voltage capacitors so they can only be used in line-level stuff. The value of 1uF would be low for CD output unless the preamp input is 100k or so, which prevent people from using it. I have not such a problem because I can design my preamp to suit. The largest value available in the same series is 1.1uF which is not much larger.
Regards,
Bill
I can ensure you that the blue box makes a substantial difference and the sound is far more cleaner and high-end. If you are not convinced, come to my home and we can do an (a bit time-consuming) experiement and swap in and out for a few times.
About a year ago, I ordered 8 of them with many other parts from Digi-key. They are not expensive, possibly less than $2 a piece. If you search on Vishay MKP 1uF you will find it.
I love to have another 1-2 dozens but it does not work out cheap if freight is included. The freight cost (about $65?) is more than the part cost. I don't have other parts to be ordered at the moment. However, if you want to order them, I am happy to share the costs with you. From what I remember visiting their website last time, there weren't too many left and I doubt if they will re-order them in and sell at small quantities, unless you have thousands to order. I am worried that they will be gone soon.
These blue boxes sound very neutral. They are designed for high frequency application and of course they sound good if they are accurate at MHz frequencies with tight tolerance. I prefer them over the ICW Clarity SA caps which sound veiled in comparison. I love to have them replacing my CD output caps. The trouble with them is that they are low voltage capacitors so they can only be used in line-level stuff. The value of 1uF would be low for CD output unless the preamp input is 100k or so, which prevent people from using it. I have not such a problem because I can design my preamp to suit. The largest value available in the same series is 1.1uF which is not much larger.
Regards,
Bill
JLH Ripple Eater
Hi Bill
Yes, they are available at a very reasonable price from Digikey. Unfortunately, they have 15mm lead spacing,and for me to utilise them, I would need to redesign the PCBs.The freight costs from Digikey is a killer,also, with smaller orders.
I don't doubt that they do for your system what you claim, but I have found that many improvements like this, are as a result of interaction with other circuit components in a particular system.
For example, many constructors of the JLH have tried your choice of Rubycon 2,200uF capacitors, and did not like them at all in our present application. In fact, I actually obtained a small quantity of Rubycon 2,200uF 10V YXF capacitors and tried them myself. I did
not like the high frequency emphasis (sibilance) The general consensus of several JLH constructors so far, appears to be that the Suntan (Jaycar) CD286 2,200uF 10V low ESR capacitors are the most neutral sounding in our applications. I actually sent some to the U.K. for a technically experienced Rock Grotto member to compare against his Rubycons etc. The Suntan capacitors were judged to be the most neutral performers. Another U.K. member (ex Broadcast Engineer) has now ordered a quantity of these Suntan capacitors from Jaycar in Australia, specifically for JLH use.
I would prefer to use capacitors like those, rather than try to nullify the sound of a particular capacitor such as the Rubycon,by adding series resistance of .22R. This ,IMHO,defeats part of the reasoning behind the use of the JLH. i.e. an impedance of <.02 ohms to past 150KHZ, combined with very low noise.
SandyK
Hi Bill
Yes, they are available at a very reasonable price from Digikey. Unfortunately, they have 15mm lead spacing,and for me to utilise them, I would need to redesign the PCBs.The freight costs from Digikey is a killer,also, with smaller orders.
I don't doubt that they do for your system what you claim, but I have found that many improvements like this, are as a result of interaction with other circuit components in a particular system.
For example, many constructors of the JLH have tried your choice of Rubycon 2,200uF capacitors, and did not like them at all in our present application. In fact, I actually obtained a small quantity of Rubycon 2,200uF 10V YXF capacitors and tried them myself. I did
not like the high frequency emphasis (sibilance) The general consensus of several JLH constructors so far, appears to be that the Suntan (Jaycar) CD286 2,200uF 10V low ESR capacitors are the most neutral sounding in our applications. I actually sent some to the U.K. for a technically experienced Rock Grotto member to compare against his Rubycons etc. The Suntan capacitors were judged to be the most neutral performers. Another U.K. member (ex Broadcast Engineer) has now ordered a quantity of these Suntan capacitors from Jaycar in Australia, specifically for JLH use.
I would prefer to use capacitors like those, rather than try to nullify the sound of a particular capacitor such as the Rubycon,by adding series resistance of .22R. This ,IMHO,defeats part of the reasoning behind the use of the JLH. i.e. an impedance of <.02 ohms to past 150KHZ, combined with very low noise.
SandyK
Nichicon KW
To mightydub at post #114:
I have tried Nichicon KW for decoupling the main supply in my DH110, that is, between the rectifier diodes and the 317/337 regulators.
Midrange/upper midrange were exquisite - one of the best caps that I tried for this part of the frequency spectrum - BUT! - bass response sounded like it was tapered off at the bottom, so they sounded a bit 'thin'.
I used original spec caps, 50V 1000uF.
There was something to these caps (KWs)... but not what I was looking for.
Overall, they rated higher than many other caps that I tried in the same position, but were ultimately bested by regular old United Chemi-Con LXV (which happened to be the only 'big can' cap that I tried except for the original IC cap.)
One last note,
My last parts order had a few different values of Nichicon KW, these caps in particular tested very, very close to their rated uF so they at least appear to be well made 😉
To mightydub at post #114:
I have tried Nichicon KW for decoupling the main supply in my DH110, that is, between the rectifier diodes and the 317/337 regulators.
Midrange/upper midrange were exquisite - one of the best caps that I tried for this part of the frequency spectrum - BUT! - bass response sounded like it was tapered off at the bottom, so they sounded a bit 'thin'.
I used original spec caps, 50V 1000uF.
There was something to these caps (KWs)... but not what I was looking for.
Overall, they rated higher than many other caps that I tried in the same position, but were ultimately bested by regular old United Chemi-Con LXV (which happened to be the only 'big can' cap that I tried except for the original IC cap.)
One last note,
My last parts order had a few different values of Nichicon KW, these caps in particular tested very, very close to their rated uF so they at least appear to be well made 😉
SandyK,
The Jaycar caps would possibly have a series resistence more than 0.22R! The Rubycon ZL has something like 0.02R. Adding the 0.22R to the Rubycon ZL may still have an impedance lower than that of the Jaycar caps.
If the output of the LM317/337 is not inductive, any capacitors would work here. The added resistance is to damp the LCR resonance and that is why you use high ESR caps or pad them with resistance.
I believe the high frequency emphasis you had with the Rubycon was due to the LCR resonance.
You can try padding your Rubycon YXF with 0.15R and see if you like the sound and this may save you some good capacitors.
As for the printed circuit board, if you like, I could publish my design on vero-board here. It will be perfect fit for the Vishay blue boxes.
Regards,
Bill
The Jaycar caps would possibly have a series resistence more than 0.22R! The Rubycon ZL has something like 0.02R. Adding the 0.22R to the Rubycon ZL may still have an impedance lower than that of the Jaycar caps.
If the output of the LM317/337 is not inductive, any capacitors would work here. The added resistance is to damp the LCR resonance and that is why you use high ESR caps or pad them with resistance.
I believe the high frequency emphasis you had with the Rubycon was due to the LCR resonance.
You can try padding your Rubycon YXF with 0.15R and see if you like the sound and this may save you some good capacitors.
As for the printed circuit board, if you like, I could publish my design on vero-board here. It will be perfect fit for the Vishay blue boxes.
Regards,
Bill
- Status
- Not open for further replies.