Made a few changes:
Replaced on board PSU caps with 100uF 100V Rubycon ZLH
Bypassed 22uF ceramic coupling cap
Replaced 47uF electrolytic feedback cap with 4x22uF fostex branded film caps
Drilled lots of ventilation holes in the base and sides of the case (it was sealed before)
Changes? Now you can see the modules yellow and blue LEDs glowing through the holes in the case. Sounds a bit different too
Possibly a touch leaner but upper mid and HF clarity far improved. Generally sounds a bit cleaner and more sophisticated, if that makes any sense.
Probably be fire burned at the crossed wood by a certain ******* - oh well.
Replaced on board PSU caps with 100uF 100V Rubycon ZLH
Bypassed 22uF ceramic coupling cap
Replaced 47uF electrolytic feedback cap with 4x22uF fostex branded film caps
Drilled lots of ventilation holes in the base and sides of the case (it was sealed before)
Changes? Now you can see the modules yellow and blue LEDs glowing through the holes in the case. Sounds a bit different too
Possibly a touch leaner but upper mid and HF clarity far improved. Generally sounds a bit cleaner and more sophisticated, if that makes any sense.
Probably be fire burned at the crossed wood by a certain ******* - oh well.
Sharpie,
Good build! Those yellow caps are the Fostex 22uF caps, I take it! That's a fairly major mod. Did you do bypass the ceramic input caps first & then listen or dive straight in with both mods i.e is the bypass responsible for the extra clarity at HF? What does the feedback mod do for the sound?
Iain, good moderation work! Thanks
Good build! Those yellow caps are the Fostex 22uF caps, I take it! That's a fairly major mod. Did you do bypass the ceramic input caps first & then listen or dive straight in with both mods i.e is the bypass responsible for the extra clarity at HF? What does the feedback mod do for the sound?
Iain, good moderation work! Thanks
I bypassed the coupling caps first - this cleaned up the HF and seemed to remove a degree of harshness (not that the amp was noticeably harsh beforehand).
Next I replaced the PSU and feedback caps (the big yellow caps replace the electrolytic feedback cap on the board). This increased smoothness while opening up the mids a bit.
To be honest the amp sounded good from the start. The mods improved the performance - more than subtley but not dramatically. Definitely worthwhile though.
I'm waiting for some of the new 22uF Wima MKS 2-XL caps to arrive. These are really small low voltage film caps that would have been far easier to fit in place of the electrolytic feedback cap. I used the big Fostex caps but may still change to the Wimas when they arrive. The Wimas would allow for much shorter connections which can only be a good thing.
Next I replaced the PSU and feedback caps (the big yellow caps replace the electrolytic feedback cap on the board). This increased smoothness while opening up the mids a bit.
To be honest the amp sounded good from the start. The mods improved the performance - more than subtley but not dramatically. Definitely worthwhile though.
I'm waiting for some of the new 22uF Wima MKS 2-XL caps to arrive. These are really small low voltage film caps that would have been far easier to fit in place of the electrolytic feedback cap. I used the big Fostex caps but may still change to the Wimas when they arrive. The Wimas would allow for much shorter connections which can only be a good thing.
Thanks for the report of the stages - it is even more informative & maybe Mike is listening in & could respond.
I came across these caps recently which seem to be stellar performers based on their specs & could have a role here http://www.paktron.com/pdf/Angstor_RA.pdf
They are Ultra-Low ESR Multilayer metallised polymer film capacitors. Capacitance values range from 0.1 µF to 20 µF and the voltage ratings are 50 VDC to 500 VDC.
This UK co is a distributor http://www.futureelectronics.com/en/manufacturers/paktron/Pages/index.aspx
I came across these caps recently which seem to be stellar performers based on their specs & could have a role here http://www.paktron.com/pdf/Angstor_RA.pdf
They are Ultra-Low ESR Multilayer metallised polymer film capacitors. Capacitance values range from 0.1 µF to 20 µF and the voltage ratings are 50 VDC to 500 VDC.
This UK co is a distributor http://www.futureelectronics.com/en/manufacturers/paktron/Pages/index.aspx
S
spitzer
Trying ultra low ESR paper. It good for all class d amp. Having to use high voltage kind, fold thrice. Also to can do ENABle PCB. This control to piezo cap. More can do copy tweak for Aussie Amp. Who steel half cookies? Can to wrap pizeo cap with ultra low and ENAble! Only silver, maybe black gate!! First Watt true, next comming to.
I nowing english some short only, no trnaslate. That gang racist for not agree with gullible club, bad logic. Moderator can do, no even telling what on topic. How many having to ask it for?
I nowing english some short only, no trnaslate. That gang racist for not agree with gullible club, bad logic. Moderator can do, no even telling what on topic. How many having to ask it for?
jkeny said:
John, hope your back is better. Brief reply will do a more detailed reply in the near future with plots of our amps compared to hypex should make interesting reading.
Nice to see James is making progress. We take on board the comments about caps and are aware of the more exotic low ESR caps but they are not used in our market place as standard.
We do have a new range of amplifiers coming out with more controls on for the pro audio market.
If there was enough interest we could produce a really high quality hi fi amplifier with dual current sources, dual current mirrors, dual differential input and low ESR caps throughout. This would comfortably give a THD+N of better than 0.005%. Please take note of the +N part which you will find others conveniently ignore.
Hi Mike,
Thanks for asking - back was fine for a while & on Wed I decided to fix a laptop which needed a solder on it's PS input socket - result: laptop works, back doesn't. irony is the money I saved on new laptop, I will probably spend on physio!
Anyway, I understand your sentiments about esoteric low ESR caps in your products - my advice was more for Sharpi31
Thanks for asking - back was fine for a while & on Wed I decided to fix a laptop which needed a solder on it's PS input socket - result: laptop works, back doesn't. irony is the money I saved on new laptop, I will probably spend on physio!
Anyway, I understand your sentiments about esoteric low ESR caps in your products - my advice was more for Sharpi31
I appear to be Not Charles since someone else got there before me.
I've been looking for something, second source, similar. Google seems to work. The American side of the pond were, sort of, wound component biased.
These capacitors look like the old Siemens devices which got sold off to Epcos. Just a moment.
http://www.epcos.com/web/generator/...itors/MetallizedPolyester/Page,locale=en.html
Chances are that link won't work
Home page is http://www.epcos.com/
http://www.epcos.com/inf/20/20/db/fc_05/MKT_B32560_64.pdf
are the 'unboxed' ones similar, maybe, to your equivalents.
One thing I have always liked about Siemens, and then Epcos, is the amount of data that they publish for their components. You might notice a slight difference between the Merkin data and the Epcos data.
Sigh. It means I have to go and qualify sources. Not that I wouldn't anyway but it would be nice if the data was available up front as a confidence builder.
I've got to say that as someone who is used to compensating 'analogue' systems I might feel the need to express some concern about what looks like arbitrary substitution of components in a delivered system.
I'd assume the manufacturer has design the product appropriately so messing about with things might end up breaking something. Reading about then perhaps playing with input capacitors or similar components won't hurt but when it comes to feedback components I'd be extremely wary about playing with them.
Low ESR might 'sound good' but, as a process control engineer, I've earned a bit of money replacing the bits someone bought off the back end of a lorry that were supposedly better/cheaper with the bits they should have used in the first place. The difference was something that did not work, lost production and some fruity repair bills versus something that did work.
They only ask once... Next time they ask first.
Anyway, I am sort of surprised that you consider these components to be 'esoteric' I very much doubt that they are any different to the Epcos devices.
Thanks for the link though. I'll go and grill them. In the mean time it could be important so play safe.
Charles AKA NotCharles
jkeny said:
I've been looking for something, second source, similar. Google seems to work. The American side of the pond were, sort of, wound component biased.
These capacitors look like the old Siemens devices which got sold off to Epcos. Just a moment.
http://www.epcos.com/web/generator/...itors/MetallizedPolyester/Page,locale=en.html
Chances are that link won't work
Home page is http://www.epcos.com/
http://www.epcos.com/inf/20/20/db/fc_05/MKT_B32560_64.pdf
are the 'unboxed' ones similar, maybe, to your equivalents.
One thing I have always liked about Siemens, and then Epcos, is the amount of data that they publish for their components. You might notice a slight difference between the Merkin data and the Epcos data.
Sigh. It means I have to go and qualify sources. Not that I wouldn't anyway but it would be nice if the data was available up front as a confidence builder.
I've got to say that as someone who is used to compensating 'analogue' systems I might feel the need to express some concern about what looks like arbitrary substitution of components in a delivered system.
I'd assume the manufacturer has design the product appropriately so messing about with things might end up breaking something. Reading about then perhaps playing with input capacitors or similar components won't hurt but when it comes to feedback components I'd be extremely wary about playing with them.
Low ESR might 'sound good' but, as a process control engineer, I've earned a bit of money replacing the bits someone bought off the back end of a lorry that were supposedly better/cheaper with the bits they should have used in the first place. The difference was something that did not work, lost production and some fruity repair bills versus something that did work.
They only ask once... Next time they ask first.
Anyway, I am sort of surprised that you consider these components to be 'esoteric' I very much doubt that they are any different to the Epcos devices.
Thanks for the link though. I'll go and grill them. In the mean time it could be important so play safe.
Charles AKA NotCharles
I've been using both the NX200 and UCD180HG power amps for a while now. I had been using a grounded grid preamp, then built a buffered passive using a cheap fet buffer board from ebay. With both preamps I liked both amps - the hypex was a bit better but there was very little in it.
Last week I finished building a modern twist on John Curl's JC-2 preamp. The board is available on ebay, and I have no idea how close the schematic is to a proper JC-2 or how it compares one. I first connected it to the NX200 amp and found the sound very detailed but a bit bright and brash (especially in comparison with the fet buffered pre I'd been using most recently). I changed power amps and found a far greater degree of difference than I had noticed previously. The hypex was so much smoother and warmer - for the first time I'm not tempted to swap amps back and forward.
It may well be that my dodgy JC-2 is far happier with the 100K input impedance of the hypex, and that the sound suffers when connected to the 12K input of the NX200. Whatever the reason, there is no doubt that the hypex is preferrable to me in this system.
Last week I finished building a modern twist on John Curl's JC-2 preamp. The board is available on ebay, and I have no idea how close the schematic is to a proper JC-2 or how it compares one. I first connected it to the NX200 amp and found the sound very detailed but a bit bright and brash (especially in comparison with the fet buffered pre I'd been using most recently). I changed power amps and found a far greater degree of difference than I had noticed previously. The hypex was so much smoother and warmer - for the first time I'm not tempted to swap amps back and forward.
It may well be that my dodgy JC-2 is far happier with the 100K input impedance of the hypex, and that the sound suffers when connected to the 12K input of the NX200. Whatever the reason, there is no doubt that the hypex is preferrable to me in this system.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.