Hi Andrew,
no, caps have not been reformed so the possibility of a difference due to the lack of is a possibility.
But the capacitor model I'm talking of (Nichicon KZ) has been ordered from Mouser so it should be fresh made, this evening I'll check date code to be sure.
BTW a fresh Nichicon ES bought from Mouser and one I've had at home for several years (I would say 3-4) sound exactly the same.
While reforming will improve a lot the capacitor restoring the original performance I don't think it will affect its sound signature/coloration but never made any test about it, I will do.
leakage through a capacitor WILL CHANGE the performance of an amplifier.
All electrolytics gradually degrade from time of manufacture.
Electrolytic manufacturers specify that their capacitors must be reformed shortly before testing (I seem to recall a maximum time of 24hours).
After reforming, all electrolytics gradually degrade.
All electrolytics gradually degrade from time of manufacture.
Electrolytic manufacturers specify that their capacitors must be reformed shortly before testing (I seem to recall a maximum time of 24hours).
After reforming, all electrolytics gradually degrade.
Those Nichicons were made in December 2015, no way they need reforming or leaks... Nichicon is a premium and trusted manifacturer.
Absolutely but C9, the cap under test, is somewhat peculiar, it's there as a filter to block DC (some tens of millivolts) not to decouple so leakage is not so important, in fact I was comparing the different caps to a plain wire mounted in C9 position in place of the capacitor...what's more leaky than a wire? 😉
Quality rating for C9 alternates
Since there are some alternate to choose from I think it's better to rate clearly them.
First choice: Cerafines 220uF 50V or Black Gate PK 220uF 4V (PK slightly better)
Second choice: Black Gate Standard 220uF 16V and up
Outsider: Elna Silmic II 220uF 50V Black/Gold, colored but interesting
Since there are some alternate to choose from I think it's better to rate clearly them.
First choice: Cerafines 220uF 50V or Black Gate PK 220uF 4V (PK slightly better)
Second choice: Black Gate Standard 220uF 16V and up
Outsider: Elna Silmic II 220uF 50V Black/Gold, colored but interesting
Even if the electrolytic was made last week, it should be reformed before testing.
Absolutely, I'll start desoldering/reforming and resoldering caps at least one time a month in all my gear 😉
Seriously, an electrolythic cap used as coupling/filter like C9 will never be reformed in circuit as happens for power supply decoupling ones, I don't think reform a capacitor, unless extremely old, would make any difference in use in that role.
Maybe, finally, the 'right' Mouser part for C9
Just now I'm listening to, and burning in, the candidate new part for C9.
In the early days of first FE versions Lehmannhill suggested the possible use for C9, as a Blackgate replacement, of AVX niobium caps.
I did buy, tried and finally discarded them (May 2013) since BlackGate Standard sounded so much fuller and richer, they were not bad at all, though.
In these days I couldn't manage to find a really transparent cap available from Mouser so I've started searching all 220uF caps I've had anywhere and those Niobium caps (NOJ and low ESR NOS series) jumped out.
Yes, they're SMD parts but really BIG (they will sit perfectly on 7.5mm pads) so I expect no problem for GB partecipants soldering it.
Tried NOJ, that at time I considered the best sounding of the two, not bad at all but slightly closed and too much warm, then NOS...
That's it!
Pretty dead neutral, near perfect timbre and imaging but a bit drier than reference on harmonic content.
Did a pass of the IsoTek System Enhancer full range track and harmonics appeared, almost... 🙄
Mmmh, some more days of burn-in and listening but it sounds really promising.
In the meanwhile an interesting reading about them:
https://www.google.it/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0ahUKEwiEmNSL9_PMAhUJ1hoKHVfGBRQQFggcMAA&url=https%3A%2F%2Fwww.digikey.com%2FWeb%2520Export%2FSupplier%2520Content%2FAVX_478%2FPDF%2FAVX_Caps.pdf&usg=AFQjCNFK3VzrXZr-Y9BR6b1ievzilvrCzg&sig2=F9v8UIP2T1lidlLirg2uWw&bvm=bv.122676328,d.d2s&cad=rja
https://www.google.it/url?sa=t&rct=...u-itneF35Kg7OtfIw&sig2=wQ55_RLZtw69eP8wJcIsvQ
Just now I'm listening to, and burning in, the candidate new part for C9.
In the early days of first FE versions Lehmannhill suggested the possible use for C9, as a Blackgate replacement, of AVX niobium caps.
I did buy, tried and finally discarded them (May 2013) since BlackGate Standard sounded so much fuller and richer, they were not bad at all, though.
In these days I couldn't manage to find a really transparent cap available from Mouser so I've started searching all 220uF caps I've had anywhere and those Niobium caps (NOJ and low ESR NOS series) jumped out.
Yes, they're SMD parts but really BIG (they will sit perfectly on 7.5mm pads) so I expect no problem for GB partecipants soldering it.
Tried NOJ, that at time I considered the best sounding of the two, not bad at all but slightly closed and too much warm, then NOS...
That's it!
Pretty dead neutral, near perfect timbre and imaging but a bit drier than reference on harmonic content.
Did a pass of the IsoTek System Enhancer full range track and harmonics appeared, almost... 🙄
Mmmh, some more days of burn-in and listening but it sounds really promising.
In the meanwhile an interesting reading about them:
https://www.google.it/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0ahUKEwiEmNSL9_PMAhUJ1hoKHVfGBRQQFggcMAA&url=https%3A%2F%2Fwww.digikey.com%2FWeb%2520Export%2FSupplier%2520Content%2FAVX_478%2FPDF%2FAVX_Caps.pdf&usg=AFQjCNFK3VzrXZr-Y9BR6b1ievzilvrCzg&sig2=F9v8UIP2T1lidlLirg2uWw&bvm=bv.122676328,d.d2s&cad=rja
https://www.google.it/url?sa=t&rct=...u-itneF35Kg7OtfIw&sig2=wQ55_RLZtw69eP8wJcIsvQ
Attachments
Last edited:
Waiting the tutorial
I'll publish some key pictures and notes for who is building in the next days, since I have yet to start updating the tutorial.
SMD Resitors with correct orientation:
My_Evo Rev A Mod (SMD part)
Wima capacitors can have straight or reversed markings
All Wimas have a small sign, the small ones on one side, the bigger on front/back:
That sign should be on the left side (small) or front (big), otherwise markings are reversed.
The one in the pic is reversed.
If markings are straight they should be on the side where the small empty dot is.
Through hole resistors orientation:
(note: C10 is void, unpopulated since it will be a My_Evo Rev A Mod build)
Indication on PCBs for R11 is wrong, it should be mounted oriented this way (resistor is raised only to compare orientation):
Transistors orientation:
(note: R3 is 0.33R instead of 0.47R since it will be a My_Evo Rev A Mod build)
A suggestion before mounting input cap:
That gummy adhesive will dampen vibration on the big input cap and will make easier to solder it since it will kept in place by the adhesive.
You can find it under several brands and names, like 'Blue Tack'
Input cap orientation:
Finished boards with through hole C9 (Nichicon KZ):
Correct orientation of the possible new C9 (but it should be mounted on the bottom side):
I'll publish some key pictures and notes for who is building in the next days, since I have yet to start updating the tutorial.
SMD Resitors with correct orientation:
My_Evo Rev A Mod (SMD part)
Wima capacitors can have straight or reversed markings
All Wimas have a small sign, the small ones on one side, the bigger on front/back:
That sign should be on the left side (small) or front (big), otherwise markings are reversed.
The one in the pic is reversed.
If markings are straight they should be on the side where the small empty dot is.
Through hole resistors orientation:
(note: C10 is void, unpopulated since it will be a My_Evo Rev A Mod build)
Indication on PCBs for R11 is wrong, it should be mounted oriented this way (resistor is raised only to compare orientation):
Transistors orientation:
(note: R3 is 0.33R instead of 0.47R since it will be a My_Evo Rev A Mod build)
A suggestion before mounting input cap:
That gummy adhesive will dampen vibration on the big input cap and will make easier to solder it since it will kept in place by the adhesive.
You can find it under several brands and names, like 'Blue Tack'
Input cap orientation:
Finished boards with through hole C9 (Nichicon KZ):
Correct orientation of the possible new C9 (but it should be mounted on the bottom side):
Attachments
Last edited:
That's a weird world
Today I did a 5 hours Isotek repeated pass and harmonics still are not enough... AVX NOS discarded.
Instead it's possible that the KZ problem was a false one...
In the last days before populating the 1.6 boards I've changed my DAC with a new better one based on Amanero + AK4495SEQ and I used AD797 as LPF opamp.
Today I've connected again the old DAC, a XMOS+DSD1796+ OPA1652 one, and KZ were again sounding 'right'...mmh
AK4495 again, changed decoupling caps, same problem with KZ, changed opamp, problem apparently disappear.
I've tried OPA627, LT1028, LME49710, NE5534DD, LM318 with all of them KZ seemed to sound right, apparently the problem is only with AD797 and KZ that do not match, somewhat.
Now it's too late to make any critical listening at a high enough volume (1 AM here), tomorrow I'll repeat listening tests and if confirmed KZs will be back in BOM.
So, possibly, a lot of unnecessary confusion...😡
In these last days I've had to do it too much... I'm sorry
That's it!
Pretty dead neutral, near perfect timbre and imaging but a bit drier than reference on harmonic content.
Did a pass of the IsoTek System Enhancer full range track and harmonics appeared, almost... 🙄
Mmmh, some more days of burn-in and listening but it sounds really promising.
Today I did a 5 hours Isotek repeated pass and harmonics still are not enough... AVX NOS discarded.
Instead it's possible that the KZ problem was a false one...
In the last days before populating the 1.6 boards I've changed my DAC with a new better one based on Amanero + AK4495SEQ and I used AD797 as LPF opamp.
Today I've connected again the old DAC, a XMOS+DSD1796+ OPA1652 one, and KZ were again sounding 'right'...mmh
AK4495 again, changed decoupling caps, same problem with KZ, changed opamp, problem apparently disappear.
I've tried OPA627, LT1028, LME49710, NE5534DD, LM318 with all of them KZ seemed to sound right, apparently the problem is only with AD797 and KZ that do not match, somewhat.
Now it's too late to make any critical listening at a high enough volume (1 AM here), tomorrow I'll repeat listening tests and if confirmed KZs will be back in BOM.
So, possibly, a lot of unnecessary confusion...😡
In these last days I've had to do it too much... I'm sorry
Dario,
I can tell you that in a very similar dac configuration (4490, with JLsounds, + in the dac Mirand) the new OPA 1688 is very very good, not a small difference.
Ciao, George
I can tell you that in a very similar dac configuration (4490, with JLsounds, + in the dac Mirand) the new OPA 1688 is very very good, not a small difference.
Ciao, George
Hi George,
I've already bought it after reading one of your posts on NextHardware forum but I've yet to try it.
I can't on this particular DAC, tough, since it uses single opamps.
Attachments
Last edited:
Another reliable choice for those in Europe that want to try Black Gate Standards. HiFiCollective has the 16V variety listed for 2.02 pounds.
I have been sorting through parts bin and found I have 4 of the Black Gate Std 16V that I don't need. They are the real parts and unused. To Andrew's point, they have been sitting for at least 4 years, so it would be worth reforming them before use. If you are in North America and want them, send me a PM.
Glad to hear that the Niobium caps worked out. They are a bit of a pain because they have to be kept sealed away from air until you are ready to solder them. But it is fun to say the name over and over. Niobium, Niobium, niobium..... 🙂
Jac
Nichicon KZ back in BOM (with a note)
Repeated tests at much higher volume and I can confirm that KZ problem is mostly with AD797 in my DAC.
With other opamps it fares really well but still not as well as Cerafines or Blackgates which are more neutral, refined and involving.
The Mouser cap for C9 has always been a good sounding one but not as good as it could for optimum performance (just for availability from Mouser) and KZs are much better than old BOM's FGs so as before the upgrade to the suggested best part is recommended but this time less stringent.
If I should rate the various alternates for C9 I would say:
BlackGate PK 102%, Elna Cerafine 100%, BlackGate STD 97%, Nichicon KZ 88%, Silmic 84% and Nichicon FG 75%
I've been so critical because I though, deluding myself, that a cap available from Mouser with full quality was finally found... well, simply this is not the case.
A My_Ref Fremen Edition made with Mouser BOM will sound fantastic and better than previous ones so build with confidence, I'm enjoying it just now 🙂
As it was before suggested upgrades over Mouser BOM will improve performance, timbre, detail and balance further.
A Cerafine for C9 is the first upgrade and I would buy it from the start. 😉
Repeated tests at much higher volume and I can confirm that KZ problem is mostly with AD797 in my DAC.
With other opamps it fares really well but still not as well as Cerafines or Blackgates which are more neutral, refined and involving.
The Mouser cap for C9 has always been a good sounding one but not as good as it could for optimum performance (just for availability from Mouser) and KZs are much better than old BOM's FGs so as before the upgrade to the suggested best part is recommended but this time less stringent.
If I should rate the various alternates for C9 I would say:
BlackGate PK 102%, Elna Cerafine 100%, BlackGate STD 97%, Nichicon KZ 88%, Silmic 84% and Nichicon FG 75%
I've been so critical because I though, deluding myself, that a cap available from Mouser with full quality was finally found... well, simply this is not the case.
A My_Ref Fremen Edition made with Mouser BOM will sound fantastic and better than previous ones so build with confidence, I'm enjoying it just now 🙂
As it was before suggested upgrades over Mouser BOM will improve performance, timbre, detail and balance further.
A Cerafine for C9 is the first upgrade and I would buy it from the start. 😉
Last edited:
So, I'll look for a pair of Cerafines 🙂 I'm not so good at desoldering, so I'm planning to go straight with upgrades.
I've seen many using extra-BOM C13 cap. Any upgrade worth with it? I've been told that Vishays in that place are equal or even better than alternatives.
Fine 😉
I still have to make some critical listening comparison among the CDE in BOM and the Mundorf Supreme.
What I can say so far is that the CDE940 is a really good cap and not much a limit.
The Supreme, though, will let pass more 'emotion', it's more involving but also a bit darker.
Any way desoldering C13 is really easy since holes are really big, a legacy of when the suggested cap was Audyn True Copper.
Most of the good Vishays are no longer stocked by Mouser and BTW the Supreme was clearly better.
Dario,
If you have a chance, give the CDE942 a try too. Mouser now has the lower voltage version available that should fit on the board (I only have heard the high voltage one). The 942 isn't a perfect choice either, but my sense is that it is slightly less neutral than the 940, but better at 'emotion' in a way similar to the Supreme.
ghiglie,
If you are uncomfortable at unsoldering (even using solder wick) and you want to try alternatives to see what sounds best to you, try sockets or even fastons. Sockets are better than fastons, but with fastons I use an alligator clip to hold a capacitor lead to the faston, so you can handle larger, heavier components. The sound may not be exactly the same as when soldered in place, but it is close enough to make comparisons.
Jac
If you have a chance, give the CDE942 a try too. Mouser now has the lower voltage version available that should fit on the board (I only have heard the high voltage one). The 942 isn't a perfect choice either, but my sense is that it is slightly less neutral than the 940, but better at 'emotion' in a way similar to the Supreme.
ghiglie,
If you are uncomfortable at unsoldering (even using solder wick) and you want to try alternatives to see what sounds best to you, try sockets or even fastons. Sockets are better than fastons, but with fastons I use an alligator clip to hold a capacitor lead to the faston, so you can handle larger, heavier components. The sound may not be exactly the same as when soldered in place, but it is close enough to make comparisons.
Jac