advertisement


Supercapacitor Power Supply tour

Please can I join in? I can start with the 12v, unless you have a 19v solution (that would be ideal). Will be comparing to the Farad PSU (which does have a regulator in front of its caps) hence curious. UK based.
 
Please can I join in? I can start with the 12v, unless you have a 19v solution (that would be ideal). Will be comparing to the Farad PSU (which does have a regulator in front of its caps) hence curious. UK based.
I hadn't heard of Farad PSU before & see that they use voltage regulators on the output of the supercaps which is what I see most battery or supercap PSes use & is a mistake, IMO compared to direct power from supercapacitors - it will still improve the sound but not to the same level as direct aupercap power.

Comparison of both approaches would be interesting & demonstrate if I am correct :) Do it blind to avoid my priming :)

I see Farad say their supercaps are 15-50F for their 12V supply
I use 350F supercapacitors so 3 in series for 5V gives ~120F; 4 in series for 7.7V gives ~90F; 6 in series for 12V gives ~60F & so on - I stay away from the 2.7V max V limit of the supercaps

Having said all that, the current tour is a 5V supercap PS - I haven't yet put 7.5V, 12V or 19V into circulation yet

However, PM me your address & will see if I can accommodate you in upcoming 12V/19V PS tour
 
Anybody else interested in auditioning the 5V ISO-PS supply now that Bruce is finished with it?

Just a note that I've changed the design from a 2 cap to 3 cap one now in a bigger case - the same size as the ISO-DAC case

I'm also ready to send a 7.5V ISO-PS on tour if anyone is interested in trying this.
 
As one of the Ciunas power supply tour recipients, I’ve been reviewing the Supercapacitor ISO-PS, 5VDC output powering my Calyx 24/192 DAC. Because we ran into an issue with the older version (2-cap) and its SMPS, John sent me the new version with three capacitors, plus I’ve tried some different SMPS feeders. One advantage of this is that I’ve got to listen two versions of the Supercapacitor PS and three different SMPS feeding them. However, it seems the problem had been an unstable and slowly failing SMPS. My reference has been a battery powered supply with 2x18650 batteries feeding 4xLT3045 regulators in parallel to handle up to 2A. This replaced an ifiAudio iPower.

DC cable caveat: it became rather apparent that different DC cables have different sonic performance. So using the supplied DC cable with the ISO-PS gave rather different results than the DC cable I’ve been using heretofore. Best to first compare ones existing PSU to the ISO-PS using the same cable if possible.

Listening:

The Supercapacitor ISO-PS has excellent dynamics. In fact, right out of the box, once charged up, the new ISO-PS sounded almost vise-like in its control of dynamics and especially in the bass. But the soundstage presented more as a flat, two dimensional plane, rather than a more ‘living’ three dimensional stage and this seemed to come with a touch of coldness. After leaving things playing for 48-96 hours, things opened up and become much more natural in all these areas.

I’ve tried AC mains cables before where the bass weight and dynamic attack seemed a bit hypercharged, but the Supercapacitors ISO-PS does not have that generally. It has a sense of attack and pace that works well and feels natural while being rather present. Note that the supplied DC cable did push things in the direction of overdone dynamics, but most of the DC cables I used had great extension, weight, pitch control, and staging without seeming at all bloated or hyper. The Supercapacitors ISO-PS portrays music with a good sense of texture and fine detail and can produce a realistic 3D soundstage, provided one's other equipment and room setup allow. The current 3-cap version is even a little stronger in dynamic control, pitch definition, and stage depth than the older 2-cap version… in a good way. Otherwise the two versions are very similar.

Comparatively, my battery PSU with 4xLT3045/2x18650 has a little lighter touch on the pace of the bass line and lacks just that bit of attack and pace. It is not noticeable on a lot of music but one some it comes across as a light touch, slightly polite, and therefore has less apparent drive, attack, and emphasis in the note sequence of some musical passages.So, the Ciunas ISO-PS qualities help some music ‘flow’ better. Otherwise the overall quietness, fine detail, definition, sound stage depth, texture, etc. seems close.

All of this is influenced by the DC cable, its gauge, is geometry, type of shielding. I have six different DC cables, each of which interact with the ISO-PS and music differently and as significantly as different power supplies.

I fed the Supercapacitor ISO-PS with three SMPS:

Unbranded 12V 48W shunted SMPS: Supplied by Ciunas, the sound has a fat, round tone notable in the bass, and portrays the edges ever so slightly softer than the Meanwell, but imparts a breath of life to the overall sound with a natural stage depth. Good detail and staging, and is musical. But my copy sometimes became unstable, which did not affect the sound quality, so much as introducing a 2Hz pulse-noise underneath it. When the noise was not there, the sound was great.

Meanwell 9V 54W shunted SMPS: The sound has tighter definition overall but heard well in the lower bass notes. Slightly more dynamic impact. Good sound stage depth, with fine detail and nuance in HF and LF. This had no 2 Hz noise.

Gophert CPS-3205ii Adjustable power supply (0-32V 0-5A), shunted: This had the best of both worlds, albeit at a cost 2-3x that of the other SMPS. It had all the definition, detail, texture, and sound stage of the Meanwell, but revealed the Meanwell as just a hair on the cool side of neutral. By comparison, the music just flowed more organically with the Gophert feeding the Supercapacitor ISO-PS. It had the breath of life and natural musical flow whilst maintaining all the definition and HF/LF nuance of the Meanwell.
 
Thanks Barry, great review
I'm interested in talking to you further about DC cables - which do you find have the best sound?

I have some Meanwell SMPS coming - they are medical grade so very low leakage current. If you want to hold onto the ISO-PS until I send this to you? I would be interested in how it compares against the other PSes
 
Sounds good on the Medical grade SMPS.
All my DC cables are home brew:
1 Excel Cat6A 23AWG UFTP (PIMF) solid core: colours terminated as one, white terminated as one, screen drain terminated to sending side, combined 1.04 mm2
2 Canare 4S6 Starquad with braided screen added, terminated to sending side
3 Canare 4S6 Starquad spkr cable, combined 1.04 mm2
Van Damme XKE Starquad mic cable, combined 0.52 mm2
Lindy Cat6 24AWG UUTP stranded, terminated as above, braided screen added, term to sending side. Actually feels more like 26AWG.
Your supplied, coax 1.5mm i assume.

My preferences numbered. So far, I like Excel CAT6A. Best staging, HF and LF detail and definition, better bass pitch definition. Is it the solid core? Between the solid cores and the LSOH jacket it's stiffer than everything else and so a little annoying, but workable. I need to listen more to the starquad screened cable. It's nicely flexible, but requires the extra time and expense of installing a screen and heat shrink over it and terminating.

Regards,
Bruce

  • I always find that terminating a screen to sending side works better than no screen, or floating screen, etc., with a similar effect on SQ to the DC neg/Earth shunt in the SMPS.
  • Since starquad is supposed to have 3x better RFI rejection than twisted pair (acc to J Swenson), I thought than unshielded it might do as well as the screened Excel Cat6A. Unfortunately in my system, not so far.
  • Definitely, everything with higher gauge handled bass and dynamics better.
  • I've also tried floating the braid screen (unterminated) and setting as JSSG with return loop. But this gave a fuzzy effect for me. One day I might try the JSSG screen plus termination at the sending end and see how that does.
  • The supplied coax was very dynamic and bassy, but this obscured a bit of HF detail and nuance, and felt a bit bloated in comparison to my others.
 
Thanks for the comparisons
The supplied DC cable is a 5Amp 22AWG shielded coax with OFC (oxygen free copper) wire.
Some thing to try, if you are interested - if you use both voltage output ports with the two supplied DC cables &
- either use one for power & one for ground
- or connect both together effectively doubling the wire gauge

Heavier cable will deliver more current with less voltage sag so your results are not that surprising although it does show the importance of the quality of the power supply in audio setups

I agree about shield connected on sending side
 
I should be ready to send on tour a new design PS - one which has dual outputs - one output can be switched to ground isolated or non-isolated (grounded) - the other output is always isolated.

These two outputs can be used independently or combined in two ways:
- both outputs combined together to give double current output at the same voltage
- both outputs can be combined together to give double voltage

So if the ISO-PS is dual 5V then each 5V output can be used independently or combined together for 5V@>3A or 10V@2A

A ISO-PS with 5V + 7.5V outputs can be used independently or combined for 12.5V@2A
A ISO-PS with 7.5v + 7.5V outputs can be used independently or combined to give 7.5V@>3A or 15V@ 2A
A ISO-PS with 10V + 10V outputs can be used independently or combined for 10V@>3A or 20V@2A

Cables will be supplied to allow independent outputs & parallel or serial combinations of the two outputs.

Anyone interested in a tour of this ISO-PS?
 
Effectively, they are
- a 10V PS split into two independent 5V outputs
- a 12.5V PS split into 5V & 7.5V independent outputs
- a 15V PS split into two independent 7.5V outputs

All have, of course, supercapacitor direct power at the outputs - no voltage regulators on the outputs
 
Finalised the new ISO-PS for tour
It's 5V + 5V both isolated outputs & one switchable to be non-isolated i.e. negative output grounded
The charging circuitry has been moved to its own case to allow the heat produced to be kept away from the supercaps - so now two larger cases with this ISO-PS
  • each 5V can supply 3A but this produces a lot of heat & would need a heatsink on the charging case for better heat dissipation (5V@3A + 5V@3A is 30Watts)
  • the switchable 5V output can be
  • both 5V supplies can be combined together in parallel to give an even higher current output for a single 5V output
  • other ways of combining the two 5V isolated outputs are:
    • combined in series for 10V @ 3A output
    • combined in series for +/- 5V @ 3A but only if both outputs are isolated (i.e negative switchable output is not switched to ground)
  • Two of these 5V+5V supplies can be combined together in series to produce 15V & 20V outputs
Cables are provided to allow the various combinations above

I should be shipping this ISO-PS along with charging SMPS & all cables, to ZK by end of the week
 
Hi John, can the old ISO-PS (twin 5V O/P) be upgraded to the new design? Having an AlloDigiSig, I'm curious as to whether isolated supplies (as Allo recommend) has any effect over the already excellent SQ!
 
The new ISO-PS is in two larger cases - it is effectively a 10V ISO-PS split into two 5V supplies & will cost more
Why not try the tour version first to see if it makes a difference?
 


advertisement


Back
Top