advertisement


Supercapacitor Power Supply tour

Right butit's not "lack of regulation" - it's where the regulation happens - I use regulation before the supercaps because outputting power direct from supercapacitors (or batteries) makes a significant difference to having a voltage regulator on the output. It's not just about being limited by the power handling of the regulator but rather about the way regulators handle dynamic current delivery.

As regards capacitors, AFAIK,
- The Uptone unit uses two banks of 7 10F Nesscap supercapacitors in series to increase the voltage handling capability of each bank to 7 * 2.7V i.e about 19V but this is regulated down to 12V max, I believe @ 1.1Amp max current output

-The downside of this is that the capacitance is decreased from 10F to 1.4F. More importantly each 10F supercap impedance is 26mohm & 7 * 26 gives 182mohm

- For 20V PS, I use 8 * 350F Nesscap supercapacitors in series to give a total of 47F & 20V max. The impedance of each 350F supercap is 3.2mOhm so 8 * 3.2 is 25mohm

- For 5V PS output, I use 2 of these 350F supercaps giving 6.4mOhm

- the importance of impedance is that the lower the impedance the faster current is delivered.

One has to bear in mind, however that these low impedances are probably moot as the power has to travel through pcb tracks, wiring & cabling (& usually other voltage regulators) before the power reaches the actual components (In my audio devices I avoid, as much as possible, regulators between power & components). Having said all this, raw supercap power seems to make a difference when powering audio devices which is audible.


The supercap PS must use a charger voltage rating which is 2V greater than the rated voltage output of the PS. More powerful is better just to ensure that the supercaps are always fully charged when in operation i.e there's no possibility of a sag.

The other aspect is that when supercaps need to charged up from low charge (initially or when unit is off charge for a week or more), a more powerful PS will ensure the PS comes up to its rated output voltage in less time - we are only talking about 5 mins or so, anyway. The recommended operation is to always keep it on charge.

I plan to use as a charger my most powerful PSU in the voltage range with a 2.1 plug and use a 2.1 male to male lead with a 2.1 female to micro USB adaptor for the output to the Qutest. I assume the connections on your PS (in and out) need 2.1 male plugs and are centre positive?

This is a post from the Qutest thread on Head-fi by Rob Watts the Chord DAC Designer. I'd be interested in your comments:-

"One benefit of super-caps is that it lowers the PSU impedance, and hence reduces the distortion crosstalk feed from the PSU. But this problem is eliminated with Hugo 2 and Qutest's PSU arrangement of regulating out from a boost voltage - so now the gain section is fed with clean power, eliminating the distortion that comes through the power rails. Of course, you have the benefit of huge current delivery from super caps - but that is not needed for a DAC."
 
Do we take it as gospel because the designer said it? I'm not arguing that he's wrong, but he's not considering the effects of that little smps spewing crap back into the mains on the rest of my kit.

How many of us run gear from pre smps days, hell even pre filtered iec socket days?
 
I plan to use as a charger my most powerful PSU in the voltage range with a 2.1 plug and use a 2.1 male to male lead with a 2.1 female to micro USB adaptor for the output to the Qutest. I assume the connections on your PS (in and out) need 2.1 male plugs and are centre positive?
Yes, that's correct
This is a post from the Qutest thread on Head-fi by Rob Watts the Chord DAC Designer. I'd be interested in your comments:-

"One benefit of super-caps is that it lowers the PSU impedance, and hence reduces the distortion crosstalk feed from the PSU. But this problem is eliminated with Hugo 2 and Qutest's PSU arrangement of regulating out from a boost voltage - so now the gain section is fed with clean power, eliminating the distortion that comes through the power rails. Of course, you have the benefit of huge current delivery from super caps - but that is not needed for a DAC."
OK, I don't want to get into one manufacturer pitted against another so I'll just talk in general. I said before that the extremely low impedance of the supercaps may well not be the reason for their superiority as a PS - I believe it has something to do with their lack of noise when delivering dynamic current. Some internal components within DACs & digital devices have dynamic current demands. This dynamic noise seems to be an issue with voltage regulators but this is just an educated guess.

When using battery power, I've done this experiment a number of times to have drawn that conclusion above. I've used battery power to power a device in the audio chain & noticed an audible improvement. If the device was suitable, I then bypassed any regulators in the device & powered directly with batteries - this brought an even bigger improvement in sound. These regulators were not LM317 or old regs but some of the modern, low noise regs, (TPS7A4700, LT3045).

BTW, this principle also applies in other areas - for instance I find that ASRC (resamplers) on the input of DACs also limits the sound. It improves low quality digital signal but doesn't allow top quality digital signal to reach it's audible level - remove this ASRC & that limitation is removed giving more audible improvements with good quality digital signals being easily identified

So the internal treatment of power within any device can limit the benefit that supercap power can achieve. Not knowing the internals of a device, the only way to evaluate this is by trying it.

You said you had already formed the opinion that the PS into the Qutest made a difference - care to say why? Have others tried different PSes into the Qutest?
 
Yes, that's correct

OK, I don't want to get into one manufacturer pitted against another so I'll just talk in general. I said before that the extremely low impedance of the supercaps may well not be the reason for their superiority as a PS - I believe it has something to do with their lack of noise when delivering dynamic current. Some internal components within DACs & digital devices have dynamic current demands. This dynamic noise seems to be an issue with voltage regulators but this is just an educated guess.

When using battery power, I've done this experiment a number of times to have drawn that conclusion above. I've used battery power to power a device in the audio chain & noticed an audible improvement. If the device was suitable, I then bypassed any regulators in the device & powered directly with batteries - this brought an even bigger improvement in sound. These regulators were not LM317 or old regs but some of the modern, low noise regs, (TPS7A4700, LT3045).

BTW, this principle also applies in other areas - for instance I find that ASRC (resamplers) on the input of DACs also limits the sound. It improves low quality digital signal but doesn't allow top quality digital signal to reach it's audible level - remove this ASRC & that limitation is removed giving more audible improvements with good quality digital signals being easily identified

So the internal treatment of power within any device can limit the benefit that supercap power can achieve. Not knowing the internals of a device, the only way to evaluate this is by trying it.

You said you had already formed the opinion that the PS into the Qutest made a difference - care to say why? Have others tried different PSes into the Qutest?

Thanks for this. Very interesting.

I've only ever used the supplied SMPS with the Qutest. My major experience with power suppliers was when I bought a cheap streamer (£150 Innuos Wave - cubox based) to feed the original Hugo (it's only 1's and 0's, right?).
I tried their battery upgrade and that made the sound less grainy compared with the supplied SMPS. I then tried a £250 linear PSU and that gave a good all round improvement. Finally I got a £500 Linear PSU from the same company Custom Hifi Cables. The improvement to the cheaper PSU was dramatic, like getting a more powerful amplifier. I was pretty amazed. Bear in mind we are talking about a very cheap, minuscule piece of computer equipment and the linear supplies were designed by the same person - Dr Sean Jacobs who does the power supply design for the Innuos reference streamer.

I have a lot of respect for Rob Watts' kit but it seems a very big claim that with a bit of clever design he can neutralise the effect of the power supply feeding the equipment. It seems contrary to what the vast majority of the industry believes so I'll keep an open mind. It will be a good test to see if the super capacitor PS makes a difference to the Qutest especially as there are two Qutest owners on the list.

If your PS meets USB spec it won't invalidate the warranty. In that regard if the power to the charger is accidentally lost would the subsequent voltage drop from the super capacitors be likely to damage the Qutest?
 
Thanks for this. Very interesting.
.........
If your PS meets USB spec it won't invalidate the warranty. In that regard if the power to the charger is accidentally lost would the subsequent voltage drop from the super capacitors be likely to damage the Qutest?
This 5V supercap PS outputs 5V @ upto 3Amps current - this is > USB spec of 500mA for USB 2 or 1Amp for USB 3

If the charger is pulled from the PS there is no over-voltage condition - the voltage will just gradually decline. This should not cause any difficulty with any DAC - my battery powered DACs simply sputtered & stopped when this condition arose. So if pulling the current SMPS from the Qutest causes no damage then the same applies to this PS
 
This 5V supercap PS outputs 5V @ upto 3Amps current - this is > USB spec of 500mA for USB 2 or 1Amp for USB 3

If the charger is pulled from the PS there is no over-voltage condition - the voltage will just gradually decline. This should not cause any difficulty with any DAC - my battery powered DACs simply sputtered & stopped when this condition arose. So if pulling the current SMPS from the Qutest causes no damage then the same applies to this PS

Thanks, JK. Pulling the plug is the only way of powering down a Qutest!

Has the PS tour started? I'd be interested in Owl's view given the impending Brexit omnishambles date.

I currently have linear PSU's for my server (Aries Mini) and streamer (SOTM Ultra
Neo). I know the linear psu makes a big difference to the streamer's original SMPS. How likely is it that your SC PSU will represent an improvement on the linear supplies (Custom Hifi Cables DC2 andDC3)? Does supplying dual voltages compromise quality? One option would be to skip the tour and order a 5v/12v dual (Qutest and SOTM) and use the 30 return period if there is no significant improvement.
 
Thanks, JK. Pulling the plug is the only way of powering down a Qutest![/quost]Ah, right - I meant if the SMPS powered down for any reason while still plugged into the Quest

[quoest]Has the PS tour started? I'd be interested in Owl's view given the impending Brexit omnishambles date.
Hasn't started yet - just some delays at my end with parts

I currently have linear PSU's for my server (Aries Mini) and streamer (SOTM Ultra
Neo). I know the linear psu makes a big difference to the streamer's original SMPS. How likely is it that your SC PSU will represent an improvement on the linear supplies (Custom Hifi Cables DC2 andDC3)?
Hard to know without trying?
Does supplying dual voltages compromise quality? One option would be to skip the tour and order a 5v/12v dual (Qutest and SOTM) and use the 30 return period if there is no significant improvement.
I'm waiting on some parts for supercap PS output with mixed dual supply - at the moment I'm only sending out dual supplies of the same voltage i.e 5V + 5V or 12V + 12V
 
Hasn't started yet - just some delays at my end with parts

Hard to know without trying? I'm waiting on some parts for supercap PS output with mixed dual supply - at the moment I'm only sending out dual supplies of the same voltage i.e 5V + 5V or 12V + 12V

Qutest, quost, quoest? Are we conjugating Latin verbs here?

I'm very happy to try. When do you anticipate the mixed dual supply will be available?

You didn't respond to the quality question re mixed dual supplies re single supplies so I guess I need to deduce from a trial?
 
Qutest, quost, quoest? Are we conjugating Latin verbs here?

I'm very happy to try. When do you anticipate the mixed dual supply will be available?
Not sure, atm, I will let you know

You didn't respond to the quality question re mixed dual supplies re single supplies so I guess I need to deduce from a trial?
It is difficult to answer these questions but some people determine that isolated power supplies are important yet I suggest that it might depend on how stable & low impedance the power supply is i.e how supplying current to one device interferes with supplying the power to another device from the same supply - how they interact. The superacapacitors, like batteries, have an enormous capacity for absorbing noise from the load & being impervious to noise/ripple from the PS charging them. To me, this is the issue with voltage regulators - they seem to not have this enormous sink for noise from the load & when power i supplied to two device from such regulated power they seem to interact.
 
Thank, JK. I certainly will give the dual voltage PS a try. I've tried the other types of power supply so it will be interesting to try the supercapacitor version especially if you can get 'two for the price of one'!

I'll pm you with my details so you can let me know when the PS is available.
 
The picture on the website for the ISO-PS shows 5V DC against the input socket. Presumably the new models will show the voltage range? I plan to use a laptop PS which is 19V and 3.9amps.

What is the estimated life of the supercapacitors and what happens when they degrade?
 
The picture on the website for the ISO-PS shows 5V DC against the input socket. Presumably the new models will show the voltage range? I plan to use a laptop PS which is 19V and 3.9amps.
Yes, I'm looking at the best way to do this, atm

What is the estimated life of the supercapacitors and what happens when they degrade?
Super caps don't wear out like batteries on charge/discharge cycles, but they do lose capacitance with time and temperature and applied voltage. Suppliers indicate 100,000 hours of constant usage. Then the capacitance will drop by 30%; but 100,000 hours is 11 years constant use; for a 50% reduction you would be looking at 30 to 100 years (30 years at 45 deg C). I run the capacitors below their 2.7V rating & trickle charge them, which improves on these estimates - (I've had my battery powered devices in the field for 10+ years without any problems).

Even if the capacitance drops by 50% I doubt there will be any noticeable effect if 350F capacitors reduce capacitance, in 30 years, to 175F capacitance? (Some other supercap power supplies use 60F capacitors)
 
Yes, I'm looking at the best way to do this, atm

Super caps don't wear out like batteries on charge/discharge cycles, but they do lose capacitance with time and temperature and applied voltage. Suppliers indicate 100,000 hours of constant usage. Then the capacitance will drop by 30%; but 100,000 hours is 11 years constant use; for a 50% reduction you would be looking at 30 to 100 years (30 years at 45 deg C). I run the capacitors below their 2.7V rating & trickle charge them, which improves on these estimates - (I've had my battery powered devices in the field for 10+ years without any problems).

Even if the capacitance drops by 50% I doubt there will be any noticeable effect if 350F capacitors reduce capacitance, in 30 years, to 175F capacitance? (Some other supercap power supplies use 60F capacitors)

I'm 66 so they will outlast me!

Any news on likely shipping dates for the mixed fixed voltage model?
 
Having PMed with Dan, it seems like a 25V version is feasible for Naim HiCap users.
I propose to put something together along these lines for some user(s) to evaluate to see if it's worth proceeding with looking at the ergonomics of half case size.
Having looked into the HiCap a bit it looks like it has 3 input sockets on the back for signal input (CH1 & CH2 & signal common) but looking at the inside shots of the HiCap it appears that CH1 has a wire looped to each of the 3 input sockets (same for CH2 & signal common) - which means all three sockets are the same i.e only 1 input can be used? This couldn't be right - what am I missing?

The CH1, CH2 & signal common then loop into a 5 pin DIN? socket which also outputs two 24V power lines
Is this nearly correct?
If you’re going to get into the Naim preamp PSU market, you’ll be competing against others such as Teddy Pardo and Avondale, but if you think you have something to offer, I’d say go for it.
Worth bearing in mind that some Naim preamps use the NAPSC to separately power the digital stuff. So there might be some interest in an alternative, or a combined Hicap/NAPSC to keep the box count low.
 
I'm 66 so they will outlast me!

Any news on likely shipping dates for the mixed fixed voltage model?
I'm having some supply issues at the moment so some delays on all devices, nor just dual mixed power supply.
Shouldn't be too long before shipping standard PSes
 
If you’re going to get into the Naim preamp PSU market, you’ll be competing against others such as Teddy Pardo and Avondale, but if you think you have something to offer, I’d say go for it.
Worth bearing in mind that some Naim preamps use the NAPSC to separately power the digital stuff. So there might be some interest in an alternative, or a combined Hicap/NAPSC to keep the box count low.
I'm testing this Naim niche as it was suggested to me - tour users will determine if my PS is worthwhile & where
 
Keep watching :)

I've ordered a 12v/5v version. I'll be very interested to see if it improves my Qutest over the SMPS and wheter it can improve on the SOTM Ultra's linear supply.

Has the tour started yet?

Given the advantages of SC PSU's (cost, size, efficiency and performance) I'm surprised they are not widely used especially for phono, pre and power amps
 


advertisement


Back
Top