Dear Members,
I'm wondering if the huge capacitance of supercapacitors could be any good for DIY audio amplifiers in linear PSU-s. I know a voltage stabilizer (e.g. LM317T) already has a good PSRR (power supply ripple rejection) but how about using 1-2 supercapacitor(s) after the usual bridge rectification ? With current limiting of course to protect diodes (depending of type of course), that little bit of extra charging time is no problem, the main amp circuit is switched after PSU measured OK by the microcontroller anyway.
What do you think ? Has anyone done such a design ? I could also imagine 1-2 supercap(s) and then normal usual "chain" of caps like normal (low-ESR) electrolytes and finally the film ones on top of them. This way "speed" would be assured too, at least this is what I think.
Any advice, pro, cons, thoughts ?
I'm not a pro in electronics so I thank you for your patience 😉
I'm wondering if the huge capacitance of supercapacitors could be any good for DIY audio amplifiers in linear PSU-s. I know a voltage stabilizer (e.g. LM317T) already has a good PSRR (power supply ripple rejection) but how about using 1-2 supercapacitor(s) after the usual bridge rectification ? With current limiting of course to protect diodes (depending of type of course), that little bit of extra charging time is no problem, the main amp circuit is switched after PSU measured OK by the microcontroller anyway.
What do you think ? Has anyone done such a design ? I could also imagine 1-2 supercap(s) and then normal usual "chain" of caps like normal (low-ESR) electrolytes and finally the film ones on top of them. This way "speed" would be assured too, at least this is what I think.
Any advice, pro, cons, thoughts ?
I'm not a pro in electronics so I thank you for your patience 😉
Having played a bit with supercapacitors for power supplies their drawback is primarily their low maximum voltage (about 2.7V). So for any reasonable amp you need several in series, together with charge balancing circuitry to keep them all up to max charge. When you connect several supercaps in series the resulting ESR gets to be far higher than for normal 'lytics.
The AVX supercaps I tried were easy to install, came with integrated balancing resistors etc.
2,7 volts, 3,3 and series up to 5,4 volts iirc.
All these types are around a few Farads, and fit well in a power supply for e.g. digital circuits.
The ESR is very low, and reaches down to a few milliohms in the very low frequency range (LF performance like a battery, maybe better, dunno!).
That is also part of its main drawbacks: having a 5 or 3 volts power supply and these kind of impedance and capacitance draws as much current as your supply can deliver. Or more. So a soft start, or way to limit the initial charging current might come in handy(or is absolutely needed!).
I use them in digital systems, from switches to dacs and they work well. Not always easy, as mentioned.
Just try, here they are:
https://eu.mouser.com/new/kyocera-avx/avx-supercaps/Good luck!
2,7 volts, 3,3 and series up to 5,4 volts iirc.
All these types are around a few Farads, and fit well in a power supply for e.g. digital circuits.
The ESR is very low, and reaches down to a few milliohms in the very low frequency range (LF performance like a battery, maybe better, dunno!).
That is also part of its main drawbacks: having a 5 or 3 volts power supply and these kind of impedance and capacitance draws as much current as your supply can deliver. Or more. So a soft start, or way to limit the initial charging current might come in handy(or is absolutely needed!).
I use them in digital systems, from switches to dacs and they work well. Not always easy, as mentioned.
Just try, here they are:
https://eu.mouser.com/new/kyocera-avx/avx-supercaps/Good luck!
Hi,
Not to be rude to hijacker this thread but I’m also pondering to use Supercapacitors for my upcoming Aleph-X. These voltage rails will be 20V and the Supercapacitors are 2.7V/400F.
I would like to stack them up in series to 9pcs which results in 24,3V/44,4F.
However I know internal resistance and ESR come into play.
Spec-Sheet :
Eaton Supercapacitor
calculation for internal resistance is
3,2mOhm x 9 = 28,8 mOhm.
Can anyone verify or convince me otherwise not to use Supercapacitors?
Best regards,
Remco
Not to be rude to hijacker this thread but I’m also pondering to use Supercapacitors for my upcoming Aleph-X. These voltage rails will be 20V and the Supercapacitors are 2.7V/400F.
I would like to stack them up in series to 9pcs which results in 24,3V/44,4F.
However I know internal resistance and ESR come into play.
Spec-Sheet :
Eaton Supercapacitor
calculation for internal resistance is
3,2mOhm x 9 = 28,8 mOhm.
Can anyone verify or convince me otherwise not to use Supercapacitors?
Best regards,
Remco
I suspect that supercapacitors would be overkill in that application. Also, the supply that charges these caps will have a longtime overload condition at turn-on as it grunts to fill the supercapacitor up.
Thank you for your answer.
Just asking since I have these already in stock.
I know it is overkill, but I have no other use for it.
A soft start circuit should be added for sequential start up. LED power source acting as a current source until the caps are sufficiently charged.
Just asking since I have these already in stock.
I know it is overkill, but I have no other use for it.
A soft start circuit should be added for sequential start up. LED power source acting as a current source until the caps are sufficiently charged.
Just turn it on day ahead of your listening session. Then never turn it off, just disconect the amp from power supply.
Indeed, soft start and leaving on.
Seems to me overall, it’s not just too crazy thinking.
I will just have to try and see what happens.
Thank you.
Seems to me overall, it’s not just too crazy thinking.
I will just have to try and see what happens.
Thank you.
I was building tube vu meters and needed ~250 volts dc, but i did not have caps with such high voltage. Instead of waiting for order, i put a dozen of 2700uF/25V in series, with resistors for charging, all worked out great.
One issue you probably neglected is the fire risk with such high levels of energy storage - if the circuit fails all the energy in the supercaps will be dumped as heat somewhere, which might be enough to start a fire even with FR4 circuit board material (flame retardant doesn't mean it won't eventually burn given enough heat).
You certainly have a point.
Actually, charging capacitors way too quick, the internals of a capacitor also heats. Because of the amount of energy flow. This needs to be taken into consideration.
Actually, charging capacitors way too quick, the internals of a capacitor also heats. Because of the amount of energy flow. This needs to be taken into consideration.
I recently increased the capacitance of a power supply for a class d amp, since I'm using a small transformer I used a 15 ohm resistor to supply the caps, with the intention of adding a bypass switch for when that caps were up to voltage. I measured the voltage drop across the resistor whilst playing music loudly, since it was only 0.8 volt ( that equates to 0.9 watts consumption for both channels ) I don't think I'll bother with the switch, that I might forget to switch back to " charge ". I've used a 4 farad cap on an amp, and I'm planing on using 4 x 4 F caps in two monoblock amps. I'll need a lot to make up for the low ( 5v ) voltage.