Power Supply Electrolytics - Low Z, High Ripple?

[tvr2500m]

I'm futzing about with one of the DIY Pete Millet headphone amps. I'm curious about power supply capacitors. The stock BOM for this amp calls for a general application Nichicon VY-series aluminum electrolytic cap. I recently built a Bottlhead Crack + Speedball, which I'm enjoying enormously. Th Crack uses a Nichicon CS-series high ripple switching power supply capacitor.

Looking at capacitor lines and design applications, some capacitors are placed under Low Z for switching supplies, some high ripple current.

Which type might be a better performing switching supply capacitor? Low Z for switching supplies or one designed for high ripple current?

Thanks.

 

[Paul Birkeland]

When I have to select a power supply capacitor, I will model the supply (or measure the existing supply if I have a prototype) to get a measurement of the ripple current required for a given cap.  There's no penalty for buying a cap with a higher ripple current rating, but be careful that you aren't buying a 3,000 hour at 85C cap just because it says it can handle 3A of ripple current.

For SMPS, the ripple frequency will be much, much higher than what you're going to get from a transformer/rectifier combo.  The ESR/ESL of an electrolytic cap may be enough of a problem under these conditions that putting a MLC capacitor in parallel with the electrolytic will help reduce power supply ripple at the switching frequency. 

If the project in question is Pete's NuTube design, you'll notice C19 and C18 are performing that function.  You can always ask Pete, he's replied to every e-mail I've ever written him. 

 

[tvr2500m]

Thanks, PB. I may ping Pete. Yes, he's always helpful and informative.

The headphone amplifier in question is the DIY Butte. I'm experimenting to see whether I might wring out just a bit more refinement.

 

[Paul Birkeland]

There is no switch-mode power supply in that project, so there's no need to focus in on that.  You can model the power supply in PSUD (free software) to look at the ripple current that needs to be handled by those power supply caps.

 

[tvr2500m]

Looking at the schematic and Pete's description, yeah, it uses linear regulators... I sent a note to Pete.

 

[Paul Birkeland]

It uses linear regulators and an AC wall transformer.