UCD snubberizing, already done?

Status
Not open for further replies.
Jump to Latest
#2
Carlosfm, how would you do this on the UCD. Remember Bruno saying you don't need to because of bypassing done on the UCD already.
I use single bridge rectifier(normal diode) + 2off 10000uF BC caps type 154 per amp. 1 Toroid 300 or 500VA for a stereo amp.
 
#3
Hi,

What "snubberizing" is and what is already done on the UCD is very different I believe.

Do it the same way you would any other supply.

Actually they're not entirely different, Bruno uses an electrolytic and relies on the ESR as well.

Doesn't mean you can't try improving on it but it may not be worth it you'll have to try and see.
 
#4
Bgt said:
Carlosfm, how would you do this on the UCD. Remember Bruno saying you don't need to because of bypassing done on the UCD already.
I use single bridge rectifier(normal diode) + 2off 10000uF BC caps type 154 per amp. 1 Toroid 300 or 500VA for a stereo amp.
Click to expand...

Try this:
100nf bypass + 0.1R + 3.3nf (MKT) right after the 10,000uf caps.
 
#11
Hi,

If you omite the 100nF bypass it would be ~3Ghz?

Obviously I need to research this more (take the time to do the math anyway) but at least I got the point of it 🙂

I'll never say it's pointless or pretend it doesnt' have any beneficial effects because it simply makes sense. Howsoever, this is followed by the onboard bypassing which I think would impart its own bandwidth, and do it where it happens to matter most.

I think this step would make your supply more transparent but you'll still be listening to that electrolytic bypass cap on board which is doing the same thing, only perhaps less well?

Sorry if I missed it what's your recommended wattage on that resistor? I'm curious to try it for all that the parts will cost.

Thanks,
Chris

PS: .1 ohm seems like a most excellent choice here 🙂
 
#13
classd4sure said:
I think this step would make your supply more transparent but you'll still be listening to that electrolytic bypass cap on board which is doing the same thing, only perhaps less well?
Click to expand...

I don't know the UCD in full detail, but if there's a big electrolythic on-board (1,000uf or more), snubberize it too.

This is an experiment you can make, and costs next to nothing.
I'm out of UCDs now. 😀 😀 😀

classd4sure said:
Sorry if I missed it what's your recommended wattage on that resistor? I'm curious to try it for all that the parts will cost.
Click to expand...

I use 0.1R / 1W, because it's the only wattage rating I can get here at such low impedance without going to 5w+ wire-wound resistors. 1W is fine, even 0.5W would do.
They are bog standard 5% carbon resistors.

For the caps I use Philips/BC MKT, 100V rating.
You don't want big (and inductive) caps here, 100V MKTs are fine.

classd4sure said:
PS: .1 ohm seems like a most excellent choice here 🙂
Click to expand...

Ya. 😉
 
#16
3 GHz.........?

You are joking, right?

How many guys here can accurately simulate PSU impedance that high? Trace width.......thickness........length.........etc.........bends.....twists............proximity to other traces.............

Nope, doubt anyone comes close at 3 GHz. I have my doubts about 100 MHz, but I'll behave today.

Jocko
 
Status
Not open for further replies.
Top Bottom