What size Capacitors to use?

Paul Dimaline · 25th November 2004, 03:24 PM

Im building a pre-amp PSU and i'm using 2 off 75VA laminated transformers with 24-0-24v secondaries and half wave rectification.
I have a couple of 15,000uf caps but would think that is to much for the relativly small transformers. I'm considering either a pair of 4700 or 6800uf caps, does this seem about right?
I've looked on the net to try and find any advice on transformer size in regard to capacitor size but didnt come up with anything.

TIA,
Paul.

markp · 25th November 2004, 03:45 PM

You might want to try a few 1000uF caps in parallel as they are cheaper and have lower series resistance all together than single larger caps will. Also, bypass your diode bridge with small film caps in parallel with the diodes.

hlyytine · 25th November 2004, 04:33 PM

Quote:

Originally posted by markp
You might want to try a few 1000uF caps in parallel as they are cheaper and have lower series resistance all together than single larger caps will. Also, bypass your diode bridge with small film caps in parallel with the diodes.

Actually, you should not bypass those diodes but place a series RC filter across the rectifier bridge output. Please look at http://www.hagtech.com/pdf/snubber.pdf.

richie00boy · 25th November 2004, 04:43 PM

Firstly I have to ask why you are only halfwave rectifying not full. Fullwave will give you less ripple for a given size cap.

As for cap size, it won't hurt to have as big a cap as you can get.

Paul Dimaline · 25th November 2004, 05:18 PM

I'm building a Naim style PSU, hence the halfwave rectification.

See here for layout.

If the output from the caps are identical I want to try hanging just the single reg from the caps to make a totaly mono supply.

Paul.

lpm76 · 25th November 2004, 05:19 PM

Quote:

Originally posted by hlyytine

Actually, you should not bypass those diodes but place a series RC filter across the rectifier bridge output. Please look at http://www.hagtech.com/pdf/snubber.pdf.

Well, I for one can't figure out how to calculate the optimal snubber values. I get lost in the math every time I read that paper. Isn't there another place where it is explained in a better way?

markp · 25th November 2004, 05:42 PM

Quote:

Originally posted by richie00boy
Firstly I have to ask why you are only halfwave rectifying not full. Fullwave will give you less ripple for a given size cap.

As for cap size, it won't hurt to have as big a cap as you can get.

A single big cap is not as good a few smaller caps that add up to the same value. The smaller caps will also add up to less money.

hlyytine · 25th November 2004, 06:25 PM

Quote:

Originally posted by lpm76

Well, I for one can't figure out how to calculate the optimal snubber values. I get lost in the math every time I read that paper. Isn't there another place where it is explained in a better way?

Forget the advanced math, page 9 gives you an example on how to calculate the exact values. Doing the maths is not that hard; measuring the leakage inductance of the transformer is more complex. I can't think of any method how to do this with a DVM (with the most expensive true-RMS-meters it could be possible). The oscilloscope helps a lot.

PRR · 26th November 2004, 06:58 PM

> 15,000uf caps but would think that is to much for the relativly small transformers.

Why?

Yes, it takes a while to fill 15,000uFd. But once filled, they don't impose any extra load on the transformer. I have a box with 15,000uFd fed from a 24VA core, has been running happy for 25 years.

And my motto has always been what Richie says: "it won't hurt to have as big a cap as you can get."

Over-capping does raise peak current. Peak current depends a little on load, but for low-ripple supplies it depends much more on series resistance, which depends on transformer rating. Rectifiers should never be less than transformer rated current, plus 2 to 5 multiplier safety margin. Your 1.5A windings suggest 3A to 7.5A rectifiers, and go higher when you have big caps. Since 15A is cheap in plain silicon, I'd go that high. Yes, even if you only really need 0.1A to the load. If you like exotic rectifiers, cost and reliability are something to ponder.

If you do like the idea of dozens of smaller caps, split them up as C-R-C. 3,000uFd, 1 ohm, 12,000uFd. This helps clean some of the rectifier spikes from the supply rails. While I understand the point Hagerman makes, and have seen it, personally I like brute-force cleaning rather than specific tricks, at least in my own DIY. But there are many ways to skin cats.

75VA seems like a lot for a preamp. I confess I've done similar over-kill.

Paul Dimaline · 29th November 2004, 09:16 AM

Thank you for the reply's, even if some of them were WAAAYYYYY over my head!!
I will use the pair of 15000uf caps that i've already got and see how I get on with them.
Have any of you had any experiance of pairing up the regs? Im using Avondale TPR's so it is possible to trim the outputs to match each other. The idea is that I can end up with a dedicated mono PSU for each rail, all be it at some cost in space and cash!

Paul.

25th November 2004, 03:24 PM	#1
Paul Dimaline diyAudio Member Join Date: Feb 2004 Location: Lincolnshire, UK	What size Capacitors to use? Im building a pre-amp PSU and i'm using 2 off 75VA laminated transformers with 24-0-24v secondaries and half wave rectification. I have a couple of 15,000uf caps but would think that is to much for the relativly small transformers. I'm considering either a pair of 4700 or 6800uf caps, does this seem about right? I've looked on the net to try and find any advice on transformer size in regard to capacitor size but didnt come up with anything. TIA, Paul.

25th November 2004, 03:45 PM	#2
markp diyAudio Member Join Date: Oct 2003 Location: L.A., CA	You might want to try a few 1000uF caps in parallel as they are cheaper and have lower series resistance all together than single larger caps will. Also, bypass your diode bridge with small film caps in parallel with the diodes. __________________ If it sounds good... it is good!

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
The size of electrolytic capacitors?	roger-k	Pass Labs	36	14th September 2008 08:07 PM
what is the difference between silver mica capacitors and ceramic capacitors ??	prorms	Solid State	1	6th May 2008 12:38 PM
what size capacitors to use?	mgriff	Power Supplies	6	16th August 2006 11:04 AM
Speaker size v Room size	cliffy	Multi-Way	6	19th February 2005 08:34 AM
differences between computer grade capacitors and small can capacitors	LBHajdu	Pass Labs	0	22nd February 2002 01:29 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

25th November 2004, 04:43 PM	#4
richie00boy Did it Himself diyAudio Member Join Date: Nov 2003 Location: Gloucestershire, England, UK	Firstly I have to ask why you are only halfwave rectifying not full. Fullwave will give you less ripple for a given size cap. As for cap size, it won't hurt to have as big a cap as you can get.

25th November 2004, 05:18 PM	#5
Paul Dimaline diyAudio Member Join Date: Feb 2004 Location: Lincolnshire, UK	I'm building a Naim style PSU, hence the halfwave rectification. See here for layout. If the output from the caps are identical I want to try hanging just the single reg from the caps to make a totaly mono supply. Paul.

26th November 2004, 06:58 PM	#9
PRR diyAudio Member Join Date: Jun 2003 Location: USA	> 15,000uf caps but would think that is to much for the relativly small transformers. Why? Yes, it takes a while to fill 15,000uFd. But once filled, they don't impose any extra load on the transformer. I have a box with 15,000uFd fed from a 24VA core, has been running happy for 25 years. And my motto has always been what Richie says: "it won't hurt to have as big a cap as you can get." Over-capping does raise peak current. Peak current depends a little on load, but for low-ripple supplies it depends much more on series resistance, which depends on transformer rating. Rectifiers should never be less than transformer rated current, plus 2 to 5 multiplier safety margin. Your 1.5A windings suggest 3A to 7.5A rectifiers, and go higher when you have big caps. Since 15A is cheap in plain silicon, I'd go that high. Yes, even if you only really need 0.1A to the load. If you like exotic rectifiers, cost and reliability are something to ponder. If you do like the idea of dozens of smaller caps, split them up as C-R-C. 3,000uFd, 1 ohm, 12,000uFd. This helps clean some of the rectifier spikes from the supply rails. While I understand the point Hagerman makes, and have seen it, personally I like brute-force cleaning rather than specific tricks, at least in my own DIY. But there are many ways to skin cats. 75VA seems like a lot for a preamp. I confess I've done similar over-kill.

29th November 2004, 09:16 AM	#10
Paul Dimaline diyAudio Member Join Date: Feb 2004 Location: Lincolnshire, UK	Thank you for the reply's, even if some of them were WAAAYYYYY over my head!! I will use the pair of 15000uf caps that i've already got and see how I get on with them. Have any of you had any experiance of pairing up the regs? Im using Avondale TPR's so it is possible to trim the outputs to match each other. The idea is that I can end up with a dedicated mono PSU for each rail, all be it at some cost in space and cash! Paul.