RC-Monster Forums - View Single Post

BrianG · 07:21 AM

A little more in-depth information about this project for those interested. The information below DOES involve opening the case, so please heed the warning in my first post about potentially hazardous voltages!

I made another supply with an extra PS I had in my computer "junk" box (this one) just for some testing. It is an Enlight with a rating of 15A on the 12v line. This brand is generally pretty good and gets good reviews.

When I took it apart, I noticed there were a couple of small trim potentiometers near the output wires. So, with the PS on and voltmeter on the 12v line, I slowly adjusted them and one boosted the 12v from 12.1v to 12.38v. Nice. It actually boosted the 5v and 3.3v lines a little as well, but I don't care about those. The other pot didn't do anything noticable. You have to be careful here because these adjustments may be for something totally different (and potentially harming), like the PWM circuitry. So, if you decide to play with them, make sure they are put back exactly like they were if they don't affect the output. Also note that many power supplies do not have these (the first one I converted did not).

Then, I saw there were two output coils (which is usually where the circuits take a amperage sample), one on the 5v line and the other on the 3.3v line. This is a good hint that one or both are used in the load regulation circuitry. I put a 12 ohm resistor first on the 5v and then the 3.3v lines seperately. Each time, the 12v line came up a bit. However, no gains were noticed if I loaded both at the same time. I'm not saying this is true for all units, but it was for this one.

I got a couple 10 ohm 10w resistors from RadioShack and hooked them in parallel to the 5v line. This creates a 5 ohm load rated for 20W. It's a good thing too, because even though the 5v line, drawing 1A, only generates 5W, the resistors got quite warm. I mounted the resistors on a heatsink using some Arctic Silver Epoxy and that helped tremendously. I chose the 5v line because I wanted to draw at least 1A with the resistors I had. Any less and the regulation circuits may not work as well.

Now, for the test. At no load, the supply delivers 12.7v (thanks to a combination of the trim pot adjustment and the 5v load). When loaded with a 0.75 ohm load (a combo of four 3 ohm resistors in parallel), the 12v line dropped to 11.63v. This equates to 15.5A and 180 watts. Not too bad at all.

	(#12)
BrianG RC-Monster Admin Offline Posts: 14,609 Join Date: Nov 2005 Location: Des Moines, IA	09.17.2006, 04:40 AM A little more in-depth information about this project for those interested. The information below DOES involve opening the case, so please heed the warning in my first post about potentially hazardous voltages! I made another supply with an extra PS I had in my computer "junk" box (this one) just for some testing. It is an Enlight with a rating of 15A on the 12v line. This brand is generally pretty good and gets good reviews. When I took it apart, I noticed there were a couple of small trim potentiometers near the output wires. So, with the PS on and voltmeter on the 12v line, I slowly adjusted them and one boosted the 12v from 12.1v to 12.38v. Nice. It actually boosted the 5v and 3.3v lines a little as well, but I don't care about those. The other pot didn't do anything noticable. You have to be careful here because these adjustments may be for something totally different (and potentially harming), like the PWM circuitry. So, if you decide to play with them, make sure they are put back exactly like they were if they don't affect the output. Also note that many power supplies do not have these (the first one I converted did not). Then, I saw there were two output coils (which is usually where the circuits take a amperage sample), one on the 5v line and the other on the 3.3v line. This is a good hint that one or both are used in the load regulation circuitry. I put a 12 ohm resistor first on the 5v and then the 3.3v lines seperately. Each time, the 12v line came up a bit. However, no gains were noticed if I loaded both at the same time. I'm not saying this is true for all units, but it was for this one. I got a couple 10 ohm 10w resistors from RadioShack and hooked them in parallel to the 5v line. This creates a 5 ohm load rated for 20W. It's a good thing too, because even though the 5v line, drawing 1A, only generates 5W, the resistors got quite warm. I mounted the resistors on a heatsink using some Arctic Silver Epoxy and that helped tremendously. I chose the 5v line because I wanted to draw at least 1A with the resistors I had. Any less and the regulation circuits may not work as well. Now, for the test. At no load, the supply delivers 12.7v (thanks to a combination of the trim pot adjustment and the 5v load). When loaded with a 0.75 ohm load (a combo of four 3 ohm resistors in parallel), the 12v line dropped to 11.63v. This equates to 15.5A and 180 watts. Not too bad at all. Last edited by BrianG; 09.17.2006 at 07:21 AM.