(click on all pictures to zoom them)
So I posted this power supply in the "PSU discussion thread" some time ago...
This is a 100w power supply made by Delta, which I pulled from a HP Vectra system 486 66 Mhz about 5 years ago.
The basic idea is to convert this psu (well, add to it as there's room inside the case) to make it adjustable power supply with current limiting.
It's a cool psu, almost silent as fan doesn't spin much (there's thermal sensor on the schottky diodes so it probably adjusts fan as well but I'm not sure, may just be protection)
As the label says, there's 5v @ 13.5 A, 12v @ 4.5 A, 100w in total. There's also -5v @0.1A and -12v @ 0.3A , but they use linear regulators for those L7905CV and L7912CV which I believe are rated for max 1A so who knows... anyway, not interested in those.
Decent mains filtering, main capacitor is Rubycon USB 270uF , 400v - enough for the wattage, with ~1.7 ripple and 2000 hours @ 85c if i remember the datasheet values correctly.
Bridge rectifier with one of those things (the proper name escapes me now) glued to it that I assume will limit current when the rectifier overheats)
The Schottky diodes are s30sc40m (40v 30a) which seems to be used for the 5v and a much smaller one that has BY208 written on it, for which I couldn't find a datasheet (but I guess it's enough for 12v 4.5v) ... there's a thermal sensor attached to the small metal heatsink and the metal heatsink is screwed to the case so there's plenty of heatsink.
Can't determine the main switcher chip due to heatsink, but there's a 1-3w resistor in shrinkwrap almost glued to it that looks to be a bit burned. On the right of the transformer there's another to220 chip that I can't tell what it is due to the heatsink clip.
The controller is uc3842an, 500Khz frequency.
Front pictures:
Back :
The capacitors are all UCC SXE, KME... the large ones are 2200uF 16v (1) and 10v (2) in the mains section, in the secondary there's another set of 2200uF 10v for the 5v lines, a lonely Rubycon 1000uF 10v, and a Taicon 1000uF 16v for the 12v...
The -12v has a UCC 220uF 35v and the -5v I think has... nothing, don't see any capacitor by the regulator larger than 10uF... the 5v stand by with the "huge" 5 mA rating has a 47uF 25v.
All the other small capacitors are UCC and a few Taicon spread around.
I started it by shorting the ps/on wire to ground and the 5v line measures 5.3v but the 12v line measures 10.96v. 5v standby was 5.03v. Obviously 10.96v on the 12v rail is bad, but right now I can't tell if this is due to no load or for other reasons.
Your thoughts?
Like I said, I started it with no load, and I don't really have something to put load on it right now (besides 12v fans which is not enough to be sure). Planning to buy some 12v 5-10w light bulbs from some car parts store so that there's gonna be some load.
So basically, I'm planning to replace all the 2200uF 10v and 16v with UCC KZE 3300uF 16v as I have about 40 of those here and I don't really use them - they're larger than the common 10mm diameter so don't have a chance to use them much.
I don't think it would affect the stability of this power supply to go from 2 x 2200 to 2x 3300, right?
I'll have to test the capacitors ESR after desoldering them from the board.. wouldn't surprise me if they have high esr, considering they're from around 1995.
Other suggestions for solving the 12v rail issue?
So I posted this power supply in the "PSU discussion thread" some time ago...
This is a 100w power supply made by Delta, which I pulled from a HP Vectra system 486 66 Mhz about 5 years ago.
The basic idea is to convert this psu (well, add to it as there's room inside the case) to make it adjustable power supply with current limiting.
It's a cool psu, almost silent as fan doesn't spin much (there's thermal sensor on the schottky diodes so it probably adjusts fan as well but I'm not sure, may just be protection)
As the label says, there's 5v @ 13.5 A, 12v @ 4.5 A, 100w in total. There's also -5v @0.1A and -12v @ 0.3A , but they use linear regulators for those L7905CV and L7912CV which I believe are rated for max 1A so who knows... anyway, not interested in those.
Decent mains filtering, main capacitor is Rubycon USB 270uF , 400v - enough for the wattage, with ~1.7 ripple and 2000 hours @ 85c if i remember the datasheet values correctly.
Bridge rectifier with one of those things (the proper name escapes me now) glued to it that I assume will limit current when the rectifier overheats)
The Schottky diodes are s30sc40m (40v 30a) which seems to be used for the 5v and a much smaller one that has BY208 written on it, for which I couldn't find a datasheet (but I guess it's enough for 12v 4.5v) ... there's a thermal sensor attached to the small metal heatsink and the metal heatsink is screwed to the case so there's plenty of heatsink.
Can't determine the main switcher chip due to heatsink, but there's a 1-3w resistor in shrinkwrap almost glued to it that looks to be a bit burned. On the right of the transformer there's another to220 chip that I can't tell what it is due to the heatsink clip.
The controller is uc3842an, 500Khz frequency.
Front pictures:
Back :
The capacitors are all UCC SXE, KME... the large ones are 2200uF 16v (1) and 10v (2) in the mains section, in the secondary there's another set of 2200uF 10v for the 5v lines, a lonely Rubycon 1000uF 10v, and a Taicon 1000uF 16v for the 12v...
The -12v has a UCC 220uF 35v and the -5v I think has... nothing, don't see any capacitor by the regulator larger than 10uF... the 5v stand by with the "huge" 5 mA rating has a 47uF 25v.
All the other small capacitors are UCC and a few Taicon spread around.
I started it by shorting the ps/on wire to ground and the 5v line measures 5.3v but the 12v line measures 10.96v. 5v standby was 5.03v. Obviously 10.96v on the 12v rail is bad, but right now I can't tell if this is due to no load or for other reasons.
Your thoughts?
Like I said, I started it with no load, and I don't really have something to put load on it right now (besides 12v fans which is not enough to be sure). Planning to buy some 12v 5-10w light bulbs from some car parts store so that there's gonna be some load.
So basically, I'm planning to replace all the 2200uF 10v and 16v with UCC KZE 3300uF 16v as I have about 40 of those here and I don't really use them - they're larger than the common 10mm diameter so don't have a chance to use them much.
I don't think it would affect the stability of this power supply to go from 2 x 2200 to 2x 3300, right?
I'll have to test the capacitors ESR after desoldering them from the board.. wouldn't surprise me if they have high esr, considering they're from around 1995.
Other suggestions for solving the 12v rail issue?
Comment