I'm planning on recording the upcoming gig of the band i'm in with my Eos camera, without the worry of the battery crapping out. As such, i sprang for one of these, i guess, "generic" supplies that come with a "coupler" that fits into the battery compartment.
I didn't manage to keep Dave's EEVblog motto "to the letter" (ie. "Don't turn it on - take it apaaaart!.."), and i did plug the bugger into the mains - works fine, 7.495v with no load.
Next step - cracking it open. Your usual cheapo sealed casing (like you'd find on laptop power bricks), although fortunately, they hadn't gone crazy with the glue. Medium pressure every half-inch or so, along the seams, with a small flathead, and i could already hear the tell-tale "pop"s of glue giving in
Photo 1: "before"
Mildly impressive, i must admit - it COULD have been (much) worse.
1A fuse, no NTC or any sort of noise filtering on the AC input. 4x 1N4007's doing the rectifying, 10u/400v "Chongx" cap (seen above the board, next to its "heir"). PWM chip is an SDC606 - never heard of'em before. Definitely an uber-cheapo - its built-in switching element is a 450v BJT
Surprisingly enough, the primary-to-secondary noise-reduction cap (CY1, the "fatter" blue ceramic south of the transformer) ***IS*** in fact a mains-rated cap - go figure
Secondary's quite simple - SDC5100 (5A 100v) schottky rectifier, a pair of Ketuo (R) 470u/16v caps, and that's about it. SOT23-cased TL431 on the bottom, with the associated passives, and PC817 knockoff photocoupler.
Photo 2: "after"
Couldn't help myself - i "had to" recap this thing, for peace of mind and longevity. 22u/400v Rubycon WXA after the rectifier bridge - this one needed a bit of creativity, to shoehorn it in there
I figured i'd "do the right thing" and heatshrink the leg-extensions, and Kapton-tape the top of the cap, due to the proximity to the diodes.
I had to relocate the far end of R6 (start-up resistor from the rectified mains into the PWM chip), but that was trivial. I might consider slipping in an inductor into the intended footprint, but i couldn't be bothered scrounging one up today. 47u/25v Ketuo Vcc cap (for the PWM chip) got replaced with a 47u/50v Rubycon YXF, for good measure.
Secondary caps got replaced with a pair of Rubycon ZL 330u/25v, and i decided to also squeeze in an inductor. Found one in my "collection", measured in at 10uH - oughtta do the job. Had to drill an extra hole for is input; its output fit in perfectly in the hole where the V+ wire used to be. Drilled an extra hole for that too.
Now, not wanting to upset the feedback circuitry, since that had a trace coming from right next to the output wire, i had to "jumper" that to where it used to be - before the inductor.
Photo 3: "after", bottom side
Just for good measure, i also tacked on a 4.7u/25v 1206 ceramic cap right at the (relocated) output wire. If something's worth doing, it's worth doing right, eh?
You'll notice some scour grooves - one's under the inductor (the bigger one, near the axial center of the board), and one on the "southern" edge (voltage coming from the output, into the feedback circuitry).
I didn't manage to keep Dave's EEVblog motto "to the letter" (ie. "Don't turn it on - take it apaaaart!.."), and i did plug the bugger into the mains - works fine, 7.495v with no load.
Next step - cracking it open. Your usual cheapo sealed casing (like you'd find on laptop power bricks), although fortunately, they hadn't gone crazy with the glue. Medium pressure every half-inch or so, along the seams, with a small flathead, and i could already hear the tell-tale "pop"s of glue giving in
Photo 1: "before"
Mildly impressive, i must admit - it COULD have been (much) worse.
1A fuse, no NTC or any sort of noise filtering on the AC input. 4x 1N4007's doing the rectifying, 10u/400v "Chongx" cap (seen above the board, next to its "heir"). PWM chip is an SDC606 - never heard of'em before. Definitely an uber-cheapo - its built-in switching element is a 450v BJT
Surprisingly enough, the primary-to-secondary noise-reduction cap (CY1, the "fatter" blue ceramic south of the transformer) ***IS*** in fact a mains-rated cap - go figure
Secondary's quite simple - SDC5100 (5A 100v) schottky rectifier, a pair of Ketuo (R) 470u/16v caps, and that's about it. SOT23-cased TL431 on the bottom, with the associated passives, and PC817 knockoff photocoupler.
Photo 2: "after"
Couldn't help myself - i "had to" recap this thing, for peace of mind and longevity. 22u/400v Rubycon WXA after the rectifier bridge - this one needed a bit of creativity, to shoehorn it in there
I figured i'd "do the right thing" and heatshrink the leg-extensions, and Kapton-tape the top of the cap, due to the proximity to the diodes. I had to relocate the far end of R6 (start-up resistor from the rectified mains into the PWM chip), but that was trivial. I might consider slipping in an inductor into the intended footprint, but i couldn't be bothered scrounging one up today. 47u/25v Ketuo Vcc cap (for the PWM chip) got replaced with a 47u/50v Rubycon YXF, for good measure.
Secondary caps got replaced with a pair of Rubycon ZL 330u/25v, and i decided to also squeeze in an inductor. Found one in my "collection", measured in at 10uH - oughtta do the job. Had to drill an extra hole for is input; its output fit in perfectly in the hole where the V+ wire used to be. Drilled an extra hole for that too.
Now, not wanting to upset the feedback circuitry, since that had a trace coming from right next to the output wire, i had to "jumper" that to where it used to be - before the inductor.
Photo 3: "after", bottom side
Just for good measure, i also tacked on a 4.7u/25v 1206 ceramic cap right at the (relocated) output wire. If something's worth doing, it's worth doing right, eh?
You'll notice some scour grooves - one's under the inductor (the bigger one, near the axial center of the board), and one on the "southern" edge (voltage coming from the output, into the feedback circuitry).
Attached Files
) I have a Corsair CX750M (Model 75-002019) that I picked up for free about 4 years ago. This is the PSU:
Comment