Re: Power supply build quality pictorial DISCUSSION
Here was my PM on this topic to Starfury:
These two ICs are Delta proprietary parts. Unfortunately, I don't have specs or pin-outs. The one on the primary side (1001?) combines a PFC controller and a PWM (functionally similar to the UC3854 and UC3842). If the controller bd. is only half populated, the PFC section is not used. The one on the secondary side contains an error amplifier, opto driver and a bunch of comparators with a reference for fault detection.
My suggestion would be to send an e-mail to Delta's Sales and Support office, either in the UK or the US which can be found on this webpage. Ask them to e-mail you a pin-out diagram and a block diagram of the two ICs ...
The specific suggestion of the US or UK offices has to do with Starfury being from the Land of OZ.
PeteS in CA
Power Supplies should be boring: No loud noises, no bright flashes, and no bad smells.
****************************
To kill personal responsibility, initiative or success, punish it by taxing it. To encourage irresponsibility, improvidence, dependence and failure, reward it by subsidizing it.
****************************
Re: Power supply build quality pictorial DISCUSSION
Re that 160W Samsung P/S, the PCB has "Samsung" in the silkscreen, so I'm guessing Samsung actually designed it. The first four characters in the S/N suggest it was built in February, 1998. IIRC, Samsung built late 90s - early 00s iMacs, including the P/S (which, IIRC, was built on the MB).
PeteS in CA
Power Supplies should be boring: No loud noises, no bright flashes, and no bad smells.
****************************
To kill personal responsibility, initiative or success, punish it by taxing it. To encourage irresponsibility, improvidence, dependence and failure, reward it by subsidizing it.
****************************
Re: Power supply build quality pictorial DISCUSSION
Re the Touch SP9627, that is a flyback design, where the "output inductor" is the primary of the transformer. Those little ferrite rod inductors are just for high frequency spike suppression. Clues: low power, single switch device, 400V I/P lytic (probably wide-range I/P voltage or 230VAC only), not active PFC (age and no PFC inductor) and no toroidal or E-core O/P inductor. Probably a UC384X family PWM or maybe a UC290X. The +5V Stby uses a Power Integrations TopSwitch (TOP 220, I think). Looks pretty solid.
That "200W" P/S looks scary. Unapproved X- and Y-caps, really skimpy heatsinks, rather small main transformer, primary and secondary wires bundled together using the fan wires, without reinforced insulation. Probably a half-bridge inverter design with a TL494//KA2500B PWM. Probably uses clones of Motorola's MJE13007 or MJE13009 bipolar transistors, with no DC-blocking cap to keep the transformer core from walking into saturation due to slight pulse width variations. The smaller of the two transformers provides base (or gate) drive; no obvious drive transistors, so the internal drive of the PWM is used for that function, which means the base drive is kind of wimpy.
PeteS in CA
Power Supplies should be boring: No loud noises, no bright flashes, and no bad smells.
****************************
To kill personal responsibility, initiative or success, punish it by taxing it. To encourage irresponsibility, improvidence, dependence and failure, reward it by subsidizing it.
****************************
Re: Power supply build quality pictorial DISCUSSION
Originally posted by PeteS in CA
That "200W" P/S looks scary. Unapproved X- and Y-caps, really skimpy heatsinks, rather small main transformer, primary and secondary wires bundled together using the fan wires, without reinforced insulation. Probably a half-bridge inverter design with a TL494//KA7500 PWM. Probably uses clones of Motorola's MJE13007 or MJE13009 bipolar transistors, with no DC-blocking cap to keep the transformer core from walking into saturation due to slight pulse width variations. The smaller of the two transformers provides base (or gate) drive; no obvious drive transistors, so the internal drive of the PWM is used for that function, which means the base drive is kind of wimpy.
The transistors are NEC 2SC2335, and the DC blocking cap (1uF) is a bit hard to see in the photo. There are only 4 transistors; two for the output and the other two for the driver transformer.
An output transformer that size would be typical in most supplies that age.
The overcurrent detection circuit is similar to modern ATX supplies which use a single transformer for output trasistor drving and current sensing.
My first choice in quality Japanese electrolytics is Nippon Chemi-Con, which has been in business since 1931... the quality of electronics is dependent on the quality of the electrolytics.
Re: Power supply build quality pictorial DISCUSSION
That Kobian is DEER. I have 6 PSUs (and more are coming) branded as "PREMIER LC-B300ATX" which look about the same inside and they all had exactly this problem.
Re: Power supply build quality pictorial DISCUSSION
Re the "Kobian Mercury ATX PSU", what a POS! Wimpy I/P rectifiers, wimpy heatsinks, wimpy transformer - 200W would be scary. WTH is that leaded dual-diode-kludge?! What O/P is it on? I see two TO-220 rectifiers and that kludge ... and the high current O/Ps are 28A (+3.3V), 40A (+5V) and 17A (+12V). Right! Does it toast marshmallows if you try to draw that much current from any of those O/Ps?
Power Supplies should be boring: No loud noises, no bright flashes, and no bad smells.
****************************
To kill personal responsibility, initiative or success, punish it by taxing it. To encourage irresponsibility, improvidence, dependence and failure, reward it by subsidizing it.
****************************
Re: Power supply build quality pictorial DISCUSSION
Yep larry. Antec's SP-400. The outer shell kind of gave it away, but this, probably for the 1000th time, reinforces the fact that Fujjy's are far too unreliable to be recommended.
What do y'all think about the Astec unit?
Q6700 @ 3.6 GHz
Zippy GSM-6600P
Curcial Ballistix PC6400 (4 x 1Gb) Micron D9GMH
Abit IP35Pro
ATi HD4870
Re: Power supply build quality pictorial DISCUSSION
Looks like a bigger version of the 200w ones Astec built for Gateway midtowers (late socket 7 and slot 1). They were pretty bulletproof for me - full of Chemicons mainly. The Newtons they used sometimes later on weren't bad either though.
What're the maximum output ratings on that one?
You know there's something wrong when you open up a PSU and are glad to find Teapos. Why I don't buy cheap cases!
Comment