Power supply build quality pictorial. part 2

[mariushm]

Re: Power supply build quality pictorial. part 2

Oh so many components inside! I think it might work with just 1 rectifying diode, why spend money on all 4?

It is not even switch-mode.

This sets a new standard for BAD.

It's a f**ing 3.7v 0.34A power supply ... that's a <sarcasm> HUGE </sarcasm> 1.25w of power...

The phone has a battery charger IC inside anyway, so it doesn't need very regulated voltage.

They kinda need 4 diodes so that there won't be a big drop in voltage for half the time... the phones probably have about 1uF or more of capacitance at the input which is enough to smooth out the voltage coming from such charger enough to be usable (keep in mind at low loads that transformer will give up to 10-20% more voltage so that also helps)

[Th3_uN1Qu3]

Re: Power supply build quality pictorial. part 2

That's an old style Nokia charger, from the time they still made them with good ole mains transformers. Not sure why you don't like it, they were very reliable. All the charging circuitry is in the phone, so there does not need to be much in the adapter.

[c_hegge]

Re: Power supply build quality pictorial. part 2

I don't see a half bridge coupling capacitor on the primary side so I'm assuming that's forward topology. That means the secondary side's current ability rectifier wise is derated by 43%. With the PFC choke blocking airflow to the secondary heatsink and the secondary heatsink being too hot to touch for so long, it's no wonder the rectifiers shorted at 40A. Unless, of course, the schottkys in question are lying about their current ability. Those BH capacitors on the secondary side probably aren't helping anything, neither are the HEC primaries.

It doesn't normally de-rate that much in forward topology. It's more like a 25-30% reduction, so 45A would be a more realistic maximum under normal circumstances. Of course, though, that's if you let the rectifier heat up to its maximum temperature (175C in this case). With half decent cooling, pulling the rectifier's maximum rating is normally no problem.

I doubt if they were overrated. That's usually only something you would see on no name parts. The rectifiers used in this power supply were ST Micro

[Wester547]

Re: Power supply build quality pictorial. part 2

It doesn't normally de-rate that much in forward topology. It's more like a 25-30% reduction, so 45A would be a more realistic maximum under normal circumstances.

Not sure why those 30A rectifiers in parallel shorted at 40A, then, assuming they can conduct 42A-48A continuously up to 150*C. It was to my knowledge that the freewheeling diode in a double diode package, in forward topology, is capable of only 1/3rd of the current that it would be if it was fully conducting, which means the schottky would be capable of around 70% of its rating, usually.

[Pentium4]

Re: Power supply build quality pictorial. part 2

Not sure why those 30A rectifiers in parallel shorted at 40A, then, assuming they can conduct 42A-48A up to 150*C. It was to my knowledge that the freewheeling diode in a double diode package, in forward topology, is capable of only 1/3rd of the current that it would be if it were fully conducting, which means the schottky would be capable of around 70% of its rating, usually.

Maybe something arced. c_hegge, How was the soldering quality on it?

Are those HEC primary caps 1000uF?

In this case it doesn't seem worth it to have the PPFC choke because it's obviously blocking a lot of airflow, which is the last thing those BH and GL caps need. I posted about a Macron Power that had bad GL caps never been used, just sitting in a box for years.

Those heatsinks look good but at 500W with possibly low efficiency, that's a lot of heat to dissipate, added with the poor airflow. Maybe those rectifiers did just overheat

[c_hegge]

Re: Power supply build quality pictorial. part 2

They were only TO-220 parts and the plastic washer on the failed rectifier had melted. When that happens, a failure is guaranteed

[c_hegge]

Re: Power supply build quality pictorial. part 2

Not sure why those 30A rectifiers in parallel shorted at 40A, then, assuming they can conduct 42A-48A continuously up to 150*C. It was to my knowledge that the freewheeling diode in a double diode package, in forward topology, is capable of only 1/3rd of the current that it would be if it were fully conducting, which means the schottky would be capable of around 70% of its rating, usually.

It's more to do with how the current is distributed between the two diodes. Basically, two thirds of the output current is supplied by the main diode and the rest by the freewheeling diode, so if you pull 30A, 20A will be cooking from the main diode and 10 from the freewheeling diode, so a 40 A rectifier (2 x 20A diodes) would be required.

Some PSUs, like my Dell H305P-00, have a separate rectifier pack for the freewheeling. In its case, the main rectifier is rated at 20A and the other at freewheeler at 10A. It would be perfectly capable of 30A on the 12V.

[Behemot]

Re: Power supply build quality pictorial. part 2

A linear phone charger? There must be a regulator in the phone or something.

Regulator is inside battery pack.

[Behemot]

Re: Power supply build quality pictorial. part 2

…the PFC choke…

Is it only my imagination, or is the thing rusty?

[lti]

Re: Power supply build quality pictorial. part 2

A linear phone charger? There must be a regulator in the phone or something.

I haven't seen a phone that had the charging circuit in the "charger" and the battery connected directly to the power input jack.

That's an old style Nokia charger, from the time they still made them with good ole mains transformers. Not sure why you don't like it, they were very reliable.

I don't know why it is so bad either. Nokia chargers look crappy at first, but they are reliable and work fine with the phone they were designed for. The charging circuit can handle stuff like this unfiltered power supply or the later switching chargers.

[cheese007]

Re: Power supply build quality pictorial. part 2

Is it only my imagination, or is the thing rusty?

I have a sneaking suspicion it's not a real coil...

[c_hegge]

Re: Power supply build quality pictorial. part 2

^
It is real. Just a tad undersized IMO. They are usually a little larger than that one.

[cheapie]

Re: Power supply build quality pictorial. part 2

I haven't seen a phone that had the charging circuit in the "charger" and the battery connected directly to the power input jack.

I actually have seen that, but what I meant was that when the battery finishes charging, I'm sure that the phone wouldn't like suddenly getting a higher voltage, so there must be a regulator before the power even gets to the other stuff in the phone, well before the charging circuit (I don't think the charging circuit would like an unstable input much either).

[DJduck]

Re: Power supply build quality pictorial. part 2

The guy I got it from said that if he remembers correctly then this is an Olivetti Intel 286 power supply. I have NO idea how much it is rated for, but the 12v rail was able to drive a Sony car stereo so that's a start, I guess... The 5v rail is crazy however, has a total of 12200uf of filtering on it and the 5v filter coil is just massive, about 10 AWG or so. Input caps are Panasonic200v 680uf. Should I scrap it for parts or keep it?

[PeteS in CA]

Re: Power supply build quality pictorial. part 2

DJd, that P/S is nearly 29 years old. Unless you do hobby projects (or have friends who do) or can donate it to a school/university, it's ready for the recycler or a museum. Even if those electrolytic caps are in good shape despite their age, technologically they are ancient. Nichicon's oldest and least great low impedance series, the PS series (or PM series) should be better than the output caps on that P/S.

The DO-3 or DO-4 stud rectifiers should be OK, but the package style is hardly used anymore. You'd have to custom fabricate heatsinks to be able to use them.

If the input caps are still OK, they could be usable, assuming they have standard snap-in leads. But modern series will use the same size package for 1000uF-1500uF, if not higher.

I'd be worried about the X caps and Y caps, though you could probably test them for leakage if you have access to a 200-400VDC source.

So, as for parting it out, the input rectifiers might be worth keeping, as might lower current output rectifiers. The inlets and switch should be OK. If the switch device(s) is an MJE13007, MJE13009 or equivalent, it/they may be worth keeping. If the PWM is of the SG3524 or TL494 type, it's worth keeping. ICs of the LM320, LM340, uA78xx, uA79xx LM10, LM358, LM392, LM393, LM324, LM339, TL430, or TL431 type are worth keeping.

[Agent24]

Re: Power supply build quality pictorial. part 2

I'd keep it as a shining example of pure quality.

You won't see any of the current crap still running 29 years from now...

[DJduck]

Re: Power supply build quality pictorial. part 2

The switchers are some Motorola T436.

[PeteS in CA]

Re: Power supply build quality pictorial. part 2

I'm not familiar with such a part from Motorola. I''m familiar with a lot of Motorola P/N prefixes - e.g. MJE, MJ, 1N & 2N (JEDEC), MM, MTP - but "T436" doesn't fit any numbering scheme I'm familiar with.

[DJduck]

Re: Power supply build quality pictorial. part 2

The second number on the silicon is BUS47P. That is all that is on there. I think i am going to keep some of the caps and stuff from it because I don't really have any uses for it.

[PeteS in CA]

Re: Power supply build quality pictorial. part 2

BUS47P is the part number. It's an NPN transistor of Mot's Switchmode II series. "P" means it's a plastic package rather that TO-3. You should be able to find a datasheet online.