Power supply build quality pictorial. part 2

[ReeceyBurger123]

Re: Power supply build quality pictorial. part 2

I recommend Samxons RS 3300/16 D10 I have on stock, they are jsut te best for such old unit.

Your wattmeter measures BS as this has no PFC whatsoever.

BTW thats some far-east version, I cannot even find that in database.

You really dont like Watt meters do you :L

[coreAngel]

Re: Power supply build quality pictorial. part 2

I recommend Samxons RS 3300/16 D10 I have on stock, they are jsut te best for such old unit.

Your wattmeter measures BS as this has no PFC whatsoever.

BTW thats some far-east version, I cannot even find that in database.

Actually no the p3 Kill-a-watt-meter as it's called,is not measuring the psu then i realized the mistake i made.
It only measures the wattage draw out of the outlet not the PSU itself....So i forgot to mention that mistake i made.

And this is the meter

http://www.p3international.com/products/P4400.html

[coreAngel]

Re: Power supply build quality pictorial. part 2

That's about equivalent to Chemicon KY and exactly equivalent to Nichicon PW, both of which are a good choice for PSUs. I'm not familiar with KMF, but KMY is another good choice.


Oh no, we DO actually .
Just please put some punctuation in your future posts. It's hard to read anything like that.

I wanted to post the long story but was afraid to just in case someone got bored and would get mad at me.

I used to work for a very small chain of retail stores, that catered to baby stuff, cribs,clothes,etc. I used to work in the warehouse ,everyone in the store would use new computers, for baby shower lists, i would use an old computer for inventory.

Unfortunately around late 2004, there was a massive lightning strike nearby that fried all of the electronics in the store (the owner had no UPS on any of the computers, only in the main office), what hurt the most was these computers were IBM PS/2's and they were fried,(unfortunately it happened during a day i wasn't there so i could take the computers) he had them recycled.

In early 2005 the owner contacted his IT tech person (who was a supposed "expert" with a 4 year college degree-he kept repeating it to everyone) and asked him to get computers for this store,about 6 in total. He suggested it would be better and cheaper for him to build the computers and would bring them in within a week.I had a new PC, and we were all were working with new PC's which were much faster than before.

Unfortunately six months later the PC's were freezing up,not starting,and odd smell would come from them (mine still worked ill explain later here) and stop working ,mine would still work and was heavily used by everyone in the store (as they came and went from the warehouse) the owner however was not happy, the computers had their motherboards replaced twice, video cards twice,and ram 2 twice.The IT tech could not explain why they had so many problems.

The owner did not like this and fired the tech (considering how much money had been put in and other private reasons) less than a day later, and hired someone else within a week, that tech suggested Dell computers with long warranties ,it would be more cost effective.

This time no recycling, they decided to dumped the computers in the dumpster outback.They thought no one would want them so i went (on my lunch) and scavenged 2 full pc's and some other parts. I took them home, i opened it up and i noticed one of the motherboards had all the caps bulging (OST's) the other one had a motherboard replaced almost the same 2 days when they were dumped, so it had never been used or pluged in.

What i did notice was a LOT of dust, and then i checked the PSU's ......DEER,L&C (whic came with the case) on the two pc's i opened both of those up ... the main caps were Cheng X, and all secondaries were Rulycons which ALL had exploded..(rulycons) no wonder the motherboard had exploded from the abuse of that PSU...

I didn't understand why mine lasted so long and to this day in 2015 it's still in use in the office as a basic machine, a week later after taking apart the pc's i went to the office and saw inside the case, the PSU was seasonic SS-300FP.. I quit the job for other reasons about 5 yr's later, and still kept those pc's i recycled some of the parts, but kept one of them..

it originally had

PC Chips motherboard
AMD Duron 1.1ghz processor
512mb of ram
Modem
Sound Card(cheap)
Ati radeon 8500

and the dreaded powersupplies..

I eventually upgraded the PC's and it still works ( with the thermaltake PSU i mentioned)

sorry for the long story but it was some time ago thanks

[Per Hansson]

Re: Power supply build quality pictorial. part 2

it has a voltage selector switch though and newegg says it is active PFC maybe its passive

The voltage selector switch means either passive or no PFC. And I don't see a large toroid on the primary side, so this is definitely not active PFC. Passive? Maybe... but unless there is a large transformer coil that is not in view on any of the pictures, then this is not a passive PFC either. Just regular PFC-less PSU. But no worries - simple PSUs like that tend to last a very long time .

Seems like it was sold both with and without PFC, here is a review (in Swedish) and there is only one letters difference in the model number.
This one does have full range and Active PFC...
http://archive.64bits.se/recensioner..._pfc/index.php
http://archive.64bits.se/recensioner...pfc/index2.php

Anyways, something tells me your P3 Kill-a-watt is somehow getting tricked (not sure what/how, though... perhaps you incorrectly mistook the VA consumption for the power draw). The most power-hungry AMD Duron CPU is the 1.4 GHz Morgan core, with a max TDP of about 60 Watts. The Radeon 9200 SE is no more than 10W (if even that much). RAM, modem, and sound card - another 15 W. Biggest load is your HDDs. Let's say each one pulls about 20W (a gross overstatement, as most usually pull half of that)... that's 100 Watts total for them. The optical drives barely pull anything when they are not used, so I will just throw in 5 W for them. So in the absolute worst case, that PC will be drawing 190 Watts DC power max. Assuming your PSU has an ugly 60% efficiency, that's 310 Watts from the wall. But I doubt that is the case, since even most old PSUs will get about 70% (which turns that 190 Watts DC power into 271 Watts from the wall, which is quite far from the 380 W you measured).

Agreed, I have a Etech PM-300 power meter and just a few weeks ago figured out it's completely useless.
It was showing 32w consumption from a PSU with nothing connected to it when it was off, and sure the 5VSB can be inefficient on old PSU's but not THAT bad
I bought a Voltcraft Energy Check 3000 instead ofter checking some reviews, it shows a more sensible 3.7w under the same conditions
Turns out my original meter does not handle VA at all, yet the display shows "watt"
Yea right, watt if you load it with a purely resistive load like an incandescent light bulb, but I think I can read the rating on those myself so don't need a meter for that

[Stefan Payne]

Re: Power supply build quality pictorial. part 2

I really really miss those ghetto modded PSUs...
Like with those seperate aPFC PCBs and other stuff...

Today everything is so clean and neat, boring...

[coreAngel]

Re: Power supply build quality pictorial. part 2

see momaka no one cares for long boring stories that anyone posts, or im really horrible at telling stories.

Anyway i decided that i should more to this thread

This powersupply was purchased from the remaining tigerdirect.com store that was in the process of closing down (only 2 stores remain in the nation) i saw this Ultra X4 power supply that started at 249.99 and i waited until it was 39.99 till i purchased it.

It's an Andyson built unit for Ultra ,750 watt 80 plus bronze active PFC with all teapo caps, and boy is the primary cap HUGE... it's really freaking huge , all caps on the secondary is Teapo (green caps) made in late 2012 it has 4 y caps,2 x caps,2 coils,and an MOV, a huge 400v 470uf primary cap, even caps on the connector board to reduce ripple, im planning to use this on a pentium 4 system with all sata hdd's etc is the primary cap up to the task? thanks

[Behemot]

Re: Power supply build quality pictorial. part 2

You really dont like Watt meters do you :L

what

[ReeceyBurger123]

Re: Power supply build quality pictorial. part 2

what

What as in Watt or what as in Watt Watt ?

[Stefan Payne]

Re: Power supply build quality pictorial. part 2

It's an Andyson built unit for Ultra ,750 watt 80 plus bronze active PFC with all teapo caps

Hm, looks very HECish though....
Reminds me of the Cougar 600GX I have lying around here somewhere (with a broken fan controller *ARGH*)

€dit:
Yip, look here
Looks the same, does it?
Well except for the PCB Color and the heatsinks...

[momaka]

Re: Power supply build quality pictorial. part 2

It only measures the wattage draw out of the outlet not the PSU itself....So i forgot to mention that mistake i made.

Right. But even 380 Watts from the wall is too much for that system.

And this is the meter
http://www.p3international.com/products/P4400.html

I have the same meter.

After doing a bit of Googling, it seems that perhaps some of these P3 Kill-a-Watt meters aren't calibrated too well. I haven't tested mine for accuracy yet, but it seems to be alright.

Seems like it was sold both with and without PFC, here is a review (in Swedish) and there is only one letters difference in the model number.

That's how my ThermalTake TR2-430W was - the W0069 model has "PFC function" whereas the W0070 does not. I got the latter.

Agreed, I have a Etech PM-300 power meter and just a few weeks ago figured out it's completely useless.
It was showing 32w consumption from a PSU with nothing connected to it when it was off, and sure the 5VSB can be inefficient on old PSU's but not THAT bad

Well, it is weird, because I have the same Kill-a-Watt as coreAngel, but I haven't seen it get tricked by non PFC PSUs yet. Actually, only APFC units tend to trick it. In particular, I have one PSU that shows to be drawing something like 8 or 9 Watts with nothing on the 5 VSB. Compare that to less than 2 Watts for my PPFC HiPro PSUs and about 2 to 7 Watts for my other standard 2-transistor designs (all without load and in "soft-off" mode).

I really really miss those ghetto modded PSUs...

Would you like me to send you some interesting ones then?

By the way, if there is anything more boring than today's PSUs, that's today's motherboards. They are all black (or some other dark color) and you can't trace and traces on them at all. Hard to do oldschool wire and jumper modding like back in the early 2000's. Or fix anything for that matter.

see momaka no one cares for long boring stories that anyone posts, or im really horrible at telling stories.

I read it and I thought it was interesting. Kind of similar to how I acquired some of my PC gear as well. It's fun when you get an otherwise decent computer that just needs a recap and possibly better PSU.

is the primary cap up to the task?

In terms of specs - yes. But it is a Teapo, so don't expect to get more than a few years out of it, simply because this PSU has active PFC. If it was a PSU with passive PFC, I wouldn't worry about it. But with APFC, I wouldn't use anything else but good quality Japanese brand caps.

[coreAngel]

Re: Power supply build quality pictorial. part 2

Right. But even 380 Watts from the wall is too much for that system.


I have the same meter.

After doing a bit of Googling, it seems that perhaps some of these P3 Kill-a-Watt meters aren't calibrated too well. I haven't tested mine for accuracy yet, but it seems to be alright.


That's how my ThermalTake TR2-430W was - the W0069 model has "PFC function" whereas the W0070 does not. I got the latter.


Well, it is weird, because I have the same Kill-a-Watt as coreAngel, but I haven't seen it get tricked by non PFC PSUs yet. Actually, only APFC units tend to trick it. In particular, I have one PSU that shows to be drawing something like 8 or 9 Watts with nothing on the 5 VSB. Compare that to less than 2 Watts for my PPFC HiPro PSUs and about 2 to 7 Watts for my other standard 2-transistor designs (all without load and in "soft-off" mode).


Would you like me to send you some interesting ones then?

By the way, if there is anything more boring than today's PSUs, that's today's motherboards. They are all black (or some other dark color) and you can't trace and traces on them at all. Hard to do oldschool wire and jumper modding like back in the early 2000's. Or fix anything for that matter.


I read it and I thought it was interesting. Kind of similar to how I acquired some of my PC gear as well. It's fun when you get an otherwise decent computer that just needs a recap and possibly better PSU.

I know i was kidding about it, i know people read it it's just a story
Unfortunately lately it's been getting harder and harder to get older PC parts even at garage sales ,craigslist. People want high prices for old used components or even any component.

In terms of specs - yes. But it is a Teapo, so don't expect to get more than a few years out of it, simply because this PSU has active PFC. If it was a PSU with passive PFC, I wouldn't worry about it. But with APFC, I wouldn't use anything else but good quality Japanese brand caps.

I thought i can get a couple of years out of it, i was thinking of recapping that teapo with this

http://www.mouser.com/ProductDetail/...0alEv55wwWA%3d




as the teapo is a general purpose cap and is rated for 2000 hours at 105c if i can trust it.

[kaboom]

Re: Power supply build quality pictorial. part 2

as the teapo is a general purpose cap and is rated for 2000 hours at 105c if i can trust it.

A very big if...

More like "not at all."

[Behemot]

Re: Power supply build quality pictorial. part 2

Agreed, I have a Etech PM-300 power meter and just a few weeks ago figured out it's completely useless.
It was showing 32w consumption from a PSU with nothing connected to it when it was off, and sure the 5VSB can be inefficient on old PSU's but not THAT bad
I bought a Voltcraft Energy Check 3000 instead ofter checking some reviews, it shows a more sensible 3.7w under the same conditions
Turns out my original meter does not handle VA at all, yet the display shows "watt"
Yea right, watt if you load it with a purely resistive load like an incandescent light bulb, but I think I can read the rating on those myself so don't need a meter for that

You read what I said about SMPS without PFC?

[kaboom]

Re: Power supply build quality pictorial. part 2

You read what I said about SMPS without PFC?

500 VARs is still 500 VARs, no matter how many watts are involved; an actual wattmeter (or W/Hr meter like the one your POCO uses) does not respond to VARs.

With just a cap across the AC line, you can have 10A at 120VAC, yet there will be minimal watts indicated; the only power registered in this case would be the I2R losses in the wires, and (minimal) losses in the cap. Effectively, this is 1.2KVAR with perhaps less than ten watts.

If the device used to measure the above shows 1200 "watts," it's inaccurate- showing VA as watts. No good.

The PFC booster actually increases the total consumption of the power supply in question; it's another stage of active semis, and inductor(s).

[Behemot]

Re: Power supply build quality pictorial. part 2

What's the relevance?

[kaboom]

Re: Power supply build quality pictorial. part 2

What's the relevance?

This:

Your wattmeter measures BS as this has no PFC whatsoever.

It's not if the PS has PFC or not; it's if that's an actual wattmeter, not a VA meter.

[Behemot]

Re: Power supply build quality pictorial. part 2

You are trying me right?

First, I have never ever said anything about some VA meter, if someone is using such crap he is fool in the first palce.

Second, I am telling you it bloody matters if the unit has PFC or not. Most wall meters simply do not have fast enough sampling rate (which should in terms of MHz at least) to catch all the current harmonics of unit without PFC. As a result, they measure bullshit. Ranging from about -50 % of actual power to +50 % of actual power. Often it actually fluctuates between these values in real time.

To measure that you need power analyzer which is built for such task in the first place.

[c_hegge]

Re: Power supply build quality pictorial. part 2

Galaxy Power SPX 250W

https://www.badcaps.net/forum/attach...1&d=1443662650
Internals. Looks like an old AT design with an add-on 5vsb PCB

https://www.badcaps.net/forum/attach...1&d=1443662650
Primary side. 330uF Jee Primary caps and non-approved input filtering parts.

https://www.badcaps.net/forum/attach...1&d=1443662650
Secondary side

https://www.badcaps.net/forum/attach...1&d=1443662650
Caps are all Rulycon, but surprisingly, none are bulged. I can't remember what the silicon was, though.

[momaka]

Re: Power supply build quality pictorial. part 2

Rulycons that haven't bulged yet?! In 2015????? These must have been the golden engineering samples, or something.
I wonder who made this one. I guess this is from the very early 2000's, as JEE and Rulycon were still quite popular back then.

[Wester547]

Re: Power supply build quality pictorial. part 2

Rulycons that haven't bulged yet?! In 2015????? These must have been the golden engineering samples, or something.
I wonder who made this one. I guess this is from the very early 2000's, as JEE and Rulycon were still quite popular back then.

They must be fake Rulycons. Or maybe resleeved Rubycons.

Okay, maybe not. I spy a "1M9828" datecode on one of them - yet another way in which Rulycon imitates Rubycon. So they are probably from July of 1998. As far as "bad electrolytics" go, I find that their "failure modes" make manifest in different ways. I think it's a race between poor quality seals (low quality cans and vents, synthetic rubber bungs... they just dry out and won't likely bulge or leak in any manner that's obvious) and poor quality electrolyte and aluminum foil (the most likely scenarios to cause them to vent). Those Rulycons may look good but their actual condition may tell a different story.