Bestec revisited

[momaka]

Re: Bestec revisited

I remember reading somewhere that Wan Nien is a PCB manufacturer.

I believe you're right.
I've also seen Wan Nien in almost every L&C PSU.

everell - I think the Bestec and PowMax use 2 different 3.3v designs, thus explaning the lack of the second toroid coil on 3.3v for the PowMax. From what I can tell by the pictures, the Bestec seems to use a saturation coil before the 3.3v rectifier and another coil after the rectifier for smoothing the output on 3.3v. As for the PowMax, 1 of the components on the secondary heatsink could be a 3.3v regulator, most likely using 5v as a source. Check and see.
...
And the different sized diodes - that's a Leadman/PowMax/Raidmax trademark . Normaly, they only use 2 sizes, with the smaller connected to neutral. On yours, however, seems they just used whatever they could find.

Made by HongKongFlyApart Company, LLC.

[everell]

Re: Bestec revisited

Toasty - glad you decided to give that Bestec 12Z a second chance and recap it. Keep me posted on your results. Otherwise I was going to say - just send it my way.

After re - engineering that first Bestec, they don't seem to be all that bad for smaller machines. One PSU I worked on a few weeks ago had three roach bodies between the PC board bottom and the metal case. Yes - all three were black and well done. In the process it took out the main switching transistor and fuse. So there are sometimes strange reasons for a PSU blowout! Ever had that happen to you?

[severach]

Re: Bestec revisited

The eMachines issued ATX-250-12E Rev. P7 is what gets all the bad press for destroying motherboards. Unless someone takes the time to separate the good from the bad then we just throw them all out or give them away to people who deserve early death computers.

Here is a Bestec ATX-300-12E Rev D from an eMachines. It was replaced because the AC power jack was damaged. The chips are TPS3510P, LM358N, UC3842B, and A6351A. Caps are all Jamicon and Ltec. The thermistor will under report the temperature since it is not actually touching the heat sink.

[i4004]

Re: Bestec revisited

that looks ok.
not crap instead of proper 5vstby circuit...

btw. can you photograph
ATX-250-12E Rev. P7
?

[Toasty]

Re: Bestec revisited

severach >>The thermistor will under report the temperature since it is not actually touching the heat sink.<<
Yes. I've seen that "flying goop thermistor" on several brands, not just the Bestec's. My most recent exposure with the 12Z is like Everell's pic. A nice metal cap over the thermistor screwed to the heatsink. Wow! Design improvement!

I also liked their soldering job on the underside of the board. Nicer than what I'm used to. Very clean and sharp.

@Everell -
No roaches in any computers, so far...
It's one of the reason every unit is blown out outside before it comes into shop.
TV's from years ago, yes. Tube sets were great for roasting them!

Toast

[hardwareguy]

Re: Bestec revisited

Wan Nein is indeed a PCB house...I've seen their name on a lot of stuff besides PSUs. I seem to recall seeing them on an amplifier board for some computer speakers.

I'm glad to see Bestec has revised their killer 5VSB circuit. The new one also seems to be using all Teapos....not great but a definite improvement.

I can confirm without any doubt at all that that Bestec is better than that Powmax.

While the early Bestec might be a mobo hazard, the other would set your house on fire if given the chance. I've seen loose wires, improper safety caps and even bypassed fuses in Powmax before.

One model has a jumper where the fuse holder was supposed to be. Its rare in the US for obvious reasons.

But with that said, if either improved, I still wouldn't buy.... I don't trust them. The revised Bestec might make a good bench test supply, but I still wouldn't trust that for day to day use...its a Bestec! As for Powmax.... those make great target practice.

[bgavin]

Re: Bestec revisited

Is there a model number or general year of production that identifies the re-engineered BESTEC?

I had a client machine in last week, Gateway GT5056.
Board has seven bulging 1000/6.3 no-names, and a BESTEC supply.
I've told the client the board needs recapping and BESTEC replaced.
The machine appears to be 2006 vintage.

[Toasty]

Re: Bestec revisited

I don't think it has much to do with years of production. I do think the problem(s) arise from that specific model ATX-250-12E Rev. P7. It being the one with the dangerous 5VSB circuit.

However, the cap problem is across multiple brands and models. I see few from that 1999-2006 (even 2007 - Thanks Antec!) era that are exempt.

[severach]

Re: Bestec revisited

Here's all the Bestec I have so far. I've had a Bestec board destroyer with the ticking near the board burn spot which was probably an ATX-250-12E Rev P7 but didn't have a reason to photograph it.

All of the 12Z have a greased metal cap on the thermistor. The 12E is a gluon.

I'm inclined to not trust Bestec unless research can narrow the problem down to a few easy to remember revisions and the good reports continue on the remainder. Good date codes would help determine whether it was an old stock sucker punch to eMachines or Bestec was intentionally manufacturing the bad ones at the same time as the good ones and if so, why do the bad supplies mainly appear in eMachines?

On Google I see bad reports on:
ATX-250-12E Rev P6
ATX-250-12E Rev P7
ATX-250-12E Rev F1
ATX-250-12E Rev F2

Fried motherboard reports for HP Compaq issued Bestec are hard to find.

[Toasty]

Re: Bestec revisited

Were all of those pictured dead or had caused/were causing problems? Or, just pulled because of whose they are?

[i4004]

Re: Bestec revisited

all of these pictured have "normal" 5vstby circuits...
ie not the extra-cheap&ugly sort with a transistor and no feedback...

[kc8adu]

Re: Bestec revisited

emachines specs the cheapest parts they can get for the price point needed to sell at walmart.

Here's all the Bestec I have so far. I've had a Bestec board destroyer with the ticking near the board burn spot which was probably an ATX-250-12E Rev P7 but didn't have a reason to photograph it.

All of the 12Z have a greased metal cap on the thermistor. The 12E is a gluon.

I'm inclined to not trust Bestec unless research can narrow the problem down to a few easy to remember revisions and the good reports continue on the remainder. Good date codes would help determine whether it was an old stock sucker punch to eMachines or Bestec was intentionally manufacturing the bad ones at the same time as the good ones and if so, why do the bad supplies mainly appear in eMachines?

On Google I see bad reports on:
ATX-250-12E Rev P6
ATX-250-12E Rev P7
ATX-250-12E Rev F1
ATX-250-12E Rev F2

Fried motherboard reports for HP Compaq issued Bestec are hard to find.

[370forlife]

Re: Bestec revisited

emachines specs the cheapest parts they can get for the price point needed to sell at walmart.

Its true, all the computers at wal mart are emachines or gateways.

Though, older p4 era intel emachines used intel motherboards, but the AMD variety always used ECS.

So~~~they used good motherboards and bad power supplies, now they are using =better= psu's and worse motherboards?

Anyways, heres a worstec from a emachine T3624 (I believe I mentioned it earlier,) it seems to be a bit better built than other worstecs.

Those are Panicsonic primaries, Jamicon and TK secondaries, and a bridge rectifier instead of the usual 4 diodes. This one is a Worstec ATX-250-12E. The soldering is actually pretty decent, but one thing concerns me, and that is the bridge rectifier is not soldered in flat, it is on a angle, so the very last lead is barely in there.

[kc8adu]

Re: Bestec revisited

here is the "killer cap" circled. i use a 22@50 fc.

Its true, all the computers at wal mart are emachines or gateways.

Though, older p4 era intel emachines used intel motherboards, but the AMD variety always used ECS.

So~~~they used good motherboards and bad power supplies, now they are using =better= psu's and worse motherboards?

Anyways, heres a worstec from a emachine T3624 (I believe I mentioned it earlier,) it seems to be a bit better built than other worstecs.

Those are Panicsonic primaries, Jamicon and TK secondaries, and a bridge rectifier instead of the usual 4 diodes. This one is a Worstec ATX-250-12E. The soldering is actually pretty decent, but one thing concerns me, and that is the bridge rectifier is not soldered in flat, it is on a angle, so the very last lead is barely in there.

[severach]

Re: Bestec revisited

Were all of those pictured dead or had caused/were causing problems? Or, just pulled because of whose they are?

Mine were all field failures, a cap or two. I didn't get any burned boards with them so they probably survived.

[Toasty]

Re: Bestec revisited

@severach - Thanks for the info.

None of them show a bloated cap, so these are all non-obvious fails.

So the conclusion is any 12E series, regardless of revision number is a killer. Yes?

[Wizard]

Re: Bestec revisited

Tread in with your tools and confirm what is what are bad first. No assume.

After that, you can reasonably expect which series is known to have this issues but still you go in and check or replace.

Cheers, Wizard

[Toasty]

Re: Bestec revisited

My conclusion is based on the discussion here and elsewhere on the forums and the web. Seeing that the 12E series has the poorly designed 5vsb circuit, I'll replace them on sight. Failed or not. If it has the chance of dealing death to a customer's computer, it's money well spent to replace it before it fails.

Yes, it "might" run forever and not fail. But, if I knowingly let it go out of here with a 12E PSU in there, and it does fail, I will feel as though I have failed my customer.

The 12Z series, as shown and demonstrated here again, seems to not have the poor 5vsb circuit. Its reported failures seem to be primarily failed caps. For me, it will probably still be a replace on sight as I did back in the beginning of this thread. At least with this 12E vs. 12Z knowledge I can be more comfortable that it didn't hurt the mobo. I also can offer the choice of recapping it to a customer. The only problem there is cost of doing so (labor & parts) versus putting an inexpensive replacement in the computer, which again, I did.

Toast

[bgavin]

Re: Bestec revisited

As a paid servicer, I don't have the luxury of wasting a lot of time on a losing proposition. It is much more cost effective to discard the BESTEC immediately and replace it with something else.

Most of my board recapping is done on discarded boards for the same reason.
It takes me an hour to remove, recap, replace a typical board and vacuum out the system. I then have to test it for integrity. This all takes time that I have to bill to the customer, or give away for free. Times are much too hard to give things away.

I recap the discarded boards and add them to my stock if they test out.
When the client needs a replacement, I sell him the board and take his in exchange as a potential core.
This has value when the client's down level investment in CPU and RAM would otherwise be lost.

Non-proprietary power supplies do not fit this category, so I toss 'em instead of recap them.

[everell]

Re: Bestec revisited

Yes, looks like the 12E is the mobo killer. But I don't trust labels on anything coming from China. Best to look inside. If there are two 8 pin integrated circuit chips between the two heat sinks, it is the good power supply with the improved 5VSB circuit. If it only has one 8 pin integrated circuit chip between the two heat sinks, AND has two transistors on the heat sink nearest the two big input capacitors, it is a mobo killer.

I have just started work on another PSU with 5VSB fluctuating around 12 volts, having a similiar two transistor design. It is a SEASONIC Model SS-250FS. I thought everyone said SEASONIC was the good stuff! To me, this didn't look much different from the Bestec 12E. Both use the UC3842/UC3843 series PWM chip and the TPS3510 supervisor chip. Take a look at the pictures. Notice that the PC board around the 5VSB is darkened due to burning and heat.