which one to use...

[Uranium-235]

7 year old 250w bestec working nonstop, with no bloating caps, nice and heavy, heavy tranny and decent caps (from HP business computer)

brand new raidmax, 450w from a raidmax case. Lighter than a mouse. This is probably the lightest POS i've ever seen. Doesn't even have a void sticker, probably cause there is no warranty. Might post pics later for sheer humor

took a lady from an 85w P4 3.06 775 to 65w C2D E6600 /w DDR3. Didn't seem to mind the PSU that much, though i'm sure the board was drawing more, even if the CPU power was drawing less (though i'm not entirely sure, it did have a 4 phase vrm)

was stable. And a small thumbs down to MSI for actually not setting the multiplier to what the E6600 is by default. Was running circles around her old one, and I realized it was running it at 1.86ghz, not even the full 2.4. Had to manually set the multiplier

[everell]

Re: which one to use...

Since I like Bestec, I am biased. What model Bestec is it?

[turbozutek]

Re: which one to use...

was stable. And a small thumbs down to MSI for actually not setting the multiplier to what the E6600 is by default. Was running circles around her old one, and I realized it was running it at 1.86ghz, not even the full 2.4. Had to manually set the multiplier

Hey, my GigaByte G33 does exactly the same thing with my E6600... I dunno why, but it always sets the multiplier to 6 when it should be 10. Something about newer chips not playing nice with older boards, I'm not sure?

That system also writes errors into the event log sometimes about APM having a firmware issue, so I guess there's a nice wee CPU microcode clash there somewhere.

Chris...

[Uranium-235]

Re: which one to use...

Since I like Bestec, I am biased. What model Bestec is it?

the one that came in an HP DX2200. lemme look. hmm found one, its probably not the exact same model but close

http://erecycleronline.com/media/cat...6_800x602_.jpg

[c_hegge]

Re: which one to use...

I'd use the bestec in either of those PCs. It's probably capable of more than the raidmax anyway.

[jamesbo]

Re: which one to use...

My DX2200 had a Bestec ATX-250-12Z, too. It works fine, but my handheld PSU tester says the Power Good timing fails, it is too fast.

[LDSisHere]

Re: which one to use...

My DX2200 had a Bestec ATX-250-12Z, too. It works fine, but my handheld PSU tester says the Power Good timing fails, it is too fast.

I also have a Bestec ATX-250-12Z with a "PowerGood" problem that is driving me crazy. When assembled it has all the voltages it is supposed too, but the 3.3 is a little low (~3.2xx) and fluctuates some and it will not give a "power good" signal. If you disconnect the fan from the supply it will give a power good and the 3.3 rail goes slightly above 3.3V. In case you are thinking the fan transistor is bad, it is not, as I replaced it just to be sure.

If I put a 20 ohm load on the 12V rail or the 5V with no fan there is no change, voltages are good and I get a "power good" signal. If I put a 20 ohm load on the 3.3V rail, the voltage drops to 2.7V but I still get a "power good" signal. If I use a different fan I still loose the "power good" signal.

The two Op-Amps (384 chips), supervisor chip (3510), transistors and small diodes all seem good. (No obvious shorts or opens) The voltage on the LED side of the optoisolators does not change fan or no fan. The voltage on pin 1 of the 3510 chip does drop from 2.5V to 1.8V when a fan is connected but this should still be good enough to turn on the "power good" signal. This voltage drop starts from the diode D10 near the big transformer, the voltage from D10 goes from ~39V to ~30V with a fan. I have not pulled the heat sink to test the diodes attached to it since the 12V rail voltage itself does not drop. So far I have only changed some of the small 50V caps that I thought could be part of the problem. There are no visibly bad caps so I have not yet changed any of the big caps.

Another thing I found odd was the 384 Op Amp that is directly above the 3510 chip has ~22V on the output pins 1 and 7 all the time. This may be normal but I thought it was strange. I pulled the other 384 chip out of the board and checked it for shorts with my DMM but I did not pull this one since it did not show any shorts in circuit. (The other 384 did not show any shorts in circuit either I just pull a lot of different part out to test them out of circuit.)

Any help on where I need to focus my troubleshooting would be appreciated.

[Behemot]

Re: which one to use...

Hey, my GigaByte G33 does exactly the same thing with my E6600... I dunno why, but it always sets the multiplier to 6 when it should be 10. Something about newer chips not playing nice with older boards, I'm not sure?

That system also writes errors into the event log sometimes about APM having a firmware issue, so I guess there's a nice wee CPU microcode clash there somewhere.

Chris...

SpeedStep?

[turbozutek]

Re: which one to use...

SpeedStep?

At BIOS level and in POST? I'd have thought SpeedStep would be off at that point and the cpu running at max clock. The board supports SpeedStep and CPU-Z reports it's enabled and working OK from within the OS.

/Shrug

Chris...

[Behemot]

Re: which one to use...

Why on earth it should not work in BIOS? It should POST right name/frequency, but that's all, everywhere else it can work.

Also having the last launched BIOS version is usually the right way to go…

[everell]

Re: which one to use...

ldsishere: back to that Bestec 12Z, I am thinking you have a 3.3 volt problem, and maybe a power good problem also. I have only made a complete study on one Bestec output circuit, but not the 12Z model. It is probably close. So for starters, make a drawing of the 3.3 volt output circuit on your 12Z and post it. That way we can all see what is there. Here is a drawing of another Bestec which you can use as a general guide. I think you will find that the 3.3 volt rail has an independent voltage regulator using a 431 chip.

As for problems, be sure that the 3.3 volt line has a load resistor when testing it. This circuit may have a flakey 431 chip, or a leaky pass transistor. And of course, crappy capacitors!

[LDSisHere]

Re: which one to use...

ldsishere: back to that Bestec 12Z, I am thinking you have a 3.3 volt problem, and maybe a power good problem also. I have only made a complete study on one Bestec output circuit, but not the 12Z model. It is probably close. So for starters, make a drawing of the 3.3 volt output circuit on your 12Z and post it. That way we can all see what is there. Here is a drawing of another Bestec which you can use as a general guide. I think you will find that the 3.3 volt rail has an independent voltage regulator using a 431 chip.

As for problems, be sure that the 3.3 volt line has a load resistor when testing it. This circuit may have a flaky 431 chip, or a leaky pass transistor. And of course, crappy capacitors!

I was able to fix this supply. I replaced Op-Amp IC7 first and that got the voltage up on pin 1 on the 3510 chip and I would get a "Power Good" signal for a split second before it would go back out. I then found another problem by accident. I was checking voltages at various points on the board while turning the supply off and on with the fan attached. For some reason when I had my probe on pin 1 of Q13 the supply would start and give a "Power Good" signal, without the probe no signal. I pulled the transistor and it would only pass a diode test in one direction, so after I replaced it the supply started giving me a "Power Good" signal every time.

I then checked the 3.3V rail by putting a 10 ohm load in addition to my ATX tester on it and I got ~3.33V. It was fluctuating some, just a few hundredths of a volt, but a recap would probably fix that issue. Overall it was a very good learning experience, I just wish I understood the function of the two Op-Amps in the circuit much better. Thanks for the schematic, it will help increase my understanding of what is happening on the secondary side much better.

[everell]

Re: which one to use...

My DX2200 had a Bestec ATX-250-12Z, too. It works fine, but my handheld PSU tester says the Power Good timing fails, it is too fast.

I dug thru my notes and found this drawing of the "power good" input circuit for this power supply. Adjusting the time constant of the resistors or capacitors can improve that timing.

[everell]

Re: which one to use...

I also have a Bestec ATX-250-12Z with a "PowerGood" problem that is driving me crazy. When assembled it has all the voltages it is supposed too, but the 3.3 is a little low (~3.2xx) and fluctuates some and it will not give a "power good" signal.

I found this drawing of the 3.3 volt output for a Bestec ATX-250 12E or possibly 12Z. I made it more than two years ago, but it does give the general idea of what is going on. Looks very close to the other drawing I posted earlier.

[Shocker]

Re: which one to use...

7 year old 250w bestec working nonstop, with no bloating caps, nice and heavy, heavy tranny and decent caps (from HP business computer)

brand new raidmax, 450w from a raidmax case. Lighter than a mouse. This is probably the lightest POS i've ever seen. Doesn't even have a void sticker, probably cause there is no warranty. Might post pics later for sheer humor

Are you seriously asking this???

[turbozutek]

Re: which one to use...

Why on earth it should not work in BIOS?

Because I switched it 'off' and in so disabling the feature it shouldn't occur, ever.

Also having the last launched BIOS version is usually the right way to go…

Yes, that's why I have the most up to date BIOS installed... My thinking is that they just stopped working on these BIOS updates as that CPU came out so there's some incompatibility they will never ever address. Ah well!

Chris...

[Behemot]

Re: which one to use...

Sometimes they just leave it be, but usually they at least add new BIOS even after two years, if there are new processors for the board…