The Bestec Mystery Solved

[everell]

Re: The Bestec Mystery Solved

@i4004 - I read the jonnyguru article on the Bestec ATX-250 12E (for the first time) and really enjoyed his testing and conclusions. First - he concluded that the really big and bad problem was the 5vsb circuit. That is the same conclusion I came to. What I have really learned from all this is .............if a psu doesn't have IC 5vsb circuit, don't buy it. PSU with old two transistor design is dangerous fire hazzard and could damage other computer components.

Jonnyguru testing also proved that the 12E would handle its 250 watt rating (at room temperature), so the label attached to the psu wasn't a deliberate lie. Looking at the size of its transformers compared to other gutless wonders, it doesn't surprise me that it did handle the power it was supposed to.

The problem with ripple on the +5 volt rail he correctly attributed to a bulging filter capacitor which was failing. This was an older power supply, not a new one. So it is not surprising that an "economy" capacitor was failing, and needed replacement.

His tests did not kill the psu - and his tests proved that the protection circuits worked well when circuits overloaded. No fire! The 5vsb circuit is the problem that caused him to give it such a poor grade. The 5vsb circuit is also responsible for killing the mother boards. Looking at the next model, the Bestec ATX-250 12Z, we see that Bestec changed the 5vsb to an IC circuit. So far I have not heard anyone show proof that the 12Z killed a mother board. Instead, it has the same problems as other economy power supplies, mostly bad capacitors.

i4004, I really enjoy your posts and comments. Hope you don't mind if I add a post to say something positive about the "killer" 12E.

[everell]

Re: The Bestec Mystery Solved

The name of this section is Power Supply Design and Troubleshooting. So I want to talk about the design side of the Bestec ATX-250 12E.

In looking at some of the other gutless wonders, many are using 494 or 7500 for PWM chip. They also have three transformers. The large one is for the output of the main switching transistors. The one in the middle is the input driver between the PWM chip and the main switching transistors. The PWM and supervisor chips are all on the output side of the power supply, whereas the main switching transistors are on the input side of the power supply, isolated by the input driver transformer. Here is what I find interesting. If you plug the power supply into the wall outlet (computer turned off) and measure the VCC voltage on the PWM chip, it is only 5 to maybe 9 volts. Turn the computer on, and this voltage goes up to 12 to 17 volts (depending on PWM chip requirements). So where does this extra voltage come from when you turn the computer on?

These older designs use two bipolar transistors in bootstrap type circuit. Follow the pc board circuit traces from the emitter of the top transistor to the collector of the bottom transistor, and you will find a winding on the driver transformer between them. This means that the driver transformer is providing drive current to the base circuits of both transistors, but at the same time ......... the emitter/collector current is putting switching energy into this same "input" transformer, and thereby backfeeding energy to the pwm circuit. Rectify it and filter it, and you have some added voltage. The disadvantage I see is that the PWM must start at a lower voltage, then get the switching transistors working, then the feedback must be transferred back thru the transformer, rectified, and filtered. It all happens very quickly, but it does take some time.

The Bestec 12E uses a newer design PWM chip which operates on the primary side of the power supply, and drives a MOSFET - better efficiency than the bipolar transistor arrangement. To power this PWM chip, a winding on the 5vsb develops 20 volts, goes thru a regulator for a stable 17.3 volts. This should improve PWM turn on time and stability AND make the psu more sensitive to problems. This design concept continues to the Bestec 12Z, and I think is a good one.

Next, when the computer is not on, the 5vsb circuit is still putting out 5 volts on the 5vsb rail AND 20 volts on the other rail for the PWM circuit. A bleeder is needed to remove this additional power from the 20 volt rail. Bestec engineers accomplished this by what I call a dynamic load circuit. When computer is off, this circuit inserts a 1000 ohm resistor. When the computer is turned on, it takes the 1000 ohm resistor out of the circuit. It works quite well.

The Bestec 12Z does NOT have this dynamic load circuit. WHY????? Those 5vsb ic chips do such a good job of regulating, they seem to be able to do this job without extra help. Another proof that the two transistor circuits are inferior as well as unsafe.

Bottom line is that Bestec was using some new designs in the Bestec ATX-250 12E at the time they were first introduced, and all the bugs were not worked out. So some mother boards went BOOM. This is the price of progress when working with managers that don't allow extra testing time for engineers and insist on using cheaper parts.

So why do we still find these killer Bestecs on the market?...............I will let you answer that question for yourself.

[370forlife]

Re: The Bestec Mystery Solved

Very interesting conclusion there. So basically the bestec should not be the only one shunned as a motherboard killer, as virtually all cheap 2 transistor based power supplies will do the same thing.

Basically, when someone blurts out that a bestec atx-250-12E is the worst motherboard killer in the world, thats not true. It's just has special attention because it was used by OEM's.

[everell]

Re: The Bestec Mystery Solved

The Bestec ATX-250 12E probably IS the worst mother board killer - but not the only. I think those people who recap the 12E and think they now have a great psu are in for a surprise. The 12E does have design flaw. Play with it or trash it, but don't put it back into service. The risk is too great.

If you have a cheap psu with inadequate capacitor filtering powering a super duper mother board - what happens? The capacitors in the mother board take the load of the filter problems and have to work harder than they should. Extra work means shorter life. No wonder so many mother boards need recapping. So who gets the last laugh on the matter? The Chinaman who made that cheap psu - which probably didn't get looked at or replaced when the mother board was recapped!

My opinion .... and ONLY MY OPINION....use a psu with IC chip 5vsb circuit AND a fan that runs while computer is off and 5vsb is still on. Then look for crap caps and replace them.

[i4004]

Re: The Bestec Mystery Solved

>They are both good psu's and fine after a recap,

nope, fortron doesn't need recap.
my 350w atests it.
and its 5vstby is always on, without the fan.
teapo and ost inside.

good design will work fine even with not the best caps...
bad design won't.


>stick with seasonic, enermax, delta, antec


these are not really budget psus. at least not here.
delta(chieftec) is 20-30$ more expensive than fortron here.
fortron is 20-30$ more than cheap chinese junk.

>Well, lets see, it could kill your hard drive, motherboard, graphics card, pci cards, etc... and as for time period? That depends on the aforementioned parts.

sure it could...but did it? ever?
any proof?
talking about fortrons mainly here.

bestec and fortron is not the same thing: compare same vintage fortron and bestec designs and that should be instantly obvious.

bestec 12e is(was) probably simillar design that fortron used prior to 2000.
they probably never used exactly the same design as they never got to be mobo killers that 12e was(is)...heh...
----------

>First - he concluded that the really big and bad problem was the 5vsb circuit. That is the same conclusion I came to.

i dunno if you're aware of it, but oklahoma wolf reads this forum. he probably got the data from you.
i mean we know they like 1kw and big names, so why would he even wanna test bestec if he didn't know some things about it that could be interesting.

ie i think some portions of the article were directly inspired by your work.