Sign in with
Sign up | Sign in
Your question

Abit uses bad/cheap capacitors on their M/B?

Last response: in Motherboards
Share
July 15, 2002 4:48:21 AM

Seems like Abit, Soyo and Epox uses cheap capacitors that fail after a couple of years. Here's the list from that M/B Repair site: <i><b>BE6, BE6 II, BE6 1.2, BE6 2.0, BH6, BX6, BX 133, * KA7 *, SA6, SA6R, SE6, SH6 ,VL6</b></i>.

I hope they use better caps now so if you have an Abit M/B (current models), can you check the brand of the caps and compare it with what the guy's listed?

<A HREF="http://www.overclockers.com/tips047/" target="_new">A Case of Crappy Capacitors?</A>
<A HREF="http://www.overclockers.com/tips054/" target="_new">A Bit InCapacitated</A>
<A HREF="http://home.att.net/~garyheadlee/" target="_new">M/B Repair</A>


<i>"The high failure rate for the boards manufactured over the past 2 years seems to be faulty components made by:

<b><font color=blue>1) Teapo
2) Tayeh.
3) Jakcon</font color=blue></b>
4) Teapo,Tayeh and Jakcon manufacture the filter capacitors that are just behind the CPU slot/ socket.

5) <b><font color=blue>Weldon</font color=blue></b> brand capacitors used on Soyo boards look like the same crap."</i>


:eek:  :frown: :mad:  :eek:  :redface: :cool: :lol:  :tongue: :wink: - What do you want to feel today? :) 
a b V Motherboard
July 15, 2002 5:36:29 AM

I don't know if it's do to poor capacitors or not, but I cannot overclock my Abit BE6-II at ALL. In fact, I can't even run it at the specified speeds for all the components!

Here's how it went down: Had an Epox BX3, worked perfect but wouldn't allow speeds beyond 133FSB in BIOS, replaced with Soyo SY-BA6+III.

Soyo was a little less stable but allowed me to push my bus speed to 150 with fair stability. Overclocking didn't seem to reduce stability. I liked the board, but two capacitors blew. The board was only 2 months old and Soyo keeps "loosing" my RMA information. I've already spent $15 on long distance calls to them, but to no avail.

Switched to Abit BE6-II. Now I have a lot of gripes about this board, especially when it comes to configuring cards, but the worst part is, it wouldn't even run my memory at the specified speed as the others had. This is Crucial Cas2 PC133 in all systems, in fact, the same two sticks. On the BE6-II, it overheats. In fact, just to run at 133FSB, I had to increase Cas Latency to CAS3, reduce all performance settings to lowest, and use only 1 stick of RAM to reduce heat between sticks. I had never seen overheating memory at stock speed before, and never had problems with this memory. Using the necessary lower settings has reduced the performance of my system by around 12% in 3DMark-2001, and yeilded similar results in games and video editing, as compared to the previous two boards. I am using the latest version of BIOS, and my voltages are all dead on spec. The system won't even BOOT with the memory at Cas2, it hangs at POST.

The processor is a PIII 1000EB. I can say with certainty that the Epox board had good capacitors, the Soyo had bad capacitors (they blew). But I am only guessing that the memory overheating on the BE6-II is caused by weak capacitors

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>
July 15, 2002 8:18:53 AM

capacitors would not cause bad memory temps.

The capacitors are used for:
Coupling and Decoupling
Smoothing erratic PSU's

As Coupling and decoupling causes a "lag" they are not used fast data lines, this is one reason why you may find people who do electronics complaining about bad waveforms due to capacitance. Basically feed a square wave into a capacitor and it will come out rounded off.

Capacitors of the wrong size (ie wrong value) would cause overclocking limits as the nice digital waveform would be messy and horrible. This is the manufacturers fault badly designed.

Yes capacitors can explode if the clock through them is too fast or the voltage across them is too high. (probably the core voltage too high for the board) once again the Manufacturers fault, badly designed.

I think it is unlikely about your memory temperature is cause by bad cap's. The memory may have been damaged when the cap's blew on your previous board, when a capacitor blows it can cause small voltage surges of all kinds. Because the voltage is allowed to fluctuate.
Related resources
Anonymous
a b V Motherboard
July 15, 2002 1:59:54 PM

Oh & what the matter?!? the motherboards should be replace after a couple of years to follow the IT evolution.


i've plugged my home blower to my case ... dunno what happen ... that works?!?
a b V Motherboard
July 15, 2002 6:04:59 PM

My memory could get hot if it's seeing voltage spikes, which could be caused by weak capacitors. The spike wouldn't even should up in the readings do to their brevity. One thing I didn't mention, the memory only runs at 3.5v on this board. Yes, I have retested the memory on other systems, without problems. But if the 3.5v setting results in peaks of 4v and valeys of 3v, we would have exactly what I'm seeing, a problem of instability and heat. And the monitoring software would report the middle number, 3.5v

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>
July 19, 2002 12:29:29 AM

Another interesting <A HREF="http://www.overclockers.com.au/article.php?id=76562" target="_new">read</A> about capacitors and the term "3-phase power solution" commonly seen in M/B reviews.



<i>Editor's note: This was originally sent to me in an email in response to my comments about "3-phase power solution" in motherboard reviews. Lots of motherboard manufacturers use this technology as a selling point, but I wasn't sure exactly what it meant. I thought Sean's comments were worthy of making into an article as there's a lot of confusion about this topic. Thanks, Sean! -- Agg

What "three phase" power is referring to in this case, is a switching voltage regulator, which switches power in three phases.

Today's motherboards require all kinds of different "DC" Voltages. To get these different voltages, switching voltage regulators are used because they can be made very efficient.

Switching supplies work by taking chunks of power from one voltage source and providing it to hardware at a different voltage. The differences in the input and output voltage aren't of much interest here. The interesting issue is that the power is transferred from source to destination in chunks. By analogy:

Imagine your MB runs on air and has an air intake hose. You put air into the intake and the MB keeps draining it out. To keep it running, you have to keep the pressure in the hose, higher than atmospheric pressure. A single phase switching power supply would handle it as follows. You take a big guy (say Pavarotti) with a big balloon. He blows up the ballon while keeping the hose end closed. Then he attaches the balloon to the intake of the hose and lets the air drain in. Then he blows up the balloon with the hose end closed again... over and over and over.

Now it might occur to you that this is an awfully lumpy way to provide power. You'd be right. Another part of the picture is another balloon attached to your motherboard. The purpose of this balloon is to hold a reservoir of air (power) and smooth out fluctuations in incoming power. (As well as fluctuation due to the MB drawing different amounts at different times.) So when Pavarotti is busy blowing up his balloon, the balloon on the MB is getting smaller. When Pavarotti attaches his balloon to the hose intake, his balloon gets smaller, but the one on the MB gets bigger. (The balloons on the MB are known as Capacitors) The bigger the balloon on the MB is the smoother the power is. (Same goes for the amount of Capacitance on the MB.) As you might notice, there is "all kinds of AC weirdness going on" regardless of whether a MB has a "3-phase power solution".

So what about three phase power? Pavarotti is big and expensive and a rather lumpy source of air pressure. Plus with modern technology we can use a more complicated scheme if there is a benefit to it. So instead of Pavarotti we get three choir boys with three small balloons. Two of the choir boys are blowing up their balloons while the third one has his connected to the hose intake. Then they switch off in a three phase sequence so that there is always a balloon connected to the hose intake.

Phase 1: Albert inflates, Billy inflates, Charlie's balloon deflates into the intake.
Phase 2: Charlie inflates, Albert inflates, Billy's balloon deflates into the intake.
Phase 3: Billy inflates, Charlie inflates, Albert's balloon deflates into the intake.

This has several advantages:

Number one; since a balloon is always connected to the input hose, the input power is smoother. (ALL ELSE BEING EQUAL)

Number two; three choir boys and three small balloons may fit more easily into the space available.

Number three; three choir boys and three small balloons may costs less than Pavarotti and a big balloon.

Number four; with three choir boys and three small balloons you can keep the power adequately smooth with a smaller balloon on the MB. (Reduce the cost in capacitance.)

Is three phase power better? Well it might be, but it might also be cheaper while being equally good or even worse.

If you believe your home stereo sounds better when the speakers are hooked up with 4 Gauge welding cable instead of ordinary 18 Gauge zip cord, then you certainly ought to get yourself a MB with a "3-phase power solution".

If you'd prefer getting a few more watts per channel, to bragging about the size of your cable, then whether or not a board has a "3-phase power solution" is probably not going to be an important issue for you. It isn't total BS though. Power system design for motherboards is sophisticated stuff, and is an area that can make a serious difference in the profitability for a manufacturer. There is ALWAYS a trade off between power quality and cost. When a motherboard manufacturer says something along the lines of, 'Our new MB with a 3-Phase power system has the capability of powering the latest generation of processors with far better [Insert Miscellaneous Category Here] than other manufacturer's 2-Phase systems.', they are just blowing smoke. I can design a "single phase power system" that beats theirs in whatever category they choose. You may not want to pay what it costs though. It's all tradeoffs.

<b>It sounds like Abit recently went a hair too far on the cheap side, and has had capacitors blowing up on their boards. I'm not going to knock Abit though, they may well have gotten screwed by a capacitor supplier. I'm writing this on an Abit MB that has been running 24/7 for five years with the processor overclocked 36%. (BH6 board Celeron-300A clocked at 450. Not much MHz by todays standards, but you don't often see 50% overclocking these days either.) I may well buy another Abit motherboard in the next month. However, Abit's Capacitor issues illustrate the fine line that all MB manufacturers have to walk to make a profit.</b>

Sean "Yes, I am an engineer" Hershberger

P.S. And no, I do not work for a motherboard manufacturer.</i>



:eek:  :frown: :mad:  :eek:  :redface: :cool: :lol:  :tongue: :wink: - What do you want to feel today? :) 
!