![]() |
|
|||||||
| Cases, Power Supply Units (PSUs), Cooling and Overclocking All things related to aesthetics, powering, cooling and overclocking your system. |
![]() |
|
|
LinkBack (1) | Thread Tools | Display Modes |
|
||||
|
Re: Great...a PSU issue...
...
Rather than examine the burned molex connector, note that it's the red (+5V) wire, make the quick calculation in your head that it would take a spike something to the magnitude of 10 times that to burn plastic, and entertain the idea that a malfunctioning PSU could do just that, you just replaced the connector and went ahead and plugged it right on back in. ??? What were you thinking?!?! Normally, this level of amateurism wouldn't surprise or dismay, hey it's a newbie forum...except for the fact you added that you build and sell systems. Incidentally a chain of power connectors is not a rail; every single red wire is on the same +5 rail. Since a large majority of components in the system make use of +5, you're playing Russian roulette. Perhaps, if you're lucky, just that cable is now burned beyond functioning. If you're unlucky, your PSU no longer has a +5 rail and if you're supremely unlucky, your graphics card no longer has functioning voltage regulation circuitry. And why do so many people have problems with the 9800 Pro/XT? Cheap parts. LiLong capacitors, PSUs that weigh roughly the same as a cubic metre of air, tiny little heat sinks and what little airflow there is obstructed by knots of cables. Under conditions of strain is when parts fail. PSUs show their weakness with unstable current under high drain. What current does it proved on the various rails, and what wattage is it, out of curiosity? |
|
||||
|
Re: Great...a PSU issue...
Ok...
First to answer your questions: My PSU:
) and connected it to a different molex. It was fine for about 4 months when it happend again (now..in foreground). But I have detected a pattern...This always occurred after I ran dual monitors for long periods of time...say 2 months, although that doesn't make much sense, and while using these stupid extension cables. For all the other PC's I've built, the PSU cables were always long enough to directly connect to the card. I've never read on any other forum about this problem happening while being directly connected and, on the contrary, I've read that it works. I read on another forum that this is more common when the 9800 Pros are overclocked. Guess I won't be overclocking for a while. So, now on my list of parts to get:
Right now, my computer is operating fine with a clean molex connected directly connected after a bit of stretching. The burned molexes still have full 5v power, so the molexes aren't ruined. After measuring the voltages/ currents of the rest of the system they are all fine. But why would this happen? BTW: So far, the PCs that I've custom built for people I know are working as expected. Its only my PC.
__________________
If you have any questions you can email me at zack.schmid@gmail.com. |
|
||||
|
Re: Great...a PSU issue...
I forgot to add, that the acrid smell wasn't comming from the PSU afterall...the only reason it seemed like it was was because the fan under the PSU was directing the flow through the inside, then out the exaust fan. It seems like this is more the fault of the card, more that my PSU...
__________________
If you have any questions you can email me at zack.schmid@gmail.com. |
|
||||
|
Re: Great...a PSU issue...
This may sound rather contradictory, but you need to be a little careful about what you read and believe on forums; this inherently includes me of course. Any know-nothing can pretend they know what they're talking about on a forum, so credibility should be critically judged.
So, believe this or not at your discretion. _________________________________________________ For a 500W PSU, you have rather weak rails. 450W should crack 20A on +12. My 450W (albeit, carefully selected) provides 24A on +12 (and over 20A on all others). As I tried to protray before, weak components show their colours under high stress. The PSU is one of the last things people put money into, so many many Radeon 9800-series owners will have low grade PSUs. The longer the power cable, the more resistance, so if there's going to be some sort of surge, it's going to really heat up at the point of greatest resistance; i.e. the connector at the end of the chain, where the extension is. I highly, with the most cynicism I can summon, doubt that ATI has supplied "bad power extention cables"; one cable is much like another, in that it's basically copper/aluminum stranded wires. And why would it occur more often when 9800-series cards are overclocked? This is no mystery that requires the assistance of a detective; over clocked cards, flabbergasting enough, consume more power. If you want to continue using your PSU, then go ahead. Don't be surprised if something dies though. Especially if you go and add a GeForce 6800 GS. If you think the 9800 was a hot and heavy part; just wait for double the transistor count. |
|
||||
|
Re: Great...a PSU issue...
Don't worry, my power supply is going to be replaced promptly. This is my first priority, next is the 7800GS. But right now, the PSU has to stay, for it seems I am losing money faster than I am gaining it :
.But the questions still stand: How could +5v become so hot that it melts plastic while the other are perfectly fine? And another, why would my PSU provide only 19A on +12, yet provide 40A on +5? What would be the point of providing that much power to 5v?
__________________
If you have any questions you can email me at zack.schmid@gmail.com. |
![]() |
LinkBacks (?)
LinkBack to this Thread: http://www.itsallpc.co.uk/cases-power-supply-units-psus-cooling-overclocking/7148-great-psu-issue.html
|
|||
| Posted By | For | Type | Date |
| College Loan - loan subnet college, college arm | This thread | Refback | 16/11/07 07:37 |
| Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
| Thread Tools | |
| Display Modes | |
|
|