Antec phantom 350 problem?
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee, Devonavar
Antec phantom 350 problem?
I currently have a quite silent system, but lately I've been experiencing random shut downs for no apparent reason. I discovered though that it must be somehow related to either the PSU og me MSI 6600gt card.
On a couple of occasions I got an error message from the nvidia control panel saying that the card isn't receiving enough power while running 3dmark05, although I have a wire connected to the 12v socket. Moments later the computer shut down :/ I tried with another cable from the PSU, and the same happens.
I checked with a multimeter to see if the 12v voltage is correct, and it says 12.07v so it's well within limit. I still think the power supply is somehow to blame, my system isn't exactly demanding:
Aopen i855gmem-lfs with a 1.8Ghz Pentium-m processor. Stock fan attached.
2x SATA ide drives, Samsung Spinpoint and Seagate Barracuda
1x Papst 12cm fan, normally running at 7v.
I'm going to let 3dmark run in a loop for some time to stress test the system. Later today I'll probably dig out a conventional 300w psu.
I'm just wondering if anyone has experienced similar behaviour? I have a strong suspicion that the phantom is to blame. The PSU has worked fine for almost 5 months now, but maybe during the hot summer here something happened.
On a couple of occasions I got an error message from the nvidia control panel saying that the card isn't receiving enough power while running 3dmark05, although I have a wire connected to the 12v socket. Moments later the computer shut down :/ I tried with another cable from the PSU, and the same happens.
I checked with a multimeter to see if the 12v voltage is correct, and it says 12.07v so it's well within limit. I still think the power supply is somehow to blame, my system isn't exactly demanding:
Aopen i855gmem-lfs with a 1.8Ghz Pentium-m processor. Stock fan attached.
2x SATA ide drives, Samsung Spinpoint and Seagate Barracuda
1x Papst 12cm fan, normally running at 7v.
I'm going to let 3dmark run in a loop for some time to stress test the system. Later today I'll probably dig out a conventional 300w psu.
I'm just wondering if anyone has experienced similar behaviour? I have a strong suspicion that the phantom is to blame. The PSU has worked fine for almost 5 months now, but maybe during the hot summer here something happened.
I'm going to try to connect the 6600gt on it's own 12V cable. I just read the review once more, and it says that 12V1 should be good for 18A. Does anyone one know which cable goes to the 12v1 rail? I just disconnected a cable with 2 12v connectors (one of the molex is dual with a short floppy cable).
-
- Friend of SPCR
- Posts: 2887
- Joined: Mon Feb 28, 2005 4:21 pm
- Location: New York City zzzz
- Contact:
that cpu can run a pentium system and two sli'd 6800GT's.
youre lil 6600GT while being a great card, isnt exactly a wattage beast at all.
Remember: some 6600GT's can be run near passively cooled, so that cant be pulling all that much heat.
I would think it's the card. I own that psu and well, its a very good piece of equipment. (my fav piece of gear i have bought in a few years actually) And it gives me stable voltage and remains only mildly warm under load in my system playing BF2 for 4 hours.
youre lil 6600GT while being a great card, isnt exactly a wattage beast at all.
Remember: some 6600GT's can be run near passively cooled, so that cant be pulling all that much heat.
I would think it's the card. I own that psu and well, its a very good piece of equipment. (my fav piece of gear i have bought in a few years actually) And it gives me stable voltage and remains only mildly warm under load in my system playing BF2 for 4 hours.
The PSU is only mild to the touch, wouldn't say it's even hot. Yeah, I'm actually still hoping for that the 6600gt is the culprit but:
Last night I used another cable from the PSU and even double checked that the connector on the card received voltage from the PSU. I measured the contact points on the 6600gt printboard, and it still showed 12v. So the connection is at least alright.
I put speedfan in the background to monitor voltage on all lines and log every 3 seconds. Then I set 3dmark05 run on repeat.
Right before the crash it looks like this:
The 6th columns shows 12v voltage. Seems it drops a lot? Further up it's even as low as 10.6v The system only ran for a couple of hours, and looking from the log the voltage was already at 11.4v 51 min. before the end. Anyway, this night I'm digging out another PSU and again run a 3dmark.
Last night I used another cable from the PSU and even double checked that the connector on the card received voltage from the PSU. I measured the contact points on the 6600gt printboard, and it still showed 12v. So the connection is at least alright.
I put speedfan in the background to monitor voltage on all lines and log every 3 seconds. Then I set 3dmark05 run on repeat.
Right before the crash it looks like this:
Code: Select all
Seconds Temp1 Temp2 Fan2 Vcore +12V 3.3V Vcc 5Vsb Vin2 Vbat
19459 43,0 47,5 1329 1,339 10,7008 3,344 4,9067 5,12 0,0 0,0
19462 43,0 47,0 1329 1,339 10,7008 3,344 4,9067 5,12 0,0 0,0
19465 43,0 47,5 1318 1,339 10,7008 3,344 4,9067 5,12 0,0 0,0
19469 43,0 47,5 1318 1,339 10,7008 3,344 4,9067 5,12 0,0 0,0
19471 43,0 47,0 1278 1,339 10,8832 3,344 4,9067 5,0933 0,0 0,0
19474 43,0 47,0 1278 1,339 10,8832 3,344 4,9067 5,0933 0,0 0,0
19477 43,0 47,0 1288 1,339 10,8832 3,344 4,9067 5,0933 0,0 0,0
19480 43,0 47,5 1288 1,339 10,8832 3,344 4,9067 5,0933 0,0 0,0
19483 43,0 48,5 1288 1,3342 11,0048 3,328 4,9067 5,0933 0,0 0,0
19487 43,0 47,5 1288 1,3342 11,0048 3,328 4,9067 5,0933 0,0 0,0
Run the machine with another psu just to test that the voltage sensor is working properly. If it really is that low, you should be able to get the unit replaced. There's no reason for it to be that erratic.The 6th columns shows 12v voltage. Seems it drops a lot? Further up it's even as low as 10.6v
Just tell them the machine is intermittently restarting and the voltage levels are far too low. You tested with another power supply, and the machine works fine and the voltage levels are fine.Wonder if it will be easy to convince the store about the erratic behaviour.
That's why you must test with the other psu, so you can tell them that.
One more thing, definitely don't void the warranty. If there's a sticker, don't break it. The explanation will be enough.
-
- SPCR Reviewer
- Posts: 8636
- Joined: Sat Nov 23, 2002 6:33 am
- Location: Sunny SoCal
Well, it's just my opinion of course, but I'd think is would be real funny if you got the shit shocked out of you while you were doing something like that. The term "poetic justice" comes to mind...~El~Jefe~ wrote:here's a tip:
take a small screwdriver and fish around the inside of the psu. when you successfully dislodge a wire or two, bring it back to the store. this ensures it being "DOA".
of course that's ny style. shrugs.
-
- Friend of SPCR
- Posts: 2887
- Joined: Mon Feb 28, 2005 4:21 pm
- Location: New York City zzzz
- Contact:
justice is a new psu.
or anything free. just ask plato.
-----------------------
"One more thing, definitely don't void the warranty. If there's a sticker, don't break it. The explanation will be enough."
see, thats the thing about screwdrivers or even better, a thin wooden dowel I guess. no sticker problems. that's why i said it.
or anything free. just ask plato.
-----------------------
"One more thing, definitely don't void the warranty. If there's a sticker, don't break it. The explanation will be enough."
see, thats the thing about screwdrivers or even better, a thin wooden dowel I guess. no sticker problems. that's why i said it.
Hehe, good idea Well, I've now had my system up and running with a 340W Chieftec PSU (I swapped it from another PC with the antec) for a couple of days and it's 100% stable. I got another confirmation that the PSU is defunct since the second system also turns itself off randomly. This even happens while idling. It's just a P4 with one drive and a gf4200ti card.~El~Jefe~ wrote:here's a tip:
take a small screwdriver and fish around the inside of the psu. when you successfully dislodge a wire or two, bring it back to the store. this ensures it being "DOA".
of course that's ny style. shrugs.
Will be sending the unit for RMA next week. I will of course note the serial number and make an entry in the Phantom feedback thread.