Jump to content
Corsair Community

HX620W & EVGA 8800GT KO : insufficient power?!!


0031nek

Recommended Posts

UPDATED INFO, PLEASE READ!

 

 

This was posted over in the EVGA forums...

 

http://www.evga.com/forums/tm.asp?m=265744&mpage=3&key=

 

Hi all,

 

Here is the official messaging from NVIDIA:

 

Thanks,

Jacob

 

 

“Several users have reported an issue on GeForce 8800 GT/GTS G92 which can cause the NVIDIA System Sentinel to display a false low power warning message, which causes the GPU to operate in low power mode, resulting in low performance or application crashes. This issue has been reported on various power supplies. NVIDIA has reproduced the issue and is working on a fix with a new version of our drivers. We will update the community promptly when a fix is available. We apologize for any inconvenience or frustration gamers may be experiencing with this issue.”

 

Hi all,

I really need this help very very much!

Bought HX620W yesterday and thought I am 1 proud owner and I'm not!

 

mobo: GA-P35-DS3P

cpu : intel E2180

HDD : WD and Maxtor

Sound card: Bluegears b-enspirer

Cooler: ASUS Silent Knight

ram: Corsair C4DHX 6400 1gbx4

 

VGA: EVGA 8800GT KO edition 512mb 675mhz

 

 

Now, my previous PSU was Coolermaster extreme Power Duo 600w, then i changed to Corsair HX620W.

 

When my windows booted, an error displayed from the NVIDIA, telling me i got not enough power supply!

 

I mean i do not have any problem with the coolermaster PSU, but this AUD$189 psu give me the "insufficient power" problem!??!!

 

I brought it back to the shop and i do not know how they tested it, and they said "well, it's all no problem, we can't provide any refund or warranty if we tested it ok"

I asked about their system they said "we tested on a 8800GTS, no problem" i asked again "don't you have 8800GT?"

"Nope" they replied.

I then try to asked them what they can do for me and they said "Well, it's nothing we can do, since it's tested ok here"

 

I am very very very upset!

 

I edited this 1st post, added in the readings from speedfan. I think there's something wrong with the voltage at the +12v. Please help.

 

Please help!

popuperror.thumb.JPG.4a5c06508896597ad3719960017ec861.JPG

speedfan.JPG.f77d31e781d0cf9cee7f78d53f7fcce6.JPG

Link to comment
Share on other sites

  • Replies 129
  • Created
  • Last Reply

Thats really anoying, your system should be running quite happily with that PSU, realisitly your system shouln't be using more than 350w even at an absolute peek load. maybe a little more if your CPU is heavily overclocked.

 

I beleve this PSU has 3 12v rails so it might be worth re-jiging your loads over those rails just incase you have overloaded one of them or somthing. maybe find a 4pin milex to pci-e adapter to help you move the graphics card to another rail.

 

This power calulator here > http://www.extreme.outervision.com/psucalculatorlite.jsp is very handy.

Link to comment
Share on other sites

Thank you very much for the 1st feedback and suggestion, well, the 4-pin molex to pci-e had been tried, nope, not helping at all.

:(

 

Only other thing I could posably suguest is a change of driver,( you seem more than compedent enough to have tried this already) but if your older PSU is still giving you no greif then it sounds like somthing is not right with your corsair PSU.

 

Might be worth testing the voltage lines are roughly within 10% of their specs with a multimeter. 12v lines your graphic card is using as well as the others( you can test for 5 and 12v through a molex connector even better if your gfx card is on the same bit of cable), also the 3.3v with software monitoring for the 3.3v unless you know exactly where to test it on your mother board. if your scared of using a multimeter just use a software utilty though these arn't know for their accuracy. Hopefully you'll see if anything is obviously wrong with it. Also goes without saying be very carful not to short anything with your multimeter probes.

 

 

Hopefully one of the corsair guy's has a better suguestion.

Link to comment
Share on other sites

The HX series have one giant 12V rail. Try one 6-pin PCIe cable connecting one to the 8800GT and the other to the psu. If the msg. doesn't go away try the other cable and other PCIe output on the HX620. If you still have a problem test the +12V rail with a DMM.

 

Otherwise it's safe to say you have a compatibility issue since your HX620 works fine in another pc with an 8800GTS, and a different psu works in your rig.

Link to comment
Share on other sites

Tried to swap all types of pins and cables, to different types of output (different rails) from the psu and not helping as well.

Tested using multimeter at the shop by a guy, and he said nothing's wrong with it.

And 1 thing i noticed, this psu do not have problem if i turned off my pc overnight and starts it the 1st time, however, after some time using it, the problem will be back when i restart the pc.

 

Maybe izzit the temperature affecting the performance? my highest recorded temp recorded via "core temp" during the error was 42celcius.

Link to comment
Share on other sites

I've got a HX520 on order to replce my noisey TX750 which was ment to be quieter than my currect hyper 580watt psu. I'm running a simlar Gigabyte board too p35 ds3. my mates got a very simlar setup to me except a different case ang hd with a hx620, but the same board as i. he's had no probs like that voltage reading on speed fan.

 

if your dareing enough try to get a multimeter voltage reading from the pci-e power connector connected to the graphics card. the yellow wires should be +12 volts and to prevent a short touch the ground probe against any exposed peice of metal on the case. ocationaly speedfan can get these readings very wrong. if the voltage is low their then it does explain the message you see on screen. you should do the same on the atx power connector. if the pce-e connector is low.

Link to comment
Share on other sites

actually i am very fadeup to have this psu. because i spent $189 to get an expected high end psu, and wanna get rid of the noise.

When this problem occurs, only hope that the shop or corsair can at least do something for me, because i do not have multimeter or whatsoever tools to test my psu, as well as will not be spending more money to test out my psu that in the early place shouldn't have any problem.

 

Sigh.. :(::mad:

 

Oh, by the way, is it normal that our psu will run 1stly run on full speed (with very loud noise), before our CPU fan runs and before the bios load, everytime we switch on our pc? Mine will always hizzzzz!!!! before my CPU fan spins, and the in 1second, my CPU fan will spins and the led will on. then only i can see the bios info.

Link to comment
Share on other sites

I cannot provide an explanation for the power issue but the noise you are hearing when you start the pc is the gpu fan. The only explanation I can think of and someone correct me if I'm wrong, is that the card is trying to lose settled dust. sounds stupid I know.

 

I'm sure by your replies that you are not new to this. try reseating your card. make sure the 6-pin pci-e power cable is inserted correctly. this is the basic stuff I always try in a situation like this. If all else fails, take the card to a friend and test it. There is another option which may or may not have relevance to your situation but if your bios provides a setting for assigning IRQ to VGA - enable it and disable any on-board graphics.

 

When you actually do get the thing to work, your worries are not over. I have a 88gt KO as well and there are a lot of corrupted displays issues. Evga actually sent me a rom to flash the video bios to lower the performance of the card - which actually makes the KO part of the card redundant.

Link to comment
Share on other sites

i really can't believe how could they do something like this, i mean lowers your card's performance?

We paid some extra money for them to factory overclock the gpu for us and then they provide us the bios to disable the overclock?

THIS IS NOT PROBLEM SOLVING AND CUSTOMER SERVICE.

Link to comment
Share on other sites

Yes well, business is as business does. Evga also had an issue with certain cards being distributed with the wrong bios. In the case that I looked at, it was to the gain of the client as the 'faulty' bios actually provided better performance than the card's advertised specs.

 

The problems I experience seems to be heat related. Heaven knows how they can ship a card that every second client experiences lockups with because of overheating. Their solution is to 'underclock' the card to just above their 'standard' 88gt range and improve the fan settings.

 

The new bios was adapted for me by one of their support guys as I was and am not willing to change anything related to the specs of the card. It lowers the clock speed but (on par with the official bios release for the sc range) "improves fanspeed auto sensor".

 

Then again, there are a lot of people out there who have only good things to say about the card like how well it overclocks.

Link to comment
Share on other sites

  • Corsair Employee
Yes you can submit an RMA from Australia, however you would normally have to pay shipping charges to Corsair, and we would pay shipping back to you. If you have any questions regarding the RMA process, please contact our customer service at 888-222-4346 and dial "0", (510) 657-8747 or email rmaservice@corsairmemory.com for RMA status.
Link to comment
Share on other sites

after a series of trial and error here and there, i found out some conditions for my system to display

 

the sentinel insufficient power msg.

 

1) Cold boot : my PC will NOT encounter this msg problem if started from a cold boot (i.e turn on the PC

 

after an overnight off)

 

2) Hot boot: If my PC started from a cold boot, no msg will appears, however once i used or turned on my

 

pc for at least an hour, the msg will come out after a SHUT DOWN and ON again, within 1minute.

Let's say if i started my PC from a condition where there's no sentinel msg, there is 70%possibility that

 

the msg will NOT APPEAR as well if i only do a Windows RESTART.

Msg might only will appear if i TURN OFF my HOT pc and then it ON again while it's still HOT.

 

3) Ambient temperature: It's summer now in Australia, therefore the last few days weather quite HOT, and

 

reached 29celcius outdoor, while my home is just slightly a bit lower temperature ranging from 26-

 

28celcius. And the message appear almost every time i turned my PC on.

 

3) GPU temperature : checked my GPU temperature at idle all the time using Rivatuner, turns out to be

 

60celcius on average and just 3celcius lower sometimes if i turn on my PCI slot cooler to exhaust the hot

 

air in the case. The msg appeared very randomly and NO DIRECT RELATION between the chances of the msg to

 

popup, and the TEMPERATURE OF GPU.

 

4) I overclock my E2180, running at 3.1Ghz, with Asus Silent Knight cooler, speed set to auto, and my

 

core temp according to CORE TEMP is ranging from 32idle to 49load and 42idle to 58load for the ambient

 

temperature 18celcius to 29celcius respectively.

 

Ambient 18celcius, the core temp 32idle, 49load

Ambient 29celcius, the core temp 42idle, 58load

 

the last few days quite hot, ambient ranging from 24to29celcius, the msg appears almost everytime i turn

 

on my PC.

I assume that the insufficient power msg is directly related to ambient temperature and since so, the msg

 

is direct related to the temperature of my system, casing, or whatsoever, and definitely NOT RELATED TO

 

THE GPU.

 

5) PSU temperature: ok, now i come to my biggest concern, the temp of my PSU. As i mentioned, the longer

 

i run my PC, i assume of cause my PSU is getting hotter, therefore, if there's not enough time for it too

 

cool off, the message will appear after i boot windows (turn on) my PC if i turned off my PC.

 

And 1 thing i notice: the Corsair HX620W PSU dissipate more heat out from the behind of it compared to

 

the Coolermaster extreme power duo 600W.

 

Although i couldn't agree more that the Corsair HX620W is really the most quiet PSU i ever encountered,

 

however, i assume that it also dissipate more heat, or is the ADDA 120mm bearing fan can't do it works

 

efficeintly?

 

What can i say? The efficiency of this Corsair HX620W seems to drop when the temperature increases.

 

Very upset to know about it.

 

This were all my assumptions after a few days trial and error with my PC in order to know exactly what

 

would cause the Nvidia Sentinel Power Indicator to appear and telling me my PSU is not providing enough

 

power to my GPU, and tell you what, if this msg appears, my PC will lockup if i run 3dMarks06.,dare not

 

to play Crysis with it of cause.

 

If no msg appear, this my new EVGA 8800GT KO scores 11311 3DMarks06, with my intel E2180 @ 3.10Ghz with

 

C4DHX 1gb x 4 RAM.

Dunno this is consider good or bad.. :(

 

Cold PC---> turn on---> NO msg appear ---> use until my PC hot---> SHUT DOWN -->Hot PC --> turn on again

 

immediately -->msg appeared.

 

Or sometimes

 

Cold PC---> turn on---> NO msg appear ---> use until my PC hot---> RESTART the PC--->Hot PC --> NO msg

 

appeared.

 

So, i'll leave the rest to anybody to help now, of cause trying to keep everything cool to make sure this

 

msg do not appear is not the only solution because if i am still with my coolermaster PSU, i do not have

 

to worry about 24/7 always on PC or hot summer or high ambient temperature, but with this Corsair HX620W,

 

i could only play Crysis when it's winter? or stop in an interval of 30mins to ensure my Corsair HX620W

 

PSU is not running "too" (although i know it's not TOO) hot?

 

Very very disappointed!

Link to comment
Share on other sites

and do you think that contacting the supplier of corsair or contacting corsair directly is a better solution for RMA? Because i contacted the supplier and they said they need to wait for corsair to confirm this issue with them, then only will take further steps. It's already 2days.
Link to comment
Share on other sites

I've just received an email from the supplier of corsair in australia, and they said they contacted Corsair, corsair replied that they didn't have issue in this product with my 8800GT.

Therefore, they requested me to bring my GPU, together with the PSU to tested it out at the shop, to show them the problem and if there is, can claim for warranty.

While in the shop, they'll need 7days to test these and pay Aud50 to test.

I mean what the hell in the world they're doing all these crap? I just want to ensure i get what i paid, and i still need to pay for them to make sure i can get what i paid earlier?

Oh f***!

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...