• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Dual Power Supplies?

vjcsmoke

Supreme [H]ardness
Joined
Dec 5, 2006
Messages
4,511
Right now I have a HAF 932 case, and it has the unsual design feature of allowing you to mount a power supply at either the top or bottom.

But let's say I want to put two 5870x2s in my system in the future. And my current PCPC PSU is rated at 750W. This isn't quite enough juice to run two 5870x2s in Quad-fire along with my overclocked i7 and other system components.

What about the idea of mounting a second PSU (500-750W) just to power the video cards and leave the first PSU to power the rest of the system. Would that work? Or would there be some difficulties with the uneven power loading? Also I'm concerned that mounting a PSU up top might block some of the airflow to the top 230mm exhaust fan. Thoughts?
 
I'm sure it could work. The only hurdle I can see is making sure the second psu turns on when the main one is. If there is a delay between the first and second, you may have a problem booting. Of course, I don't know anything for sure, just making semi-educated guess. As far as interrupting the airflow to the top fan, I can't see it hurting your actual temps.
 
What about the idea of mounting a second PSU (500-750W) just to power the video cards and leave the first PSU to power the rest of the system. Would that work? Or would there be some difficulties with the uneven power loading? Also I'm concerned that mounting a PSU up top might block some of the airflow to the top 230mm exhaust fan. Thoughts?

It'll work and shouldn't be any difficulties with uneven power loading. Some airflow will be blocked but shouldn't be that much of an adverse effect. The only real major problem with a dual PSU setup is lower efficiency but even then that's a minor concern.

Just make sure to get this adapter to avoid the problems that SicKlown42012 mentioned:
http://www.performance-pcs.com/catalog/index.php?main_page=product_info&products_id=21193
 
Right now I have a HAF 932 case, and it has the unsual design feature of allowing you to mount a power supply at either the top or bottom.

But let's say I want to put two 5870x2s in my system in the future. And my current PCPC PSU is rated at 750W. This isn't quite enough juice to run two 5870x2s in Quad-fire along with my overclocked i7 and other system components.

What about the idea of mounting a second PSU (500-750W) just to power the video cards and leave the first PSU to power the rest of the system. Would that work? Or would there be some difficulties with the uneven power loading? Also I'm concerned that mounting a PSU up top might block some of the airflow to the top 230mm exhaust fan. Thoughts?

188x2= 376watts, roughly half what your PSU is rated at, still plenty left over to power an i7 processor, even an overclocked one. I don't think you need to mess with dual PSU's.
 
188x2= 376watts, roughly half what your PSU is rated at, still plenty left over to power an i7 processor, even an overclocked one. I don't think you need to mess with dual PSU's.
He said dual 5870X2s, not dual 5870s. Big difference.
 
Something to consider with the dual PSU's though. Unless they've changed the designs (and by designs I mean more than just added power) then you'll need to put some load on the 5v rail if you want full power available on the 12v rail.
 
Something to consider with the dual PSU's though. Unless they've changed the designs (and by designs I mean more than just added power) then you'll need to put some load on the 5v rail if you want full power available on the 12v rail.
Unless the PSU is group-regulated, a modern unit shouldn't have any problem running without a minimum load on the minor rails. Especially if it's a newer DC-DC design, since those derive the minor rails straight from the +12V output anyway.
 
Thanks for the input. Yeah I'll be definitely getting that handy power supply daisychain cable in the future then. Just another though, but I heard that there are drive bay PSUs out there. If I install that in one of my free 5.25" bays, that could solve my possible airflow blockage up top and provide me enough juice for the aforementioned setup. Any recs on what drive bay PSU to use? (Obviously I'd daisychain it to the PCP&C unit.)
 
Thanks for the input. Yeah I'll be definitely getting that handy power supply daisychain cable in the future then. Just another though, but I heard that there are drive bay PSUs out there. If I install that in one of my free 5.25" bays, that could solve my possible airflow blockage up top and provide me enough juice for the aforementioned setup. Any recs on what drive bay PSU to use? (Obviously I'd daisychain it to the PCP&C unit.)
The best drive bay PSUs are still mediocre compared to the standard form factor units that are out there. I wouldn't recommend it since you'd probably end up spending as much or more for an inferior PSU. If you have the space for a standard ATX power supply, you should use one.
 
I recently tried out a multi PSU configuration with my RIG.

Used dedicated Corsair VX450W for my HD4770s.

Used dedicated Corsair TX650W for CPU, MB, hdds, fans.

I did this by connecting the VX450 to a separate MB without a CPU. I would first power up my MB powered by the VX450W.
Then I would immediatly power up my CPU MB powered by the TX650W

Ran OCCT PSU test and discovered my power ripples were less which is better. Not by much though.
Only performance gains I noticed were with the improved ripple. RIG can still OC just as good.

So in short IMO it's best to get a decent single PSU that can power every thing right. Less money and hassels with wiring and potential extra heat from wires in case.
 
Ran OCCT PSU test and discovered my power ripples were less which is better. Not by much though.
Only performance gains I noticed were with the improved ripple. RIG can still OC just as good.
Do you have an oscilloscope? I'm guessing not, in which case you have no way of measuring ripple which occurs at a far higher frequency than other measuring equipment can sample. And software voltage readings are worthless, including those that you get from OCCT. Only way to check your voltages properly is with a digital multimeter.
 
Do you have an oscilloscope? I'm guessing not, in which case you have no way of measuring ripple which occurs at a far higher frequency than other measuring equipment can sample. And software voltage readings are worthless, including those that you get from OCCT. Only way to check your voltages properly is with a digital multimeter.


No, just OCCT PSU test.

However it makes sense that adding a dedicated power source to the GPUS would take some stress off of the rest of the system so the test results did follow the theory of improved power quality.
 
No, just OCCT PSU test.

However it makes sense that adding a dedicated power source to the GPUS would take some stress off of the rest of the system so the test results did follow the theory of improved power quality.
The test you performed was meaningless as were the results you obtained. Like I said, any voltage readings you get from software are inaccurate. You must use a digital multimeter to check voltages. And it is impossible to measure ripple without an oscilloscope.
 
One thing to consider is that the top 230mm fan has to be removed to fit a top mounted PSU, that will do more than just interfere with airflow.
 
Back
Top