• 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.

Crazy Idea?

LesterOfPuppets

Limp Gawd
Joined
Nov 14, 2005
Messages
143
This is hypothethical at this point, since I've got a new PSU on the way, but...

I recently got a DFI Lanparty mobo and 7600gt, moving up from my Asus A7n8x-e dlx/6600gt. I run an Ultra X-connect 500 with only a 20-pin mobo connector.

Since It'd cost $15 and several days to get a new Ultra 20-24pin cable, a little less dollarwise for an ugly f-m 20-24 pin adapter, I considered using a secondary old AT PSU to power the 4 extra pins. If one of them wasn't 3.3 volts, I'd have just powered the extra pins with available 12v and 5v leads.

The plan was to keep the AT psu running 24/7 powering the 4 extra pins (12v, 3.3v, 5v and ground), and my Ultra X-connect would power up the remaining 20 with the computer's power button.

Any electrical wizards here have any opinions on how well this would work. Would spikes from powering up the AT power supply bother the mobo, PCIe bus, anything like that?
 
I see a couple of problems here. Most PSUs aren't designed for paralleling the outputs which is what would happen in this scenario. The problem is that one PSU rail may be 5.00 volts while the other might be 4.92 volts. The result is that each PSU will try to regulate the other with a large current flowing between the two PSUs.

The other problem would occur in the event the 20 pin PSU turned off for some reason while the PSU powering the four pins was still powered. The MOBO would then try to power itself through just four wires. You would probably end up with a burned connector.
 
Frank4d said:
The other problem would occur in the event the 20 pin PSU turned off for some reason while the PSU powering the four pins was still powered. The MOBO would then try to power itself through just four wires. You would probably end up with a burned connector.

Oh! That would suck. Didn't even think of that.
 
Aha, I see now that the AT's orange cables aren't 3.3v. I was just brainstorming, looking at that old AT PSU, never looked up the color codes.

As for Frank's scenario, wouldnt' the mobo shut down if the ATX psu supplying the 20 pins lost power, as soon as no juice on the green wire, I think that would happen anyways.

I did consider the upgrade kit, but I'm just going to keep the stiff as hell round cables, and keep the old Ultra with my olde mobo.

I got a new Ultra, non-modular, for just a little bit more than the cable upgrade kit, after MIR.

http://www.zipzoomfly.com/jsp/ProductDetail.jsp?ProductCode=373060

I suppose I do have a spare ATX PSU around, in the future, if I come by a sacrificial 24-pin mobo/vid card/ram I'm going to have to experiment, see how well those regulators work. I can see how it would be a problem reading Franks description.

Thanks for the input y'all.
 
LesterOfPuppets said:
As for Frank's scenario, wouldnt' the mobo shut down if the ATX psu supplying the 20 pins lost power, as soon as no juice on the green wire, I think that would happen anyways.

No, unless the MOBO controls the green wire on both PSUs. If the four wire PSU is completely on it's own then the problem I described can occur.

You can splice the green wire and one black wire of the four wire PSU into the green and black wires of the 20-wire PSU. In that case the problem I described would not occur.
 
Frank4d said:
No, unless the MOBO controls the green wire on both PSUs. If the four wire PSU is completely on it's own then the problem I described can occur.

You can splice the green wire and one black wire of the four wire PSU into the green and black wires of the 20-wire PSU. In that case the problem I described would not occur.

You're right, I would've had to do that. constant power on the 4pins would likely be a bad thing.

The splicing you mention has worked well for me in the past.
On this build with a couple of low power PSUs (180+200).
 
Back
Top