Jump to content
Corsair Community

vx450w won't power old system


katspaugh

Recommended Posts

My new vx450w doesn't react when I hit my system's power button. Passes the 'paper clip' test (PSU's fan turns on), gives 5V on standby. Tested in system with just the 20- and 4-pin connectors (split from 24 and 8) in the motherboard, nothing. Tested old PSU with same setup, powers on fine.

 

RMA'd with the seller, second unit does the same thing.

 

System is an HP pavilion 753n -- maybe a decade old? Motherboard unknown, probably Xenon but not the one HP says it is (visibly different layout).

 

Fan went out on last power supply ( coolmax ap-400x ). It still powers the system, but gets hot eventually and the screen starts blanking during HD reads. Coolmax supposedly ATX 2.03, though it has a -5v wire so who knows. 15A and 10A on two 12Vs, 21A on 5V, 22A on 3.3V. Coolmax was 350w max.

 

Original power supply was 163W from Delta Electronics -- 10A on the 12V, 23A on the 5V, and 10A on the 3.3V. No -5V wire.

 

vx450w gives 33A on the 12V, 20A on both the 5V and 3.3V.

 

Is that 1A on the 5V (difference between coolmax and vx450w) keeping things from even powering on?

 

Also strange - the power on pin gives 5V when off, then dives down when the power switch is hit, then goes back to 5V. The coolmax is 0 when off, then 12V when on. Did the method for powering on change between ATX 2.03 and 2.2? This was on the one I RMA'd, haven't tested it on the new one yet.

 

 

Also, the support page ( http://tsxpress.corsairmemory.com/default.htm ) tells me I have to enable session cookies, when I'm pretty sure I have.

 

Update: Bah -- found another link to tech support, submitted there.

Link to comment
Share on other sites

This is seriously mysterious -- shorted with a paper clip, the PSU gives proper voltage on all the pins. Plugged into the (stripped of all other load) motherboard, it fails to turn on.

 

Unless the mobo + CPU sucks down 21A on the 5V, the PSU would at least power on, right?

 

Is the 20/24 connector somehow unique on this PSU / my board? Maybe the pins aren't touching at all.

 

(It's so quiet when on, I have to double check the fan when I short it -- inconvenient for this test, but really nice if I can get this in my case.)

Link to comment
Share on other sites

  • Corsair Employees
Older versions of the ATX spec used the 5 volt rail to power the major components such as the CPU and graphics card. Judging from the specs of the original power supply, I would suspect that the board is built to an outdated ATX spec and is looking for something that does not exist on our PSU. I would contac tthe manufacturer of the system to find out if the system is designed to work with either an outdated ATX spec, or possibly a proprietary one.
Link to comment
Share on other sites

CPU is an earlyish Pentium 4, 2.53 Ghz, so it's at very least ATX 1.0, likely ATX 1.3; all of which need more 5V, yeah . . . but 1A off?

 

Did I get lucky with the coolmax in the interim, then? (I at least know the spec's aren't proprietary, normal ATX PSUs work.) It has more of an ATX 1.x loadout than the 2.03 it claims, I agree, but it's still only one more amp on the 5V than the vx450w. Maybe the coolmax lacks some sort of protection present in the vx450w, and it was running outside of specs for a few years.

 

Is there more to the difference in ATX spec than the amps on the 5V rail (and no -5V, that's not the issue)? Is that 1A enough to keep it from powering on at all?

 

Obviously something is incompatible; I'm really as interested in solving the mystery of what as I am getting the system back up at this point.

Link to comment
Share on other sites

Archived

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

×
×
  • Create New...