AX860i pin 20 (3V Sense): what should be there?

[Kagetora]

I am having an issue with my AX860i. When I connect the 24-pin ATX cable to it, and try the self-test button, it fails, just shutting down. I thought it might be the cable, as otherwise the PSU self-tests fine, so I constructed an entirely new (shorter, as the original is WAY too long for my needs) cable, and got the same results.

 

But, if I remove pin 20 (3.3V Sense) from the 10-pin connector, it self-tests fine. This is doubled up with pin 13 (3.3V) at the mobo connector, so, supposedly, it can sense the 3.3V output and the PSU can fine-tune it.

 

I used my DMM to check every output from the PSU, all voltages check fine. And, as I mentioned, it self-tests fine, right up until the point pin 20 is inserted into the 10-pin connector. Then it fails.

 

Checking pin 20 on the PSU to case ground, I show +3.28V. Is there supposed to be actual voltage on the 3.3V Sense pin? Am I missing something here? What am I doing wrong, or is there something wrong with the PSU, the cables, or what have you?

 

Thanks in advance.

[WesternDreaming]

you wired the cable yourself and tested the dmm

 

can i ask the positive and negative 12v numbers?

[Kagetora]

+/- 12.01, IIRC. When I was checking the voltages, I just made a list of what went on what pin and checked off the list. THe only ones I actually wrote down were "PS_ON," which was 4.50V, and "PWR_ON," which was 4.96. All of the +5V were spot on at 5.00, the 3.3V were all 3.29 (except the one in question, pin 20, which was 3.28). All the COm were 0. It seemed to DMM-out just fine. Just wondering about pin 20.

[WesternDreaming]

so really its just the 3.3 that is undervolting

this is not an issue for the forums

 

this is an issue for techs

 

its an "i" psu, there is no reason for the undervolt

i bought the 1200i to have guaranteed power for my parts

[Kagetora]

Well, the question was "Is there supposed to be 3.28V on the 3.3V Sense rail?" I was under the impression there shouldn't BE a voltage here, that it should be a connection from the 3.3V pin 13 back into the PSU so it can "sense" what it is outputting on the 3.3V rail. Is that incorrect? Is there supposed to be voltage on pin 20?

 

But you are saying that the 3.28V on pin 20, or the 3.29V on the 3.3V rails, is out-of-spec? Is that it? I'm not sure I understand.

[WesternDreaming]

i used to be told that there was some allowance for out of spec voltages

the sense and the 3.3v are linked byo ne of the pins i dont remember which

 

i bought an "i" to fix this problem

id be asking why it is out of spec even a bit

when i OC even a .1v increment can be the difference between posting and not

 

its just my thought, no reason to be under volt at all

[Kagetora]

Got it. I don't know what the tolerances are for the 3.3V rails, but 3.29V is only .3% out of tolerance, or something like that. Probably within manufacturing specs, but not perfect.

 

The 3.3V pin 13 and the 3.3V Sense pin 20 are linked together to pin 13 on the mobo connector, with pin 20 on that connector empty. I'm just not sure if I should have +3.28V on pin 20.

[peanutz94]

Got it. I don't know what the tolerances are for the 3.3V rails, but 3.29V is only .3% out of tolerance, or something like that. Probably within manufacturing specs, but not perfect.

It's 5% on all rails. So you are well within spec for that volatge.

 

The 3.3V pin 13 and the 3.3V Sense pin 20 are linked together to pin 13 on the mobo connector, with pin 20 on that connector empty. I'm just not sure if I should have +3.28V on pin 20.

Pin 20 should be blank , no wire at all. That used to be a -5v line that was removed when the AtX specs were updated a few years ago. Pin 16 should be your sense line. And pin 13 is indeed 3.3v and pin8 would be your Power Good line.

http://www.google.com/imgres?imgurl=http://image.pinout.net/pinout_network_rj45_files/24-pin-atx-pinout.gif&imgrefurl=http://image.pinout.net/pinout_network_rj45_files/&h=414&w=536&sz=15&tbnid=Yspi0t0GxiIaPM:&tbnh=97&tbnw=126&zoom=1&usg=__CCkpoVoOYFViLi5BH1GNvD7lFn0=&docid=AFNJgyTTVmc6ZM&sa=X&ei=3YyfUc-HNOX8yQGi3oGICA&ved=0CDMQ9QEwAA&dur=1178

[Kagetora]

It's 5% on all rails. So you are well within spec for that volatge.

 

 

Pin 20 should be blank , no wire at all. That used to be a -5v line that was removed when the AYX specs were updated a ew years ago. Pin 16 should be your sense line. And pin 13 is indeed 3.3v and pin8 would be your Power Good line.

http://www.google.com/imgres?imgurl=http://image.pinout.net/pinout_network_rj45_files/24-pin-atx-pinout.gif&imgrefurl=http://image.pinout.net/pinout_network_rj45_files/&h=414&w=536&sz=15&tbnid=Yspi0t0GxiIaPM:&tbnh=97&tbnw=126&zoom=1&usg=__CCkpoVoOYFViLi5BH1GNvD7lFn0=&docid=AFNJgyTTVmc6ZM&sa=X&ei=3YyfUc-HNOX8yQGi3oGICA&ved=0CDMQ9QEwAA&dur=1178

 

Yes, thats the connection diagram for the motherboard. Pin 20 as "NC," no connection. Thats fine. The problem is that the PSU connector has pin 20 (listed as "3.3V Sense") running to pin 13 on the mobo connector, which is ALSO connected to pin 13 on the PSU connector ("3.3V"). With both connected, the PSU fails self-check. With pin 20 removed from the PSU connector, it self-checks fine. I have 3.28V coming out of the PSU on pin 20. Is that correct for a "3.3V Sense" rail? Or is it supposed to have nothing on it?

[peanutz94]

The problem is that the PSU connector has pin 20 (listed as "3.3V Sense") running to pin 13 on the mobo connector, which is ALSO connected to pin 13 on the PSU connector ("3.3V"). With both connected, the PSU fails self-check. With pin 20 removed from the PSU connector, it self-checks fine.

I'm certain that is correct. The PSU self check should only be run with NO cables attached to it. It will fail otherwise.

[Kagetora]

I'm certain that is correct. The PSU self check should only be run with NO cables attached to it. It will fail otherwise.

 

It would be REALLY nice if Corsair printed that in the absolutely worthless manual they include with their PSU's.

 

I've been using self-test to test individual components like cables/voltage outputs, fans, lights, etc. It works fine with things plugged in, actually...or it seems to. For example, if I plug the SATA power cable in, and self-test, the DVD players power up fine. Ditto if I plug in the fan controller...self test spins up the fans and activates the lighting. It didn't say anything anywhere about not doing this, and since it was working fine, I figured it must be designed that way.

 

Is that not the case? :bigeyes:

[peanutz94]

It didn't say anything anywhere about not doing this, and since it was working fine, I figured it must be designed that way.

 

Is that not the case?

Sorry, that's not the case. Like I said the self test must be run with no cables attached to it.

 

quote from RamGuy:

You will need to disconnect all cables from the PSU to perform the self test. You should not connect anything to the MB while doing the self test

http://forum.corsair.com/v3/showthread.php?t=116854&highlight=test

 

Here is another thread where he stated that information was in the owners manual for the PSU. *shrugs*

http://forum.corsair.com/v3/showthread.php?t=112526&highlight=test

[Kagetora]

Sorry, that's not the case. Like I said the self test must be run with no cables attached to it.

 

quote from RamGuy:

 

http://forum.corsair.com/v3/showthread.php?t=116854&highlight=test

 

Here is another thread where he stated that information was in the owners manual for the PSU. *shrugs*

http://forum.corsair.com/v3/showthread.php?t=112526&highlight=test

 

RamGuy is wrong, to put it bluntly. The manual that comes with the AX1200i, AX860i, and AX760i (same manual, lists all three on the front) doesn't say anything about not having cables connected for a self-test. I just re-read every word in the English-language section...it wasn't hard, its only 12 pages. And the first 3 are the front cover, a page with nothing but a Corsair logo, and a list of what page each language section begins on. So, its really only 9 pages. The next three (4-6) congratulate me on my purchase and advertise the features of the PSU. 7-9 are various cable configuration lists and efficiency graphs for the 3 PSUs. :laughing:

 

FINALLY, on page 10-11, I get two whole pages of actual information. One quarter of one page is dedicated to the self-test feature, with no mention of leaving everything unplugged. Page 12 is a bunch of warnings about not opening the PSU, getting it wet, using it outdoors, etc. Then we move on to Francais.

 

Totally, utterly useless. The "manual" is essentially toilet paper. Not that I need a lot of actual instructions or anything, but there is a lot of technical data that could have been included...like a pin output diagram, for example. :sigh!:

 

So, if that bit about the self-test is in there, I apparently keep missing it, glossing over it, or blanking it from my memory. :p:

 

I'm starting to wonder if I don't have a bad motherboard, though. If the PSU is actually installed, and all the connections are made (to the mobo, drives, GPU, etc.), the front panel "ON" button won't do anything. I've DMM'd the button mobo connector, "infinite" resistance dropping to .02 ohms when the button is pushed, so the button, cable, and plug work fine. Shorting the PWR_ON pins on the mobo directly with a DMM probe also resulted in absolutely nothing. And I mean NOTHING. The mobo has Standby Power (LED is lit), but apparently something else is not functioning on it. I know its getting the power from the PSU, because as I mentioned, I checked every voltage coming out of it, and it does self-test fine as long as I don't have the ATX cable connected with pin 20 (3.3V Sense) in place...as a side note, its not really pin 20 on the PSU itself. Its pin 6 on the 10-pin connector...it would be pin 20 on a 20+4 connector. Thanks for the odd proprietary connector, Corsair.

 

Anyway, I'm waiting to hear back from Corsair about what voltage, if any, is supposed to be on the "3.3V Sense" rail of an AX860i, and if they get the same results from the self-test button with cables attached, before I RMA anything back for replacement. But getting responses from them is literally like pulling teeth. At this point I don't expect to hear anything back until Tuesday or Wednesday. Ugh.