AX1200i - Intermittent hard shut downs, protection tripped

To keep it short and sweet, my system will inexplicably shut down after varying power-on hours. Sometimes it lasts 3-5 days. Rarely, it'll last less than 1. Each time I have to flip the power supply main off, wait a few seconds, and then flip it back on for the case power button to work and the PC to boot. From what I understand from reading around the net, this is indicative of some sort of current or voltage protection being tripped in the PSU (AX1200i).

I've used several different monitoring programs to see if there are fluxuations in any of the rails. Unfortunately, I haven't had a single one give me what I'd consider ACTUAL readings. Most tell me the 12v and 5v are 12.000v and 5.000v (which I personally believe to be BS, as I've never heard of a rail constantly hitting the EXACT voltage down to 3 decimals). The CorsairLINK2 software tells me the 12v is 12.03-12.06v and bounces are on that. 5v also reads similarly. 3.3v is another story. It seems to be read properly by several programs, at 3.312v. Asus Suite II is pretty unreliable for logging events on pretty much anything, because it's been ridiculous since day one. About twice an hour, SOMETHING will jump to some ridiculous reading (mobo temp -63*C for instance) and then jump back to normal. However, I'll note that all three rails jump low and high in this log (Probe II). A typical example below:

[05/24/2013 at 02:45 am] +5V 0.000 Abnormal

[05/24/2013 at 02:45 am] +5V 5.000 Normal

[05/24/2013 at 03:51 am] +12V 0.288 Abnormal

[05/24/2013 at 03:51 am] +12V 12.000 Normal

[05/24/2013 at 04:31 am] +3.3V 0.032 Abnormal

[05/24/2013 at 04:31 am] +3.3V 3.312 Normal

[05/24/2013 at 04:33 am] +5V 0.120 Abnormal

[05/24/2013 at 04:33 am] +5V 5.000 Normal

[05/24/2013 at 05:07 am] +5V 0.120 Abnormal

[05/24/2013 at 05:07 am] +5V 5.000 Normal

[05/24/2013 at 05:45 am] +12V 24.480 Abnormal

[05/24/2013 at 05:45 am] +12V 12.000 Normal

[05/24/2013 at 06:33 am] +5V 0.000 Abnormal

[05/24/2013 at 06:33 am] +5V 5.000 Normal

[05/24/2013 at 07:05 am] +5V 10.200 Abnormal

[05/24/2013 at 07:05 am] +5V 5.000 Normal

[05/24/2013 at 07:09 am] MotherBoard -128.0 Abnormal

[05/24/2013 at 07:09 am] MotherBoard 29.0 Normal

[05/24/2013 at 07:18 am] +5V 0.000 Abnormal

[05/24/2013 at 07:18 am] +5V 5.000 Normal

[05/24/2013 at 07:19 am] +3.3V 2.224 Abnormal

[05/24/2013 at 07:19 am] +3.3V 3.312 Normal

I'd also like to note that this log doesn't seem to log any abnormal events close to the times it's powered off. It almost always does it when I'm not home / the PC isn't being actively used.

Any troubleshooting anyone can provide would be excellent. If you have any questions, fire away. Thanks!

I've used several different monitoring programs to see if there are fluxuations in any of the rails. Unfortunately, I haven't had a single one give me what I'd consider ACTUAL readings.

If you have access to a multimeter, i would pull the PSu out and start it using the paperclip method found in the sticks on the main PSU forum page. You can get actual readings from your molex,sata and PCI cables.

Do you by any chance have someone else or another system you could test the PSU with?

I have a multimeter (father has been in the auto repair industry his entire life) and someone to show me how to use it properly. I have 2 other desktops. One is my old one I gave to my brother and has problems of its own I have to sort out. The other is an A10 APU system I just built for my parents with it's own inexplicable problems. So I can't reliably use either to test.

So I jump the PSU, what pins in the molex (maybe not this one, kinda obvious :P), sata, and PCI cables am I touching power and ground to? Is there a trustworthy diagram you know is correct you can link or direct me to? And what rails are each of those running off of? I've never had to troubleshoot power before so this is all new to me.

For the record I've done the typical testing one does when they build a new system. RAM is good, CPU seems good, and I have no GPU cards installed. The motherboard has been a pain in my rear since day one, but I'd like to rule out the PSU first due to the protection circuit being tripped.

Sure, here is a good link that has all your connectors wih pin locations and present voltages.

http://www.playtool.com/pages/psuconnectors/connectors.html#eps8

This shows your 24pin and 4 pin molex pins.

http://en.wikipedia.org/wiki/Molex_connector

And SATA pinouts.

http://www.allpinouts.org/index.php/Serial_ATA_%28SATA,_Serial_Advanced_Technology_Attachment%29

Thanks for the pinouts! I figured I'd try a few things while the cables were still neatly run in my case.

I made sure I didn't have any missing mobo connections. This led me to notice a few things I never had before on previous motherboards. There is an 8pin AND a 4pin CPU plug on the Asus Maximus V Forumula. My research into the purpose for this turned up nothing important. They are there for compatibility reasons and only one needs to be used. I've been using the 8pin. Some people believe you could plug in both for use when doing some extreme CPU overclocking to pull some extra power off the 12v rail. There was also a molex connection on the bottom of the motherboard labeled "EZ Plug". This is also apparently only needed for overclocking PCI-E cards. I don't plan on overclocking anything, and I don't have anything in the PCI-E slots yet. Nevertheless, I plugged in both of these "extra" power connectors. I also double checked to make sure all plugs were seated correctly by giving each a good push.

The comp has been on for over 6 days without incident. I'm really hoping something wasn't seated right. At stock clocks with no add-on cards, these extra connectors should not be needed. I plan on shutting down the comp in a few more days (to be absolutely sure the problem has disappeared) and removing them to see if it was a seating problem.

Would a slightly loose connector cause the circuit protection to trip? Or is this just false hope?

If the problem returns, I'll pull the PSU and test the rails to make sure it's not just weak and the extra connectors were providing a boost.

I'll also note that the 5v rail now alternates 5.000v and 5.040v readings. Thanks for all the help thus far, I'll keep you posted on my findings.

Would a slightly loose connector cause the circuit protection to trip? Or is this just false hope?

And you are correct about the extra power connectors on your bopard. They should not be needed for stability unless you extreme overclocking. However, if you find that your hard restarts return after removing them , then you should get the board replaced and not the PSU. The single 8pin EPS plug should be all you need.

if you find that your hard restarts return after removing them , then you should get the board replaced and not the PSU.

Absolutely. Thats the plan. The whole point was to narrow it down between the mobo or PSU. While the board is out for replacement though I might as well make sure the PSU is kosher as well. I'll let you know what happens.

Good to know. AI Suite II has quite a few issues, and is known not to play well with other diagnostic software (i.e. AIDA64, which I'm also using / use primarily). It is also known to get stuck at 40*C on the motherboard and CPU temperature sensors, despite the actual temp being completely different. I essentially use it for the Fan Xpert II, which also has it's own issues. I have a fan that drops in RPM, and sometimes when I adjust the speed to ramp it back up, it'll spin up for a few seconds and then drop to 0. I believe this to be, in part, due to weak 200mm+ fans that spin at 700rpm max. I disable Q-Fan in BIOS and let the software handle fans. Q-Fan doesn't allow as much control, and the fans seem a lot more stable this way.

If I pull each of the extra power connectors out and it's still busted, I'll disable AI Suite and AIDA64 and see what happens. Thanks for the help guys! I'll post what happens for googlers. I hate when people come back and say "I got it. Thanks!" and never post a solution.

So two things I would check the fist of which is if this is an ASUS MB and you are running AI-Suite turn it off and see if the issues go away. If so I think you know what you need to do..

If you have more than one application running to check the system health turn them off and only run one application at one time and I would strongly suggest only using the application provided by the MB maker.

Ummm, if the MB maker is ASUS, then that wouldn't exactly apply... would it?

Hello,

From what I understand from reading around the net, this is indicative of some sort of current or voltage protection being tripped in the PSU (AX1200i).

Yes so your best point of call would be to take everything out of the case and power the system outside, assuming you have not done so already. Place the motherboard on a flat non-conductive surface (Motherboard box inside out).

Connect just the essential items and see how what happens.

• 1 hard drive with OS
  
• Memory module x1
  
• CD DVD drive x1
  
• ATX connector
  
• Only the 8pin +12v CPU connector. (You are correct, leave all the optional extra connectors out)
  
• Press your start button to power the board.

What this may rule out

• Case causing a short
  
• Motherboard spacers causing a short
  
• USB headers causing a short (faulty USB sockets on case)
  
• Firewire headers causing a short
  
• Front panel Audio connectors causing a short
  
• Accidental placement of headers, EG: Firewire connected to USB

If your problems still occur under these conditions then perhaps consider testing with another power supply.

Unplugged the EZ Plug first and it was on for 5.5 days without incident. Had some power trouble in my area involving a transformer and some power dips cut the computer off at that point. Plugged it back into my UPS to prevent this (had it in a regular surge protector to rule that out). While it was off I unplugged the 4pin CPU and within 8 hours it tripped.

It has currently been on 3 days with AI Suite disabled (after a lot of reading I've come to believe this prog and some of its components to be poorly written) and AIDA64 disabled. No problems yet.

Found this. The second page gets into the meat of the issue. RAM GUY may be absolutely right about a software conflict being the issue. I honestly did not think software (that isn't for changing voltages) could cause something like this. I stand corrected and educated. I've learned to hate computers over the last few years haha.

If it lasts the week this way I'll be uninstalling AI Suite (which is a pain in the butt apparently), using the bios for fan control, and AIDA64 for diagnostics.

This is great news considering I don't have the Mobo box anymore, the screws and whatnots have been combined with others I had, and I don't know if that's ok with Asus for RMA. I plan to watercool my gtx 780's so the comp is coming apart when I get all of that together. I'll test the PSU then, as well as Agent-Sharp's suggestion, just in case. Don't want to fry $1400 in graphics cards.

Again, I really appreciate the help you guys have been giving me.

Kinda sucks, I could have waited for a Haswell cpu and mobo.

hmm, hi

the higher than "normal" voltage readings on your 1200i are completely normal and absolutely true. if mine ever dropped below 12,5 and 3.3 more than a 3 OR 4 point decimal, i would call for an RMA or never buy a corsair product again

i would put a corsair psu under 100% load against any other manufacturer and put money on it, ive been oc'ing since PIII and would trust no other power supply

the 8pin on the mobo is for cpu stability under OC( 1.5v+), the 4 pin is for extra power to gpus under OC, if this wasnt there, under the right circumstances, a high OC would draw from your mobo (usually only rated at 3 amps) and fry your mobo

AI suite is a piece of junk, if CL2 says something then it is right

dont ever attempt to OC using ai suite as your "monitor" is is a pos

please dont have multiple monitoring software, this will lead to heartache as each one tries to override/counter/double the efforts of another (doubling efforts are really really bad :P )

please post the mobo probe-it point measurements with a multimeter if you can and start the computer with just the psu and mobo.

AI suite is a piece of junk, if CL2 says something then it is right

dont ever attempt to OC using ai suite as your "monitor" is is a pos

 

please dont have multiple monitoring software, this will lead to heartache as each one tries to override/counter/double the efforts of another (doubling efforts are really really bad :P )

 

please post the mobo probe-it point measurements with a multimeter if you can and start the computer with just the psu and mobo.

I'm not a fan of CorsairLink2 either. It may provide more accurate and correct voltage readings, but I've read just as much negative feedback on it as Asus Suite II. It doesn't help me with anything else other than voltage readings so I don't even use it.

I don't OC. I'm just not into it. I never have, I probably never will. I like a system to run efficiently. That means very cool at factory clocks. I just buy better hardware so stock clocks are higher ;):

I've learned my lesson on monitoring software. However, I disabled everything except Fan Xpert II and it still shuts off. The forum thread I posted above comes to the conclusion that its the Asus HM Com service after Fan Xpert II changes it, since after you uninstall AI Suite, the issue persists, and if you remove this service and reinstall AI Suite WITHOUT Fan Xpert II, the problem is gone.

I don't know what you mean by "probe-it point measurements" or exactly what I'd need plugged in to power both the psu and mobo on without damaging anything. That is, just the 24pin? Do I have the CPU in/powered?

I don't know what you mean by "probe-it point measurements" or exactly what I'd need plugged in to power both the psu and mobo on without damaging anything. That is, just the 24pin? Do I have the CPU in/powered?

Your MB has points that you can use the probes of a multi-meter to get accurate readings straight from the MB . You don't really need to be concerned about it.

But to power the system outside the case you would need both the 24pin and 8pin EPS cable attatched to the board. It will not start with out it.

This is great news considering I don't have the Mobo box anymore, the screws and whatnots have been combined with others I had, and I don't know if that's ok with Asus for RMA.

I have been through ASUS RMA several times. You will need the original box and the plastic cover that is used to cover the CPU socket. They will reject it if it is missing .

http://farm6.static.flickr.com/5089/5305869499_c114ef3735.jpg

located next to your 24 pin header

these are the true readings of what is going to the mobo at an electrical level

and are used by OC'ers as the most accurate real time info you can get

this will decide right away if it is the psu or mobo

Woah nice. I never noticed these before. I've got the inner black box it came in, and the plastic CPU socket cover. Just not the main box that it all goes in. Oh well. I don't think my mobo is borked though. Seems to be alright with the software disabled. But we'll see for sure when I pull it apart for the watercooling setup.

Unfortunately due to AI Suite II being a suckfest, I can't seem to uninstall it. Let alone REinstall the components I DID manage to get rid of. It's currently in a broken state. Looks like I'll have to reinstall windows.. and redownload 100GB of games (on 3Mbit DSL no less) I pre-installed in preparation for the beast graphics setup. Bummer.

I'll still chime back in when I can definitively point a finger at the software as the culprit. I've learned to never make assumptions based on what something seems like with computers.

Edit 7-16-13:

Reinstalled Windows 7 x64. It's been on completely glitch free for 10 days. AIDA64 now sees all of the sensors it didn't see before. The fans don't fluctuate or drop out. Mobo and CPU temps don't get stuck at 40C. It's been running with the same AIDA64 sensorpanel setup I had before, plus a few extra sensors it wasn't picking up with AI Suite II installed. I'd venture to say that AIDA64 is more stable than AI Suite II, and a conflict between these programs was causing low level electrical issues as outlined in the forum posts I mentioned above. Jumped the PSU and all is as it should be as well! Thank you for all of your help guys.