Jump to content
Corsair Community

H110i v2 - Can a broken PWM Cable stop PC from Posting?


B77WGE115B

Recommended Posts

My computer has not been posting for the past few days. All of the lights come on and the fans spin up, but then, a few seconds later, everything restarts.

 

I opened up my PC and noticed that one of the cables on the splitter attached to the H100i v2 is broken (not the fan cable, but one of the cables that plugs into the unit itself).

 

I was wondering if there is some login within the unit that tells the computer to automatically shut down if there is no current between one of the PWM cables as a safe measure.

 

As a test, I stripped back some of the wire and tried my best to tie the two ends of the wire together and put some electrical tape over it. I still had the same problem, but I'm not the best handyman, and I'm not sure if the connection is good (I don't have a multimeter).

 

Additionally, I noticed that when I try to boot, the two fans connected to the unit spin up for a second and then stop (while the case fans continue spinning). Is this normal behaviour pre-post since not too much cooling is required at that point.

 

Also, if the problem lies elsewhere (motherboard, ram etc) would I be able to use a PWM splitter and put both fans into one of the inputs on the unit, knowing that I would only be able to control the fans as one unit?

Link to comment
Share on other sites

Does you computer constantly restart? From what you have described it sounds like it could be a PSU issue. When the computer turns on does the pump appear to be working?

 

Yes, it constantly restarts. I thought it was a PSU issue too, but I tried connecting an power supply from older PC (Corsair CX500) to my machine and the same thing happens. When I put the CX500 back into the old build it works fine. Unfortunately, the old build has an older motherboard spec and DDR3 ram, so I can't use it as a test bench.

 

All of the lights on the waterblock come on. How can I tell if the pump appears to work? I felt the radiator and waterblock and none of them got hot.

 

It might be helpful to add that I started having this issue when I tried to apply 1.4V for an overclock and enabled the motherboard's XMP Profile. I've also tried resetting CMOS and resetting the CMOS battery.

Link to comment
Share on other sites

Take off the XMP profile and try again. If there were some kind of short, you would probably get a Over Current Protection error from the board. If the H100i v2 pump was not getting its power, you would get the classic CPU Boot error. This kind of smacks of memory setting issues and 1.40 Vcore and whatever frequency XMP invokes may be a step too far - at least at those specific settings.
Link to comment
Share on other sites

Take off the XMP profile and try again. If there were some kind of short, you would probably get a Over Current Protection error from the board. If the H100i v2 pump was not getting its power, you would get the classic CPU Boot error. This kind of smacks of memory setting issues and 1.40 Vcore and whatever frequency XMP invokes may be a step too far - at least at those specific settings.

 

 

I would love to be able to do that, but I cannot even get into the BIOS. I try to boot and nothing comes up on screen, and then the computer restarts itself and the process repeats.. Even if I press the delete key while booting up, I can't get into the bios.

Link to comment
Share on other sites

Are you sure the clear CMOS is "taking?" Normally, I would also suggest unplugging the PSU from the wall for a while, but that should have triggered a reset on the PSU change, both with the new PSU and putting the old back in.

 

Do you get any kind of error code on the motherboard itself?

Link to comment
Share on other sites

Are you getting any beep codes on boot? Any kind of error indication? Check your motherboard manual for any troubleshooting codes that the motherboard may provide.

 

You may also want to clear your CMOS. That's a jumper on the motherboard; again, check the manual to confirm the process.

Link to comment
Share on other sites

Are you sure the clear CMOS is "taking?" Normally, I would also suggest unplugging the PSU from the wall for a while, but that should have triggered a reset on the PSU change, both with the new PSU and putting the old back in.

 

Do you get any kind of error code on the motherboard itself?

 

Are you getting any beep codes on boot? Any kind of error indication? Check your motherboard manual for any troubleshooting codes that the motherboard may provide.

 

You may also want to clear your CMOS. That's a jumper on the motherboard; again, check the manual to confirm the process.

 

 

I don't have a motherboard speaker, so I am not hearing any beep codes. After looking through my motherboard manual (Gigabyte Z170X UD3 Ultra), it seems like it does not have any error indication lights, and since nothing comes up on screen, I'm not seeing anything there either. I tried clearing the CMOS both using the onboard pins (shorting them with a screwdriver with power off and PSU unplugged as stated in the manual) and by taking out the CMOS battery. In addition, I left the unit unplugged with the CMOS battery out overnight on Saturday (at least 10 hours).

Link to comment
Share on other sites

I find it difficult to accept that this is the cooler - as c-attack mentioned earlier, this isn't they typical behavior that we'd see from a failed cooler. But you do have the cooler plugged in to the fan header, yes?

 

Honestly, it sounds like a short or a component failure. So try booting with a single stick of RAM (and try different sticks if you have them). And see if you can grab a motherboard speaker from an old build to see if there are beep codes. That'll narrow down issues pretty quickly.

Link to comment
Share on other sites

I find it difficult to accept that this is the cooler - as c-attack mentioned earlier, this isn't they typical behavior that we'd see from a failed cooler. But you do have the cooler plugged in to the fan header, yes?

 

Honestly, it sounds like a short or a component failure. So try booting with a single stick of RAM (and try different sticks if you have them). And see if you can grab a motherboard speaker from an old build to see if there are beep codes. That'll narrow down issues pretty quickly.

 

Good to know that its not the cooler. Yes, I do have it plugged into the cpu fan header. I tried booting with a single stick of ram as well, I tried both of my sticks individually in each slot (i.e. stick one only in slot 1, 2, 3, 4 and the same with stick 2). I don't know about a short since this problem started before I even opened my pc (its was working fine for 2 years before), but its possible. If it isn't a short, do you think it would likely be a motherboard or ram failure, or even possibly both?

Link to comment
Share on other sites

So back to the original query... where exactly is the break in the wire? It is on the fan splitter coming off the H100i v2? If so, this is a fairly easy premise to test. You don't need fans to boot up or even run the system at low level. Disconnect the fan splitter and make sure it's not making conductive contact with anything. Try again.

 

If it is the power line coming from the motherboard header to the pump, that is a little trickier but can be done. You can take a fan off a radiator fan and connect it to CPU fan to get by the boot error protection. Leave the pump unpowered. You can go that way for 60-90 seconds -- enough time to get past the current sticking point. Not sure how you would feel about loading up without moving coolant, but this is essentially what happens when a pump fails on boot. Nearly everyone survives unscathed.

 

I do think we are looking for something else, but you have already covered most of the first recommendations anyone would make. If we eliminate the cooler, PSU, and RAM, that leaves two things you hope are not responsible... CPU and motherboard.

Link to comment
Share on other sites

So back to the original query... where exactly is the break in the wire? It is on the fan splitter coming off the H100i v2? If so, this is a fairly easy premise to test. You don't need fans to boot up or even run the system at low level. Disconnect the fan splitter and make sure it's not making conductive contact with anything. Try again.

 

If it is the power line coming from the motherboard header to the pump, that is a little trickier but can be done. You can take a fan off a radiator fan and connect it to CPU fan to get by the boot error protection. Leave the pump unpowered. You can go that way for 60-90 seconds -- enough time to get past the current sticking point. Not sure how you would feel about loading up without moving coolant, but this is essentially what happens when a pump fails on boot. Nearly everyone survives unscathed.

 

I do think we are looking for something else, but you have already covered most of the first recommendations anyone would make. If we eliminate the cooler, PSU, and RAM, that leaves two things you hope are not responsible... CPU and motherboard.

 

The break is between the cooler and the fan splitter. Tried making sure it the wires were isolated and not conducting against anything. No luck. I ordered a motherboard speaker, hopefully the beeps can tell me more.

Link to comment
Share on other sites

I find it difficult to accept that this is the cooler - as c-attack mentioned earlier, this isn't they typical behavior that we'd see from a failed cooler. But you do have the cooler plugged in to the fan header, yes?

 

Honestly, it sounds like a short or a component failure. So try booting with a single stick of RAM (and try different sticks if you have them). And see if you can grab a motherboard speaker from an old build to see if there are beep codes. That'll narrow down issues pretty quickly.

 

I got my hands on some new DDR4 RAM that I could use in my build. If I get a beep code that indicates a possible memory error, would it be safe to insert the new ram as a test? I'm slightly paranoid that my (multiple) CMOS resets didn't "take" and if its a ram problem, then inserting the new ram could destroy it as well.

Link to comment
Share on other sites

It seems unlikely your motherboard has turned into a DRAM eating monster, pushing heavy current through the sockets. However, I am worried it would not boot up with one stick of the old. Nevertheless, this seems like a settings issue and either way you still need current to do actual damage. That should not be possible, XMP or not, with any of your settings.

 

You were pretty thorough in your attempts to reset the board. Suddenly I am more appreciative of my simple push button reset.

Link to comment
Share on other sites

It's certainly possible. It's tough to say without those beep codes.

 

Just tested my system with my a motherboard speaker. I hear five beeps. My motherboard manual says that my system as an AMI UEFI Bios. According to this website https://www.lifewire.com/amibios-beep-codes-2624543

that means that there is a processor error. The computer shut down right after the fifth beep.

 

Does this mean that something went wrong with my CPU? Did putting 1.4V through it cause a big problem? Is this enough information to rule out the RAM? I did make sure that the RAM was seated properly.

 

I haven't tried the new ram yet. I'd prefer to keep it in the box so that I can return it with amazon if its not necessary.

Link to comment
Share on other sites

I think you can leave the new RAM in the box, especially at the current prices. I would try and find a second confirming source for the beep code, preferably a Gigabyte one. While probably the same, we are getting into really difficult testing part. Checking your CPU or motherboard is a pretty expensive obstacle. I suppose the obvious next step would be to reseat the processor, if you have not done so already. Perhaps Dev has some more ideas.
Link to comment
Share on other sites

I think you can leave the new RAM in the box, especially at the current prices. I would try and find a second confirming source for the beep code, preferably a Gigabyte one. While probably the same, we are getting into really difficult testing part. Checking your CPU or motherboard is a pretty expensive obstacle. I suppose the obvious next step would be to reseat the processor, if you have not done so already. Perhaps Dev has some more ideas.

 

According to this link from Gigabyte, 5 beeps on an AMI bios means "CPU Error" https://www.gigabyte.com/Support/FAQ/816

 

I resat the CPU and had no luck. In the process, I snapped some pictures. Does anything look wrong? I didn't see any burn marks or anything obvious, but maybe I'm missing something.

 

Imugr link in case pictures don't work: https://imgur.com/a/GuhgJIB

 

VLQp57z

 

X5tGPgS

 

And so it seems that we know its not PSU, RAM or expansion cards (due to beep codes) so if its not the CPU that only leaves the motherboard? Could updating the BIOS possibly help? Can I even flash the updated bios if the computer does not post?

Link to comment
Share on other sites

Oh man. That sucks.

 

At this point, it's either the CPU or the motherboard. The best way to test is either a known-good CPU or a known-good motherboard. I doubt you have either laying around. BUT ... all you would need to check the motherboard would be a ~$40 Skylake Celeron. If that posts fine, then you know that your CPU is shot. If it doesn't, then it's likely the motherboard.

 

1.4V, especially for a short period, really shouldn't kill your CPU. Neither of the pictures really showed anything as an issue that I could see but ... if this started and you hadn't remounted the CPU, then the CPU is probably the issue - it's mounting the Intel CPUs that can damage motherboards; those pins are super-fine and easy to mess up.

Link to comment
Share on other sites

BUT ... all you would need to check the motherboard would be a ~$40 Skylake Celeron

 

Seems like they are kind of hard to get a hold of these days. I've called every computer shop in the area and no one seems to have them available. Getting a Kaby Lake Celeron might be easier, and my motherboard supports Kaby Lake, but I'm not sure

if my bios is up to date enough to support them.

 

Oh man. That sucks.

1.4V, especially for a short period, really shouldn't kill your CPU. Neither of the pictures really showed anything as an issue that I could see but ... if this started and you hadn't remounted the CPU, then the CPU is probably the issue - it's mounting the Intel CPUs that can damage motherboards; those pins are super-fine and easy to mess up.

 

 

I didn't remount the CPU at all. As soon as I hit "save and exit" in the bios and rebooted, I had the issue. I'm still trying to figure out how all of this happened as well, I really don't understand how 1.4V and enabling XMP fried either the motherboard or CPU.

 

I'll try breadboarding my build when I get off work. I was saving up for a GTX 1080Ti, but I ended up using the cash to get a new 8700k and motherboard. Hopefully the problem lies only with the motherboard/cpu and the new components will sort everything out.

 

I'm assuming that my testing has ruled out PSU, graphics card, RAM and cooling so the issue must be the motherboard and cpu, and if I were to replace both, the problem should be solved? Any other tests I should be running to figure out if other components are faulty as well? I would hate to lose my ability to exercise Amazon's excellent return policy for the new parts only to figure out it was a graphics card problem all along.

Link to comment
Share on other sites

Seems like it is narrowed down. Tough to recommend "go replace your CPU and board" as a troubleshooting strategy, but the alternatives probably involve shipping the those two components off to some other location for an extended period of time. My current motherboard brand infers they will charge you for service if they don't find the answer, even when their tech support recommends you send it in. Read yours carefully if you go that path.

 

I leaning toward the motherboard, but mostly based on the idea that 1.40v on boot is not enough to damage a CPU and if the CPU was somehow damaged this way, then the motherboard improperly passed on a level of current beyond what it should have. Also, while I can't speak for every board, the "bad CPU" usually kicks back a code but not necessarily a shutdown. This power on, power off stuff seems like a power issue, which is why this started on the PSU. The motherboard is the next step in the chain.

 

Dev's recommendation seems like the least expensive option, certainly more so than a new board or even mailing both components off.

Link to comment
Share on other sites

I leaning toward the motherboard, but mostly based on the idea that 1.40v on boot is not enough to damage a CPU and if the CPU was somehow damaged this way, then the motherboard improperly passed on a level of current beyond what it should have.

 

Maybe my fingers slipped and I accidentally set 4.1V? At this point, I'm not sure. I'll call around local technicians and see if any of them have a skylake processor that they are willing to let me try in my build.

Link to comment
Share on other sites

Oh man. That sucks.

 

At this point, it's either the CPU or the motherboard. The best way to test is either a known-good CPU or a known-good motherboard. I doubt you have either laying around. BUT ... all you would need to check the motherboard would be a ~$40 Skylake Celeron. If that posts fine, then you know that your CPU is shot. If it doesn't, then it's likely the motherboard.

 

1.4V, especially for a short period, really shouldn't kill your CPU. Neither of the pictures really showed anything as an issue that I could see but ... if this started and you hadn't remounted the CPU, then the CPU is probably the issue - it's mounting the Intel CPUs that can damage motherboards; those pins are super-fine and easy to mess up.

 

So I was looking some things up on my train ride home, and realized that many people had issues with 5 LONG beeps on Gigabyte boards. Since gigabyte only publish error codes with short beeps, I was quick to assume that my beeps were "short". However, upon closer inspection, they are quite long. Similar to this video. [ame]

[/ame]

 

Looking online, it seems like a lot of people believe that 5 LONG beeps on a gigabyte board is a RAM problem (even though gigabyte doesn't publish this themselves). Maybe enabling the XMP profile caused a problem with the RAM and is causing the boot loop. In fact, if I remove the RAM from the motherboard completely, I get the same 5 LONG beeps!

 

Do you guys think that applying the XMP profile could have done something to my RAM? Is it possible that my system is currently "bricked" and stuck in an XMP mode, even though I reset the CMOS (multiple times). If I tried my new RAM (Corsair Vengeance LPX rated at 3000MHz), is there a good chance that it would work, or would I need RAM rated at 2133MHz for any chance of getting my computer to boot?

Link to comment
Share on other sites

Archived

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

×
×
  • Create New...