cancel
Showing results for 
Search instead for 
Did you mean: 

CH V Formula-Z CPU Fan Speed totally erradic...

xenonexia
Level 7
Hi all!

Sort of going crazy over here. Seems my totally new mobo goes totally heywire while trying to check for values on CPU Speed, for no reason at all.

Got an old, comfy 1100T CPU (at least for the moment) beneath a CM Extreme Freezer fan that has worked perfectly on other boards. But for some reason, when I connect the fan to its designated header (I've even tried optional header to rule out problems at preferred one), the fan speeds meassured in both BIOS/UEFI and in Win/FanExpert goes totally bonkers!

One second it reads 200rpm. Next, it reads 0. One second after that it says 28000. Goes down to 1800 or such. And keeps looping a totally weird, totally random and hysteric rpm curve. The graphs look like the NYC skyline, for heaven's sake...

It won't give any correct readings. I can not control it at all. Whatever I do, the system seems to love sending me warnings every other second that the rpm is at 0. And I have had a fair few botched system starts because of this.

I have now turned readings off, and it is a bit quieter. System seems stable while in Windows, but totally freezes in BIOS/UEFI after a few seconds. But I cannot control the fans at all. And I cannot get any readings that are even remotely correct while having readings activated.

For some reason, even though I run chassis fans on a Scythe controller, off the mobo, I get the same weird patterns in readings from fan speed headers that aren't even connected.

Anyone have any clue as to what one is supposed to do? Am a bit terrified about running anything that will need a bit more from the CPU or anything else at the moment. I'm not sure if the CPU fan or any other fan will react to any signs that the system is actually under extra pressure. No BF4 for me. Big win buying an advanced gamer keyboard when I won't be able to use it for what it's built for.
5,611 Views
5 REPLIES 5

MeanMachine
Level 13
Hi xenonexia and welcome to ROG
First check you have correctly installed all fan headers satisfactorily according to your installation manual,
There are several headers on the MB so try different ones. First test and set Bios fan monitor and speed controls without Scythe controller. Test and set controls for each fan individually.
Your main consideration is what shows in the Bios from the Monitor Menu. Don't worry about Windows readings atm.
You can add other headers as you go till you find the culprit.
Could also be an incorrectly seated or poor thermally pasted CPU.
If you re-seat your CPU, check for bent pins using a magnifier.
Also make sure both 4 and 8 pin MB sockets are occupied.
We owe our existence to the scum of the earth, Cyanobacteria

My System Specs:

MB:ASUS ROG Crosshair VII Hero/WiFi GPU:EVGA GTX 1080 sc PSU:Corsair AX-1200i
CPU:
AMD R7 2700X Cooler: Corsair Hydro H115i Case: Corsair Carbide 780t

Memory:G.Skill TridentZ F4-3200C14D-16GTZR SSD:Samsung 500GB 960 EVO M.2


[/HR]

xenonexia
Level 7
Thx, MeanMachine! 🙂

So... I'll just be going through the list, step by step and look at what I've done and what I can do that I haven't. But I've already gone through most of it a few times now, sadly without the results I was hoping for. :mad:

I have built more computers than I can remember, for at least two full decades, and have been developing, servicing and done mech repairs on everything from computers to high end mobile phones as well, so I just hate being thrown off the horse by something like this. I kinda feel that I should be able to see what is going on here. But I don't. 😉 :confused:

Connector is right where it should be. Might be a lot of headers onboard, but this is not my first build. I pretty much can manage to find a cpu fan header in my sleep. There are two cpu 4-pin connectors on the board. I've tried both, checked the headers for damages, et c. It is all as it should be.

BIOS crashes instantly or within half a minute when entering it. All I can see in BIOS/UEFI is those hysterically crazy numbers going totally heywire in all directions with just milliseconds between 0-28000 rpm and everything inbetween. The sensor does not pick up anything, and is just randomly showering me with numbers. Therefore, there isn't even anything to control. The board doesn't know what to control or look for in the first place. 😕
The random, looping, werid numbers from the fan connector is the same in BIOS as it is in Windows. No difference. Except windows is way more stable than BIOS is. It only freezes up totally once a day, while BIOS crashes as soon as I look at it, basically. And I don't like that. It's something I expect from an old Asrock pile of rubble, but not a Crosshair board. Know what I mean?

The Scythe controller is a mechanical switch with its own physical temp readers on its own cables and also 6-channel fan controller with its own wiring that doesn't ever come into contact with the motherboard connectors. That way, I usually control all chassis fans via physical controls at front panel and I can check manual readings for the entire case, at 6 selected places around the chassis. All, as I said, totally manual from front panel, with no connection to the motherboard, except the main 12V molex power connector. And, as the PSU is totally new, stable as anything and one of the most raved about PSU's on the market (EVGA SuperNova G2), plus the rest of the 12V line of molex connectors work just fine, I don't think there is any problems with that either. Which is a bummer, since it would have been easily fixed, obviously.

Since there is only one motherboard fan header connected at all, which is the cpu fan standard connector at top of board, and it gives me the same random numbers and non-function as the optional cpu fan header, also after replugging it all, it seems there aren't that many culprits to look out for. Choices are pretty slim. But since I get the same random numbers from headers that aren't connected, it looks to be either hardware malfunction or a very large problem within BIOS. I am not sure which, at the moment.

The cpu fan connector is not connected to the Scythe controller, obviously, which is, as I said, physical - not software. And, since I use the Scythe controller for all chassis fans (six of them, carefully placed from front to top to back to produce a standard design air flowthrough, and with clear paths throughout the chassis via extensive cable managing), there are no other motherboard fan headers occupied. Still, some of the - very unoccupied - headers on the motherboard give me the same very weird random numbers that go dead one millisecond, shows anything between zero and heaven above the next - and keeps on going... Even though they aren't connected. A few of the others don't. They just give me the usually expected N/A. So... with me, I sense that there is something really fishy going on in motherboard suburbia.

There are no problems, as far as I can see, with the physical Scythe controller, though. It works just fine. Does its thing. But that doesn't help the CPU fan or its sensors and controllers onboard the motherboard. Controllers work just like they were built for. The temp readers are showing parts of the chassis being around 20-28C and the highest value being below 30C if no load on cpu and just above 40C while under load.

The pins on the processor are totally okay. And I've been meticulous with the thermal paste used. Could be that it still isn't good enough, but I have never yet had any difficulty with adding a perfectly smooth layer of high-end paste on those darlings before. Which makes me kind of wonder.

As I installed the processor onto this new mobo, I was very careful while doing it, but I did replace the thermal paste and so on, since the processor had already been used on a different mobo for a good while previous to this install. I am kind of wondering if this is partly because the new paste hasn't been "burnt in" just yet. Which would be probable, though. Oh, and while installing my entire games listing (a fair amount of games) on several paralell channels, the hard drives are at about 45C after a few hours of constant writing writing writing. Which is pretty good for being under pretty massive load, after all. So temps shouldn't be a factor either. Except that the cpu fan header and some of the cpu fan headers seem to be living in another dimension, or a few, entirely. 😞 Bummer... They aren't even living in Andromeda anymore. ^^

Both 8-pin and 4-pin connectors, plus the 4-pin for extra power to GPU are all definitely seated correctly, too. And the PSU should be way beyond what I need for this hardware configuration, as well. So powering the motherboard in itself shouldn't be a problem. If there isn't a problem within the motherboard itself. Or problems with drivers/BIOS, compatibility with the cooler itself, or such.

I think I've gone through all steps, but it still isn't any of these things, as far as I can see. :confused: 😞

I could reinstall latest BIOS, again, of course. But I ain't sure what the reason is behind it not working its magic to begin with. It seems like those random numbers are trying to tell me that the sensors just ain't able to pick up any life at all, and that they are totally confused about it. Which seems weird, but that's the best I can make out from this.

Sorry for the WOT, TL;DR thing. Just trying to figure things out, and mapping my own comings and goings while at it.

Really thankful for any input on this.

MeanMachine
Level 13
xenonexia

As you know its a matter of elimination and as you have covered all aspects as suggested, it may be time to RMA the MB or try a different CPU.
i dont think a Bios update will fix it, but you can try.
I have the same board as you and have not experienced this kind of problem.
Maybe other ROGers will offer other sudgestions who have had this kind of issue.
We owe our existence to the scum of the earth, Cyanobacteria

My System Specs:

MB:ASUS ROG Crosshair VII Hero/WiFi GPU:EVGA GTX 1080 sc PSU:Corsair AX-1200i
CPU:
AMD R7 2700X Cooler: Corsair Hydro H115i Case: Corsair Carbide 780t

Memory:G.Skill TridentZ F4-3200C14D-16GTZR SSD:Samsung 500GB 960 EVO M.2


[/HR]

HiVizMan
Level 40
I would strongly suggest doing a fresh BIOS flash using either the USB flashback method or EZFlash in the tools section. A corrupted bios will certainly exhibit some of the stuff you have listed.

Plus what have you go to to loose. If that does not work and you have checked for bent pins in the socket RMA time.
To help us help you - please provide as much information about your system and the problem as possible.

Dr__Zchivago
Level 12
HiVizMan's statement seconded; you'll gain nothing by submitting RMA without flashing - just like having corrupted drivers in OS.

I would add in also that whenever I get stumped like this, I not only flash, but also completely break down the assembly and re-assemble. I would also think about the possibility that, while the CPU worked fine on other boards, is there a possibility that it was damaged between the last board and this? Or, maybe the CPU is finally at the end of its life? I don't know if wear in the CPU can cause these issues, though.

Definitely try a flash, at least - maybe to an older version that lists that CPU as supported?

Z