Crosshair VI Wifi, fan control / sensors are broken

[weare]

Hello.
I'm using c6h wifi motherboard with 3700x CPU and currently experiencing a lot of problems with fan control / sensors on recent bios versions. I did expect some problems when i was building this system (bleeding edge bios, etc), but i never thought there will be so much issues with such a basic functionality as temperature control.

I have 1 CPU fan and 3 PWM case fans (Noctua A14/A12). #1 and #2 case fans should change RPM depending on CPU temperature, and #3 depends on PCH temperature. Basically it works fine for some time (usually 2-10 hours), but then the expected behavior breaks in many different ways:

1. This is the most common issue i'm experiencing right now. Usually fans should spin around 800-900 RPM when the system is in idle state. When this problem happens, RPM decreases to 400-500 and the more i load my CPU the less RPM i get until fans completely stop. And it happens to all CPU-related fans. This is so ridiculous i even made the video — https://streamable.com/i3379 where you can see the idle behavior (~400 rpm) and what happens when i load my CPU using 'stress' utility. I've tried different fan profiles (silent, standard, etc), different low speed, delays and nothing changed. I've checked and can confirm that fans are really stopping, so it's not just monitor tool issue. The fix is to reboot the system.

2. Motherboard temperature sensors just stop working and hang up on the last value it was, even though RPM sensors still work. Once more, it's not the monitoring tool related problem because fans aren't changing the speed if i load my CPU. For instance, if CPU temperature hangs up on ~50C temperature, fans will still have ~800 RPM even if i stress my CPU (usually in this case RPM should increase up to 1500). The fix is to reboot the system.

3. Fan is just completely ignoring the component's temperature. I had this issue on my fan #3, which wasn't changing idle RPM even though my PCH temperature was already around 80. The fix is to reboot the system.

4. Fan gets stuck at 100% RPM. This usually happens on my fan #3 when i cold boot my system. Rebooting the system didn't help me though, i had to clear CMOS (or make some changings in the bios) to fix this.

So basically my system is barely usable right now, i have to constantly monitor the sensors and regularly reboot my system when something breaks again. Otherwise this behavior can possible damage my CPU/motherboard due to overheat in high-load scenarios. As far as i can see, other people are experiencing pretty much similar fan-related problems i do, so it's not just defect motherboard i have.

I've been using ASUS motherboards for 17 years starting from P4PE and you can imagine how disappointed i am right now. In my opinion this kind of issues are critical and should be investigate ASAP. I hope my report can be useful, and this situation won't end with changing motherboard to another vendor.

[Ryo_z]

Richard_Dower wrote:
Can I bet that 7302 is a pre ryzen 3000 bios ?

no, it is the first version of test bios posted by Shamino for c6h.

[Alex656]

Ryo_z wrote:
no, the issue is still occur even i manually set the correct dram and dram boot voltage.

fyi, this issue also present in bios 7403/7401, but not on version 7302.

Same here!

[htf123]

This fixed the fan bug totally for me and my C7H Wifi. Thanks for this bios with the fix, I was a day away from buying a new motherboard to get away from this issue despite loving my C7H so much.

[Tubss]

The newest 7302 bios has this same issue. So far at random times either my PWM CPU fans or the AIO pump stop working, and my cpu temps rise to 95-105 and cpu freq drops to 500mhz. Gonna have to downgrade to this beta bios, but have to say im not really happy with my C7H purchase...

[mikev190]

Shamino wrote:
yes f9 is dram related
just checking if it actually boots with the right dram voltage, set both dram vboot voltage (under digi vrm) and dram voltage to whatever is needed / used.

When are you going to take the issue with some CPUs not boosting to max clocks ever more seriously? The same CPU can be put into a competition mobo with ABB bios and boost properly, but not in the Crosshair VI. I appreciate you communicating with the community but actions speak louder than words and outright blaming AGESA isn't cutting it, otherwise, the issue would persist on ALL mobos and it is getting tiring repeating myself. It is affecting many other users across reddit, these forums and OC.net too.

My 3600 hits 4.1Ghz max with the C6H when gaming it goes to 4075MHz and lower depending on temps. Yet in 2 separate brand boards with ABB it goes to 4.2Ghz max and around 4.2Ghz to 4150Mhz in games. So I am loosing 100mhz somewhere. Also, PBO and autoOC doesn't work and makes performance worse. Yet the same tests done on the other mobos gave me 4.3Ghz spikes and 4225Mhz - 4.2Ghz in games.

Alternatively, on my C6H I can run a BCLK overclock with the default profile enabled in Ryzen master to allow stock boosting and voltage works perfectly. I can basically run it above MAX boost this way 100% stable. I just don't see why we have to use BCLK to get round UEFI issues. I'm considering buying a different brand mobo and leaving ASUS for the first time since 2003. But... With that said I am willing to help you resolve the issue and test any bios. I just want the correct performance. 4.2Ghz isn't breaking Zen2 limits and the BCLK overclock with boost proves it is capable with the C6H and the other boards prove it works properly without BCLK on the current ABB AGESA.

[YOLO___]

after 21 hours of uptime my fan still working, fingers crossed 😄
remain the issue at boot that my RTX 2080 fans goes to 100% for few seconds

[Richard_Dower]

YOLO!!! wrote:
after 21 hours of uptime my fan still working, fingers crossed 😄
remain the issue at boot that my RTX 2080 fans goes to 100% for few seconds

The RTX 2080 fan issue you noted, also on my setup. See the recent videos I posted to visually see if yours is similar to mine.

[gupsterg]

UEFI 2703 & 0002 with SMU FW 46.34.00 for C7H & C7HWIFI in this post.

[neotokyo]

I had this glitch (finally) happen on Sunday with bios 2703 while I was packaging applications for deployment. The CPU cooler went down to 0RPM, then a couple of minutes later it's the case fans that started acting, like they were going down then ramping up almost immediately. The computer was back from sleep state when it happened, it didn't happen again for the couple of hours it was running it to package stuff after I rebooted. I couldn't test Sunday, but I'll try with 0002 during the week, I need to do CPU intensive work at some point before Friday.

Edit: The case fans ramping up is because they're connected to the fan hub of the case, and, apparently, when the motherboard lowers the PWM to 0RPM, or below a certain threshold, the hub ramps up the fans to full speed instead.

[mikev190]

Can anyone reupload the C6H non wifi ones please. The dropbox links don't work any more