[SOLVED] Weird DC fan behavior on ASUS B550-E Gaming

[hyperreal]

Hi all,

My s/o just got a new PC and I promised to set all the fan curves to maintain positive pressure. So I downloaded Fan Control and ran the automatic detection process.

The three front fans (Fractal Dynamic X2 GP-14) went completely nuts after the setup, randomly showing 0 RPM for brief moments while constantly starting and stopping. I could run them stable at some speeds but even between those speeds they would just shut down.
I then booted into BIOS and tried the Q-FAN tuning. It ended up setting the the header's minimum speed at 90%. But the fans stll work fine if I switch the curve back to standard.
I'm sure something is not right here.

Other fans (all PWM) work fine. Here's the complete case fan setup:

• Front: 3 x Fractal Design Dynamic X2 GP-14 -> CHA_FAN3 on mobo (through splitters)
• Top: 3x be quiet! Light Wings 140 PWM -> Fractal Nexus+ 2 fan hub -> CHA_FAN2 + SATA power
• Rear: 1x be quiet! Light Wings 140 PWM -> CHA_FAN1 on mobo

The motherboard is ASUS ROG Strix B550-E Gaming and the control mode for CHA_FAN3 is set to DC. I tried googling this issue and bumped into some posts where people have had this problem with various ASUS boards.

Can anyone tell if this is a known BIOS bug or should I start troubleshooting the fans and how?

Also, we didn't build this one ourselves, but it was built to order by a reputable PC store. They provide (paid) tech support but I decided I'll ask around the web first.

 

[JohnBonhamsGhost]

3 x Fractal Design Dynamic X2 GP-14 -> CHA_FAN3 on mobo (through splitters)...
the control mode for CHA_FAN3 is set to DC

with no option for changing it to manual PWM in the BIOS?

it's very likely caused by having too many fans connected to a single header without providing external power.

I downloaded Fan Control...
ASUS ROG Strix B550-E Gaming

what is Fan Control?
this ASUS board should be using AI Suite III's Fan Xpert for settings up fans connected to it's fan headers.

 

[hyperreal]

with no option for changing it to manual PWM in the BIOS?

it's very likely caused by having too many fans connected to a single header without providing external power.

It can be changed to PWM but that just makes the fans go 100% all the time, since they apparently are DC fans.
Can you elaborate on the power draw issue? I know that too many fans in one header is bad - can be dangerous even - but three fans at max 0,2A each shouldn't be a problem as far as current goes. Then again, I'm not an electrician and I'm not sure how wattage and voltage play into all this.

what is Fan Control?
this ASUS board should be using AI Suite III's Fan Xpert for settings up fans connected to it's fan headers.

It's an insanely great, lightweight, customizable open source fan controller software for Windows.

 

[JohnBonhamsGhost]

It's an insanely great, lightweight, customizable open source fan controller software for Windows...
(Fractal Dynamic X2 GP-14) went completely nuts after the setup, randomly showing 0 RPM for brief moments while constantly starting and stopping.

i would have to say obviously it isn't if it has caused this issue with your fans.

Can you elaborate on the power draw issue?

headers usually can only handle two fans through a splitter at once without an external power source; SATA, molex, etc.

they apparently are DC fans...
posts where people have had this problem with various ASUS boards.

the only issue i've ever run into with DC fan modes on ASUS or other boards is them sometimes only offering two speed modes, ~33% or ~99%.
when there should be more levels to choose from.

 

[hyperreal]

i would have to say obviously it isn't if it has caused this issue with your fans.

I'm 99% sure the software didn't cause the issue. It just revealed it. The exact same problem happens in the BIOS fan settings. The RPM reading for CHA_3 in the EZ Mode view randomly drops to 0 (even when the fans are running stable) and just like Fan Control's setup process, the Q-FAN tuning has no idea about actual min/max RPM or start/stop thersholds.

headers usually can only handle two fans through a splitter at once without an external power source; SATA, molex, etc.

I've understood that headers can handle as many fans as there is room for amperage and that's why I asked you to elaborate. The header can't count the fans. It can just deliver power according to certain specs but the only spec I practically understand in this case is max current delivery which is 1A according to the manual, and the fans draw 0,2A each, which makes a total of 0,6A.

the only issue i've ever run into with DC fan modes on ASUS or other boards is them sometimes only offering two speed modes, ~33% or ~99%.
when there should be more levels to choose from.

My previous board was an ASUS and I also never had any problems with fans. It's anecdotal of course, but leads me to believe this is an electrical fault rather than a mobo issue.

I'm now considering buying three Arctic P14 PWM fans and the Arctic PWM hub. It would be a cheap and easy solution.

 

[hyperreal]

I called the retailer and they adviced me to check that the 3-pin connector is properly connected, if it is then try another fan header. If that doesn't help, then test the fans one by one without the splitter. Possibly there is a problem with the RPM reporting and that's causing erratic behavior when automatically detecting min/max speeds.

 

[JohnBonhamsGhost]

I've understood that headers can handle as many fans as there is room for amperage

it also has to do with properly "splitting" the signal for motor control & RPM stats.
i've fixed many user's fan issues by limiting their headers to two fans at max.

 

[hyperreal]

I see. In the case of DC fans, when connected through a splitter, how is the RPM reported to the motherboard?

I looked at the mobo manual and there are no free chassis fan headers left, unfortunately. So if it's a header problem, I'm pretty much screwed with the current fans unless I just take out one on the front fans, but that will also mess up the case pressure.

Or maybe I could connect the rear exhaust to CPU_OPT, since it's a PWM fan, to free up a chassis fan header? It would then be in sync with the CPU fan. Or remove the rear exhaust completely since the NH-U12A's rear fan will push air out from the back.

 

[JohnBonhamsGhost]

maybe I could connect the rear exhaust to CPU_OPT, since it's a PWM fan, to free up a chassis fan header?

usually this header clones the value of the CPU_FAN header.
if you are okay with any connected fan following the instructions sent to the CPU_FAN header it will work fine.

if PWM fan on CPU_FAN & 3pin fan on CPU_OPT i imagine the DC fan will just stay at 100%..?
test and see, i'd like to know if that scenario works with no issue.

 

[hyperreal]

if PWM fan on CPU_FAN & 3pin fan on CPU_OPT i imagine the DC fan will just stay at 100%..?
test and see, i'd like to know if that scenario works with no issue.

I was talking about the rear fan which is a PWM fan, and it following the CPU fan curve would be logical in terms of getting heat out. Especially since the cooler is overkill so that it can be kept at low speeds even under gaming loads.
So I'll plug the rear fan into CPU_OPT if I need to free up another header for the front fans. The front fans to CPU_OPT would just make them go at steady 100% since they're DC fans.

 

[JohnBonhamsGhost]

The front fans to CPU_OPT would just make them go at steady 100% since they're DC fans.

i would image the voltage would remain steady for the 3pin CPU_OPT fan while the PWM signal was adjusted for the CPU_FAN
but i've never actually tried this type of scenario.

it's also possible that the current varies a bit and the 3pin fan could just stop & go repeatedly.

i haven't worked with 3pin fans in 10+ years and always avoid them
while also making customers pay to switch if they are using them so i can configure detailed fan curves for their systems.
never liked depending on DC power variables for control.

 

[hyperreal]

I just tried switching the front fan splitter connection to CHA_FAN2. The problem remained: In BIOS the RPM reading alternates between the actual reading and "N/A".
I'll now test the fans one by one.

 

[hyperreal]

If you or anyone else would be interested, here's a video of the issue as seen in bios fan monitor with a quick glance at the fan curve right before the end: https://www.dropbox.com/s/wbp5bpy5d81sgcz/IMG_2219.MOV?dl=0

 

[JohnBonhamsGhost]

I just tried switching the front fan splitter connection to CHA_FAN2. The problem remained...
I'll now test the fans one by one

would definitely be odd to have all three fans malfunctioning at the same time, almost impossible i would think.

there's always the chance that for some reason this motherboard can't supply a steady voltage through the three pins but that would also be very odd.
could be some issue with detection and BIOS control but then you should be able to override that through OS software and since they continue to malfunction no matter what software you use it just makes the case even more odd.

those AC Arctic P14 fans you mention earlier are pretty decent with good airflow & air pressure.
if they are also quiet enough, those with their hub would be a better idea than these GP-14s anyway.

 

[hyperreal]

Turns out you were right.

I first tested one fan straight to the same mobo header. Worked fine. Also gave me an opportunity to overwrite the ridiculous 90% minimum speed setting in Q-fan. Then I tested the same fan through the splitter, going through all three connectors. Each one worked fine.
Then the other two fans through the splitter, but just one at a time. Worked fine.
Two fans at once through splitter, worked fine.
Three fans though... The RPM number freaked out just like before.

So the header can't handle three fans, just like you suspected. I'm surprised the retailer, who also built the PC, didn't notice, since they pride themselves on extensive testing of custom builds.

I'll leave the bottom fan disconnected for now and look into buying new fans if and when the actual owner of the PC so desires. They're already angry their partner is optimizing fans when they just want to play RDR2.

And yes, Fractal fans... I don't like them either but since they came with the case, I thought "oh well" and asked the builder to fit the stock fans in the front. I don't have any experience with Arctic fans but I know they're cheap and they have a 10 year warranty, so there's that. My old build had Fractal fans and they make this annoying, inudstrial-like humming noise at certain speeds. Arctic fans can get really loud at higher speeds but at lower speeds they're acceptably quiet.

Anyway, thanks for your help! Much appreciated.

 

[JohnBonhamsGhost]

I'm surprised the retailer, who also built the PC, didn't notice, since they pride themselves on extensive testing of custom builds.

you may be surprised to find that a large portion of pre-built systems have similar or even stupider mistakes in place.

many will come to forums complaining about 100°C CPU temps at boot because coolers aren't plugged in,
no boot because CPU 8pin cables are not connected,
'no signal' errors because PCIe cables are not connected to GPUs,
drives not connected to SATA data or power,
memory not fully seated in DIMM slots, etc..the list goes on forever.

a lot of these systems are never tested before shipping out, no matter what the retailer or distributor claims.
in the past i've read some articles about larger companies supplying pre-builts even having migrant workers just following diagrams for assembling even though they've may have never even handled a mouse before.

i would never recommend anyone buying a pre-built unless they are custom ordering from a local shop.
some companies like NZXT seem to have quality builds but their component selection is extremely limited.

 

[hyperreal]

i would never recommend anyone buying a pre-built unless they are custom ordering from a local shop.

That's what I did, except it's not exactly local, but a nationally renowned PC component retailer & custom build shop with excellent service. I'll of course call them and let them know about the situation. But that's already very much offtopic so let's just leave it at that. Have a nice weekend!

 

[JohnBonhamsGhost]

that's already very much offtopic so let's just leave it at that.

not off-topic at all since this thread is actually partially about pre-built distributors causing issue(s) with customer's systems, like yours.

it's not exactly local, but a nationally renowned PC component retailer & custom build shop with excellent service.

how can you call it nationally renowned with excellent service if they cause simple issues like this for their customers?
and who determined that they are that?
if some internet ad or some statement by the company, that doesn't add up to anything.
if by customer reviews those can easily be manipulated and only posted when verifying some good statement about the said company.
even the worst companies on the planet claim they are something special for advertising/marketing purposes.

the purpose of these forums and any threads we create here is to offer solutions and help for other users and those searching for help with similar issues in the future.
if this company causes issues like this, that take even what i would call at least a somewhat intelligent owner to determine and fix, it is good for others to see it and avoid them.

 

[hyperreal]

how can you call it nationally renowned with excellent service if they cause simple issues like this for their customers?
and who determined that they are that?

Ok, that's a valid argument.
I'll begin my answer by disclosing that I live in a small country with a relatively small population base, so anything or anyone commercially associated with a fringe hobby - such as custom PC building - and doing a bad job or acting in a particularly disrespectful manner will be unanimously known as someone you don't deal with. The word goes around. That's because every hobbyist scene here is very small, everyone knows each other and most commercial actors - such as entrepreneurs and investors - related to various hobbies actually come from the said scenes. Even some of the bigger ones such as the place that built the PC we've been talking about.

So who determined they are renowned and offer excellent service? The PC geek scene in my country. People like you and me who argue on internet forums because we are passionate about building, using and maintaining computers.
It's not some second-hand part salvage and resale scam operation nor a global corporation posting tons of fake reviews and paying for Instagram influencers to praise them, although they do sponsor various tech YouTubers who make content in our native language.
They aim to make a profit, sure. But I'm not buying into the idea that they just try to screw their customers over with deceptive marketing without providing actual substance of their expertise.

I will be calling them just to let them know that what happened with my fans is an issue they should remember to keep an eye out for and hopefully they will be open to my criticism and take it into account in the future. And if they won't, then that's a bad anecdote for the whole scene to know and it will bite them back. And they know it.
And for the record, using the three-way splitter was my idea, as I'm the one who ordered a custom build and even had a lengthy conversation about the fan connections and the fan configuration in general with the sales rep. What they failed was to check if the connection ordered by the customer really works as intended.

So that's my take on this.
Now, I know what you're trying to say and I agree with you. Businesses should be held accountable for deceptive marketing, review manipulation etc. Not to mention even worse stuff like poor working conditions or cutting corners so that it affects workplace safety or product safety. But I get the feeling that we're coming from very different places.

 

[JohnBonhamsGhost]

I will be calling them just to let them know that what happened with my fans is an issue they should remember to keep an eye out for and hopefully they will be open to my criticism and take it into account in the future.

yeah, hopefully they would find a new solution to their fan setup for this particular build line.

 

[hyperreal]

it also has to do with properly "splitting" the signal for motor control & RPM stats.

Hey, I'm returning to this subject since a thought just occurred to me.
The splitter we got is built so that it first branches into two wires, and then one of the two wires again branches into an extra wire. So it's like an Y-shape but then one of the branches is another Y-shape.
Could this be what caused the RPM signal instability and is this the kind of stuff you are referring to when saying the signal needs to be split properly?

 

[JohnBonhamsGhost]

So it's like an Y-shape but then one of the branches is another Y-shape.
Could this be what caused the RPM signal instability and is this the kind of stuff you are referring to when saying the signal needs to be split properly?

could be.
once that signal got split, and then split again, it could have further reduced the ability for the fan(s) to get proper constant voltage for their DC modes.

is this splitter 4pin and you're only using the 3pins or designed specifically for 3pin?

 

[hyperreal]

It's 3 pin all the way. And it has been working fine for now with only two fans connected in regard to both fan stability and overall temperatures. We won't be needing extra cooling until summer arrives.