I have a feeling the fans are replacemen fans, as they only have GND and +24DC cable going to a 3 pin connector. The control board has 3 pins, I haven't yet measured what the third pin is for, but I would think it's a tach input signal?
Does anyone know what the original fans are for this unit? Schneider support doesn't want to comment on it.
Or if anyone knows if the third pin is for Tach, then I will obtain fans that match the CFM but also provides tach.
I think you are OK going for similar power (either CFM or roughly going on voltage x current). We sometimes have scrap units that have OK fans If you cant find decent fans from somewhere then contact us and we'll leave a note to email you when we next get some.
My feeling is that changing the fans is not going to change the fan behaviour on the UPS - I think you are aiming to have the fans slow/stop or just be less noisy.
Depending on load / ambient temperature we have had customers experiment with quieter fans but you are off into the unknown with that so you need to test seriously to make sure it does not overheat.
However, my confusion is why the fans are running at full speed, continuously even when there is less than 70W load and batteries are fully charged. They never change speed, from my understanding, the SMX units should adjust speed depending on demand, but this doesn't seem to be the case.
Have you come across SMX units that use the third fan pin on the controller board? This must be tach signal right?
Nope! I've just gone to check a UPS in stock and both fans are none tacho 2 pin type. They must be driven on a temp/load map. The unit I have is sitting idle and the fans are completely stopped. As before I dont think changing your fans is going to change that yours are on full all the time.
There is a risk in flashing the firmware but you could give that a go. I think something in the UPS has gone bad, but in a none critical way to normal use.
Did you ever figure out why the fans were stuck on @satanski? I'm having the same issue with an SMX2200RMHV2U I got recently. It seems to work but the fans run all the time even when there's no load and it's on mains power. Not sure if they are on full or not but I gather they shouldn't be on at all...
Edit: I have tried updating the firmware to the latest but this has made no difference.
I did not, I believe it's due to a fault that this situation occur. Internal temperature wasn't high for the fans to trigger, nor was the load levels or charging. So most likely something to do with the sensor or inverter?
I ended up buying a core UPS without bezel, battery etc from ups trader which tested the unit to ensure fans were not running at full speed. This unit has been working fine for me since I bought it.
Just to check - the fans always run their fastest when running from battery power, so if you remove the input AC cable you know what high fan mode is like. I think you will have done that test, but some people do not and just dont like any fan noise.
We are learning something here too. If you check the UPS temperature from a web card that would be interesting. Its extremely rare for it to be broken, and they are normally in agreement with sibling UPS within a degree or two, so presumably accurate.
You've flashed the firmware. Are the cells old? I think the fans can run at high speed when initially charging from a complete discharge cycle.
Bit of a brain storm from me as so far I cant pin down what causes the difference from one SMX 4u UPS to the next.
In lateral thinking - we'd never as a company be able to recommend swapping fans far from original OEM or direct replacements of different brand, but if this was my UPS in my room and under fairly constant supervision I'd be tempted to replace with a known quiet fan type (Noctua?) - and I'd then test at a higher than normal load level, on the hottest day, and partially blocking the air ingress grill to ascertain if the flow rate is able to deal withthe worst case scenario.
I had an amplifier at home that had a fan which would cycle on as needed, from no flow to loud enough to bother me, so I stacked a second fan behind this, which gave a constant and low flow rate - once I'd done this it maintained a cool enough case that the main fan never came on again, but was still there and would come on if needed.
Proceed if you are expereienced and know the risks only.
Ah that is useful info, thanks. I didn't notice any difference in noise when going on battery power so I guess that confirms they are running at full speed all the time!
The internal temperature shows as 59 degrees on the PowerChute UI. This is over serial as I don't have a net card, I assume this temp is available over serial and they don't just print a bogus number. If so then I guess that explains why the fans are on full blast...
The cells are new, from UPS Trader :D. They show as 100% full on the display. I think they are good. I'm not sure what the expected behaviour is when the batteries are unplugged but the fans are on full then as well. When powering on from cold with no battery connected they turn on full after a few seconds.
I did the same for my POE+ rackmounted switch, swapped out the fans to Noctua, triggered the low fan speed alert but temperature has been stable even under heavy load. Which is why I also thought this would be a good idea for the UPS before I decided it's actually just more practical to change the core unit as they aren't that expensive and saves me hassle in the future.
There was a Sunon EE80252B1-0000-A99 and Y.S.TECH NYW08025024BSS in my unit from my notes. I believe the YS TECH was mounted infront of the inverter board. The Sunon has 41CFM while YS TECH is at 59.2CFM (might be wrong, just checking my notes from when I tinkered with this last). The YS TECH spins at 4500 RPM and produces 41 dB.
Looking at for instance Noctua NF-A8 PWM, it has 32.67 CFM which is essentially half of what the YS TECH is, which is a significant decrease. Also, the fans in my unit are 24VDC, which I doubt Noctua would produce (as they are targeted PC market which runs 12V rails)
In my unit, I could fit a bit thicker 80mm fans, and the one I was eyeing up was Sunon ME80252V1, 41CFM at 33 dB. Still a reduced airflow but for my usecase it would work.
However, to manually manage expected load and permanently reducing air flow throughput is a bit of a gamble, specially when dealing with batteries of this nature. A user would have to consider the temperature, load and test charging from empty every time they make a significant change to what devices are plugged in to the UPS. I don't think this is reasonable but someone might consider this as a acceptable operational cost if their infrastructure doesn't change much.
Though, the problem here is more there is something triggering the units to place the fans on full speed rather than variable rate based on need, I never managed to get an answer to this and my old unit won't be helpful as I managed to brick it as well.
Is this 59 degrees celcius? That's quite warm and it should be expected that the fans run att full speed. I would expect fan's to rnu at full speed at 40 degrees celcius.
Yeah 59 celcius sorry. The room is 24C and the 59C was with the UPS only having been on for a few minutes with no load. I will check with my thermal camera later but I don't think the 59C is correct. So my current guess is something is wrong with the temperature sensor. Not sure which component on the board is the internal sensor, maybe TH303 which is bolted onto the heatsink? If I can figure out what part that is I guess I can check if the room-temperature resistance is what it's supposed to be.