Hard drive temperature - very hot

Do you have a concern about the fan, or was your question 'interest only'?
I think that you will find that most folk here run the custom firmware with the fan package, which keeps the fan running all the time at lower revs. The benefits of this is that the HDD runs cooler and more temperature stable, so does not reach the temp at which the Humax switches it on to full revs causing it to be pretty noisy. I run mine at 50% which keeps the HDD at or below about 40C.

I have just replaced the HDD which was beginning to fail, probably not helped by the large temperature cycling. I have now also installed the "fan" package and set the minimum duty cycle to 34%. This keeps the drive at 50C, but I do have a concern over the life of the fan when running continuously. The 34% was chosen as a compromise between HDD temp and fan speed. Fan life is dependant upon speed and temperature. Also, of course, noise is also a function of fan speed.

So the Nidec fan is likely to be slower, move less air, but be less noisy than the Delta. My Humax was quite an early sample, so did Humax change the fan to improve cooling? It may also explain why some owners complain about fan noise and some don't.
 
Last edited:
So the Nidec fan is likely to be slower, move less air,
I still think that's rather a leap of assumption. Yes, if everything were absolutely like-for-like, power consumption would be proportional to volume of air moved per unit time... but it's not like-for-like. Two different manufacturers publishing specs according to their own internal processes and standards does not equivalence make, and you have no idea whether their efficiencies are the same.

I find it hard to be concerned with fan life - my several units have been running pretty much continuously, one of them since 2010.
 
Wow 34% = 50C? Mine is 50% = <40C
Can't remember when I got mine, but it was a long time ago before they discontinued them. Manager's special, I think it was.

What is the MTBF for both fans? Is it a lot different?
 
I still think that's rather a leap of assumption. Yes, if everything were absolutely like-for-like, power consumption would be proportional to volume of air moved per unit time... but it's not like-for-like. Two different manufacturers publishing specs according to their own internal processes and standards does not equivalence make, and you have no idea whether their efficiencies are the same.

I find it hard to be concerned with fan life - my several units have been running pretty much continuously, one of them since 2010.

If there was a way of accessing the fan tacho signal we could set the fan controller to maximum duty cycle and measure the fan speed for a machine using the Delta and another using the Nidec. That would tell us whether Humax changed to a higher output fan sometime during its production life.
 
Wow 34% = 50C? Mine is 50% = <40C
Can't remember when I got mine, but it was a long time ago before they discontinued them. Manager's special, I think it was.

What is the MTBF for both fans? Is it a lot different?

50C in itself is not really a problem for HDD's. Reliability is more related to the temperature change, but more specifically to the rate of temperature change. Some drives run a lot hotter than others. Some Seagate drives do run hotter than WD.

The spec for the Nidec fan is not known but since they are both fairly reputable (Chinese!!) manufacturers and both use fluid dynamic bearings I doubt whether there is much difference in reliability.
 
Wow 34% = 50C? Mine is 50% = <40C

My intention is to leave it at 34% for a few days to ensure that the system is fully stable under all the normal usage conditions and then to increase the duty cycle in stages, probably up to 50%, recording the resulting steady-state temperatures. I will post the results in a couple of weeks time.
 
I fail to see why you are getting so concerned about fan speed etc. It's a fan. It goes round for quite a long time. It shifts air. It keeps the HDD cooler than it would be otherwise.
Humax obviously recon that 55C is too hot though.
Look forward to your results. Will make interesting reading, but you will also need to keep a record of the ambient room temperature as well to work into your findings.
 
It would be interesting to follow up the Delta/Nidec thing, I'm sure some of us could measure it (I don't have a Nidec sample).
 
Interesting to note that the temp charts never exceed 40C, and only occasionally exceeded 35C.
 
I have now completed measuring the relationship between minimum fan speed and HDD temperature, the graph is attached.

Following the tests the fan speed is now set to 55% duty cycle to give a maximum HDD temperature of about 4oC.

I have also determined that the Delta AUB0512LB fan is almost certainly an equivalent of the Nidec H35755-58. The Nidec fan is marked with a current draw of 0.04A, whilst the Delta fan is marked with 0.11A. However, I eventually found a specification for the Delta AUB0512LB which clearly states that the 0.11A is a "safety rating", and the normal current draw is 0.04A - the same as the Nidec H35755-58.
 

Attachments

  • Humax HDD Temp v Fan Speed.pdf
    372 KB · Views: 18
Back
Top