1

I have two Seagate hard drives exactly the same model, same size and from the same batch. Both of them pass with flying colours using Seagate Tools,

  • Smart Test OK
  • DST Test OK
  • Short Generic OK
  • Long Generic OK

However when using such tools as HD Tune Pro to read the smart data one of the drives reports an issue with the Airflow temperature which has:

  • Current: 38, Worst: 45, Threshold: 45.

While 38c doesn't seem high, looking at the data value indicates 639238182 which I believe at some point in the past its got very hot and that's why HD Tune Pro is recommending to replace the drive.

Reading the advice of others on Super User, I now have a bigger gap between the drive and better airflow going in and out of the case. I've also read that some sensors, generally older ones are prone to going faulty and giving incorrect values, which brings me to the question!

Question: Where on the PCB or within the Drive is the sensor located physically? and what does it look like? or if its calculated from another sensor, where can one find that?

Simon Hayter
  • 284
  • 4
  • 17
  • When I stress my SSD hard, it shows 51 or 52 C. I don't think 48 is high. – Aganju Aug 09 '17 at 23:15
  • You can de-case SSD's pretty easy :P – Simon Hayter Aug 09 '17 at 23:28
  • Are you sure you don't have the current and worst numbers above backwards? The worst number should be the hottest it has ever gotten. That big number you posted is the RAW data and that has been converted in to the temperatures you see. The drive got a little warm at some point. I wouldn't worry about it personally. 45c isn't even that high of a threshold. Kind of surprising, many drives run hotter than that on average. – Appleoddity Aug 09 '17 at 23:33
  • Hi @Appleoddity its actually suppose be 38, sorry. – Simon Hayter Aug 09 '17 at 23:39
  • @Appleoddity Sorry but I left something out whcih is puzzling me... the drive is around 5-7 years old but the drives were unused and sealed when I received them. The Power on Hours count is only 41 and was under 1 when I received them. A number that high with only 41 hours usage? It is now 857079859, 49 curr, 45, 45. Temp is 51, 55, 55 with 858993445971. Everything is pointing to a faulty sensor I, but I have heard the drive click around 6 times. – Simon Hayter Aug 09 '17 at 23:43
  • Reason I'm asking the question is maybe I can rub some 99% Isopropyl over the sensor and see if that helps, I know its a long shot but worth a try I guess. – Simon Hayter Aug 09 '17 at 23:47
  • Well, I don't think the hours of usage is related to its operating temperature. It isn't like a fan that gets clogged. It's probably just that you had it installed somewhere where it wasn't getting enough airflow. If the current, under load, temperatures are good I don't see any reason to be concerned. HD tune shouldn't be reporting smart failure based on a temperature threshold alone. 45c is really not that warm, not sure why the manufacturer set the threshold so low. Seagates are actually the hottest running drives I'm aware of and often fail because of it I think. – Appleoddity Aug 09 '17 at 23:47
  • Ah I see, most likely because I tested the drive in an enclosure with the fan not working :) – Simon Hayter Aug 09 '17 at 23:49
  • Probably. To answer your question, the temp sensor is probably a component on the controller board. But where or what it looks like is not something I can answer. – Appleoddity Aug 10 '17 at 00:01
  • Make sure you keep plenty of air flow over those seagates. I'm telling you, those things run hot and fail often. I don't trust them at all. Probably a barracuda right? Out of a stack of bad hard drives 80% to 90% will be seagate barracudas. If not barracudas, maybe you'll have better luck. – Appleoddity Aug 10 '17 at 00:03

2 Answers2

2

To answer the original question, for many hard drives (ie older Maxtor) and some WD drives, one of the sensors is located on the read head actuator near the base. In the case of some Maxtors, this is a 100k NTC SMD chip mounted on the flex cable usually near the preamp for the read heads.

Ben Smith
  • 21
  • 2
0

The value is most likely derived from a sensor on your board. I stand corrected and erroneously derived my conclusion from the situation at iMacs.

Here is a very interesting article clarifying a lot of the questions regarding this matter including the question how reliable SMART sensors are. Following the given notes there it's not unlikely that you suffer from incorrect thresholds, wrong evaluation of the values or actual sensor errors.

In this light it makes now much more sense to me why external HDD heat sensors are used even when there are internal SMART sensors.

wp78de
  • 1,662
  • 10
  • 25
  • Sorry but this is not true. Hard Drives have multiple sensors and the SATA spec does not include passing temperatures from the motherboard. See: https://superuser.com/questions/588878/how-do-software-programs-determine-the-hdd-temperature and in regards of iMacs, this is true, they used probes because they decide not to use S.M.A.R.T to obtain the temperature. – Simon Hayter Aug 10 '17 at 10:37