Simple cooling solution for the T2

I just checked the drive data sheet. The operating temperature range is 0C to 75C. So I am operating it at the midpoint so happy with that.
Also the drive is designed for 24/7 use, as in a PVR. So keeping it at a constant temperature should help to prolong its service life by reducing component wear
that would normally occur as result of the cyclic expansion/contraction that occurs with the temperature changes in normal use.
Also the service life of most of the internal HDR components may well be improved as a result of the lower internal temperatures inside the unit.

I have just had another look at sysmon. The temp has remained at 36C/37C for the last six hours.
 
Last edited:
Now you have convinced me.:D

Consider the fire well and truly golden showered on and put out :):D:laugh::roflmao:

And to convince you some more. I just taken another five hour plot, and as you can see below it's not bad at all and pretty steady, and as you can also see. It has even fallen to 35C, but that is probably due to the ambient temperature falling.

I would say that it is a pretty damned effective method of keeping any Humax PVR with a USB connector running a hell of a lot cooler and keeping it at a constant temperature. It becomes the primary cooler and the Humax PVR internal cooler becomes the backup should the primary fail. In the case of the T2, overall a much more effective method than the very good fan package (its limitation is the crappy little fan it controls), but used in conjunction with the fan package gives you a fully backed up very efficient cooling method.:thumbsup: What more could you want? Oh the laptop cooler was £15 from asda some 5 years ago. :thumbsup::laugh:


chart (2).jpeg

I'm now going to put my second T2 and 2000T on laptop coolers after carrying out this very interesting exercise. :D
Gone will be the sound and song of the Hummy fan. :laugh:
In my case the choir of the Hummy fans, with the three of them. :laugh::roflmao:
 
Last edited:
Oh I've also removed the CAM slot cover at the rear to give another exhaust port for the air over the mainboard side of the unit as well.

And before anyone thinks about it. I would not attempt to use the Humax fan output to drive anything much bigger than the type it already does. It doesn't look man enough to drive much more and you're likely to blow it up if you over load it with bigger more current hungry fans.
I have looked at the main board in and around the fan connector.

I'm going to shut the T2 down. And do a timer control power up at about 7am. Will be interesting to see what that plot looks like.

Here's a suggestion for a useful little addition to the WebIf. When you hover the mouse pointer over the "Freeview" bar. When you get the "Updating..." and the list of what's being watched/played/recorded is displayed. At the bottom of that list, have an additional line showing the current HDD temperature and the Fan status.
 
Last edited:
Heads are all voice-coil servo positioned, and not by stepper motors any more.
Heads were always voice coil operated until the sealed (winchester) drives came along and started using steppers. It is platter expansion I was commenting on.
Any variation in the location of tracks due to thermal expansion is irrelevant.
How is it accounted for now ? You can move the heads well enough but how do you know how far?
Further, low-level formatting is definitely a manufacturer-only operation.
Yes - most people seem to believe that. But what has happened is the term"low level" has been "adjusted" repeatedly.
The term has been made almost pointless.
As has the term "format" itself.
The built in "format" command on these boxes doesn't do anything whatsoever related to formatting for example.
You could even be forgiven for calling in trading standards.
 
How is it accounted for now ? You can move the heads well enough but how do you know how far?
Instead of dead-reckoning (which is what you seem to be talking about), the track spacing is now far too close and precise to hit and hope. The positioning is feedback-controlled by monitoring the signal from the read head - both to identify which track it is over and to position accurately in the centre of it. That's what I meant by "servo". Absolute position of the track (which would vary by platter expansion) is irrelevant.
 
Hi guys

Here's the plot from the standby start-up this morning. And it looks pretty much is I expected it would.

chart (3).jpeg
 
A calibration cycle isn't wanted repeatedly and you dont want to double seek at all.
- so some sort of expansion feedback seems inevitable. I just wondered what it was.
I find it hard to believe there is nothing but servo data because bi-metal just works
and improves the servo data response.

However it's done - as track to track density increases temperature of the platter must be
more and more critical in this respect - hence my suggestion to "low level" format if possible.
20 degrees off normal previous operating temperature seems significant to me.
(I could be wrong - this is just gut feeling)
Especially long term.
If you think about it - you wouldn't even do a high level dos format on a freezing cold drive-
you'd at least let it hit room temperature first. That's for good reason.

If you need a "low level" format program try hdllff - that seems to have most bases
covered in a simple package.
 
However it's done - as track to track density increases temperature of the platter must be
more and more critical in this respect - ....
20 degrees off normal previous operating temperature seems significant to me.
I'm sorry but this makes no sense to me. I understand what you are suggesting I think (that the increase in diameter of the platter with temperature is/may be as much as the width of a track).
But these discs all start and operate from cold and gradually warm up over a period of time - for a laptop left in a car in winter the temperature range could be very large over that time - so the drives must already be well able to compensate for that.
I don't see the point of reformatting at any particular temperature, high or low.
 
There isn't any, and you would be daft to attempt it on a modern ultra-high density HDD. The platters are initialised by the manufacturers using highly specialised equipment. Going back maybe 20 years (when a drive was expensive and one was desperate to get the maximum life out of it), we used to play around with low-level formatting in the hope that one could mark dodgy sectors as bad, but I am not at all convinced it did any good - and those were in the days of IDE, when the drive was more-or-less bare mechanics. A modern drive on a SATA interface or whatever has its drive controller built in, and whatever you try to do to the disk there will be some electronics in the way interfering.
 
@ Neil. Got a link to the type of cooler that you are using please?

The cooler I'm using is made by Targus and is for a 17" laptop which makes it the perfect width on which to sit the hummy. It is sloped so I got a length of wood under the rear to level it.
http://www.targus.com/uk/lap-chill-mat-awe55eu
They have a whole range at various prices.

I'd google laptop cooler and see what you get. You may find something thinner or flat, more suited to your installation. Make sure it has large fans. (They run slower so are quieter but still move a good volume of air compared to the smaller ones)

Just make sure the cooler fans are directly under the vent holes in the bottom of the hummy. You'll probably have to have the cooler back to front. Also remove the CAM slot rubber bung. The hole provides an extra exhaust port, and increases the internal airflow.

I assume you now approve of this method of cooling the hummy. :D
I've just checked the plot for all day and except for the power down and restart at 7am the temperature has not gone above 35C all day.
 
Last edited:
Thanks for all the info... but I will just keep to the excellent fan package that does much the same job without any external add ons, wiring or cost, and at 60% is quiet enough for me.

My box came on shortly before 8pm this evening to do timer simultaneous recordings until 11.30pm, the temperature rose to 36° and sat there for about 3 hours until for the last 20 minutes it rose to 37° when I was also playing back a previous recording at the same time. So I am happy. :)

HummyHDD.jpg
 
I googled and found a few with multiple fans, up to 5, and thought that rather defeats the object of the exercise.
Yes, your 'mod' seems fairly impressive, but the results do not seem a great deal better than Stummery's above. I run my fan at 50% and get a temp of about 44. I'll try upping it a bit in a while and see what happens to the temp (when it has 'warmed up').
I'm still with Stummery, insofar as it seems a bit like the proverbial sledgehammer to crack a nut when there is a simpler solution available, albeit that you do have to install the CF to achieve it.
 
I Look at it this way. I have three fans. If one fails in the cooler, there is the other and if that fails. I still got the hummy fan which is not going to be worn out from being run continuously.
Your hummy fan wraps its hand in and your likely to start cooking the HDD.
I've never really had faith in cooling fans below 50mm in size, as so many I have seen in the past have suffered bearing failure. So would much rather have a bigger fan running at a lower speed for the same airflow as a smaller one. It's going to last much longer and be a darn sight quieter.

Another thing is that my mainboard and PSU are also likely to be running much cooler than yours as I'm flooding all the bottom vent holes with a large volume of moving air. So internally I've got a lot more air moving around inside the whole of the case and exiting at the 'fan holes', the CAM slot hole (cover removed) and at the bottom edges of the unit cover on both sides of the unit.
The primary airflow with the internal fan only really covers the HDD.
 
Last edited:
Along similar lines, I have my 'Fox installed in a unit with various shelves above and below, and because I was worried about temperatures* I fitted a big (100mm) PC fan vertically behind the 'Fox, powered by an old 3V Nokia phone charger. So the big fan turns slowly, is genuinely inaudible unless you put your head right next to it but still delivers airflow, and sysmon reports a steady 45-48 deg C without the internal fan ever running. I make no judgement about whether this is better or worse than any other solution, just mention it as something that can be done.

* I should add that originally
a) it was a Topfield TF5800 on that shelf doing my TV recordings, and it has no internal fan at all
b) there was a transparent plastic "window" partially covering the front, to avoid little fingers changing settings
and so additional cooling was considerably more important than it is now...
 
Again, along similar lines, I modified my two units about two and a half years ago,with external fans as described below. The internal fans have never needed to operate since and the temperature never exceeds 20 degrees above ambient.

http://hummy.tv/forum/threads/fan-more-control-over-hdr-fan.3907/page-2#post-47447

Yeah the 20 degree differential is seen in my results too. I was thinking along similar lines to you with a ducted airflow system under the HDR blowing air up through the bottom vent holes in the unit base.
Then I looked at my laptop sitting on it's cooler. The obvious light bulb moment. My laptop now needs a new laptop cooler. :)
Not only that. When the T2 goes into full standby, being powered by the rear USB port, the cooler fans conveniently shutdown as well. :)
And overall, once you got your unit positioned so that he fans are under the vent holes, it's just Plug'n'Play. And you've also got the T2 internal fan as a back-up.

Now here's one for 2000T owners. The 500GB model has no fan! The exhaust holes in the rear panel are there and just a large number of holes in the base (Another step backward they made with the 2000T). Forced cooling from below is the only way if you are concerned that it may run too hot for your liking or are cautious about thermal issues. A small cheap laptop cooler should be just the ticket.
 
Last edited:
Back
Top