• The forum software that supports hummy.tv has been upgraded to XenForo 2.3!

    Please bear with us as we continue to tweak things, and feel free to post any questions, issues or suggestions in the upgrade thread.

[fan] More control over HDR fan

Ezra Pound said:
I personally think there is some merit in saving fan life by delaying initial 'fan on' until a certain temperature is reached,
Click to expand...
With fan speed set at 50%, I see that it takes about 30 minutes from power on for the disk temperature to reach 38 and another 30 minutes to reach 40 which appears to be the norm for me when the machine is not recording or playing but just doing background activities.

If the fan start were delayed the disk would warm up to the trigger temperature faster so I think you would only save less than 30 minutes fan use per resume from standby which is probably not going to noticeably increase the lifetime.
 
Running the fan fast wears out the bearing much more rapidly, so the current behaviour of the 'fan' package may actually extend fan life. Wear on the bearing is proportional to something like the cube or even fourth power of the RPM (my dad would be able to tell me the exact relationship).
 
Ezra Pound said:
At present, the Fan package only operates until the threshold is reached, at this point the standard 'Humax' contol takes over, so although 'minimum remains the minimum', the fans speed is now being governed by the standard 'Humax' program and will go up to a fixed 70%

I think Zekepliskin wants full control at all times, although that could be dangerous, I personally think there is some merit in saving fan life by delaying initial 'fan on' until a certain temperature is reached, but I would still allow 'Humax control' when the threshold is reached, it is the safest option
Click to expand...
I'm not sure whether you were expanding on what I said or thought that as I had not written more that it may be because I did not know.

But either way the point I was attempting to make centres around Zekepliskin stating that control is "removed". That could be a misunderstanding possibly brought on by a misunderstanding of the prior sysmon diagram where it states that the fan is no longer running, just as if the fan package control had been removed.
If that is how Zekepliskin interpreted the sysmon diagram than the suggestion of full control as an improvement could be based on a false premise.
 
And having read all that, it would be good if Zekepliskin came back to tell us what he meant, if it's not already done.
 
The idea would be to still have the fan settable between 0% and 100% but change the temp thresholds so the possibility of 100% fan speed is lower although the box would run a little hotter. A lot of people state you can't hear it on 30% but I can even though it's slight.

In this case I think the fact I'm currently running a desktop 2TB drive in there doesn't help, as in this case it's going to run hotter, not have as good a level of acoustic management and be noisier even with the fan off. Which is why I have one of these arriving tomorrow... because I'm a bit of a nerd I'm sure I'll enjoy using the command line tools to prep it for GPT then get all the videos from the current drive across as well. I've had the 2TB version in the Foxsat HDR and my old VM TiVo, they are worth the extra money.
 
The issue is that to implement a scheme like that would involve intercepting the existing Humax code and replacing it, instead of simply checking the current value in the PWM register and replacing it with a preset lower limit if too low. The Humax code is essentially a black box, undocumented and requiring laborious disassembly even to find the bit that actually runs the fan, and then analysis to work out what it's doing. The benefit returned for time spent wouldn't be justified. At the time the fan package was written, af123 wasn't disassembling the Humax code at all!

If you can hear your fan at 30%:
  • Stand the Humax on rubber feet to isolate it from a resonating shelf;
  • Turn the sound up;
  • Turn your hearing aid down.
If the fan is audible (to you) at a setting that does not stop the HDD having temperature excursions, there is nothing you can do anyway. At some point the fan will need to be increased above the inaudible threshold simply to maintain a reasonable temperature. Setting the minimum to (say) 10% will at least delay the onset compared with the default minimum of 0.
 
Black Hole said:
Setting the minimum to (say) 10% will at least delay the onset compared with the default minimum of 0.
Click to expand...
But it'll come on with an almighty roar when it does come on. Is not the purpose of the fan package to prevent that happening? (and to keep the box generally cooler than it would otherwise have been.)
 
zekepliskin said:
The idea would be to still have the fan settable between 0% and 100% but change the temp thresholds so the possibility of 100% fan speed is lower although the box would run a little hotter. A lot of people state you can't hear it on 30% but I can even though it's slight.
Click to expand...
It would be easier to control the fan from a Raspberry Pi, the RPi could get SMART data from the Humax and then control the fan in any way you fancy
 
Trev said:
But it'll come on with an almighty roar when it does come on. Is not the purpose of the fan package to prevent that happening? (and to keep the box generally cooler than it would otherwise have been.)
Click to expand...
Yes, it's a compromise between general audibility and distracting audibility. The greater general audibility you are prepared to accept, the less chance there is of the turbo starting up.

If there were a scheme to gradually increase the fan speed as the temperature went up, what that would achieve is to automatically find the best steady-state fan speed (for the current ambient temperature and usage) - which may or may not be audible. The performance would be no better* than setting that steady state level manually, except in adapting to the current circumstances rather than being fixed.

If an audible degree of fan is required to stop the unit overheating, it has to be audible - and that's that!

* And could be much worse. Control theory shows that it can be difficult to get such a system to settle properly. In simple terms: there is a large thermal lag, so bringing in the fan linearly will still allow the temperature to overshoot - so the fan overshoots, and the temperature falls (and overshoots the other way)... etc. This may be why Humax decided on their scheme - it's simple, therefore cheap, and it does what it needs to do.
 
Black Hole said:
it would come with its own set of problems (AKA hunting).
Click to expand...
And bl00dy annoying that is. I was working in an equipment room with a couple of (then) old rack mounted APC UPS units about 11 years ago and they nearly drove me mad as the period was quite short. Should've done a British Airways on 'em :)...
 
I'm not saying it can't be done from the Humax, but you would have to change (or add to), Humax's code and that doesn't appear to be an easy task, the suggestion was to disconnect the fan from Humax's control completly and control it from an RPi that gets HDD temperature data from the Humax, unlike the Humax software the RPi is fully documented, where do you see the problems?
 
It was implicit in Post #188. How else could you control the fan with any sort of external control without disconnecting it from the Hummy circuitry?

But why bother? The 'fan' package works just fine, and if you set it to a low level (quiet) and the box gets too hot, Hummy takes over and 'gives it some' to cool it down. Or does it fall into the 'because I can' category?
 
prpr said:
I can't see anything there that implies such a thing.
Click to expand...
Well, you have just agreed that you would have to disconnect the fan from the Hummy. (or didn't you mean that?). So post #188 implies that, as there is little option other than to disconnect it.
Now stop nit picking.
 
Ezra Pound said:
I'm not saying it can't be done from the Humax, but you would have to change (or add to), Humax's code and that doesn't appear to be an easy task, the suggestion was to disconnect the fan from Humax's control completly and control it from an RPi that gets HDD temperature data from the Humax, unlike the Humax software the RPi is fully documented, where do you see the problems?
Click to expand...

Thanks again to you and all who have made these boxes such good players.

I have HDR Fox T2 and like the idea of using the fan utility but I am a bit unclear if it will work or not. The notes on the package management page say "HDR only" yet on the previous page of this thread kev7387 talks about using it on an HDR Fox T2. Please can somebody clarify for me?

Chris
 
Yes it will. The terms 'HDR Fox T2' and the abbreviation 'HDR' are used interchangeably in the HD/HDR Fox T2 thread. The 'R' indicated that the box is a recorder. The HD Fox T2 is not a recorder(out of the box).

The Fan package does work on the HDR, so go ahead, install it and set the minspeed to about 40% and keep an eye on the tempmon. If the box gets too hot, the Humax default control takes over and cranks the fan up to 100%(ish)
 
You must log in or register to reply here.
Back
Top