View Issue Details

IDProjectCategoryView StatusLast Update
0001869SpeedFanFan controlpublic2015-02-23 13:58
ReporterRykoshetAssigned Toalfredo 
PriorityhighSeveritymajorReproducibilityalways
Status acknowledgedResolutionopen 
PlatformDell Latitude E6520OSWindows 7OS Version64-bit Pro
Product Version4.45 beta 
Target VersionFixed in Version 
Summary0001869: Fan control can only be achieved by quickly incrementing fans up/down
DescriptionSetting the fan manually, even to 100% does not give high RPM, and keeps temps around 55-60*C. Setting it quickly between values, such as 80%->85%->80%->85%->80%..etc spins the fans much higher, and can drop the temperature below 50*C or maybe even lower (it is tedious work...)

To be fair, fan control DOES work, per se, but not nearly enough. Setting fans to 100% is faster than if there was no fan control, but not as fast as this fan can go. I have even tried setting the desired and warning temps very low, but the fan will not spin any faster.
Steps To ReproduceIf on DELL Latitude E6520, on the main screen, rapidly change fan values in any direction, and the fan will get much faster. Other settings don't seem to matter.
Additional InformationBIOS thermal control panel can NOT be activated in this model (via FN+SHIFT+1,5,3,2,4. FN+R) so that may play a part in why this happens.

If I can help further in anything, please let me know.
TagsDELL
Motherboard ModelDell E6520 motherboard for i7-2720qm
Video Card ModelNvidia NVS4200

Relationships

has duplicate 0002136 closedalfredo Fan control Not Fully working 

Activities

ranran

2011-09-14 16:05

reporter   ~0006453

I will confirm this behaviour. I have attempted the 4.44 and the new 4.45 beta 2 and I get identical behaviour.

If I simply click the manual fan adjustment up or down I can feel the fan start to spin, but as soon as I stop clicking, the computer regains control and the fan stops.

I have just donated to have this issue resolved. <Cue Star Wars music>, "Please, Alfredo, you're our only hope!" :)

alfredo

2011-09-24 12:40

manager   ~0006466

I have a private beta where I have completely rewritten DELL SMI support.
Consider that if the BIOS, the firmware or whatever, override software settings, there is little that can be done. But... If DELL wanted to prevent softwares from changing fan speeds, it would have been enough to completely disable those calls. The fact that they are still there makes me think that there is some unknown call that could disable any further "hidden" BIOS intervention. Just my 2c.

Rykoshet

2011-10-10 01:38

reporter   ~0006523

The Dell SMI problem was evaded in the e6510 with the SHIFT+FN+1,5,3,2,4; FN+R keystroke, but no such keystroke seems to bring up the dell thermal control panel on this model.

Please let me know if I can join the private beta, and if there is a minimum donation required let me know and I will email you a paypal confirmation.

my address is venustraphobia@gmail.com

Thanks.

zrajnai

2011-10-10 09:59

reporter   ~0006525

Last edited: 2011-10-10 09:59

View 2 revisions

I have a Dell e6410.
4.45 Beta 4. unfortunately does not fix this behavior.

Rykoshet

2011-10-10 15:15

reporter   ~0006532

zrajnai, see the combination of FN+SHIFT as described in the post above yours. That should work on E6X10 models to turn the BIOS control of the fan off, allowing speedfan to work without being overridden.

zrajnai

2011-10-24 14:27

reporter   ~0006597

rykoshet: the key combination does not work.
after SHIFT+FN+1,5,3,2,4 the scroll lock begins to flash, but FN+R does not bring up the dell thermal control panel.
(We checked with an E6400 in the office, and it works on that model.)

ranran

2011-10-24 15:39

reporter   ~0006600

Hi Alfredo,

Just curious if you've had time to look into this? How is that private beta working out - do you need more testers? If so, I'd be happy to check it out on my E6520.

Also, has anyone been able to access the thermal panel control?

alfredo

2011-10-25 10:49

manager   ~0006608

@ranran
Can you contact me by email (alfredo [at] almico.com)?

ranran

2013-07-31 15:17

reporter   ~0007264

Well, it's been quite a while, and in using 4.49, I see there has been no change. I can affect fan behaviour by quickly incrementing fan up and down, but control is immediately lost when I stop.

At this point, should I assume that this issue will not be resolved?

cr08

2013-08-21 00:56

reporter   ~0007271

Last edited: 2013-08-21 00:56

View 2 revisions

I had a small thought on this given prior experience with i9kfangui and trying to solve this issue myself (I would rather use i9k but the signed driver issue on 64 bit is a PITA). Using the shift+fn+1,5,3,2,4 scares me because it seems to me without actually testing that it would shut off native thermal controls completely and put it 100% on Speedfan. If it still has an emergency takeover when things get too hot it wouldn't bother me that bad and would love to know if this is the case.

What i9k does though that I can tell and without user intervention of shutting off the hardcoded controls is simply send a speed change to the system every couple seconds or so as the system seems to go every 5 seconds or so on its own.

Could a simple function like this be added to speedfan to 'brute force' fan speeds every couple seconds? Maybe as an addition to the Dell specific checkbox as this seems to be common across many Dell models.

BTW, I run an Inspiron E1505/6400. The shift+fn keystrokes work here as does Speedfan's Dell support and fan control. Just needs this small addition to make it actually work.

P.S. I have posted a similar comment on issue 0002136 which centers around the same issue and I assume a number of others are the same.

alfredo

2013-08-21 09:33

manager   ~0007272

Hello cr08,
Are you sure that reissuing the fan speed setting command every 2 seconds would work?

cr08

2013-08-22 13:37

reporter   ~0007273

In theory it should. That is pretty much how a couple of us here have been doing it by hand just diagnosing the issue. Essentially just toggling the PWM back and forth every couple seconds manually. More or less it is just the user and the hardcoded thermal functions fighting over control but in a passive way. The system seems to be doing it the same way and issuing a speed change to the fan periodically (or maybe more likely monitoring speeds at a couple second intervals and if not where it wants to be that moment, send a change) and likewise the user is doing the same for what they want. Not interfering with each other in a bad way, just repeated fan speed changes.

What I guess the common scenario would be for most people is wanting the fan to run faster than the system wants it to, maybe say full speed full time if running the CPU full bore. It will be running the fan full speed with intermittent split second slowdowns to what the system wants. And this is, from what I can tell, how i9kfangui does it with the same fan speed symptoms. Not perfect but gets to 98% of what is desired without going through the unsure process of turning off the system's thermal controls altogether.

lhassall

2015-02-23 13:58

reporter   ~0007773

I'd love for this issue to be finally resolved as poor fan control is even more of an issue with the latest CPUs.

I have a Dell Latitude E5440 and the fan control is just awful. Fan comes in at 50C and won't switch off until most temps are below 30C. This is impossibly low in a warm room, unless you leave the machine idle for 10 minutes. The thresholds are fixed and no playing with power options will change them.

Same behaviour as above with SpeedFan, altering the percentage temporarily alters the fan speed but almost immediately returns to a droning 3,000rpm. My impression is the system is constantly monitoring the fan speed and reacts when it changes?

In researching this I've seen many people complaining so there's demand for a solution if you can figure it out! Really strange that Dell markets high-end gaming machines and yet locks this stuff down.

i8kutils controls the fan under Linux, can anything be learned from the developer of that?

Lawrence

Issue History

Date Modified Username Field Change
2011-09-12 03:47 Rykoshet New Issue
2011-09-12 03:47 Rykoshet Status new => assigned
2011-09-12 03:47 Rykoshet Assigned To => alfredo
2011-09-14 16:05 ranran Note Added: 0006453
2011-09-24 12:40 alfredo Note Added: 0006466
2011-09-24 12:40 alfredo Status assigned => acknowledged
2011-10-10 01:38 Rykoshet Note Added: 0006523
2011-10-10 09:59 zrajnai Note Added: 0006525
2011-10-10 09:59 zrajnai Note Edited: 0006525 View Revisions
2011-10-10 15:15 Rykoshet Note Added: 0006532
2011-10-24 14:27 zrajnai Note Added: 0006597
2011-10-24 15:39 ranran Note Added: 0006600
2011-10-25 10:49 alfredo Note Added: 0006608
2013-07-31 15:17 ranran Note Added: 0007264
2013-08-21 00:56 cr08 Note Added: 0007271
2013-08-21 00:56 cr08 Note Edited: 0007271 View Revisions
2013-08-21 09:33 alfredo Note Added: 0007272
2013-08-21 09:35 alfredo Relationship added has duplicate 0002136
2013-08-22 13:37 cr08 Note Added: 0007273
2013-11-20 17:32 plamen2286 Tag Attached: DELL
2015-02-23 13:58 lhassall Note Added: 0007773