|
2004-11-16 05:16
|
fan doubling.gif (14,969 bytes) |
|
|
Did you try CONFIG / ADVANCED and a higher FAN DIVISOR? |
|
|
At 1 or 2, it reads 0/double, and at anything >= 4 it reads 1x/2x. I have it set at '4' by default. |
|
|
My own computer has got a W83627HF. This hardware monitor is one of the best supported by SpeedFan. It never made a mistake. I'm afraid this is a fault on the fan itself which, sometimes, doesn't send the proper revolution signal. In 5 years you are the first one to report such an odd situation. It appears to be more likely to be an issue with your configuration rather than with W83627HF support. |
|
|
Every other tool that I have used to monitor RPMs, SpeedFan is the only one that causes this doubling. In the BIOS, it's a perfectly stable value, in SilenTek, AOpen's utility, it reads stable as well. MBM5 also consistantly reads the correct value. So, I don't think it's a hardware-only issue. |
|
|
How often do you get improper readings? Might it be you get them when you are running some other hardware monitoring tool? |
|
|
SpeedFan is the only monitoring tool that I have run for about 4 months now ;) The improper readings are there 100% of the time. |
|
|
This issue is driving me crazy :-) Did you try with a different fan? |
|
|
Fans from other manufacturers are just fine. I have only seen this with Panaflo fans. It's not as common now as it used to be, but it used to also randomly report a fan speed of 300,000 RPMs for one single sample. This of course would make the graphs useless for 15 minutes until that data point was off the graph (or close/reopen SpeedFan). |
|
2004-12-02 05:26
|
big spike.gif (11,521 bytes) |
|
2004-12-02 05:33
|
fan doubling2.gif (18,781 bytes) |
|
|
I just added 2 more pictures. One is a 350,000 RPM spike, the other is all 3 Panaflo fans (CPU, Case, modded PSU). The PSU's fan readings are much more stable, but still show spikes. |
|
|
I have never played with the PWM Clock to 'fix' this, until tonight. Default was 23KHz. Dropped it to 11.7 and no change. Raised it to 47KHz and it smoothed right out. Now, to the question: I don't know what this is changing. Can it cause any physical damage, either to the mboard, the chip, or the fans? I'm leaving it this way for now, but would appreciate any insight into what this is doing. |
|
|
I thought you were getting odd results at 100%. You can find further info on PWM control here: http://www.almico.com/sfarticle.php?id=1 |
- Anonymous
