PDA

View Full Version : non-linear delta t


Flow
10-30-01, 04:00 PM
My second rig is running at a modest FSB overclock of 114 Mz, making 6.5 * 114 = 741Mhz at the default Vcore (it's a Duron 650). For cooling I use an unidentified TAISOL aluminium heatsink with stock fan. And you know what puzzles me? In short:

1) last summer: ambient temperature in my room: 32 °C (89.6F), CPU temp: 55 °C (131F)!
Delta= 55-32 = 23°C(73.4F)
2) Right now: ambient temperature in the room: 21°C (69.8F), CPU temp: 30°C (86F).
Delta= 30-21 = 9°C (48.2F)

Why is there such a big difference in delta temperatures? I mean, if I would have to write a review on a heatsink (or anything else concerning cooling)….. well, too bad it that were in summer :D Of course, in the mean time nothing has changed in my rig, and I use MBM 5 to measure temps.

JigPu
10-30-01, 04:32 PM
That's really odd...... Your delta t should be the same (or close to the same) no matter what the ambient temp is..... Mabey you cleaned your case from dust from then till now... That would have quite an impact on it.

JigPu

Warlord2
10-30-01, 06:15 PM
sometimes thurmal past takes a while to settle in it could of just been a reasult of a good burn it

I was at mobo 32c cpu 39c idle 45c load and thin in the next two weeks my delta T got really good

currently at mobo reads 27c cpu at 27c idle 34c load

some of it was from a cooling program but not that much

Hoot
10-30-01, 06:44 PM
When's the last time you took some compressed air to the insides of your PC, PSU and HSF? It makes a big difference keeping the dust off of things. Also, consider removing the HSF, cleaning the CPU and HSF base, reapplying a quality thermal compound and putting it back on. Preventive maintenance is time well spent.

Hoot

BillA
10-30-01, 09:28 PM
thermisters are notoriously non-linear
thermisters are used for "feel good" indicators

any "reviewer" using a thermister is conning their audience,
or perhaps just REALLY ignorant

to speak of a temperature, the device must be calibrated (somehow, someway)

YSI makes the only "linear" thermister (that I know of)
it's a patented dual (3 wire) model, Series 700, at $150 a pop
and I had to get 3 to find 2 that matched (to 0.01^C, WAY beyond the factory spec)

be cool

Flow
10-31-01, 08:05 AM
Originally posted by BillA
thermisters are notoriously non-linear
thermisters are used for "feel good" indicators

any "reviewer" using a thermister is conning their audience,
or perhaps just REALLY ignorant



I am not a reviewer, That part was just a joke!
(just for the REALLY ignorant ones).

I am fully aware of the limitations of a thermistor. That's why I added the info about how I measured the temps in the first place.
So guess I was NOT "conning the audience"

BillA
10-31-01, 08:24 AM
sorry Flow, I guess I misunderstood your question
as you are "fully aware of the limitations of a thermistor" . . . .
- it's clear (to me) that I still don't understand

no sweat
be cool

Flow
10-31-01, 02:57 PM
Ok, ok. I know thermistors are far from reliable, but these measurements were a bit odd. I mean cmon, a difference of 14°C !! I never saw it before....
About the cleaning and stuff: I maintain my puters on a regular base. I'm speaking of "summer" and "right now" as kind of average temps for longer periods. Long enough to rule out the cleaning.
Right now taking off the HSF and cleaning the cpu and heatsink base is the last thing I'm going to do, with this delta. Although I don't think it has anything to do with that.
And again: the review thing was a joke!
The reason I posted this question, like I said before, is that these values struck me as being well... extreme. And thus maybe beeing of other origin than from the thermistor/MBM variation.
Sorry, but after a long day of work I can be a bit...well...itchy :)
No hard feellings though