: Temp gauge dampened? Here's the answer:

07-28-10, 04:31 PM
I've been going through some old service manuals I have. This; due to my '97 STS having an overvoltage condition that blew out the PCM....

Section 8C1-8 of the 1996 Service Manual Book 1.


Coolant temperature is broadcast from the PCM via Class 2 serial data to the IPC. The IPC compares the temperature data with a sixteen-point calibration table to determine needle deflection. A smoothing filter is used to secure a continuous response.

This says it. Think about what the smoothing filter would do provided there was quick temp spikes or changes. Think about the fact that it uses a 16 point calibration table. One degree of temperature change would be the difference between the IPC commanding the needle to move to the next 16th position or not. This gives a lot of room for variances between cars. Very easily. So after a head gasket job is complete; if the vehicle ran one degree hotter or cooler; this one degree could be the difference between the guage needle moving past 12:00 or not.

This is out of the 1996 manual; but keep in mind Cadillac retained a lot of the same electronic engineering through many years and models.

I don't know if this has ever been found by another forum member or not; but if possible; Ranger; please Sticky this somewhere for future reference. This has been of topic for a while; and short of talking to the original engineers; this may be the most in-depth information on the subject we will ever come up with.

Also the smoothing filter; since we now know one is used; may be programmed different through the years and models. Doubtful but easily possible.

07-28-10, 10:19 PM
I'm going to leave this one here for discussion and copy one as a sticky.

07-28-10, 11:27 PM
My mind is made up; don't confuse me with facts.

07-29-10, 09:04 AM
Ok. Just thought I'd post what I found. I stumbled on it; was looking through the manual for theft system solutions. One fried PCM; just because the battery wasn't accepting charge and I revved the engine once. One quick, bright flash from the interior lights and the car died...

07-29-10, 02:25 PM
Jake - FWIW I just finished the third cold start-warmup during which I used the Scangauge to log coolant temp increase to normal.

The photoshopped gauge is correct - at least for the first 5 ticks (to 225).

My gauge stays within a needles' width of 12:00 - dead straight up - from 188 to 213. Up to 12:00 from cold (90 degree line), the needle moves in 2-degree increments, and at 213 it moves incrementally to the 225 line (12:45).

I deduce that the gauge - or its electrical signal - is highly damped in the 190-212 band.

07-29-10, 05:47 PM
Thanks for your research Jim! :thumbsup: This is, IMO with your test and the service manual; 100% proof about the dampening theory. I will use this reference when I need to help a customer understand differences in temp guage fluxuations and positions.

07-29-10, 09:26 PM
In my car, from 190 to 217 degrees, as displayed on my DIC, the temp gauge needle position is exactly the same: 12:00 straight up (plus a tiny, barely perceptible bit). 188 is noticeably lower/to the left, 219 is noticeably higher/to the right.