OK, so looks like we are back to my initial question then. The gauge is a VDO and my thinking, call me wild and crazy, is that there must be a VDO sender to match the gauge. Here is a link to the VDO temp senders and their stated resistance.
http://www.vdo.net.au/se...ature-senders-switches/
320.002 is rated at 297 to 23 Ohms
320.003 is rated at 73 to 10 Ohms
320.021 is rated at 323 to 18 Ohms
These specs do not correlate well with the findings outlined in the graph from GM5735. EG VDO specs the 320.002 at 38.6Ω for 212deg F (100deg C) yet the graph has it just below 50Ω. However, the data does support GM’s observation that the 320.021 will result in a low temp reading on the gauge; a false sense of comfort if you like.
Testing from the gauge
My gauge reads from 100deg F to 260deg F, with 180 deg F noted in the middle and dashes at 120 and 220. Right, so what I did was work backwards from the gauge using a variable resistor in place of a sender and digital multimeter to see what the gauge needs to read the various temp markings on it. At 180deg F it needs 40.2Ω and 140deg F was achieved with 56.5Ω. 10Ω sent it off the high end and 128Ω sent it under the low end. What this shows us is that the sender starts with a high resistance and this reduces as the temperature rises.
The results
I plotted my gauge results together with the sender specs from the VDO site. Where I did not have a specific measurement for a given temp, I extrapolated a result using a linear change in the resistance at a given temp. Here is how it looks:
The results
For my gauge, it looks like the 320.003 sender provides the closest match to how my gauge reads. That is, the brown line (320.003 sender) is the closest match to the red line (the gauge).
Hope this makes sense and is of use to others. Let me know if I have made any errors in my testing/approach.
Edited by user Wednesday, 14 February 2018 3:40:47 PM(UTC)
| Reason: fixed a type