• Protip: Profile posts are public! Use Conversations to message other members privately. Everyone can see the content of a profile post.

Oil temp gauge sensor location questions

23 May 2020
SF Bay Area
I’ve been digging around thinking about where I could add an oil temp sensor. I’ve read a bunch of threads and the consensus seems to be: put it in a flow path or put it in the pan. Without an external oil cooler, there isn’t really an easy place to put it in the flow path.

The easiest place is at the pressure switch on the stock oil cooler. So my first question:

How bad will the reading be in this location? Would it just be slower to react because it’s just getting oil stuck in the tee? Or straight up wrong?

Or SoS sells this this oil adapter: https://www.scienceofspeed.com/inde...il-feed-oil-pressure-adaptor-nsx-1991-05.html but also not seeing much flow, just whatever oil gets stuck in that area. SoS told me I shouldn’t use this for temp, but put it in flow, or in the pan.

Question 2: how dangerous is this? Seems not ideal. It’s the setup on the car now. A bunch of brass tees seems like a break waiting to happen from vibration. It’s running the stock switch, an aftermarket pressure sensor, and what looks to be a temp sensor that was clipped.


From easy to harder, some solutions

* add a sensor where that clipped sensor is
* get a filter sandwich plate with some 1/8 npt ports, only run the stock switch at that location
* maybe that SoS adapter linked above
* a line to a manifold bolted somewhere that won’t break
* something else??

The sandwich plate seems like a reasonable option, then I could get rid of the brass tees. But there aren’t really any available that I found. Anyone know of a good one that has min 2 1/8 npt ports? (Not AN for oil cooler)
I would avoid the Rube Goldberg hardware. It just strikes me as additional failure points on something (oil flow path) that you really don't want to fail.

In my limited experience, oil temperature is a slower responding indicator of something happening in the engine that engine coolant. I personally would favor a sensor installed in the sump which would give an indication of average oil temperature after the oil has exited the 'the doing stuff' part of the engine. If you do not want to remove the pan to have a bung welded in for the sensor, I would consider a surface mount sensor. If you are using a thermocouple based sensor you could use a surface mount thermocouple on the outside surface of the pan. Cover it with insulation to prevent radiated heat from the exhaust skewing the measurement. If you are using an RTD, I don't know whether surface mount RTDs are available. Surface temperature will be lower than the temperature in the middle of the sump; but, once the engine and pan have come up to operating temperature and the oil has been sloshed around it should not be way off.

If you want to get creative, investigate whether it is feasible to have an adapter machined which substitutes for the oil drain plug and allows you to fit the oil temperature sensor in the drain hole. Its right at the bottom of the pan so the reading will be slightly lower than the center of the sump. It will also make oil changes a bit of a hassle because of the need to disconnect the sensor prior to draining.

I personally like the surface mount sensor because it requires no mechanical modification to anything. In the worst case if the bonding agent for the sensor should fail the sensor will hang loose and you will just read suspiciously low temperatures.

My two bits worth of blue sky engineering!
Last edited:
Interesting idea on the surface mount sensor. I have been looking for a place to use one of these: https://rifesensors.com/product/surface-mount-temperature-sensors/

I do actually have a pan out that I’ll install soon. It’s already got a baffle and has been powdercoated. Maybe I could send it back to crf and trade him one with a bung.

I agree the oil drain plug is not ideal. I’m also not super keen on a sensor hanging off the oil pan.

I like the surface mount idea. Though it feels like it might be slower to react than a sensor in the oil, and I’d still need to do something about my current Rube Goldberg setup near the filter..
The sensors you linked look like they still require some mechanical modification to attach them. I was thinking something more like this

TC-SA-K Type K Surface Thermocouple with Adhesive Backing – IOThrifty

I assume that the temperature range listed in the specs is for the adhesive because type K thermocouples with #30 gauge wire are typically good for over 1000 F. You would need to investigate the solvent resistance of the adhesive although if you cover it with some heat reflective insulation that might be less of an issue. The devices you linked look like they may be RTD sensors and if your display / logger is set up for an RTD a thermocouple will not be an option without an elaborate interface circuit.
Well, I made some decisions and have a path forward. I took the advice from both of you and will be measuring temp from the pan. The only sandwich plate I could find that might work is this one from Mishimoto. But I would block off the I/O ports and only use it for the sensor. I was worried it might not work or restrict flow or something given it is designed for an oil cooler setup. I'll cross that bridge if I get there. So:

* I sold my CRF pan waiting to go on, and ordered another with a temp sensor bung
* I'll run an aftermarket pressure sensor from a combination of the SoS relocation kit + a manifold I found. I dont have it 100% figured out, but based on all the measurements so far, I can make it work with easy install + have a nice mounting setup
* I'll replace the nonsense at the filter with a new pressure switch

The thermocouple is an interesting idea, but it definitely would need an external box to convert to 0-5v. A thermistor is easier to wire in, especially if the ECU input has a pull-up resistor already baked in
Last edited:
The sender included with the ScienceOfSpeed relocation kit kind of sucks, mine stopped working after a year or so and just sat at zero which was kind of scary until I was able to hook up my AEM sensor and confirm I still had oil pressure. I know of a few others who've had similar issues....if you can, it would probably be better to just relocate the factory sender to the firewall using the SoS hose if its still working. IIRC the stock sender is an M10x1 thread with an O-ring but you could use any M10x1 fitting with a crush washer, and the SoS sender is a 1/8" NPT.
Last edited:
[MENTION=18194]Honcho[/MENTION], what temp sensor are you using?

Get rid of the T, it won't be accurate or responsive. But where you have it now should be pre oil cooler.

Yep the plan now is to get rid of the tees and run only a factory switch there, then a sensor in the pan. I’m not going to run an oil cooler yet. I’m curious what the temps will be, then next year when I get it on the track I’ll see if it’s necessary
[MENTION=18194]Honcho[/MENTION], what temp sensor are you using?


If you go with an oil cooler, do the Driving Ambition/Le Mans GT1 style in the passenger air scoop. [MENTION=20915]RYU[/MENTION] or Shad can help with placement, plumbing, etc.
Yeah that seems like the move over the rear mounted SoS unit. And seems like a pretty simple system without an accusump. I was looking at the mishimoto 10 row coolers … and I’d probably want to run a thermostatic sandwich plate. There are some pretty good threads with RYU and illwillem on the topic, so have been reading those

This is a pretty deep “well, while I’m in there” rabbit hole, so imma stop at the sensor changes and heat shield + baffle on the pan (ahem, for now) :). At least then I’ll have some baseline data. I got a bunch of parts coming in the next week or so while I wait for the new pan, so I should have a rough draft together for the sensor changes soon. The pan was next up, but this shifted things…