I'm curious if it would work to test the water thermosensor for resistance but at the ECU pin rather than the sensor itself?

Reason being, to test per FSM means removing it, draining coolant, etc. Is there any reason why I couldn't start with a cold engine to test resistance at the ECU and let the engine warm up while testing?

 

its fine to test @the ecu

 

You can test either way. No need to remove the sensor. The sensor test in the FSM assumes it's removed, whereas the ECU section assumes it's in. So you can just unplug the sensor and test it there, although getting to the pins is a little tough. Have a reliable way to measure the temp, like a proper thermometer. The gauge won't really do.

The engine should run ok, albeit not ideally, since the ECU won't have a temp. input while it's unplugged. The ECU will just default to a base temp. and maybe pop out a code.

 

I was thinking I would simply backprobe the ECU thermosensor pin with positive DMM probe and ground the neg probe. All this while leaving the harness plugged into the ECU. Start the engine and let it warm up while taking readings. I've got an aftermarket temp gauge with the sensor tapped into the front of water pump housing. It's literally almost touching the OEM sensors on the backside of the housing. So long as my gauge is accurate it's seeing the same temps as the thermosensor. I also have a temp gun I can shoot at the waterpump housing.

 

Yeah, doing that means removing the UIM which is another thing I'm trying to avoid.

 

edit: Re- backprobe at the ECU - You can do exactly that. How many readings does the FSM show @ various temps? Don't have the S5 FSM section on hand.

I know the S4 FSM provides more detail in the section when testing the sensor itself. At the ECU I believe it just shows a range.

 

It's 3 different temps, I don't have the FSM in front of me but the initial one is ridiculously cold. It was low enough that I'd think even ambient temps would have it warmer. Then it moves to 70 degrees-ish and then much closer to operating temps. This is all going on memory though.

 

I'm assuming S5 turbo water sensor (don't know if they differ) and ECU.

Since you don't have the manual on hand:

http://www.rotaryheads.com/PDF/2nd_g...tems_turbo.pdf

PDF doc. page 80, FSM page 78.

Pin 2E: water thermosensor shows two specs...idle, engine cold (no temp shown): 0.4-1.8V, water temp 20deg C/ 68 deg. F: approx 2.4V

ECU connected, backprobe readings with (+) and (-) grounded.

----

Sensor test...FSM page 81 same PDF doc.

3 temps shown for specs:

-20C (-4F): 16.2 +-1.6kOhm
+20C (68F): 2.5 +-0.2kOhm
80C (176F): 0.3 +-0.1kOhm

So I feel the sensor test gives you the better test of whether the sensor is good. But the ECU test shows what it sees, so it's up to you.

 

I had some broken logic about testing resistance at the ecu. My idea won't work cause I still need to get to the thermosensor harness to check resistance.

I have a couple spare thermosensors. My other options is to simply test one out of the engine and make sure it's within spec and replace...see if there is a difference in how the engine runs.

 

That's right, you don't get a resistance reading at the ECU, you get a voltage reading. I'd check those spare sensors you have, then replace it.

But I would still double check at the ECU after, to make sure the reading is correct.

What issues are you having with the engine?

 

My idle won't stay put and it's rough. I'll set idle to under 1k and when I come back another day the idle is back to just above 0. No matter if idle is set or not it sounds/feels like it's running rough.

I'll get the engine warmed up, I'll set idle to around 850 or so RPM. I'll get TPS set to 1V and I see that it's goes through it's range of 1-4.75V on narrow. Full also sees an unbroken sweep of .5something-3.8something V. The full is a little below spec by about .3v. I come back a couple days later and start it up after installing a wideband and idle is back to just over 0.

Wideband shows that it's idling a little rich at 10-8-11.8.

So I'm going through FSM list of priorities for rough idle and checking sensor on that checklist.

I've got a separate thread about timing...which is off on trailing side,
https://www.rx7club.com/2nd-generati...g-off-1034918/
...I've been told not to worry too much about the trailing timing. But through that thread I decided to start checking voltage at the ECU. A couple things are off like A/C relay. I was expecting to see that my TPS is off replaced it with a spare one, just in case. I even thought it was bad at first, but then saw I must not have had it backprobed correctly. When I went back another day to verify it tested just fine...and has tested just fine 2 other times I tested it since.

 

How can the engine possibly run w/the rpm just over 0 rpm? How about jumpering the bullet connectors found at each coil and the tach signal will be provided by the leading rather than the trailing coil. In doing so you might find the engine is running higher than previously indicated. And have you tested the output of the alternator as a low voltage output will cause idling problems.

 

Not at just above zero RPM, but a strong engine can run at rediculously low rpm. For example, when I lost my primary fuel injector on my rear rotor...it was running at starter motor rpm. I have proof!

 

Forgive me, by zero I meant whatever the bottom line indicates...maybe 250? 150?

I'll look into both the things you suggested.

 

Our engines sound very similar. And it never actually stalls. It's idling rich at 11.8 before I took it for a drive and 10.8 after I took it for a drive.