Toyota Nation Forum banner

Faulty Water Temp Sensor - Causing Problems?

37458 Views 13 Replies 5 Participants Last post by  wasmit13
My buddy has a '90 4runner V6 and it has been having issues lately. Sometimes it will run fine and other times it will act power, missing, rpms drop...just complete crap. Sometimes it does it right away at startup and sometimes it does it out of no where when the engine is warmed up. Once the problem starts the only solution is to stop and come back to it and it will run fine. Also, he says it smells like its running rich...though this could be from the missing too...

I check for codes and the only one I pulled off the computer was code 22...water temperature sensor circuit. I asked him if his gage has been working and he said it comes on and off once in a while. I don't know if its going on and off corresponds to the engine running bad or not.

So, my question is: Would a faulty water temp sensor (or ciruit) confuse the computer and make it run crappy like this? I thought it might be a the comptuter is thinking the truck is cold when its really not, etc.

Any advice or suggestions would be great. Thanks!
1 - 14 of 14 Posts
Yes, coolant temp is a major parameter in the EFI system.
The code means the coolant temp sensor circuit is either open or shorted. Could be as simple as a bad connection, but most likely the sensor is bad. You can test it with an ohmmeter.

Check the factory manual, ENGINE - MFI SYSTEM - ECT SENSOR section. If you don't have an fsm, google 1993 toyota pickup service manual.

The ECT is one of many sensors on an aluminum appendage ("coolant block") on the top of the motor in the back - just in front of the firewall and half under the plenum. The 93 collection of sensors is slightly different than the 90, but the position of the ECT is the same. The ECT is on the driver's side, in the back row. There's another sensor a little further left, but that one is in the forward row: you don't want that one. If you get the new ECT first, you'll be able to positively identify the old one, because fortunately all the sensors on that block look different.

If the ECT is bad, it must be replaced in order for the motor to run right. Be careful when you order it to call it by its right name: engine coolant temp sensor (ECT). There's a coolant temperature sender too (for the gauge) and you don't want that one. The naming of them does vary in the catalogs of various parts suppliers, but USUALLY the ECT is called a sensor and the temp gauge sender is called a sender. They look different: the ECT has a plastic connector on top:

...and the gauge sender just has a metal tab:

The other way to know the difference is the price: the ECT costs $20-40 more than the sender.

Safest bet is a sensor from Toyota - probably has the best price, but it will still be more than aftermarket. Half the time, I can't find the part I'm looking for in their catalog, so I just call them and ask for the internet parts sales guy (forget his name, but he knows his stuff). sells an ECT that is probably Japanese-made and could well be the same part the dealer sells - cost is $51. They call it the "Water Temp. Sensor" and it's in the Fuel Injection section of their catalog. They also sell an o-ring for it but the o-ring may come with the sensor - ask them.

The cheapest ECT I would take a chance on would be the Beck/Arnley from They want $31 for it. Many of the B/A parts for our cars come from Japanese manufacturers, but not all. At $31, this one probably isn't Japanese-made. Rock Auto calls it the "Coolant Temperature Sensor" and it's in the Cooling System section. Both AutoHausAZ and Rock Auto are excellent, trustworthy vendors - I've bought parts from both of them many times.

If it were me, if the dealer part from 1sttoyotaparts was too much $$, I'd pony up for the NTC from autohausaz.

While you're helping your buddy with sensors, when is the last time he replaced the oxygen sensor? If it's over 90k miles old, it also needs to be replaced - over time it gets coated with carbon and starts to read leaner than the exhaust really is. As a result, the computer runs the motor too rich. The computer has no way to know if the readings of the O2 sensor are inaccurate - it only knows if the sensor fails completely or if it is really, really bad. sells Denso's (the best bet) at a good price, and NGK's (if $$ is tight). You can also get the Denso part number from the sparkplugs catalog and search for it on Amazon - I got myself a brand-spankin new Denso for just over $50 that way.

Good luck.
See less See more
Thank you for all the great info! That's good to hear. Hopefully that will fix his problem. I'll check the sensor with an ohmmeter (I take it I'm just looking for an open circuit in the sensor...not a specified resistance range correct?).

I just happen to live close to the dealership that runs so I can order it and pick it up (done that a few times for some parts). I did do some searching and the only part on the website that had "water" and "sensor" in it was a part called: "Air intake temp sensor, pick-up, land cruiser, van, water temperature" for $ that the right one? Kind of an odd part name...I'll have to give them a call to be sure.

So let me get this straight...there are two water temperature sensors? One for the gauge and one for the computer? That's good to know...don't want to end up replacing the wrong one.

Good call on the O2 sensor. He's had the truck for 50k miles and has never replaced it...I'm sure the sensor is way over its expected lifetime.

Thanks again for the help! I'll report back with our progress :thumbsup:
See less See more
Air intake temp sensor is in the air meter - different animal. I would call them.

You are looking for ECT resistance to be in correct range both hot and cold. See the referenced fsm section.
Can someone with a service manual for a 1990 4runner (V6) verify that this is the correct chart for checking the ECT?

I found it online but it said it was for a '95. Would they be the same?

See less See more
I checked the fsm for my 89 truck (which should be identical to a 90 runner), and the chart is nearly identical to the 93/95 chart in the center of the temp range, but the range is a little different at the low temp end, and at the high temp end the allowable range of resistance is a little tighter.

At the low end, at -20C (-4F) the 89 chart shows 14-20 k. At 10C (50F) the range is about 3-5 k ohms.

At 20C (68F) the range is 2-3 k ohms - same as the 93/95.

At 40C (104F) the range is .9k (900 ohms) to 1.5k At 60C (140F) it's .5k (500 ohms) to a hair over .7k (maybe 720 ohms).

At 80C (176F) the 89 range is .3 k (or 300 ohms) to .4 k (400 ohms). At 100C (212F) it's about 150 to 250 ohms. At 120C (248F) it's 100 - 150 ohms.
Thanks for checking on that for me. I'm going to see if we can get it running this weekend. Hopefully the sensor will fail the test completely so we won't have any question if its the sensor or not. Also, I found the 'Niehoff Ignition' sensor from Schucks for $ he'll probably go with that since its stocked at the local store and he needs his rig running.

Thanks again
Here's an update:

I removed his sensor yesterday to check it with my multimeter. It tested within the correct range after I initially pulled it out, so it wasn't looking good for finding out the problem. To be sure, I took it inside and stuck it in hot water and took a reading. Nothing. Put it in the freezer and still couldn't get a reading. After it sat around for a while I was able to get a reading off it until I warmed it up again. So thats good! Bad sensor. We picked up a new one and I performed the same tests on both sensors one more time to be sure. Anways, so that is one problem fixed.

I cleared the computer of codes and then we started it up. When it was sitting there idling, I tried giving it a little gas. You could increase the revs fine if you go easy on the throttle, but if you punched it at all the revs would drop way down and the engine would want to die would recover though after a few seconds. Well crap I didn't know what to think. But as soon as the thermostat opened up (I could see the temp gage go right up to temp), the revs dropped down to the normal idle. I tried giving it some gas again and it had perfectly fine throttle response. No dying out or anything.

So it seems that the motor only ran crappy when it was in open loop mode. I told him to let it sit and start it up cold this morning to see if it did the same thing. I haven't talked to him yet. Since this was the initial startup after it ran bad before, could that have anything to do with it? I thought maybe it was a computer problem still trying to recover from the last problem...I dunno.

But we got the sensor fixed for sure, hopefully that will get him back on the road for a little while.

Thanks for all the help.

PS. What is the temperature sensor that is next to the ECT? I accidently pulled that one out first because it has the same exact plug as the ECT but the threads were bigger and the temperature probe is larger. It doesn't look like the temp sender because of the plug. I was just curious to what it is. (it registered around 80 ohms at the outside temp).
See less See more
Congrats on getting the sensor replaced, and thanks for posting the outcome.

The stumbling that happens when the motor is cold and the throttle is opened very quickly is something my engine has always done. I've checked everything, and I think it's just a shortcoming of the ecu's programming when in open loop. Apparently the later years don't do that, so I can only guess that they got the ecu dialed in a little better after a few years. On mine, it's no problem if I'm gentle on the throttle, and it's no problem after a few minutes when the motor reaches operating temp.

You might check the tps and vafm, and check for vacuum leaks, and make sure the throttle body is clean and the plugs, wires, cap, rotor & air filter are good, and timing is set correctly. That should probably be done anyway, but if the motor acts like my 89, it will still stumble when the throttle is opened too quickly when in open loop.

I forget which sensor (other than the obviously different and smaller temp gauge sender) is next to the ECT - could it be the cold start injector time switch??
See less See more
Yeah I'll have to give his motor a once over for vacuum connections and such. His motor probably deserves a good tune up (plugs, wire, cap/rotor, etc). Those V6 are a pain to work on lol well compared to my little 22RE I am used to :D

From the sounds of it the motor never hesitated like that before this problem started happening.

I think you are right about the cold start injector time switch...I did a few searches and people said it was right next to the ECT.

Ran across this page:

The person had the same problem and it ended up being the CSI. I'm going to refrain from worrying about that until everything else is checked out. At least the motor runs fine once its goes to closed loop.
According to all the information I can find, the cold start injector usually only fires for a few seconds, and never more than 15 seconds. (And 15 seconds is when it's like -30F). The minimum fire time is three seconds (except when motor is warm - then it doesn't fire), and over freezing it probably never fires for longer than like 5-6 seconds.

Maybe the engine would run better if it DID fire for longer, but I don't think it's designed to do that, so I don't think it factors in to the stumble-when-cold problem.
could the coolent temp sensor keep my truck from kicking into overdrive
Yes it will absolutely. The thermostat will as well. Check codes and then check coolant temp. If it's not getting to half way it will not allow o/d to work.
1 - 14 of 14 Posts