I hate to post, I really do but after a week or so of searching and some money spent, I figured I'd go ahead and see if any of you have had this issue before.

So the initial problem was that my 97 civic HX (with a DX d16y7 motor swap) was overheating. After replacing the radiator, flushing and refilling and getting the air out of the coolant, and replacing hoses and verifying that there were no coolant leaks, I was able to drop the temp gauge from red hot to about 3/4s during normal operation. Now for the past 3 years the owner of the car before me claimed that it always ran that way, and when he had it tested it was just the gauge that was broken.

Sure enough I plug up a scanner and I get a reading of around 190-210 degrees F during holds of around 3k RPM for a minute or so. To be safe, I replaced the old Coolant Temp Sensor (the one that sends info to the ECU), and tested - same result. I know there are two other sensors, the fan switch (on the thermostat housing) and the sender unit (beside the coolant temp sensor) that sends to the gauge. Pulling it out, it looks just fine, but I'll replace it soon. My mechanic said he highly doubts it has failed as it's just a lump of metal stuck in the engine sending varying electrical pulses to the gauge in the cluster.

The wiring looks solid, and the gauge isn't drastically dropping and raising, it's almost as if it's just not calibrated, where 3/4's on the gauge is equal to 180-190 degrees (which is about when the fan comes on at now).

Is 190-210 normal operating temp? I know it varies by thermostat and such, but does that sound off the chart to anyone?

Also, the thermostat and fan are fine, around 180 the thermostat opens and the fan comes on, but that's more than 3/4s on my temp gauge so I will have little clue when I actually overheat.

I have done my research by the way, i've just come to these stopping points and don't want to start spending money in the wrong direction. Really just wondering if anyone here has seen that problem before or has any insight from experience :D

Thank you all!

 

210 sounds a little bit high to me.

You can resistance test the sender unit. It's not that uncommon for them to go bad.

Does the temp gauge read cold when the engine is cold?

 

Yeah I thought that might be a problem, what should it be, close to no resistance right? EDIT: I didn't even look at that image, saw it and filtered it out thinking it was a sig. Thanks!

And yes it does, it will start just below cold (like on most civics), and then within about 15 seconds rise to C and within a minute will be around halfway, then slowly as the temp rises it will get up to 3/4s within about 3 minutes or so, and then stay there. It's almost like the whole gauge is reading 40 degrees higher than it really is.

 

If the resistance readings of the sending unit are fine, then the gauge might be the problem.

 

Awesome, I'll check all of that and see if the readings at the sender and at the gauge match to make sure wiring is good.

Thanks so much!

 

I finally got my multimeter in the mail today, and did some tests:

At dead cold it's around 180 ohms, I am not sure what "Low Engine Temp" exactly means (what degree range), but at around 70F (the temp the car is at after sitting for a few days), it's 180 ohms.


Now something I noticed was that when the fan came on, I got these two readings:
Engine running: 20 ohms
Engine off: 40 ohms


I noticed that it seems to be around a 1:2 ratio for these readings when the engine is on and when the engine is off (30 while on is ~55-60 while off. That might be normal (I figure since the alternator is running that would change it), but I figured I'd say it.

EDIT:
Let me go test it again and see what temp's I can pull while testing this.

 

The information I posted about the sending unit came from the Civic service manual.

Are you making the resistance measurements with the sending unit unplugged (not receiving voltage from the cluster)? That's how it should be done.

 

I ended up getting a manual too because of your post, I just wish they would have put the corresponding temps to the resistance in the manual haha.

Yes I am testing with the sending wire unplugged. But with the car on the resistance changes. At around 190 the resistance is ~20 with the car on and ~40 with the car off. What I am going to do is get a temp reading while I run these tests and see what I get.

What I do not know (because I have no known good to test), is about where the gauge is when the fan turns on around 180-190F (which is of course a decent operating temp), on mine it's around 3/4, but I assume it should be around half way? I wish these gauges had temp number markings on them so we know what half way is supposed to be.

Thanks!

 

With the sending unit unplugged, the Ohm readings should be the same with the engine on or off.

 

Ahhh, well there is a problem, good. Now I know where to really look. Thanks for all your help Ron!

 

210 is normal....thats about when my fan kicks on

BTW....here's the continuity test for the ECT sensor. Should be the same for the temp sending probe

 

As you can see in the diagram you posted, the fan should come on long before the engine temp reaches 210F. I live in Houston and the engine temp of my Civics rarely rises above 203F.

 

Well then, after running more tests I found that the sending unit works fine, at around 35-32 ohms the fan comes on, which is when the car is around 195-200F. A little bit hot but I think a flush and bleed will do some justice.

Regardless, the cluster's temp gauge is almost max red hot at 30ohms resistance from the sending unit (after testing it and then quickly plugging it back in and checking), which if others are claiming the fan kicks in around halfway then I know my cluster has to be the issue, that or wiring. I'm going to replace my cluster (or just the temp/fuel section if I can find one around here) and see if that does it, if not I guess a rewire will be in place.

The changing resistance based on if the engine is running or not is odd though... I can't wrap my head around how that even makes sense, less resistance when the engine runs... Maybe it's just because the engine heats up from heat sink after running? That would make sense actually, because it goes back down after it cools of course.

 

replace the gauge cluster

 

You could also just try replacing the temp gauge.

 

replace the head gasket

 

Haha funny.

 

Yea I see that but in practice my fan temp switch usually doesnt kick on til around 205-210ish....

My engine temp usually goes to slightly above 210. Ive seen it in the 220's while driving....Weird....

Although, I just recently put in OEM coolant....not sure if I've checked the temp since...