T_ypEr

Does anyone know the engine temperature of a 92-95 D15B7 that kicks on the radiator fan? My dad and I have found that the car has a to be a little bit under maxed out on the gauge for the fan to kick on. We have been doing with the car stationary in our garage (with a fan pointing at the muffler to not asphyxiate ourselves). If the car were moving, would the temp gauge stay at normal operating temperature?

We have tested EVERYTHING you can possibly test. LITERALLY. All electronics related to the radiator fan, gauge, and fan motor all check out, relays and sensors and all. We did a heat test with water on both the thermostat and the thermoswitch and they all work.

Wdpfreak145

A sensor is bad. It has to be. There is a threshhold at a point in the temperature range where the switch is completed and the voltage flows to the fan. Check again, if you asks me.

s13hero

203F 95 C I believe, are you sure your gauge is reading correctly?

Former User

You risk destroying you engine by letting it get this hot^. In general never let the engine get hotter than the halfway mark on the gauge.

If you have thoroughly bled the cooling system, then the next step would be to replace the fan switch located on the thermostat housing.

T_ypEr

Thanks for the temperature. I'm sure it's correct because we just replaced the coolant temperature sending unit that sends info straight to the gauge.

My bad, I got the switch messed up. The switch we tested was the fan switch. Like I said, we did a fool-proof test on that switch and it checked out. The procedure was to suspend the switch in a pot of water on the stove and take it out right before the surge of bubbles from the water (boiling) and see if the switch then closes. We did this test and it did open and close.

Former User

So have you bled the cooling system properly?

Give some details about your overheating problem.

Wdpfreak145

Water boils at 212 degrees Fahrenheit. If your switch opens and closes around this temperature, that wod mean that it is opening and closing at the approximate operating temp of your motor. If you replaced the sending unit already, it would seem to me that the gauge is, well... improperly gauged.

T_ypEr

We haven't really driven the car yet, so this is all done stationary. We basically keep the tach at 1500-2000rpms for a while and wait for the temp. needle to go up. At about a 1/4 of an inch from overheating the fan comes on.

T_ypEr

What could cause a gauge inaccuracy?

s13hero

I'm confused about this, you boiled the fan switch? As far as I know you can see when it opens or closes.

T_ypEr

We pulled out the switch just before the water began to boil. And yes, we did see it open and close.

Wdpfreak145

Wiring discrepancy, too much torque on the back screws, etc. Id recommend swapping it with another eg cluster. They are very abundant.

T_ypEr

We're going to swap the thermoswitch on the thermostat housing with the one in my 2000 CR-V (B20Z2) since it's the same switch, and we'll see if that does anything. If the switch really is bad, this should tell us.

Former User

Bleeding the cooling system may take 20-30 minutes:

1) Park the car on an inclined driveway with the front end higher than the rear.

2) Push the dash heater lever/**** to MAX heat.

3) Follow the directions in the diagram below (hint: the bleed bolt is located where the upper radiator hose connects to the engine):



During the procedure, keep an eye on the dash temp gauge. Turn the engine off if the reading ever rises above the halfway mark.

T_ypEr

The problem I have with this picture: "the radiator cooling fan comes on at least twice". The radiator fan didn't come on at all the last time my dad bled the coolant.

Former User

Just do the procedure for at least 20 minutes exactly as described in my post while carefully watching the temp gauge in the cluster. Terminate the procedure immediately if the temp rises above the halfway mark.

T_ypEr

Ok, we'll give it a shot.

T_ypEr

So we switched the thermoswitch with the one in my 2000 CR-V and the fan kicked on when the gauge was about 3/4 of the way up. We took it for a test drive to blow out the coolant and the gauge was normal the entire time. But when we were sitting stationary the gauge would slowly, but surely, go up. We're thinking about bleeding the coolant some more (since coolant escapes when the thermoswitch is removed) and maybe even changing the thermostat itself just to be safe. Any thoughts?

T_ypEr

If the temperature gets above the halfway mark and we have to terminate, then what do we do to bleed the coolant?

Wdpfreak145

Wait on the engine to cool down, at bottom mark, then bleed until a) the temperature reaches the appropriate range and the bubbles have stopped or b) the temperature goes above the appropriate range and you terminate the engine immediately.

T_ypEr

Sounds good, we'll bleed the coolant this afternoon and post the results. We're also going to get another fan switch and thermostat (better safe than sorry).

But as for what happened when we used my CR-V's fan switch, I have a theory. My 2000 CR-V has a B20Z2, which is a DOHC 2L engine with 146hp and 123 ft-lbf of torque. The D15B7 in the '92 Civic we're working on is only a SOHC 1.5L engine with 102hp and 98 ft-lbf of torque (when it was new, at least). My theory is that the B20Z2 is designed to handle more power and heat, and that's why the fan came on a bit late while it's fan switch was in the Civic.
If I'm wrong, please let me know!

Wdpfreak145

If that were true the parts wouldnt be interchangeable.

Rayzorx

Hi,
I'm helping T_ypEr with this process, and here's the latest.
We replaced the fan switch on the thermostat housing, as well as the thermostat, this time with a genuine Honda part. The Pep Boys thermostat we had previously installed was bad. (the clip had come off of the bottom of the spring)

So we did the bleeding process as described above, and ran at idle with the radiator cap on 1st. stop. At 15 minutes the temp gauge started to go past 1/2 way, with no fan coming on. We terminated the test and are now waiting for the engine to cool.

I did notice that when I took the radiator cap off and there was a bit of air that escaped, so maybe some air came out of the system?
I repeated the steps of loosening the bleeder, filling with coolant until it streamed out. Didn't see any bubbles. Tightened the bleeder and filled with more coolant to the radiator neck.

We'll try again when the temp is at bottom and see if the results change.
If we get the same results, is there something else we should check?

Rayzorx

Success!!
We did the test a second time and the fan kicks on just before the temp gauge gets to half way. Just to be sure, we cooled the engine and did the test a third time. Same results, the fan kicks on just before half.

Thanks so much to everyone on this forum for providing us with accurate and much needed help with diagnosing this problem. Your help is greatly appreciated.

Ray