When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
I apologize if this has been beat to death, but I've been scrambling all day, have to head back out in a minute, and need the car for the morning.
99 Civic EX, stock D16Y8.
Yesterday on my commute home I noticed the temp gauge 3/4 up. I got my OBD adapter out and TouchScan showed ~235*. I pulled off and was able to limp the car home in sections, trying to keep temps down. I got the car home, and there was coolant sprayed around the radiator. (No smoke, no oil/coolant mix, no loss of power by the way. I assume I got a little lucky here, since I really don't know how long the car was running that hot originally.)
The fan works when connected directly to 12V. When I jumped the thermoswitch connector, the fan did turn on. Based on this, I bought a new radiator (old one was seeping a bit anyway) and thermoswitch (this one, to be exact). I put everything back in, burped the system a bit, and waited until it heated up. At the 203* mark I figured the fan wasn't turning on so I shut it down.
I'm going to pick up another switch while I'm out in case the one I have is defective, but is there anything else I need to check? I assume that since the jumped connector engages the fan, the relay and such are ruled out. If I'm missing something, please let me know.
I grabbed another sensor because the sensor I bought originally had a connector that fit a little oddly.
I grabbed a BWD sensor, put it in, same thing.
I then attempted to bleed the system again, by jacking up the front of the car as high as my jack would go, loosening the cap, setting the heat to max, and running the car.
The temp wouldn't go above 172* like that (~10-15 minutes of idling at roughly 172,), so I closed the hood. It still didn't go up much, so I turned the heat off for a little. It started to rise, and eventually crept to 200. Every time I turned the heat back on, however, it would drop significantly. At no point during any of this did the fan turn on. The cap was new and came with the radiator, but I did notice some steam during all of this, and some coolant bubbled over and out of the radiator. I really didn't have to add any, as the level didn't really drop significantly at any point.
The Honda fan switch turns on the fan somewhere between 196-203F, so turn off the heater and let the engine temp rise higher to see if the fan turns on in the specified temp range.
Honestly, it may have crept to 203* or so before I shut it back down. I know that the first time, I let it run up to 205*.
Is there anything else I should check here? Later today I'll put the cap back on and let it run up again and see what happens, but if the fan still doesn't engage, I'm at a loss. The lower rad hose got hot, and since closing the circuit via paper clip engages the fan I assume everything between there and the switch works correctly.
Well, the cooling system is nothing if not efficient.
I idled the car today and couldn't get it passed ~196* or so (66* outside temp). Drove it around, and the temp dropped to 190ish. I found some hills, and was able to get it to 203*, but without being stopped I really couldn't tell if the fan was coming on (I was trying to hit higher temps than that.) I then kicked the AC on and let it idle. It eventually hit 207* via TouchScan, but no primary fan movement. After a moment it dropped back to to ~203-205*, where it hovered until I just put it back in the garage.
Where does the ECU pull temp from? Is it possible that the location it's looking to for the reading is just a little higher than what the thermoswitch is seeing, which is why it's not kicking on? If the coolant near the switch actually hit 207*, then I'm not sure why the fan didn't engage.
Where does the ECU pull temp from? Is it possible that the location it's looking to for the reading is just a little higher than what the thermoswitch is seeing, which is why it's not kicking on?
The Engine Coolant Temperature (ECT) sensor (located below the distributor) measures coolant temperature at the cylinder head, that is where the ECU gets the input from. The ECT sensor is near the outlet attached to the upper radiator hose where the coolant goes to the radiator.
The Engine Coolant Temperature (fan) switch is located at the thermostat housing. The side where the lower radiator hose connects; so, it measures coolant temperature coming back from the radiator. The thermostat itself responds to coolant temperature on the other side of the thermostat housing (where recirculating coolant is coming from the head and mixes with coolant returning from the radiator).
The ECT sensor and ECT switch are measuring coolant at different areas.
When I jumped the thermoswitch connector, the fan did turn on....I assume that since the jumped connector engages the fan, the relay and such are ruled out.
Originally Posted by egsleepercivic
unplug the fan switch on the thermostat, jump the connector with a paper clip
does the fan turn on now ?, if not bad relay or wiring
Bumping this back up, although the problem is slightly different. I just didn't see the need to create a new thread, but can if necessary.
zturek, sorry, missed your request for an update.
I can't quite remember the timeline here, but I eventually replaced the thermoswitch and the radiator/cap (mine was seeping slightly, didn't catch that at first.) Bled the system, all seemed well.
Fast forward to maybe 6 weeks ago or so, and it started doing it again. I replaced the thermostat since it was under warranty anyway, reassembled and bled the system. Drove around, seemed fine. The next day, it started overheating again, but this time I noticed that the lower radiator hose wasn't getting warm, so I again bled the system, as best I could - jacked the nose of the car as high as I could get it, topped off the system, cracked the cap loose, and let the car run for 20 min. It never overheated, although it bubbled some coolant over. Lower radiator hose got hot, fan was turning on as usual, all seemed well.
Drove fine for another month or so, and then recently it started doing it again. The odd thing is that it seems random. It did it a few days ago very briefly, and then was fine driving ~44 miles round trip to work including highway and stop and go traffic up until today. Coming home from work, it started acting up. While moving, the temps fluctuated from 207-230, never staying still (as per TouchScan and my bluetooth OBD reader.) Sitting still, temps would creep up and I'd have to turn the heater on blast to drop it. A note point here - once I got home, I popped the hood and while it was getting hotter, the fan did engage and stay on. The fan system isn't the problem here.
Any idea what gives here? Parts have been replaced so it seems like air in the system, but as far as I can tell I've bled the system properly, several times, and I don't get the randomness. Is there a better way to bleed it? Is it worth flushing the system or something?
So I took the car back out tonight. Drove it a few miles out, drove back home and let it idle with the AC running to see if I could get it to overheat, and it hovered in the 203-207 range. No issues. I'll update again tomorrow, but this is sort of what it does. Overheats "randomly" and then won't show symptoms for weeks.
seems to be an annoying sensor to me.. when the temperature rises up have u tried actually seing if the engine feels hotter to you?
imo i would do it all step by step. (i know you have already done some of them)
1- check any leaks wiv those pressure pumps.
2-thermostat
3-Bleeding. sometimes when bleeding air gets caught out on the system which prevents coolant from circulating and doin its job.
4- check if the fan is actually working by connecting it directly. check all relays fuses etc
If none those worked.
then you should probly check headgasket , water pump+tensioner+belt , and any seals/bolts related to the water/coolant circuit
I'd agree with a wonky sensor sending false signals to the ECU, but the coolant overflowtank has, in the past, spit out coolant. It didn't show obvious signs of this today, but it does seem to be overheating to some degree. Compression is still fine, no coolant/oil mixing, no power loss, no smoke. Car still routinely gets 30+ MPG mixed
Pressure test the cooling system, and also rent a block tester kit to test for exhaust gases in the coolant. Blown head gaskets can be tricky to diagnose.
Grabbed both tools, although I didn't realize that the block tester didn't come with test fluid until I got home, so that'll have to wait. The system did hold 15PSI rock solid, though.
The car ended up misfiring pretty badly recently, and after limping it home (this car has never left me stranded, even when battered and bruised) I found a pretty fouled up (kind of green) #3 spark plug (12k miles on these, give or take.)
I finally got my hands on some block tester fluid, and surprise surprise, it turned a little green. At least a noticeable green compared to the dark blue it's supposed to be.
So, I yanked the head off and replaced the head gasket/bolts/t-belt/tensioner/water pump. The old head gasket looked fine and the head was awfully flat, but the one head bolt seemed to require less effort to crack loose. I don't know if that matters or not. I did notice a very very slight scratch-mark on the one cylinder wall (#2) that looked like a crack, but I couldn't feel it and couldn't see it from the water jacket side, so I'm chalking it up to a scratch. I wasn't interested in yanking the block, anyway.
Flush some things out, reassemble, fire it up, and it's idling at 1500RPM and is heating up (~220*). Let it sit overnight, disconnect the battery for 10 min, lift the nose of the car in the air to bleed it, and fire it back up. This time it actually starts to idle down, sounds smooth as usual, revs fine, fans are kicking on and the temp is normal, etc. I assume it's bled, but I'm skeptical.
I honestly don't know if the headgasket was the problem, but I'm not interested in rebuilding/replacing the motor, so this is basically the last of the work. If it stops overheating I'll continue to drive it, but if in a few weeks it overheats again I'll probably unload it.
A fast question about the timing belt - when replacing it, I did it Honda's way where I had it set to TDC, rotated the crank pulley ~3 teeth, and while holding the wrench on the crank pulley I tightened down the tensioner pulley. I did this probably 10 times, because it just didn't seem tight enough to me (based on other cars I've worked on), but after a while I figured this is what's required and just buttoned it back up. When I got the car running, it seems like there is a slight whine. Honestly, I haven't driven the car in a week and don't really recall if it used to "whine" or not, so maybe it's a normal noise and I'm overthinking this, but can you even over-tighten the belt when doing it Honda's way? I wasn't prying the tension pulley up or anything, so I honestly don't understand how I could've over-tightened anything.
Thoughts?
Edit: To answer axis11, I don't think the car was burning anything. When it would overheat, I usually ended up adding a tiny amount of coolant, but I just assumed it was being pushed out of the overflow/around the radiator cap, based on the residue left over.
That whining is commin when changing the timing belt..
in my case when i did it. i used a toda racing timing belt which is a bit thicker/stronger than the ova ones. It did whine alot in the beginning. now it barely does it i guess its just a matter of days untill the belt kinda stretches .. even though if it doesnt stop consider re adjusting
good luck buddy
align everything, slip belt on and tension, rotate crank to confirm timing marks line up, loosen tensioner 1/4 turn, rotate 3 teeth counter-clockwise (engine rotation)and tighten. that is the Honda way, if you tighten it than that is where the whine is coming from.
As mentioned earlier, I'm terrible at updating this thread, but I did want to post this up in case anyone else ran into this.
I took the upper timing cover off and there was no way the timing belt was too tight. I started taking off the other belts, one by one, and started the car each time. It was the idler pulley on the AC belt screaming like a banshee. I replaced the pulley and it's quieter now than it's ever been.
No overheating since I got it running, although I do intend on grabbing the block leak tester again from AutoZone and checking it just to make sure the head gasket really fixed the problem. I'll probably do that soon and report back at some point
04-09-2015, 01:25 PM
Radiator fan not turning on (solved)
