Fan relay wiring to coolant temp sensor?

Post by **bmdubya1198** » 12 Mar 2018, 18:43

I noticed today that I am missing a connector on the middle terminals of my fan relay, and the wires (held on with the spade-type connectors) were very loose. I plan on getting a new pigtail and soldering it on (or replacing just the connector) but I temporarily closed up the wire terminals a little bit to connect it securely to the relay.
Now, my question... I've had a "hiccup" for the longest time that happens primarily when I'm coming to a stop or waiting at a stop, and it's incredibly irritating. The engine will drop a 150-200 RPM and shake, then come back up and idle normally. I'm just realizing after noticing these wires that the hiccup may be happening at the same time the fan comes on.
The wires I'm referring to are solid yellow and yellow with a white stripe. I know they don't go directly to the ECT sensor, but from what I've read, that is where the relay gets the signal. I know the ECT sensor failing (or the wiring, in this case) can cause rough idle, so is there a chance this could cause my hiccup? I plan to replace the ECT sensor soon as well just for preventive measures.

Other than that hiccup, I have no reason to believe my ECT sensor or wiring is an issue. My engine warms up normally and the gauge stays in the middle. The gauge has never acted strangely. I guess that brings me to the second part of my issue... can the ECT sensor fail without obvious symptoms minus the roughness/hiccup?

Post by **bmdubya1198** » 13 Mar 2018, 10:51

Also an interesting note... I tested my coolant temp sensor a little bit ago using Pelican Parts’ write up, and it’s showing an open circuit.
The car has not been started today, so coolant is cold. Ambient temp is around 47 Fahrenheit.

Post by **scot850** » 13 Mar 2018, 13:17

If the ECT is not working then your engine may start when ambient temp is cooler ok, but as temp changes the car may well run lean or more often rich.

You have checked the ECT for resistance v's temp values, and if this is ok the ECT is ok. If your gauge is reading normal when warm then the ECT is probably ok too.

If the engine takes a long time to warm up, it could just be a bad thermostat.

Neil.

Post by **bmdubya1198** » 13 Mar 2018, 21:30

No issues there, the car takes a reasonable amount of time to warm up. It runs reasonably smooth except for those little hiccups, which only happen when the engine is warm.
It might have been just the low temperature that it was reading the way it was. I didn't check it after driving.

I will mention, I changed my spark plugs a couple months ago and the old ones were pretty black... I wouldn't be surprised if the car was running a little rich. I just wish I could figure this problem out. I've been dealing with it for over two years now and it's driving me insane.

Post by **jimmy57** » 13 Mar 2018, 21:54

The wires go directly to the ECM. The ECM reads ECT sensor and selects the hi or lo fan as needed. A/C high side pressure sensor is also what the ECM uses for the hi and lo speed decisions. If the ECM misses lo speed and then the high pressure or temp goes to level for hi speed then you get idle bauble. The ECM has adapted idle thinking fan is already running and the load is there. When hi is triggered the sudden load catches ECM by surprise.

Post by **bmdubya1198** » 14 Mar 2018, 09:07

jimmy57 wrote: ↑13 Mar 2018, 21:54 The wires go directly to the ECM. The ECM reads ECT sensor and selects the hi or lo fan as needed. A/C high side pressure sensor is also what the ECM uses for the hi and lo speed decisions. If the ECM misses lo speed and then the high pressure or temp goes to level for hi speed then you get idle bauble. The ECM has adapted idle thinking fan is already running and the load is there. When hi is triggered the sudden load catches ECM by surprise.

In that case, I will check the wiring to the ECU for both the ECT sensor and the fan relay. That definitely makes sense. I just put together over the past couple days that the surge in idle seems to happen when the fan would be coming on (e.g. sitting at an intersection). It might also be happening now while I'm driving, which seems to be occurring more lately, but I can't be sure that's the RPM surge or the transmission or just me running over a rock and thinking too much into it.

So could this be an issue within the ECT sensor?
Another thing to note, I think I have a small A/C leak... could that contribute to this issue, or does the A/C pressure sensor just determine if the A/C is running? I can't figure out where the leak is coming from, though. I can see the neon green stop-leak/R-134A in the area of the compressor, which is more noticeable during the summer months I actually use the A/C.

Post by **jimmy57** » 14 Mar 2018, 11:38

the charge level being low would lower high side pressure and that would make it less likely for ECM to see a pressure high enough to command high fan speed. BUT with no low fans speed it will still get to 285 or so psi needed to trigger high speed selection

Post by **bmdubya1198** » 14 Mar 2018, 18:46

Well, after taking a close look at the terminals (wire side) on the yellow and yellow/white wires on the relay, they were burnt and frayed. I soldered on new wires and terminals, that brought the resistance down to acceptable levels (0.03 ohm) between the ECU pins and the wires. Before with the bad terminals it was spiking to 50 ohms!
I think that part has been solved now.

I turned on the A/C to test the fan, and the fan didn't come on. I could hear the A/C compressor cycling on and off every couple seconds (low charge issue?) and nothing was happening. I'll check tomorrow with Vol-FCR to see if there is any code relating to that issue.
I don't think the fan not coming on is a big problem, because my cousin's '99 S70 didn't kick the fan on with the A/C either when I tried that earlier.

I tested my ECT sensor resistance again today when the coolant temp was warmer (engine was warm to touch, but not hot) and still no dice. That seems very odd to me.
Another interesting note is that an open circuit is shown when testing for continuity between the ECT sensor wires (on connector) and ECU pins that correspond to those connections. I know I was testing the correct pins because the pins for the fan relay were correct.
I might try to strip back a little of the coating tomorrow to see if the issue is with the pins in the connector or the wire itself.

What's crazy with all of this is that my temperature gauge still functions properly! And according to my multimeter, the ECT sensor isn't even functioning!

Post by **bmdubya1198** » 15 Mar 2018, 10:46

Okay, had a stupid moment. I misread the resistance specs for 300 ohms, so I was on the wrong setting of my multimeter. On the 20k setting, I got a reading of 4,520 ohms at 49 degree coolant temp, which sounds right. I am going to trip back the wires for the ECT sensor on the harness side and back probe them to the ECU terminals and see if those terminals on the connector are the issues.
After driving today, the car hiccuped again, so my repair yesterday didn't solve the problem.