ndndndnd

Hey all,

My car's throwing a code 15 - coolant temperature circuit.

I've replaced the sensor itself and soldered a new connector in it's place. Still threw a code 15.

Following the diagnostic procedure in the shop manual, I measured across the terminals on the sensor connector expecting to see approx 5V. I saw 12V! Why would there be 12V instead of 5V?

I'll admit I haven't had a chance to check the sensor properly, or pull out the ECM and check the terminals on there.... but seeing 12V instead of 5V seemed weird enough to ask if anyone knows why?!

Many thanks,

Neil

RBob

iTrader: (1)

Unless someone did some rewiring I'd say the ECM is bad.

RBob.

ndndndnd

Ugh, thanks for the answer - I was afraid that'd be the reason...

I'd better double check the connectors at the ECM anyway to check something bizarre hasn't happened with the wiring.

Is the ECM feasible to repair or is replacement the only way?

Tuned Performance

iTrader: (92)

Replace the ecm it's a 5v circuit.

ndndndnd

Can anyone recommend where to get an ECM?

I can get a remanufactured ACDelco one for Ł115 delivered to the UK from Rockauto. Can anyone suggest a better deal than that?

Does anyone know what might burn out an ECM? I had a 'friend' flip the killswitch while I had the ignition on, which I've heard isn't great for the computer...

Tuned Performance

iTrader: (92)

You don't state if this is tpi or ccc , if it's tpi you might consider swapping to the 165 from the 870. You can check the classifieds section or rock auto or even eBay for good deals.

ndndndnd

It's an E4ME - CCC. I guess I'd better start researching part numbers!

OrangeBird

iTrader: (1)

Just a quick double check , you are measuring the correct sensor , yes ? I ask because there are a couple of different temperature sensors on the engine and on at least one of them it would measure 12V (the cooling fan switch)

ndndndnd

Haha, you're welcome to ask - I've been fooled by the simplest of things before!

I've had to fanny about with both the temperature gauge sender (driver's side cylinder head) and the fan switch (passenger side cylinder head) before, so I know which is which. Plus, I think both of those are single-wire connectors, not a double-wire. The sensor I'm testing is the on the barb from the water pump housing at the front of the engine, right next to the thermostat.

I'm thumbing through the Shop Manual, but I can't find anything that seems to define a 'test' for the ECM. Can anyone suggest a way of specifically testing the ECM before I chuck it out?

Many thanks,

N

RBob

iTrader: (1)

This is for DTC-15, VIN H, 305 4-bbl, '85 Camaro. Self explanatory...

RBob.

ndndndnd

Thanks RBob,

I've got that diagram in my shop manual - I ought to dig out the ECM and check if the voltages make any sense from there. I've also found a pinout diagram giving voltages for all the pins, so I could go to town with the DVM and see if anything else doesn't tally to confirm if anything else on the ECM is off.

Cheers,

N

ndndndnd

Hello! Back again....

Well, pretty embarrassing... I should have checked those wires more carefully. When I checked back on the wiring diagram I realised the wires to the coolant sensor should have been black and yellow, not green and black...

Lo and behold, languishing hidden down beside the carburettor is a little connector with yellow and black wires and, lo and behold, 5V across the terminals.

Damn. Really shouldn't have taken a punt on that reman ECM.

Anyway, soldered the wires all up to the new connector and triumphantly took the car for a drive.

Threw a code 15.

Damn.

I've just been out to the car, checked for airlocks in the coolant (coolant is fully, absolutely full and all pipes are hot) and the sensor (measured 470 ohms at just below the cooling fan coming on), and voltage across the connector is still just above 5v.

So, why the hell is it throwing a code 15?? I'm going to pull out the ECM again when I get a chance and check the connections, but surely if the wires weren't all hooked up I wouldn't be getting the 5V reference? I'd either get nothing across the terminals, or something else?

My shop manual says if there's over 4v it can only be the coolant sensor (seems to work) an airlock (doesn't seem to be) or the ECM (the reman one could be junk).

I'll stick my original ECM back in to see if that helps, but in the meantime is there anything else I'm missing? Could anything else cause a Code 15?

Cheers for any help/jeering/pointing and laughing,

ND

Tuned Performance

iTrader: (92)

You should ohm the cts. I think this should be the chart for your cts.

ndndndnd

Thanks very much for the chart - I have a similar one in my Service Manual, but yours has more data points. Where did you get it from?

I pulled the ECM out of the car this afternoon and checked continuity of the wires from the coolant sensor to the ECM plug. All came back positive, with the coolant sensor wires clearly connected properly to the ECM.

I then swapped back my original ECM (transferring the PROM!), fired the car up and took it for a drive. It threw a code again, so I guess that rules out the ECM being the problem. (I only removed the original ECM as I thought it was the reason for 12v in the first place, whereas it was my misreading of the circuit diagram and the ECM was not at fault).

When I got back, with the car relatively warm I shut off the engine and tested the coolant sensor. It read about 360ohms, which puts it in a logical place on the scale above.

So that means the wiring is fine, the coolant sensor is fine, there doesn't seem to be an airlock and it's highly unlikely to be the ECM...

.... So why the heck do I have a code 15??

Any thoughts anyone? Many thanks for any suggestions...

ND

Tuned Performance

iTrader: (92)

I googled gm cts ohms to find the chart.
I’m not sure why you have the code 15 bad ground maybe .
The code gets thrown when the circuit is open on the yellow wire or a bad cts. Your cts seems to be fine so it might be corrosion in the cts pigtail or its ground.

sofakingdom

iTrader: (1)

The black wire should be grounded. If memory serves, there is a ring terminal that connects that wire to ground, along with a couple of others related to the ECM, on the top front of the intake manifold. Check for continuity between the connector pin for the black wire and the block; repair if found inadequate.

ndndndnd

The only thing I can think of is a bad ground too. Although the connector to the sensor is new, and the connectors into the ECM appeared to be in reasonable condition. When I measured the reference voltage across the sensor connector I got a strong 5V+ - wouldn't the reference voltage be diminished if there was an impaired ground?

Sofa - do you mean I should check for continuity from the black wire of the sensor connector to the ground on the front of the intake, as in this would check the ground through the ECM?

sofakingdom

iTrader: (1)

No.

5V is applied to the sensor through a resistance or by a constant-current source, such that the higher the resistance of the sensor, the higher the portion of the 5V that will appear across it. When it's all working right, at high temps, when the sensor has very low resistance (looks more nearly like a short to the ECM), the voltage across it will be quite low; at low temps, when its resistance is high, nearly all of the 5V will appear there. The ECM then has a way to know what the temp is, according to the voltage that is developed.

Disconnected ground = very high resistance, for all practical purposes infinite. Therefore you'll see the whole 5V source on the yellow wire, and the ECM will think the coolant temp is -40°C or whatever its lowest possible temp might be.

Yes I mean check the continuity of the block wire between the sensor connector and the engine. It has nothing to do with the ECM ground; rather, you'd be checking the sensor ground. Making sure the black wire is hooked up as well as the yellow one.

ndndndnd

I was a bit confused by your comment about high voltage with high resistance, until I realised you meant with the sensor still plugged in!

I was referring to the voltage and resistance in the wires directly to the ECM measured with it unplugged.

I've measured the resistance of the sensor wires (unplugged) to the ground at the front of the intake manifold.

Black - 1.8ohms
Yellow - 660ohms

I checked using another couple of grounds in the engine bay and got similar numbers. These were taken with the key-off (no keys in ignition).

Do these sound about right? I would expect more resistance in the yellow wire as it must run through circuitry, whereas black goes straight to ground.

Cheers,

ND

sofakingdom

iTrader: (1)

Those #s sound fairly believable. 1.8 ohms is kind of high for ground resistance but not to the point that it should be causing problems or anything like that.

Try this: Unplug the sensor and measure the ohms across it with the car cold. I'd expect something in the 3000 - 5000 ohm range at that point, assuming that it's maybe 10 -15°C in your place at the moment. Then plug it back in and measure the voltage between the yellow wire and ground. Should probably be 3.something V, maybe close to 4. Then warm the car all the way up to normal operating temp. Unplug the sensor and measure the ohms across it. It should be something in the 150 - 200 sort of range. Then plug it back in and measure the voltage, which should then be down in the .5 - 1.0V or so kind of range. If it varies approximately like that, the sensor itself and its circuitry is probably working. If not, ...

ndndndnd

Ok, I've just come back in from testing the car.

Cold:
Ohms: 440 on the 2000K scale (I think I might have missed the decimal point - 4,400ohm makes sense, given it's between 5 and 10 degrees outside).

Volts: 5V across the connector not plugged in.
5V volts across the connector with it plugged in...

Started the car and ran it up to temperature (210 on the dial)

Ohms: 220 on the 2000K scale
Volts: 5V with the connector plugged in...

So the voltage at the connector never varied, regardless of temperature or resistance, or even if it was plugged in or not... any ideas why that would be?!

The only thing I could think would be that the connector somehow isn't connecting... it seems very unlikely as it's a good weatherpak-style connector and both sensor and pigtail were bought in a packet together. I could pull the ECM again and check for continuity between pins 3 and 7 with the sensor plugged in though...

Can anyone think of any other reason for the unwavering 5V though?!

Cheers!

ND

ndndndnd

Ok,

Back out to the car with the DVM.

First, pull out ECM again and check continuity across pins 3 and 7.

30ohms. Bit of a weird number. I would have expected to see the resistance of the cold sensor.

Open bonnet. The sensor's unplugged. Oh... but how would I have had continuity if the sensors unplugged? My blood ran cold. Damn, I must have a short between the two wires...

To double check, I measure across the two poles of the connector, expecting to see 30ohm from the other direction too. Nothing. Open circuit.

I plugged the sensor in, and checked across pins 3 and 7 inside the car again. 30ohms again. What the...??

I check continuity of each wire again. Plug a crocodile clip to one terminal on the connector at a time and clip to the ground on the alternator.

Terminal 3 reads 30ohms again. And terminal 7 reads 4ohms at the same time. How can that be?! Only one pole is connected?!

Jump out of car and switch the crocodile clip from one pole to the other.

Terminal 3 reads 30ohms again! And terminal 7 reads 4ohms again!

AAAARRRRGGGHHH!!!

At that point, I packed up and went back into the house!

Tuned Performance

iTrader: (92)

Maybe rbob can answer this . I’m wondering if it’s maybe not the correct wires your ohming at the ecm.

ndndndnd

Thank you!!!

Stupid here had been reading the pin out diagram backwards!

When I stopped and looked a little closer it made more sense that the diagram read from the top of the connector, not from the face.

Anyway, I repeated my experiments again: grounding each wire of the connector in turn and checking continuity at the ECM connector. This time everything made sense. Pin 3 related to the yellow, and 7 to the black. Both wires had a resistance of approximately 4-6 ohms. Like sofa said: not ideal, but shouldn't be enough to cause a problem.

I then ohm'd the sensor, which read about 160kohms, and then plugged in the connector and ohm'd between pins 3 and 7 at the ECM, which also read 160kohm.

160kohm seems a little low, but the sensor does seem to read more reasonably at higher engine temperatures.

However, the ECM does seem to be getting the correct resistance at the connector. What's the chance I just need to clean the damn contacts?!

It could be an excuse to invest in an OBD1/ALDL scanner, then I could actually see what the ECM is seeing. I have one of those little OBD2 Bluetooth dongles, does anyone have any experience with one of those OBD1 converter cables and ALDLdroid?

Thanks!

ND

Tuned Performance

iTrader: (92)

I’m wondering if the cts is bad. You might be able to source one locally.Reading low at startup causing the check engine light.
You might try win aldl for a cheap program I’m not sure of a android application for older obd1 vehicles.

ndndndnd

Mr. Tuned Performance, I owe you an internet beer!

I swapped in a new coolant sensor and the code 15 went away. Like you said, the super-high resistance from cold must have automatically triggered the Code 15, even though the sensor seemed to ohm correctly when the engine was warm. Interestingly, my shop manual doesn't mention this behaviour in relation to the qjet, but it does mention throwing a code for detecting a super-low temperature for the TBI setup. I guess the qjet setup must behave in the same way, even if the shop manual doesn't mention it!

Anyway, my car's still running rich, but now it's properly entering closed loop I can have a proper stab at tuning up the IAB and getting it running properly!

For those interested, the equivalent part that I managed to source in the UK is a Hella 6PT 009 107-591

Cheers for the help everyone!

ND

Tuned Performance

iTrader: (92)

Refreshing lol, glad your making progress.

ndndndnd

Not sure if this merits a new thread or not...

Car is now throwing a Code 14!

It didn't throw codes for a couple of weeks, and probably only a handful of journeys, but, when I tried to use it to commute to work, it almost immediately threw up a Code 14. Still made it to work, although it was a sod to restart when I came home!

The service manual defines a Code 14 as 'short in coolant sensor circuit' or else, low resistance/high engine temperature.

Engine doesn't seem to be overheating. I re-checked the sensor cold and it ohm's as I would expect (3050 at 20-ish degrees C). I then checked the reading at the ECM harness and that matches the reading at the sensor. I then unplugged the sensor and checked continuity across the wires and found none, then checked each wire to ground and again found no continuity. This all means the sensor works, the same reading reaches the ECM, and the wires haven't shorted with each other or grounded themselves on the chassis/engine. I'm also getting 5V across the connector KOEO.

I swapped in a reman ECM, and it still threw a Code 14!!

I'm going to take the car for a short drive tomorrow and re-test the sensor at higher temperature, but it's been fine when I've checked previously at running temperature.

I'm a bit puzzled now. The sensor seems to be fine, the wiring seems to be fine, the ECM seems to make no difference and the engine isn't overheating.... why am I getting a Code 14?!

Can anyone suggest anything?!

Many thanks everyone,

ND

NowhereFast

Wiring is super easy once you do it a while. You don't even have to know how things work. Most important thing is having the RIGHT wires. Disconnect the two circuits completely leaving only the wires as part of the equation. Test the circuits from end to end. Now it sounds like you've already done that, but I didn't read anything about a load test. Testing wires without loading them will give you false positives, and you'll be chasing your tail, as you've been. Ground one end of the circuit. Take a taillight bulb/headlight bulb (I use one with a socket and the wires for ease of connecting it) and have it in series with alligator clip wires. Now put that little string to battery power. If it lights up, it's holding the load of a taillamp bulb, that should be more than enough for a sensor. If it doesn't, you have a broken wire, and now you have to find where it broke.

Reason for doing this is, all it takes is one strand of that wire to be connected for it to check out with a meter. Doesn't mean it's enough for the circuit to work properly.

Here's a picture if I wasn't super clear. Not that it's a difficult test, but usually I'm standing in front of someone when I tell them how to do it. If this checks out, and there's no short to voltage or ground on those wires, then it's an ECM or the sensor you got is bad.

Also, to be doubly sure, move the harness the wires go through while you're loading them. See if it shows a break. Just in case the movement of the engine and the harness while you're driving is causing the wire to be pulled apart.

ndndndnd

Thanks for the info Mr. Fast - I didn't know about doing a load test like that. I have one of those screwdriver-y test-light things, that I guess will do the same job as your bulb-and-wires get up. I'll pull out the ECM connectors again when I get the chance, ground the connection and test to the battery.

My primary suspect if the coolant sensor again, but I'll have to take the car out and warm it through again when I get the chance!

I was thinking about this the other night. I've been struggling to get the coolant sensor to seal. I bought a Hella one that didn't come with thread sealant like the Airtex one I had before. Thinking nothing of it, I just stuck it in without sealant and it wouldn't seal. So I pulled it out again and tried with a little bit of plumber's jointing tape, but it still won't seal - I've been very cautious of over-tightening it, so each time I've driven it I've checked for leaks and given it an extra quarter or eighth turn until the leaking stopped. I'm wonder if perhaps the jointing tape wasn't the best thing and I've started to damage the sensor... I'll have to test it to know, and I'm just thinking out loud! Bit annoying if it is, as that was quite an expensive sensor!

Cheers,

ND

NowhereFast

That's called a test light, and it won't do the job. It doesn't put enough load on the wires. Ideally you would use a headlamp bulb, but the 3157 bulb I use does just fine. I really only use my test light to check fuses at this point, because again, it will give you false positives.

ndndndnd

Well, it would seem the sensor is working perfectly. I checked the ohms with the engine hot and it read 160 ohms just as the fan cut in. Right on the money.

ndndndnd

Oops, double post!

ndndndnd

Right...

I've finally managed to carry out a load test as described by Mr. Nowherefast. I didn't use a headlamp bulb, but I did manage to find a tail-light bulb, complete with fitting, which I used for the test. And.... well, it passed. I plugged each of the two connectors at the ECM to ground, and then connected the bulb to the battery and connected it to each wire on the temp sensor connector in turn and the bulb lit up, bright as you like. I gave the loom and connector etc, a good wiggle and yank and the bulb stayed steadily lit.

I also double checked the resistance of each wire, and waggled each wire and in turn. The resistance reading didn't fluctuate. There's also no hint of continuity between the wires, or between the wires and ground. I then checked the resistance at the sensor (about 3600, which is about right for 20-ish centigrade) and I got exactly the same reading between the ECM connectors in the cabin.

Totally stumped now.

The sensor checks out, the wiring checks out, and swapping the ECM made no difference.

My next best guess is to chuck another new sensor at it, even though this one seems to test ok. I did read something somewhere that strange things can happen with anything but an AC Delco sensor, but I should think a quality Hella sensor should be ok?

Is it possible to start the car with the ECM disconnected so I can measure the resistance across the wires with the engine running?

Am I likely to cause harm to the car by driving it around with Code 14 showing? I think the Code 15 over-advances the timing, and I'm guessing Code 14 does the same as the car is very hard to start.

I'm starting to understand why people throw away the E4ME and stick an inferior mechanical carb on there instead... another summer is disappearing chasing gremlins...

ndndndnd

I guess all you guys are getting just as bored with this b******t as I am!

Does anyone at least know if I'm damaging the car by driving it around? I can't hear any obvious signs of pinking, but the car's pretty loud and I've never heard pinking before so I'm not entirely sure what I'm listening for...

ndndndnd

Well, I may as well keep on updating this, as there's nothing worse than a thread with no conclusion!

I enacted an elaborate experiment to check on the voltage of the coolant sensor circuitry while the car was running. I connected two short lengths of copper wire to the coolant sensor connector with crocodile clips to the sensor itself and my DVOM connected to the copper to keep track of the voltage. The voltage read about 5V disconnected, and 2.28V connected. I started the car and let it warm through. Voltage dropped gradually to about 1.28V when the fan cut in. No CEL.

Hmm... Why is everything working properly?

I wiggle the wiring harness about to see if that does anything. Nope. I clamber inside where the ECM is hanging in the footwell and wave that about. Nothing. Hmmm.... I turn off car, connect sensor directly and restart. No CEL. Hmmm... I wiggle the harness some more, and wave the ECM about some more. No CEL. I then try to stuff the ECM back up behind the dash...

*ping* CEL comes on.

I guess that means I really do have some dodgy wiring, which makes sense for a Code 14, and it's in the little bit of wiring the hangs down behind the dashboard. Great. I secure the ECM in place and restart the car again. No CEL. I wait and wait. No CEL. Oh well... I guess I'll deal with it when I have to!

I just read in this Months' Octane magazine about a Porsche 944 that required extensive re-wiring as the loom had simply got too old... that car was a 1983 and mine is a 1985. I guess I'll have to brace myself for an extensive re-wiring project in the near future!

In the meantime, the rear brake caliper burst a seal so the car's not going anywhere for a little while. And so it continues!

Franky B Jr

So what if my ecm on my ford focus 2015 is telling me -60c C. Anddd My car won’t start…. Itll try to turn over and even almost does when I give it gas on the pedal but then shuts down… And I’m not getting any voltage to my ECT sensor but the wires have continuity on both ends. Not to mention they have continuity to each other on the ect sensor connector itself on the two ports.

Franky B Jr

yeah I had the same problem it’s your ecm you probably somehow got more than 12v on temp sensor connector or circuit and now it’s stuck open.

sofakingdom

iTrader: (1)

This is a forum for 82-92 Camaro & Firebird exclusively, not a general automotive winner-take-all tech Q&A.

No idea whether a Focus is anywhere near the same in any way. I suggest you try a Focus forum. That said, since you already know the engine is reporting a completely wrong coolant temp, sounds like your coolant temp sensor has gone bad, or is disconnected, or the wire is damaged, or some such.