louis77 · Dec 14, 2019

Cold start problem.

It happens only in the cold season, else it starts at the first key.
When temp drops below 10C it won't start at first...it needs 2-3 attempts. No problem with the engine hot or in the summer. Also the idle stepper motor I think it's fine because I have a steady iddle 800 rpm when right temp reached , and after cold engine start, iddle rpm rise above 1000...so I assume the stepper works.

Tried to press the throttle pedal when crank...to give it more air...but nothing happens. Last winters I remember that it started at second attempt...but first, no matter how long, newer worked; more, I noticed that it was sufficent to wait 1-2 secs at first key...stop...then...and the second key it start; it is something like the first key..unlocks something; is there a way to check if the petrol it's on line..(maybe it hase some air) or the sparks?

The model is 2001 without temp gauge (blanked) under the ECT sensor and electronic sparks.

From the manual I understand that the problem could be a faulty ECT sensor, because at cold start the ECM read this sensor so see what temp it's outside and use "fast iddle" if too cold. Found a temp diagram for the termistor and checked the resistance with a ohm-meter...and it looks good. 4.4K at 7-8C, 2.2 at 20C etc. Cleaned a bit the plug for this sensor.

Last summer I remember (and now I link it to this problem) that the radiator vents started few times during a trip...although the temp gauge indicate a good water temp; the same behaviour occurs if ECT sensor is disconnected/unplugged.

louis77 · Dec 15, 2019

to test this: fuel pump will prime when you turn the key to the on position

it is possible to crack whitout first priming the pump, so that the first crank acts like priming the pump, and the motor starts at the second attempt?

The fuel pump is controlled by the Engine Control Module (ECM) via a fuel pump relay located in the engine
compartment fusebox.
When the ignition is switched to position II, the ECM provides an earth path for the coil of the fuel pump relay on pin
68 of ECM connector C0913. The relay is energised for a short period to pressurise the fuel system. When the ECM
senses that the engine has started by receipt of a valid signal from the CKP sensor, the ECM energises the fuel pump
relay for as long as the engine is running.

The fuel pump relay is located in the engine compartment fusebox which is positioned on the LH side of the engine compartment. The relay is normally open when the ignition is off.
When the ignition switch is in position II (ignition on), the ECM provides an earth path for the relay coil on pin 68 of connector C0913. With the ignition on, the relay receives a feed from the main relay which energises the relay coil, closing the contacts.
A permanent battery supply is provided to the relay contacts from fuse 10 in the engine compartment fusebox, via the fuel shut-off switch. The feed passes through the relay contacts and operates the fuel pump to pressurise the fuel system. The relay will be energised for a short time only to pressurise the fuel system.

louis77 · Dec 16, 2019

"If the ECT sensor fails or becomes disconnected, the ECM will use a default value which is based on values from the
engine oil temperature sensor. The driver may not notice that a fault is present although a fault code will be stored in the ECM which can be retrieved using TestBook. The default value will also include operation of the cooling fans in fast mode when the engine is running."

removed the plug from ECT (colant sensor) and it started at first key but after few secs of cranking...which it's not good; also..the cooling fans started; outside temp..few degrees..3-4; perhaps if I'll chenge the spark plugs it will be better, perhaps not; in this situation by the docs..the ECM got the temp from oil sensor...so if it was a ECT problem..would have been solved. which seems not.

the problem is hard to track because I have to wait for the engine to cool till next day in the morning.

Nodge68 · Dec 16, 2019

If you disconnect the CTS, then the ECU will apply a default value, unless there's another sensor it can use to get a temperature reading from. I'm sure that if it's putting the fans on, then the ECU is applying a default value, so runs the fans for safety. It won't initiate cold start enrichment, as the default value is normally for a mildly warm engine.

Ideally you need to look at the live data, so you can get a clue as to what is going on.

louis77 · Dec 16, 2019

Nodge68 said:
If you disconnect the CTS, then the ECU will apply a default value, unless there's another sensor it can use to get a temperature reading from. I'm sure that if it's putting the fans on, then the ECU is applying a default value, so runs the fans for safety. It won't initiate cold start enrichment, as the default value is normally for a mildly warm engine.

Ideally you need to look at the live data, so you can get a clue as to what is going on.

you say that if the fans run..it failed too to get the oil temp? from rave "If the ECT sensor fails or becomes disconnected, the ECM will use a default value which is based on values from the engine oil temperature sensor."

I can try to fake the sensor by using a 6.8K (0C) resistor inside the plug..also I'll have to test the plug at the EMC box side, to see if the wires are fine, with the resistor in place.

my feeling is that it starts with a setting for a middle temperature as you say...because after it ignites (finally)...in one or two seconds after I turn of the key from cracking.. I feel like the engine accellerate a little bit..to reach 1200rpm...as if it senses that it is too cold; this doesn't explain why it ignist at the second key? will the ECM switch to a lower temp combustion..if it fails at first key?

Nodge68 · Dec 16, 2019

Is there actually an oil temperature sensor? I don't the oil temperature sensor is fitted to all K series engines. So if your fans are running, when the CTS is disconnected, then this to me suggests that your K series doesn't have the oil temperature sensor installed.

louis77 · Dec 16, 2019

Ok, just came back..I did the resistor trick; I inserted 9k resistor instead ECT and the engine started at first key (1 sec); the resistor value should give to the ECM minus few degrees;

for the moment, the experiment it's irelevant because outside are 12C and at this temperature mostly it would ignite at first attempt; but...

what I noticed is that the engine start was more brutal as usual...started quickly and didn't felt that gap .. 2-3 secs after I released the key ...before throttle up by itself to 1200-1300rpm, because it was already at that aspeed, (but I must redo de experiment to be sure); also I felt some petrol smell; it is possible that this was indeed a "cold start" with more fuel enriched air ("fast iddle start")

I must check this tomorrow morning, and hope temp will drop below 5.

louis77 · Dec 16, 2019

Nodge68 said:
Is there actually an oil temperature sensor? I don't the oil temperature sensor is fitted to all K series engines. So if your fans are running, when the CTS is disconnected, then this to me suggests that your K series doesn't have the oil temperature sensor installed.

remember here

but what we do if it is also defective?

recently I changed my home refrigerator's thermistor...drilling the plastic wall inside..where nobody can reach. maybe it's the year when all thermistors fail.

and since my FL 01 doesn't have the temp sensor for the gauge (the mounting place blanked)...then..last summer, during a trip.. when vents started, why I got the temp gauge showing normal...but the vents running at full speed..like a broken CTS?!

yeah..CAN msg and some delay and fake display on gauge for short time periods, this could explain the behavior I had

"The engine coolant temperature gauge is driven by a CAN message from the ECM (the "brain" that controls the engine). The ECM derives the engine coolant temperature from an engine coolant temperature (ECT) sensor. This fitted on the aluminium hose connector to the right of and below the head when looking into the engine bay.
When the engine is hot, the gauge will display normal temperature until the engine has been running for more than 15 seconds. This prevents the gauge moving to the red sector of the gauge if the ignition is turned off and then on after a journey. The 15 second delay allows for the engine to be started and coolant circulated, allowing the gauge to display the true average temperature."

louis77 · Dec 17, 2019

This morning test, temp around +2..+4C, 9K resistor instead CTS, engine started instantly. I turned it off after few secs..I pushed back the plug into sensor..and the old problem..crank..crank..crank and almost nothing.

Ordered a new sensor MEK100060L and some locking paste. So at last, I have left 2 options. I bet it's the sensor, not the wire (which I didn't had the time to test), but now it's not a big cost to replace the sensor.

louis77 · Dec 17, 2019

I tested the wires between ECM - Coolant Temp Plug with a ohm meter on pin 7+33 (ECM connector) and a 9k resistor on plug and found no problem. Some time ago one of the locking mechanisms of the doors stopped working...I suspected the old known problem (motors inside). but not, after few hours I found that a conector inside the doors post had a metal plate cracked.

Tomorrow morning the final test...starting the car with the new sensor hanging the by the wire. (ll test this to avoid instalation for nothing, because the old sensor also shows values on ohm meter) If it won't start perhaps the software inside ECM has other values for "cold iddle". I used the 9k resistor but the problems occurs below 10C. If I think now..I should have tested for a temp of 5C equiv resistor 4-5KOhm

Anyway, the new sensor is now hanging along the motor...I'll see in the morning.

louis77 · Dec 18, 2019

nop. with the new sensor hanging by the wire this morning it didn't started at the first key.

Nodge68 · Dec 18, 2019

louis77 said:
nop. with the new sensor hanging by the wire this morning it didn't started at the first key.

Did you get the correct colour sensor?

louis77 · Dec 18, 2019

Nodge68 said:
Did you get the correct colour sensor?

yes. Facet 7.3248

maybe it is someting else working wrong...and putting that fake resistor forced the motor to start as if it was colder...with more fuel+air

if I think at the past...from last winters...there were an interval..under 10C to -2-5C degree when it doesn't start at first key..but when it was much colder..like -10C or lower...my impression was that the engine started better. so I asked myself..wtf is happening...is' going better at lower temperatures?

the new ECS has around 5.5Kohm in water with ice...aka 0C;

tested also the oil sensor resistance...and got 3.21K..while the water had 2.72K with cold engine; I belive they uses the same kind of termistors.

tested the wires between ECM and IACV and are fine; there are 4 wires and they ar connected as 50-24 and 13-39 (coils inside the stepper motor, the numbers are the pins at the ECM plug), and gives a resistance around 6hm.

louis77 · Dec 18, 2019

found a nice article for checking Engine Coolant Temperature Sensor (ECT)
I neglected the tensions

Usually the chip gets resistance by measuring the voltages. If there is an electrical problem, the voltage used by the EMC (5V) to test the resistance could be changed and would measure other values of the resistance; EMC knows that it has 5v in the circuit and that it measures a simple resistance; if there is a short on the circuit the voltage is no longer 5v .. then it obtains another value of the resistance and implicitly another temperature.

Procedure for verification the functionality of the ECT sensor
— Sensor with negative temperature coefficient —Test with voltmeter

Open the protective rubber cap on the connector of the cooling system temperature sensor.

Connect the negative wire of the voltmeter to the chassis ground.

Determine which are the signal and the ground terminals.

Connect the positive wire of the voltmeter is to the ECT signal terminal.

Start the engine form cold state.

Depending on the temperature, voltage readings should be in the range of 2V to 3V. The relationship between the voltage and the temperature is shown in Table-1.

Check whether the ECT voltage signal corresponds to the temperature. You will need a thermometer for the purpose.

Start the engine and warm it to the operating temperature. During engine warming the voltage should decrease in accordance with the values given in Table-1.

Common problem is that the output resistance (and voltage) changes incorrectly beyond its normal range. The normal value of the ECT sensor voltage is 2V at cold engine and 0.5V at warm engine. A defect sensor can indicate a voltage of 1.5V at cold engine and 1.25V at warm engine, thus causing difficulties of starting a cold engine and a presence of rich fuel mixture when the engine is warm. This will not generate any fault codes (if the onboard controller is not programmed to detect voltage changes) since the sensor continues to operate within its design parameters. If such defect is found, the ECT sensor must be replaced.

If the ECT voltage signal is equal to 0V (lack of power supply or there is a short circuit to the ground) or if it is 5.0V - we have an open circuit.

Possible sensor failures:
-- The voltage of the ECT signal terminal is equal to 0V.

Check sensor terminals for short circuit to ground.

Check the integrity of the signal wires between the sensor and the onboard controller.

If all wires are correct, but there is no output voltage from the onboard controller, you have to verify all power supply and ground connections of the onboard controller. If supply voltages and grounds are good, the onboard controller itself falls under suspicion.

-- The voltage of ECT signal terminal is equal to 5.0V
The voltage has such a value when an open circuit is present and may be obtained in one of the following conditions:

the signal terminal of the ECT sensor does not provide connection to the sensor;

sensor circuit is open;

sensor’s ground circuit is open.

-- The voltage signal or the reference voltage is equal to the car battery voltage.
Check short circuit in the wire, connected to the positive terminal of the car battery or the power supply wire.

Screenshot_2019-12-18 Engine Coolant Temperature Sensor (ECT).png

So, it is possible that the resistance is good, but measured with other than 5V..to give wrong temperatures.

Nodge68 · Dec 18, 2019

louis77 said:
So, it is possible that the resistance is good, but measured with other than 5V..to give wrong temperatures.

Start by measuring the voltage at the CTS terminals and see what you get.

louis77 · Dec 18, 2019

Nodge68 said:
Start by measuring the voltage at the CTS terminals and see what you get.

yeah, but how to do that since I cannot touch the metal aroud the connected plug

well, I'll do the test with an external resistor replacing the ECS with ignition on II.

rave shows 2 sensors for my FL 1A, one brown MEK100060L..which I found already on my car (don't know if the stock was changed sometime in the past, because I'm not the first owner)..and one blue MEK100130L; purchased the brown sensor but perhaps I need the blue one; if termistors are diffrent that could be an explanation. now I try to find the differences.

louis77 · Dec 19, 2019

found on a russian forum..
some guy needed to replace his coolant temp blue sensor MEK100130L with a equiv one..but instead finds an equiv to MEK100060L...YCB100420 . another it tells him that..ok..try that...but take care...if you notice "cold start" problems..it may not be good.

In any case, we know that both sensors have the same physical dimensions M12x1.5

The wiring from the MEK100130L sensor goes to the electronic harness and then to the computer, and not directly to the device. And what is happening in the computer with its signal? It depends on the specific task of the program that is installed there. It is likely that the computer still takes into account the signal from the sensor and makes some adjustments to the operation of the engine systems. Similar to the signal from the MEK100060L sensor. In my opinion, you can definitely find out if the MEK100060L sensor is in good working order and the engine is cold started in cold weather. If something is wrong, then the engine will behave as in the case of a malfunction of the MEK100060L sensor, i.e. start only after several attempts, or not start at all, lose much power in the unheated state, do not reduce idle speed below 1,000, pull idle speed ... i.e. there is a possibility that the computer takes into account the readings of exactly two sensors. timon1978 assures that a cold start is normal and that’s good, because removes suspicions that the MEK100130L sensor works not only for tidy ... Or not? After all, there are problems with idle speed on the timon1978 car ... Maybe because of this (abnormal) sensor an unwanted correction is made on a hot engine? It would be good for him to find out the problem of floating revolutions to the end.

so I start to belive that MEK100130L MEK100060L are 2 version usind 2 different thermistors..with different temperature/resistance diagrams; it is possible that previous owner changed the blue stock sensor with the most common, easy to find.. brown one.

they say

I understand that the differences are precisely on the warmed up engine, namely it has lower resistance at 100-110 degrees

I think this analogue is successful. But it’s interesting: when it gets cold, does the engine start normally in the morning or not? If yes, then do not bother with the “indispensable” original MEK100130L!

so the brown version has a lower resistance for a given temperature...and the russian guy complains that the fan starts in traffic under hard conditions; so it is possible that in my case at 0C extern temp...my ECM using the brown sensor (instead of the blue one) to belive that the temperature is 10C or more and not use conditions for "fast idle".

I must dig on this.

Nodge68 · Dec 19, 2019

If you only have one sensor, then I believe it should be the green/blue one. IIRC the brown sensor is for the gauge on the earlier 1.8. I'll check what my 2004 1.8 has later, as I can't remember off hand.

louis77 · Dec 19, 2019

I did the tension/voltage measurement and it's fine. 5V when the circuit is open and when I attach the 9k ohm resistor I have 4.3V; it is not short to gnd or 12V on the wires or on the installed sensor .

but take a look at how the plug and sensor look. they do not match; someone carved the plastic bump from the sensor to fit the socket; so the plug or sensor is at least one changed. In one image is the new sensor which doesn't match the plug, and I asked myself..how the hell I inserted it yesterday when I tested it on the car....but I figured out that with enough force it fits

tested again the engine start with 9k resistor and it starts instantly...with petrol smell around, so It requires the other sensor.

Nodge68 · Dec 19, 2019

louis77 said:
I did the tension/voltage measurement and it's fine. 5V when the circuit is open and when I attach the 9k ohm resistor I have 4.3V; it is not short to gnd or 12V on the wires or on the installed sensor .

but take a look at how the plug and sensor look. they do not match; someone carved the plastic bump from the sensor to fit the socket; so the plug or sensor is at least one changed. In one image is the new sensor which doesn't match the plug, and I asked myself..how the hell I inserted it yesterday when I tested it on the car....but I figured out that with enough force it fits

tested again the engine start with 9k resistor and it starts instantly...with petrol smell around, so It requires the other sensor.

You have the wrong sensor fitted. The sensor with the offset plug location key is blue. Like this one.

https://rover.ebay.com/rover/0/0/0?mpre=https://www.ebay.co.uk/ulk/itm/333296506755

Member

Attachments

Member

Member

Well-Known Member

Member

Well-Known Member

Member

Member

Member

Member

Attachments

Member

Well-Known Member

Member

Member

Well-Known Member

Member

Member

Well-Known Member

Member

Attachments

Well-Known Member

Similar threads