23 replies to this topic

[Puddles]

Hi all,

 

I'm having some idling trouble with my 1996 JDM SPI mini and would greatly appreciate any help. I have been searching the forum since this issue started about a year ago reading any threads people have made with similar issues that my car is presenting with no luck. 

 

Background

 

I purchased my mini last July. It was a hot day (I would guess about 28c) when I went to look at it and it fired up right away and it idled fine although it would bog down a bit when the AC was turned on. 

 

A few days later after completing all of the transfer papers getting insurance and registration I picked up the car and again it was around 28c. The car fired up right away and drove fine on the way home. 

 

The first time I noticed an issue with it was about a week later when I went to take the car out at night. It would have been about 15c out and the car fired right up however the idle would hunt up and down by about 200rpm until the engine was up to temperature. 

 

If i took the car out for a drive while cold it wouldn't have any power and almost stall when coming to a stop. I noticed that the PTC heater was not plugged in and after reconnecting it I get power while the engine is cold however the idle still hunts.

 

I took the car out yesterday without letting it warm up first and it ran fine while cold until I came to a stop and it would hunt for idle. This time it continued to hunt when stopped even once the engine was up to temp. 

 

The car runs perfectly, never knocks, never misfires. It just won't idle consistently when cold and occasionally but not nearly as badly when hot.

 

 

Fault codes

• Throttle position sensor faulty.
• Coolant temp sensor faulty.
• Air temp sensor faulty.

I cleared these fault and they have not come back.

 

What I have done so far

• Inspected all vacuum hoses and fuel trap.
• New spark plugs. (Old spark plugs looked fine to me if anything it might be running a little lean, although the gap wasn't set right on these plugs.) 
• Adjusted tappets. (They were fairly good but a couple had walked out)
• Reset/reboot ECU (disconnect battery for a few mins, reconnect, start car, let idle for 10 mins).
• Checked the throttle cable, it moves freely and doesn't get stuck. 
• Adjusted lost motion linkage.
• Tested the wiring loom for any issues with the wiring for the Air Temp, Coolant Temp, and Lambda. (the conductors for the air temp were frayed at the connector. A new connector was installed. The problem doesn't seem to get worse or better whether or not this is plugged in.)
• Removed inspected and cleaned ground connections in engine bay and trunk. 
• Pulled the Lambda and ensured the heater was functioning. 
• New air filter installed.
• swapped out lambda for another used one that I had sitting around. Didn't seem to effect anything.
• Swapped out ECU for another I had sitting around.

Possible causes

After doing some research on this forum, these seem to be the most common issues people have.

 

Coolant temperature sensor

Apparently this sensor can have a big impact on the car's idle speed, plus there was a fault code logged for it. My hesitation to replace this sensor is that it's a major job and the car's temperature gauge works perfectly (on min when cold, on white line when at operating temp). I have also noticed that most people have an issue with a run away high idle when this fails, not a fluctuating idle. This also seems to read fine in MEMS_Diag.

 

Air temperature sensor

This also has some effect on idle speed from what I have seen. As I mentioned above, the wires were frayed and a new connector was installed after repaired the wiring still wasnt 100% but what I would consider acceptable with several Mohm of resistance to common and between each other, however the car doesn't seem to run any differently whether or not this is plugged in. MEMS_Diag was showing the temp a little high at 22c on cold start when we only had a high of 14c today. I do have a calibrated thermocouple that I could do further testing with. Was considering jumping the connector with a resistor to force a colder reading to see of that helps at all. 

 

Stepper motor

The stepper motor is another likely culprit. This seems to be functioning correctly as far as I can tell. When the car is turned off it retracts and clicks and resets. I tested all windings to common and get about 14.9 ohms on all 4. The hunting idle is being controlled by the stepper as I can see it advancing and retarding the throttle while the engine is cold, but what is causing the stepper to act like this is where I am at a loss. Using MEMS_Diag the steps seem to be high to me from what I have read, fluctuating from 90-110 when cold and 60 when hot. 

 

Lambda/oxygen sensor

Not a likely in my opinion as I have confirmed that the heater and relay are functioning. It also is fluctuating from about 200mV to 800mV.

 

Crankshaft position sensor

Even more unlikely as the car runs well and does not knock or misfire.

 

Vacuum leak

I have ruled out the vacuum hoses and fuel trap as being an issue, but their could be a vacuum leak elsewhere in the system. MEMS_Diag shows MAP holding at 30-35kPa.

 

ECU

As much as I'd hope this isn't the case, the ECU could be faulty. I have a second ECU and when I swapped it out the car ran better for half of a day, but this could have been that the car was just warm from trouble shooting. 

 

Next steps (?)

I had a USB to RS232 cable show up today and ran a log on MEMS_Daig. I was hoping someone with a little more experience could shine some light on the results.

 

I'm trying to narrow it down to one or two issues instead of throwing parts at the problem. I live in Canada and get hammered with shipping and duties every time I order parts from mini spares.

 

I'm starting to think its a wiring issue somewhere as the previous owner seems to have made some modifications to the wiring harness, but as far as the sensor wiring go's I have tested it and have not found any issues other than the Air Temp Sensor. 

 

The only other thing I can think of is that someone adjusted the screw on the throttle that sits on the pin of the stepper motor, possibly causing the high steps. This seems unlikely but I don't know enough of its impact to say for sure. I also don't want to adjust this as everything I have read indicate that I shouldn't even be looking at it without an ACT.

 

Any advice you may have for me would be greatly appreciated!

 

[FlyingScot]

Okay when you say it fluctuates at idle what kind of variation are you seeing when cold? When the MEMS system is not at operating temperature there is some variation on idle. 
Did you check for air leaks ? Spraying some kind of solvent around the inlet (like WD40 or the like) to see if the revs rise?

 

You have checked many of the obvious candidates. For vacuum pipes and the fuel trap I have a simple mantra - if you didn’t replace it yourself with new parts then replace them. The elbows split and get porous over time and visual inspection isn’t enough (the connections onto the inlet manifold for example you can’t see and the heat from the exhaust is just below.

 

FS

[humph]

As a starter I agree with Flying Scott swap out the vac pipes.  Most seem to unavailable from Minispares at the moment & they can be pretty pricey for what they are.  Next time I plan to buy a fuel trap and replace the lines with silicone vacuum pipe (3mm I believe) without elbows, this seems to be a popular route at the moment.  The issue comes with making connections behind the inlet as access is terrible.  This is a much cheaper option though.

[Dutchdave's25]

Hi, 

just thinking out loud;

 

you have one input that changes when your engine warms up, in my opinion this could only be the coolant temp sensor.

i guess your best go is to messure the resistance of the sensor..

 

David

Edited by Dutchdave's25, 09 June 2020 - 09:44 AM.

[genpop]

and where is the log-file?

[Pops_Guild]

As has already been suggested, first port of call would be vac pipes, easy for me to say "just replace them" but they tend to be the No. 1 problem with idling issues. What FlyingScot says makes sense, if you didn't replace them you can't rely on them being ok.

 

The (coolant) Temperature transmitter/sender is the next possible candidate. It might not be faulty (not fault code) but unfortunately the water jacket in the inlet manifold that it reads from can get really silted up. Might we worth undertaking a thorough flush through of your coolant system, with extra focus on the inlet manifold. It'll be pretty obvious if a blocked coolant jacket is likely to be a problem when you remove the hoses from the inlet manifold, the pipes corrode really badly. Be gentle when you remove the hoses!

 

The other thing that might not help things during the warm up is a disconnected or broken TPC. Like the temp sender, its located on the underside of the inlet manifold. Often these aren't even connected but also, just because they're old, they fail. Most obvious sign of failure is whether it rattles when removed. In simple terms the TPC is a little heater that the ECU fires up to help the fuel/air mix get to the right temperature quicker.

 

Assuming that when you set the lost motion gap you followed the process as per the Haynes or workshop manual then turned off the ignition, then you've done all you should do with the stepper motor without having a diagnostic tool. It doesn't seem like to me that adjusting the screw-that-shouldnt-be-turned will help you anyway on this problem, just my view.

[FlyingScot]

The engine coolant sensor (CTS) is important in the sense that it is measured by the ECU and determines whether the system is at operating temperature (so switching to closed loop). The gauge is fed In turn from the ECU so if this is correct (middle dot) just under N and your software is reading correct (It takes the data from the ECU obvs) it isn’t likely to effect the running.

CTS s normally implicated in rich running as the ECU doesn’t see operating temperature (88C) due to faulty sensor, blocked water ways in the inlet manifold or broken wiring. 

The SPi system is a “wet” system in that the fuel atomisation by the injector isn’t direct to the inlet port, and is needed to help (it’s like an electronic carburettor). In cold running a faulty or broken PTC (wiring, relay or unit itself) will cause a problem as will an air leak under here. As noted by Pops_Guild often you find them “rattling” as the ceramic inside is damaged leading to the heater wire being broken or shorting to the case.

 

One final point of clarification to your original comments and because I see others repeating this myth. Removal of the battery does not remove the “learning adaption” of MEMS. In Testbook, T4 and several other of the then current aftermarket systems (Crypton Checkmate, Advanced Checkmate Terminal (ACT) and Sykes Pickavant Advanced Code Reader (ACR)  there is an ECU reset function which clears the non volatile values allowing the ECU to relearn afresh. 
 

FS

[Puddles]

Wow! Thanks for all the replies.

 

FlyingScot to answer your question, the fluctuations at cold are up and down about 200RPM every second or two from about 1000RPM to 1200RPM but not super consistent. Once warm it evens out at about 950RPM +-50. It doesn't exactly sound healthy when it is cold hunting. I would be surprised if this was functioning as intended. I sprayed some quick start on the manifold to check for leaks as suggested while the engine was at temp and didn't notice any change in RPM. 

 

Today I picked up some 1/8" vacuum hose (This was as close to 3mm as I could get locally.) and replaced the existing lines. The issue persists and MEMS_Diag show the same values for MAP 30-35kPa. I did notice it get as low as 28kPa at one point while the engine was cold hunting. 

 

As I mentioned in my original post I did notice that the PTC was unplugged and I re-attached it. This did seem to help with the drive-ability of the car when cold, as it stopped bogging down on acceleration. Testing the PTC resistance through the wiring harness I get about 1 ohm while cold. Reading a thread someone mentioned that their defective PTC would read about 7 ohm and the replacement was close to 0 ohms. I have not tested for voltage at the PTC yet. A visual of the PTC through the intake everything looks in tact. I forgot my IR thermometer at home today but I will try to energize the PTC and take a temp reading through the intake to ensure that it is working. 

 

I thought I had attached the log to the original post, but I never completed to upload process. Hopefully it works this time?  MEMS_Diag_2020_06_08_192700.txt   101.39K   6 downloads

 

 

[Puddles]

Ok, so quick update.

 

I checked the PTC heater and these were my results. 

 

I inserted a calibrated thermocouple inside the manifold so it was sitting on the PTC. With a stable temp reading of 44C I jumped 12VDC to the PTC and the temp quickly climbed to 59C. I removed power from the PTC and monitored the temp which continued to climb to 83C at this point I was satisfied that the PTC is functioning and removed the thermocouple.

 

I removed the connection from the PTC, started the car and monitored for voltage. After a 2 minute run with 0VDC I am starting to think the relay pack may be defective. From what I understand the PTC will energize shortly after ignition when the engine reaches a target RPM and turn off once the engine heats up to 75C. Please note the car was reading 33C at this point. What is the best way to open up the relay pack for trouble shooting purposes?

 

I decided to bypass the relay pack and jump 12VDC to the PTC. With the PTC now energized I started the car and let it run for a minute however the hunting idle continued. I did not want to run for to long with the PTC jumped out to prevent burning out the element. From what I understand this is a self regulating element and would be fairly hard to do, but I didn't feel like taking any chances today. Not being familiar with MEMS I don't know if it would immediately start and idle smooth again if this was the issue or if it would take time to re learn idle control.

[genpop]

Althoug the ecu switches to closed loop, your coolant temperature is never higher than 81C. As well Iac does not drop lower then 60 steps.

I would first check and flush the water hose inside the intake manifold, often they are clogged  so the water does not reach the coolant sensor. If that does not help i would change the coolant sensor.Then have a logfile again and check for the IAC-steps.At the moment your short term fuel trim is slightly over 100 that means higher fuel consumption but this may be due to your IAC.Load your logfile to excel and you can see all  that. Or download mems rosco ,there you will have a graph for this.

[genpop]

I forgott to mention the thermostat,check this too, Should be a 88 degree one.

[Puddles]

Thanks genpop,

 

This make sense to me but I would like to run this by you to make sure I understand.

 

The CTS at start up is at 28C. The car had been sitting over night and the ambient temp was 14C. As the CTS was reading higher than ambient this would have caused a lean running condition possibly causing the hunting idle. When the engine is at temp the CTS is only showing 81C which being lower than actual operating temp would cause a rich running condition possibly contributing the the rare occasions that I have a idle issue when at temp. So although the CTS is has not failed it may be out of calibration.

 

My question would be if the temp gauge is fed from the ECU than wouldn't it be reading low if the CTS is only reading 81C?

 

I did download mems rosco yesterday, however I can't load my file into it. What version are you running? I have had to resort to using mems analyser which is not ideal to the point that I mostly rely on breaking down the data in the .txt.

 

I do have a new CTS and 88C thermostat that I ordered but was hoping I wouldn't have to pull the intake manifold.

Edited by Puddles, 10 June 2020 - 10:03 AM.

[genpop]

Temp gauge shows only rough , there are no figures on it. Yes it is fed by the ecu.

I wonder why your ecu switches to closed loop already at 29 zo 30 degrees, it should only do that when temp is at 88 degree.But it seems this depends on the version of the ecu (1.3 or 1.6) Couldn`t figure that out for shure up to now.

Yes mems rosco didn`t read your log file.Download once again, it will do it now!

If you have a laptop you could try to reset the ecu with mems rosco.If you can not activate the reset button let me know there was a bug in mems rosco.But if you can connect mems -rosco to your car,do a logfile there.Besides closed loop the lambda status is important..

If reset is not possible download mems-scan from here:https://github.com/h...S-Scan/releases

You may read haros logfile in mems-rosco too.

[Puddles]

Ok,

 

So I spent the day replacing the CTS. That was even less fun than I thought it would be, and the issue persists.

 

I was not able to change out the thermostat. I had the bolt up soaking in penetrating oil for 4 hours and they didn't budge. I didn't want to put a breaker bar on them, it was starting to feel like I would just snap the bolt up instead of break them loose. 

 

I ran another log. Looks like there was a couple faults this time. The Air Temp Sensor pretty sure is my fault. I had it unplugged earlier and forgot to clear it. not sure what the Fuel Temp Sensor error is though. It looks pretty much the same to me as last time. Maybe someone else can make something of it.   MEMS_Diag_2020_06_11_191407.txt   132.69K   2 downloads

 

I didn't have a laptop with me today, but I will try to do a ECU reset tomorrow, make sure the codes are cleared and run another log.

[genpop]

Temperature is now at 85 and idle deviation at 100 but iac is still over 50 steps.Have you flushed the intake manifold? It is easy: Take off the hoses .Connect a water tube to one of the inlets and flush it by clear water! Please do a logfile with mems-rosco or mems-scan, they write more subjects.

Attached Files

•  Puddies.JPG   51.8K   9 downloads