Car Only Runs Well with MAP Sensor Unplugged

[mozayyani97]

Hello,

Picked up a 91 Elan a little bit ago that has me stumped. After bringing it back to life with a new fuel pump inertia switch, I cannot figure out why it runs the way it does. As in it barely runs, the exhaust sounds likes popcorn and you have to constantly play with the throttle to get it to stay running. It is not throwing any codes. However, if you pull the MAP sensor, it runs seemingly perfect (albeit wont rev past 4k rpm because its in limp mode). Pulls nice, revs smooth, idles good etc. Between me and the PO, the things we have tried:

multiple ECUs
New plugs, coilpacks, wires
New injectors, fuel pump, fuel filter, fuel pressure regulator
New CLT sensor, new MAP sensor, updated hall effect sensor

I've also done a timing belt/water pump job on it to get rid of a top end rattle it had as well. I managed to set the base timing to 16 deg BTDC as well with it in limp mode.

Finally got elanscan working and nothing seems out of the ordinary to me except at idle its commanding a ~4ms injector pulsewidth which seems high compared to what I've read but it runs well at that pulsewidth with the map sensor unplugged. If I let it warm up enough I can get it to idle, albeit poorly, and the MAP sensor reads like ~0.7 bar but honestly I think thats due to the poor idle quality. Otherwise, not seeing codes, all the temperature data looks fine, throttle position sensor looks fine, no knocks detected, timing doesnt look like its doing anything funky. My next guess is maybe the magnetic pickup on the CAS? If anyone knows what sensors throwing it in limp mode does NOT use, that would be really helpful. I'm pretty close to throwing a megasquirt at it honestly lol

Thanks for the help

[dapinky]

Hello and welcome.

I am not an expert at diagnosis of problems (although I can use a hammer and a spanner)....

... one thing, you say that it goes into limp mode and works when the MAP sensor is unplugged, but if you disconnect a different sensor to force it into limp mode what happens?

Also, if you short out the wires at the ALDL plug to get it into Service Mode (most sensor inputs ignored), does it run and drive in any form?

[Tuga2112]

Im surprised Dave didnt mention this already, as it was one that happened to me last year and We both were scratching our heads for weeks trying to figure it out.

I Had a rather similar scenario to your description. where the MAP being disconnected or inducing a vacum leak on it made the car just about start.

my issue was a direct consequence of incorrect instalation of the fuel pump. but the root cause was effectively not enough fuel pressure.

I suggest you plug the MAP and unplug the vacum hose from it. see if the behaviour changes. in my case, being unplugged or having a vacum leak made no difference, and there was no CEL neither codes.

if you do a fuel pressure test, i would greatly appreciate pictures of both the tools used and how you connected them to the car, as I have no idea how i would do that, and seeing an example would be very helpfull.

[mozayyani97]

Shorting the ALDL pins to put in service mode and it still runs bad with the MAP sensor plugged in. I should note that unplugging the MAP sensor I mean electrically. I wouldn't think it's a fuel pressure problem because it runs perfectly with the MAP sensor disconnect (no vacuum leaks).

A question, I noticed there is a 3 pin plug coming off the harness going to the ECU. It's got a ford logo and I can't see where it plugs into. Is this an OEM thing?

For reference this is a 1991 US market turbo car

[Simon_P]

As others have said or alluded to.unlug any of the sensors will force limp mode so you can't conclude it's the Map
The base timing should be set in service mode (don't understand your limp mode comment)

Post the scan

Post a picture of the plug.

[mozayyani97]

I'll get some scans/pictures posted when I get home today. I did some more fiddling yesterday. If I get it warmed up to operating temperature, I can shut the engine off, plug the MAP sensor back in, and have it running. When I do that, it shows I am idling at 0.45 bar with around 3ms injector pulsewidth. There seems to be a lot of hesitation below 3k rpm but above that it feels good. I did notice that after doing a lap of my neighborhood it threw a VSS code, may need to investigate that further.

Also of note, the battery light is on because of some O2 sensor wiring issues it seems. The O2 sensor voltage in elanscan seemed okay to me but my issues occur more during warm up when it should be all open loop anyways. Also of note, I also never see the closed loop flag turn to on in elanscan. ECU ID is 1499.

I'm going to try and check fuel pressure too, friend of mine says the test ports on GM cars of this era are 3 AN threads.

[dapinky]

I think the issue is possibly with the O2 sensor wiring.....

.... a 1499 chip will run 'open loop' until it warms up, then switches to 'closed loop'. It appears that your issue is when cold, so maybe it's something to do with it - but I'm crap at diagnosing issues!

[chrism]

I agree with dapinky:

In my experience the O2 sensor heater 'gets' it's voltage when the ign light is off and the engine is running - so I reckon your sensor probably won't get hot enough to go 'closed loop' as you warm up, unless the light goes out (btw I hear that some replacement alternators from one of our suppliers apparently doesn't switch off that red light...)

In that case you'll stay defaulted to 'open loop' - if you've got an SE that's been 1499 converted like mine - and if you've got a mixture 'pot' still attached I reckon it'll run at whatever that's adjusted to - which might be well out, causing your symptoms? (Successful 'closed loop' will have over-ridden this.)

Elanscan will tell you what your air:fuel ratio is. It should be 14.7:1 on 'closed loop' (btw I found the on/off info flag at 40 in the new EScan...)

When EScan recently showed my a:f as 14.5:1 I discovered my sensor m/plug had de-pinned behind the wheelarch plastic and I was running 'open loop' absolutely ignorantly at 14.5:1 on my pot adjustment for a year or so wondering why I was using a little more petrol and MoT emisions were not quite as good as before . New Escan interest prompted me to use it again or I might never have been aware (thanks Matt, Ken, Ron and any others I might have missed ...)

You can also use a multimeter to check the O2 sensor m/plug heater voltage is 'there' by earthing it with the engine running; I can't remember which of the 3 wires it was atm tho' - but there is a LEC post...)
edit: almost certain it was the brown lead that gave the ~12V heater supply - pin B apparently...someone might verify, I hope...

I wonder if this might help with your symptoms?

[Tuga2112]

why are you guys refering to closed loop when the car is a 91 ?
i didnt see anything in the previous messages mentioningin a close loop conversion ? did i miss a important sentence ?

[Fredjohn]

why are you guys refering to closed loop when the car is a 91 ?
i didnt see anything in the previous messages mentioningin a close loop conversion ? did i miss a important sentence ?

Yes you did! His last post says his ECU Prom ID is 1499. This is the required chip for closed loop.

[mozayyani97]

Ok uploading a log where I can get it to idle (poorly) with all the sensors plugged in. Like I said, if I let it warm completely up itll run decent with normal MAP readings (~0.45 bar) but with some hesitation until you get into boost and then it feels pretty good. Also attaching photos of some plugs I'm confused about. The white ones almost look like they want to plug up but theyre coming off the same bundle of wires. Would definitely appreciate a second set of eyes here.




Uploaded the log to my gdrive because the forum wont let me upload: https://drive.google.com/file/d/1rv4qwh ... sp=sharing

[chrism]

_DSF0341a.jpg

Screenshot (77).jpg

mozayyani97 I'm no way an expert but I compared your scan (rhs) to one of mine (lhs) and your prom id is 1499, your a:f ratio is 7.4 compared to my 14.7, your O2 sensor isn't cycling compared to mine (pictured) and according to your code 40 you're not 'closed loop'.

I hope this helps to confirm something that we might have suspected...

A more conventional way of sending your log file is to firstly compress it to a zip file and then attach that to your LEC post.

I can't help with diagnosis but I hope someone else can look at your log and come along with something more helpful very soon?

edit: I added my ecu pics cos I can't see that white multiplug anywhere under my dash...

[RonR]

From the photos your car appears to be Left-Hand Drive, so that would explain the white connector which is either LHDHC-B (LH Electric Window) or SHC-B (Security Harness, probably central locking related), have a look on the last few pages of the electrical manual to see the connector shapes.

The O2 sensor voltage should fluctuate between 0.1 and 1.0 Volt like on Chris' scan, yours is fairly steady at 0.43 Volts which is the Voltage generated by the ECU, so I'd hazard a guess that your O2 sensor has failed in some way that hasn't produced a fault code, probably the heater circuit. Could be the O2 sensor wiring or possibly the sensor itself has given up the ghost.There are tests you can do to check the circuits in the Closed Loop manual, EMJ.4 Page 32-33.

Edit 1: Didn't see your previous post saying the Battery light is staying on: The Battery light circuit powers the O2 sensor heater - Has the Alternator been replaced recently? Some supposedly compatible Alternators have Regulators which don't produce enough current to power the heater, resulting in the Battery light staying on and O2 sensor troubles.

Edit 2: If you need to change the regulator the original part number is - Denso 126000-0440, but 126000-0430 will also work. Choosing one at random from a g**gle search: https://www.maniacelectricmotors.com/80904381.html. The indication that it produces enough current is either "Active lamp function - with output." or "L-terminal drives the choke."

[chrism]

...and when mine went open loop w/out me knowing for a year or so it didn't throw up a fault code, I suspect because it defaulted to the original mixture adjustment pot that it was running on before I added the O2 sensor (which was 14.5:1 - so not different enough to throw up immediately noticeable symptoms)...

[mozayyani97]

Appreciate the responses. It is a LHD (US market car) so not trim pot. Makes sense on the white connectors, I found a few more throughout the car. I'm guessing they just prewire all the cars for the options and only install the modules if applicable. I guess the black ford connector is some previous previous owner hackery? I got the car with the 2 (1499) spare ECUs but weirdly the one in the car is some prom ID I don't recognize (4799?). It runs the same on all ECUs.

I've known about the O2 sensor wiring/alternator thing for a bit but I always figured it shouldn't cause my issues since I'm closed loop, the computer is ignoring O2 sensor data anyways (or at least it should be). Though I suppose it would be smart to at least fix the issues are clearly there before chasing others.
It seems like a pain to dig into the alternator to replace the regulator. Has anyone ever just ran a fresh ignition switched 12V line to the heater?

[HJ2]

If you unplug the O2 sensor completely, the car should stay open loop but otherwise perform like it should.
This could rule out MAP or O2 sensor…
But you’ve probably did that allready…

[dapinky]

I've known about the O2 sensor wiring/alternator thing for a bit but I always figured it shouldn't cause my issues since I'm closed loop, the computer is ignoring O2 sensor data anyways (or at least it should be).

To me, this looks like either a bit of mistyping, or incorrect understanding.....it sounds backwards to me (but what you say may not be what I hear!)

Yes, you are 'closed loop' - but only once the engine warms up - initially it will be 'open loop'.

The ECU is looking for an input from the O2 sensor in 'closed loop' mode - I assume it approximates a value until that point, or ignores it completely, but I don't know.

Anyway, your elanscan shows that the O2 sensor appears to be giving an incorrect value and the A:F ratio is all wrong....

.... I'm pleased that you have been able to try a different ECU (that would have been a good thing to try, but now unnecessary, unless there is suspicion that all 3 are broken in the same way!), and it made no difference.

It still comes back to a bad O2 sensor and/or wiring to it and/or alternator output.

Of course, one thing I have overlooked, and seems to actually make a difference, is the MAP sensor - I am assuming that when it was replaced it was with a 'correct' 2-Bar sensor, and not a higher or lower value one?

[Fredjohn]

Concur with dapinky.

Do a continuity check on the O² feed to the ECU, then do a voltage/continuity check on the O² heater wires. If all OK, I would suggest getting a new sensor. This is much easier and cheaper than doing the alternator.
The wiring diagrams in the manual will help doing these checks.

[chrism]

Um...imo I think I'd get the alternator and its charge lamp working first, since the O2 sensor depends on that lamp being OFF to work...



When I discovered I wasn't 'closed loop' anymore I did some checks:

With the engine ON and the charge lamp naturally OFF I used a multimeter to check for 12V on the brown/yellow heater supply lead, which I didn't get. The black lead is the heater return to earth - which is worth checking for continuity too - and the grey/green is the signal lead. (ps KenS and I - but mostly KenS - found that we think the manual might give the wrong advice with the test process...)

I assumed a faulty sensor and bought one from RAuto - but please be aware; the lead on the wiki-suggested Bosch 13277 wasn't long enough whereas the Bosch 15701 and Walker 350-33022 (I know) are a perfect fit.

Anyway, as previously mentioned the fault was eventually found to be a de-pin in the m/plug behind the drivers wheelarch liner and now I've got a shiny new Walker on the car, a shiny new Bosch 15701 and a used SJS Denso(?) as spares and I returned the 13277; bad idea...don't ask...

I went to RA because there was nothing on our UK shelves at the time (end of Feb '24) but, cor, did I take a financial hit and I'm hoping someone will learn from this and might want to buy the spare Bosch 15701 (which will certainly serve to help me out )

[mozayyani97]

To me, this looks like either a bit of mistyping, or incorrect understanding.....it sounds backwards to me (but what you say may not be what I

Ha yes you're right, what I meant was my car is stuck in open loop presumably because of an alternator with the wrong regulator or a fault in the heater circuit causing the O2 output never spitting out values that make the ECU satisfy the conditions required to go closed loop. With that said, while I know that is an issue, it shouldn't affect engine operation in open loop so I'm hesitant to dive into that repair process because I don't think it'll fix the main issue I'm having. I have probed the plug with the engine running and I do get 12V out the brown wire but I haven't tried diagnosing the sensor itself

Also, when you say I can just replace the sensor, is there a sensor that has a heater that draws less current or something?

I should maybe take a video of the car, it barely runs in open loop mode with the MAP sensor plugged.