Use of Memory Clear and Diag Check for ECU Errors instead of MUT/OBD

SkiFreak

Expedition Leader
#31
@SJI: Different countries have different setups, so it would probably be helpful if you told us where you are located.
Welcome to the forum BTW...
 
#32
turn ignition on then disconnect diagnostics check causes orange engine light to flash error codes.

the "tens" digits are flashes are 1.2 seconds then gap of 1.2 seconds, followed by "units" of 0.4 seconds with gaps of 0.4 seconds.

then gap of 2.4 seconds to next code.

if you are seeing a single flash of 0.4 seconds followed by 2.4 seconds off that would be "01" which is Normal (no error).

While ignition on and diagnostics check disconnected then disconnect memory check will flash past codes.

If both are re-connected with ignition still on the memory will be erased, all codes cleared.

To leave codes in memory turn off ignition before reconnecting the two wires.

My vehicle is Aus, 2005.

Good luck.
 

SJI

New member
#33
HI SkiFreak & Gait,
Thanks for the reply and help. I'm based in the UK, Mog on here has sent me an owners workshop manual this afternoon which I am working through at the moment, but initial it seems like I was doing it right but the truck is not giving me a code :-( , all I'm getting is a red engine light flash for about 1 second,then nothing for about 3 seconds, then this just repeats over and over again with no quick flashes, so think it is just diagnosing 10 which from service manual Mog sent me equates to nothing, or maybe even stating everything is normal from finding posts about diagnostics in Mitsubishi cars? Are the codes in the Mitsubishi service manual the only codes given out by the trucks? Is there no other codes for things like crank sensors?
Just re-reading your post Gait - you mention the orange light flashes, on my truck it is the red light giving me this code of 10? Does this mean anything else? The orange light illuminates as this is the light coming on stopping the truck go over 2k rpm and 50mph. It literally comes on and the truck goes into safe/limp mode after 10 seconds, if I keep the revs above 1500rpm it keeps it at bay until the revs drop again?
Any pointers or help on what to start looking at would be great, i know its potentially like a needle in a haystack but I'm running out of options.
Cheers Steve
 
Last edited:
#34
sorry, my bad, its the red light that flashes ..... its a couple or more years since I had problems.

Be aware that there may be different codes for different years/models.

Thus, I would also interpret your long flash as "10" but there is no "10" in my manual and "01" is my normal.

(silly aside .... it took mathematicians a few thousand years to invent zero and a few seconds for some software engineer to ignore it).

My engine is 4D34.

To be honest, I haven't a clue. Most odd. But when I meet problems like this I usually develop a few methodical in-situ tests to see if I can change the problem. Subtly different to "trial and error" which is very much a last resort and may seem to solve the problem without ever knowing "why". Replace ECU is the sort of trial and error I've seen happen. This may be a bit long winded but it will hopefully let you follow the thinking and develop your own lines of thought. very important to keep open mind and not leap to conclusions.

Apologies if its a bit slow or long, or I'm teaching grandmother to suck eggs. I don't know what you know etc....

My thinking is that the ECU seems to be telling us there is an error, but seems to provide no useful error code. There's a couple of possibilities and a couple of approaches.

First is that the ECU has tied itself in knots. That can happen in any computer, just less likely in low level control systems. Normal answer is to reboot. But there's no reset button (design failure #1). If you haven't tried already, disconnect battery, turn on ignition (to help discharge any small capacitors etc) and leave for a few hours or overnight (there may be internal small batteries). Don't ask why about the few hours, just a long ago experience with a Ford. I don't know how long it takes with out trucks. Reconnect the battery. Give it a few minutes to get its house in order then start engine and see what happens.

If something different happens that's good. If not it either means it didn't reboot or the problem is elsewhere.

Second is to give the ECU a real problem and see if it diagnoses it correctly. A note here. My view is that the diagnostics is aimed more at preventing pollution than providing for engine reliability. Either way the ECU likes to see all of its sensors, then does a reality check on the measurements it gets, then uses the measurements for control.

Giving the ECU a real problem can tell us that our use of the diagnostic switch is correct and its working, or that the ECU isn't functioning as expected.

So, disconnect a sensor and see if it creates a relevant error code. Engine temperature sensor is fairly easy to locate. On mine its near the thermostat cover. Don't confuse with boost pressure sensor which is near but in the inlet manifold. Disconnect it (remove plug from sensor). Observe the effect on "flashes".

The error code in my manual is "21". The internal ECU action is to turn on the orange light and assume the temperature is 80 deg C. The vehicle is driveable. Its a fairly benign error. It doesn't cause limp mode.

Note that I haven't tried disconnecting the temperature sensor and observe the error on my vehicle. I may be wrong.

Both of the above tests are definitive if something changes. We've made a change and the vehicle changes what its telling us. If nothing changes we have to think again. Sorry to labour the point, if nothing changes we have either taken a wrong action and our expectations are wrong or the problem is elsewhere. All the options about whether our tests/actions are valid or whether the problem lies elsewhere. An open mind.

If disconnecting the temperature sensor doesn't create a change I'll disconnect mine to check. Then think again.

I should add that there are relatively few errors that put the engine in limp mode. If we can't identify the problem by working forwards from the ECU we may have to try working backwards from the few conditions that result in limp mode. One at a time .............. did I mention its important to only change one thing at a time and observe results.

There's no magic wand with this stuff.

Good luck.
 
#35
sorry, my bad, its the red light that flashes ..... its a couple or more years since I had problems.

Be aware that there may be different codes for different years/models.

Thus, I would also interpret your long flash as "10" but there is no "10" in my manual and "01" is my normal.

(silly aside .... it took mathematicians a few thousand years to invent zero and a few seconds for some software engineer to ignore it).

My engine is 4D34.

To be honest, I haven't a clue. Most odd. But when I meet problems like this I usually develop a few methodical in-situ tests to see if I can change the problem. Subtly different to "trial and error" which is very much a last resort and may seem to solve the problem without ever knowing "why". Replace ECU is the sort of trial and error I've seen happen. This may be a bit long winded but it will hopefully let you follow the thinking and develop your own lines of thought. very important to keep open mind and not leap to conclusions.

Apologies if its a bit slow or long, or I'm teaching grandmother to suck eggs. I don't know what you know etc....

My thinking is that the ECU seems to be telling us there is an error, but seems to provide no useful error code. There's a couple of possibilities and a couple of approaches.

First is that the ECU has tied itself in knots. That can happen in any computer, just less likely in low level control systems. Normal answer is to reboot. But there's no reset button (design failure #1). If you haven't tried already, disconnect battery, turn on ignition (to help discharge any small capacitors etc) and leave for a few hours or overnight (there may be internal small batteries). Don't ask why about the few hours, just a long ago experience with a Ford. I don't know how long it takes with out trucks. Reconnect the battery. Give it a few minutes to get its house in order then start engine and see what happens.

If something different happens that's good. If not it either means it didn't reboot or the problem is elsewhere.

Second is to give the ECU a real problem and see if it diagnoses it correctly. A note here. My view is that the diagnostics is aimed more at preventing pollution than providing for engine reliability. Either way the ECU likes to see all of its sensors, then does a reality check on the measurements it gets, then uses the measurements for control.

Giving the ECU a real problem can tell us that our use of the diagnostic switch is correct and its working, or that the ECU isn't functioning as expected.

So, disconnect a sensor and see if it creates a relevant error code. Engine temperature sensor is fairly easy to locate. On mine its near the thermostat cover. Don't confuse with boost pressure sensor which is near but in the inlet manifold. Disconnect it (remove plug from sensor). Observe the effect on "flashes".

The error code in my manual is "21". The internal ECU action is to turn on the orange light and assume the temperature is 80 deg C. The vehicle is driveable. Its a fairly benign error. It doesn't cause limp mode.

Note that I haven't tried disconnecting the temperature sensor and observe the error on my vehicle. I may be wrong.

Both of the above tests are definitive if something changes. We've made a change and the vehicle changes what its telling us. If nothing changes we have to think again. Sorry to labour the point, if nothing changes we have either taken a wrong action and our expectations are wrong or the problem is elsewhere. All the options about whether our tests/actions are valid or whether the problem lies elsewhere. An open mind.

If disconnecting the temperature sensor doesn't create a change I'll disconnect mine to check. Then think again.

I should add that there are relatively few errors that put the engine in limp mode. If we can't identify the problem by working forwards from the ECU we may have to try working backwards from the few conditions that result in limp mode. One at a time .............. did I mention its important to only change one thing at a time and observe results.

There's no magic wand with this stuff.

Good luck.
AWESOME POST, some good words of wisdom on diagnosing problems that I should practice myself.
 

SJI

New member
#37
sorry, my bad, its the red light that flashes ..... its a couple or more years since I had problems.

Be aware that there may be different codes for different years/models.

Thus, I would also interpret your long flash as "10" but there is no "10" in my manual and "01" is my normal.

(silly aside .... it took mathematicians a few thousand years to invent zero and a few seconds for some software engineer to ignore it).

My engine is 4D34.

To be honest, I haven't a clue. Most odd. But when I meet problems like this I usually develop a few methodical in-situ tests to see if I can change the problem. Subtly different to "trial and error" which is very much a last resort and may seem to solve the problem without ever knowing "why". Replace ECU is the sort of trial and error I've seen happen. This may be a bit long winded but it will hopefully let you follow the thinking and develop your own lines of thought. very important to keep open mind and not leap to conclusions.

Apologies if its a bit slow or long, or I'm teaching grandmother to suck eggs. I don't know what you know etc....

My thinking is that the ECU seems to be telling us there is an error, but seems to provide no useful error code. There's a couple of possibilities and a couple of approaches.

First is that the ECU has tied itself in knots. That can happen in any computer, just less likely in low level control systems. Normal answer is to reboot. But there's no reset button (design failure #1). If you haven't tried already, disconnect battery, turn on ignition (to help discharge any small capacitors etc) and leave for a few hours or overnight (there may be internal small batteries). Don't ask why about the few hours, just a long ago experience with a Ford. I don't know how long it takes with out trucks. Reconnect the battery. Give it a few minutes to get its house in order then start engine and see what happens.

If something different happens that's good. If not it either means it didn't reboot or the problem is elsewhere.

Second is to give the ECU a real problem and see if it diagnoses it correctly. A note here. My view is that the diagnostics is aimed more at preventing pollution than providing for engine reliability. Either way the ECU likes to see all of its sensors, then does a reality check on the measurements it gets, then uses the measurements for control.

Giving the ECU a real problem can tell us that our use of the diagnostic switch is correct and its working, or that the ECU isn't functioning as expected.

So, disconnect a sensor and see if it creates a relevant error code. Engine temperature sensor is fairly easy to locate. On mine its near the thermostat cover. Don't confuse with boost pressure sensor which is near but in the inlet manifold. Disconnect it (remove plug from sensor). Observe the effect on "flashes".

The error code in my manual is "21". The internal ECU action is to turn on the orange light and assume the temperature is 80 deg C. The vehicle is driveable. Its a fairly benign error. It doesn't cause limp mode.

Note that I haven't tried disconnecting the temperature sensor and observe the error on my vehicle. I may be wrong.

Both of the above tests are definitive if something changes. We've made a change and the vehicle changes what its telling us. If nothing changes we have to think again. Sorry to labour the point, if nothing changes we have either taken a wrong action and our expectations are wrong or the problem is elsewhere. All the options about whether our tests/actions are valid or whether the problem lies elsewhere. An open mind.

If disconnecting the temperature sensor doesn't create a change I'll disconnect mine to check. Then think again.

I should add that there are relatively few errors that put the engine in limp mode. If we can't identify the problem by working forwards from the ECU we may have to try working backwards from the few conditions that result in limp mode. One at a time .............. did I mention its important to only change one thing at a time and observe results.

There's no magic wand with this stuff.

Good luck.
HI Gait. Thanks for the great email and help in trying to sort, and sorry for the delayed acknowledgement, work has been busy and just got an afternoon to look at the truck. So the evening you posted I disconnected the battery and left ignition on like you suggested, and made no difference a day or two later, so today I went out disconnected a couple of sensors and all were picked up as a fault code on the ECU , for example MAP sensor removed flashed up as 32, so guess ECU is understanding faults and errors for what it is told to look for. I looked at all the earths and appeared OK, the main chassis earth looked a bit flimsy so removed upgraded and cleaned all that up but unfortunately still no difference.

So are there any obvious parts of the truck which would cause limp mode, but would give no diagnostics? For example I disconnected crank sensor on bottom front of block and this didn't give an error code when unplugged, but the truck still started even though very lumpy on idle, I wondered if this was effecting fuelling but is obviously working as disconnected it was very lumpy and misfiring, so guessing this isn't faulty. I'm guessing the turbo/actuator wouldn't have a diagnostic code but would cause limp mode, so maybe let it stuick air through the intercooler only and see if light comes on? maybe exhaust brake stuck on would cause it by building up gases and pressure? Any other parts you guys can think of, or am i looking at these parts in the wrong way?

Thanks again for the help and posts sofar!!
 
#38
it gets harder now ...... and I'm making it up as we go,

we've sort of made two steps forward and one back. We're still not certain the ECU is working correctly as we still have limp mode but no useful error code. We do know its capable of recognising and reporting error codes and that our process for reading them is working. "Yeah". We can probably work on the assumption the ECU rebooted, while keeping an open mind.

"limp mode" is a term that arrived after our vehicles were manufactured.

in my manual, and hopefully the one from Mog that you have, the table of error codes has columns
Diagnosis Code, Fault Location, Warning Lamp, Electronic Control Unit Countermeasure(s) in the case of failure (backup mode), Drivability.

Hopefully the same as what you have in front of you.

We are interested in drivability. I have three possibilities in my table. Drivable, Undrivable, and Back-up mode driving possible.

Its the third, Back-up mode, that I believe we are interested in.

There's another bit of information we have. Its the orange engine light which is lit. Some codes that go to back-up mode turn red on, others orange.

I see only the following which cause orange light and back-up mode.

12 - pre-stroke servo and 13 - pre-stroke sensor. The ECU countermeasure for both is "pre-stroke control stop engine revolution speed limit (approximately 1800 rpm)."

22 - control rack position sensor. ECU countermeasure is "switch to the sub-control rack position sensor control"

32 - boost pressure sensor. ECU countermeasure is "set the boost pressure to 0kpa"

There is only one "undrivable" condition that has orange light, the rest are red light. That one is "engine running backwards" which we can probably ignore as it would be rather evident. :)

We still have to consider that the ECU is "lying" to us, but for now assume its working ok, just that we don't know what its telling us.

So ....... next step is to clear the past error codes we created in our previous testing so we have a clean sheet. "Should be in the manual" - with ignition on, disconnect both mem and diag the reconnect. Check that current and past codes are cleared.

Confirm that still giving us our strange orange light but no useful error code condition. This is our known reference point that we need to keep returning to as we conduct further tests one at a time.

We can probably discount "32" as from last post you've already tested. In my manual its boost pressure sensor, you refer to as MAP. I'm happy with that.

Which leaves us with 12, 13, and 22.

I would now check the above logic against the manual you have. Then locate the three sensors and, one at a time, disconnect, check error code, observe engine operation (has operation changed from our reference condition, particularly 12 and 13), reconnect. One at a time.

Can you also describe in a bit more detail the symptoms. Just checking that what you describe as "limp mode" is what we think it is. I understand the limit of 2k rpm and 50 mph but I don't understand "if I keep the revs above 1500rpm it keeps it at bay until the revs drop again". Particularly interested if we manage to create error codes 12 and 13 if it changes the engine operation.

We are really still just methodically checking, while flumoxed at the inconsistency in the ECU - an error but no error. It doesn't help that when I got error "32" in Kyrgyzstan I really couldn't tell it was in back-up mode.

All this could also be red herring, while the hope is that it gives us some clue.

Also, keep checking things like those you mentioned. Hard to check the exhaust retarder but if there's a bit of slack and you can move it a little by hand its probably ok.
 
#39
PS

I checked the blink code for my vehicle which has no problems.

Ignition on, red and orange engine lights on, pull diagnostics switch, red light off, orange light flashes. Single one second flash followed by 6 second pause. Not a recognised code. Not "1", not "10". I'm guessing "Normal".

I guess my vehicle is different to yours.

I'm now doubtful that the vehicle is in "limp mode". The "back up" emergency mode that limits engine speed on my vehicle to 1800rpm is in response to sensors 12 and 13 where the ECU uses the alternative of the engine speed sensor to change fuel flow.

Can you provide a bit more detail of how it ran before fault occurred and now?

How long have you had the vehicle? What is battery volts with engine off and engine running? When were fuel and air filters last changed?
 

SJI

New member
#40
Hi Gait.

I think mine does the same orange light flash on diagnostics check as yours 1 flash and then a long pause and then repeats, nothing like the 32 reading I got when unplug MAP, so think they maybe the same - mine is a 2002 UK model.

Thanks again for your thorougher reply, it is so appreciated!! I do have that page MOG sent me, 13-32 I have as the page reference. However all mine says is as you mention Defective Pre-Stroke or Defective accelerator pedal position sensor. Not sure how you tally these up with fault codes though unless there is another table in the book I've missed?

With the pre-stroke problem is this all built into the Bosch injection pump housing mounted to the side of the block below injectors, as this part seems very expensive, and I wouldn't know how to test that?

With regards to fault 32 cant be the problem as the boost is read by the MAP which isn't seeing this as the fault but I do have a boost sensor in the pipework which is connected to the vacuum pipework, I'm wondering if this could still cause a problem. Think I will pull the vacuum from it cap it off so the pressure switch cant make a reading and see what happens, also it has no wires to this sensor just vacuum pipework so guessing as no electrics it couldn't generate a fault code for this?

The 'limp mode' (orange solid warning light) used to come on very occasionally, sometimes if I would then slow down for a junction it would switch off, and then may come back on once I pulled away, sometimes I could switch off ignition, switch it back on carry on with journey light would go off and the fault would go, it might come back 5 on and limit to 50MPH mins later or might be a month later. It made no difference if the truck was empty or was carrying 2 ton of weight it would do it in both scenarios. Then the other day it just came on straight away (about 10 seconds after start up), and now it does this all the time. When I say limp mode you are indeed right, it wont rev past 2000 rpm and wont go above 50 MPH. When I said about keeping the revs high, what I meant was if I rev the truck and keep the revs say above 1500RPM the limp mode and light doesn't illuminate until the revs drop to idle. You can really tell that the truck is in back-up mode it feels so slow and sluggish as you constantly are changing gear at 2k RPM until you reach the 50mph. Before this happened the truck would run great, pull weight well, run really well. It was always sluggish/slow (but nothing like this) on startup but I think that was just because it was cold and hadn't warmed up, after 5/10 mins of driving it would be fine.

My exhaust brake, looks like its all vacuum controlled, with I guess a butterfly in a tiny section in the exhaust, so only way to see if its stuck I can see is to remove exhaust to get access to the housing, unfortunately no rods to adjust like a linkage which I was hoping for to see movement.

I've had the truck for about 2 years now, I changed the battery recently as thought the old one wasn't holding charge, and the fuel and air filter changed the day after this problem started thinking they may have just been clogged or excess water in fuel filter but unfortunately no quick fix.

PS
one other thing it did do very strangely before this fault came on all the time, is it wouldn't start at all one morning, dash lights came on, but no crank, no noise just nothing, and then weirdly after trying about 10 times it fired up as if nothing was wrong, that that night I did notice a 10A fuse had blown for something but with it blown it still worked when it did start so I think this may have just been another issue and not related but thought I would mention it.

On these does the crank sensor only give a reading which is needed for start up or will it have to give a reading to the ECU at all times for fueling, I am wondering if this was faulty would it be able to give a good enough signal to start the truck without a misfire, but give bad data for fueling - I guess if it can give a reading to start correctly it will be able to give a good enough signal for fueling needs when idling or driving?
 
Last edited:
#41
thanks,

I'm still making it up as I go along .....

definitely difficult with "error but no error code".

By way of explanation (your question about crank sensor) ... the way the fuel system works .... inputs are accelerator pedal position, engine speed, engine coolant temperature, idle speed adjustment and boost pressure. The control rack is moved based on data within the ECU (its map). The control rack actuator moves the control rack until the control rack position sensor says its in the right position. If the sensor fails there is a sub-control rack position sensor used instead.

The rack position sensors, actuator, and engine speed sensor are all in the fuel rack on the side of the engine.

A little bit more complicated as there is "pre-stroke" for timing and the above for rate. But all part of the the same "sub-system" within the ECU.

If both fuel rack sensors fail then fuel is delivered based on engine speed only ... limited to 1800rpm in my engine.

As far as I can tell the crank sensor is not involved in fuelling.

The consistency of 2,000 rpm you describe has to be a clue. Is it a "hard" or "soft" stop. ie when stationery, out of gear, foot on accelerator, does it rise quickly then suddenly stop rising always at the same revs or does it slowly approach the limit? Is it the same under load and not?

Something crisp and very consistent tends to be digital, ECU related, though may be mechanical. Something "soggy" tends to be fluid flow (air, fuel, exhaust) or mechanical (worn) related.

At the moment I'm tending to home in on the fuel rack as that and the turbo boost pressure are the only parts that create errors that lead to back up mode. Without discounting anything else.

Useful test would be to disconnect wiring to the fuel rack. See how the ECU reacts to not having rack position sensors, actuator, and engine speed sensor. Depends on engine. The engine will not start - don't even try - but we should see error codes. If we are lucky its something silly like a bad electrical connection in that area.

Definitely a challenge trying to find a problem that the ECU sees but doesn't tell us about. - that's a software bug. But there's also a real problem.

Also. If you have a voltmeter measure battery voltage with engine stopped (around 12.7v) and engine running (around 14v). Just picking up on the clue of battery replacement and the "won't start" morning. Do you do much driving at night?

Sorry, still more questions than answers.
 

SJI

New member
#42
Your thoughts and your process is all helpful! It may just solve it which would be great!!

Right so if it is the one of them other sensors in the fuel rack, that cant be tested individually, and it would be a large replacement part of all the fuel pump assembly mounted below the rail, (couldnt post a link so private messaged you the part I think we mean) or am I looking in the wrong place?

With regards to the fault, the truck will drive and run smoothly upto 2000RPM and then just feel like it hits a rev limiter, when it started doing it while driving I could be cruising at 70MPH, and if the fault happened it would literally drop the revs and speed to 2000RPM, and slow the truck down straight away. Once I was overtaking with acceleration and it kicked in and I thought my head was going to go through the windscreen with the jolt from it dropping the revs back down. When in limp mode it is a noisy engine, as if something has changed or opened to put it in this limp mode for safety.

Its Saturday here in the UK so am going to have a look at the truck today as not working, and I will disconnect that fuel pump (as think this is where you are saying those sensors are inside - whats in the link above) and see if a fault code is given, if I'm looking at the wrong part where these sensors you mention are inside please let me know. I will take battery voltage readings too, trying plugging the vacuum boost pipe I mentioned hopefully to bypass the sensor, I've also found in the workshop manual how to test the throttle pedal so as this was in the trouble shooting table will test this how it describes with voltmeter to see if anything obvious is there.

I dont us the truck at night, very little use during the day to be honest, maybe a couple of times a week. No problem with all the questions, appreciate your help and knowledge.
 
#43
all useful info, sounds like a very "hard" condition. Suggests the ECU or something mechanical is "switching". Just the ECU not telling us what it thinks the fault is.

I think we can discount the exhaust retarder (for now) as that would be much "softer".

the link you sent me to the fuel pump looks right. Looks like one connector for lots of things inside. Really we are still probing to see if we can change or detect something.

a sort of methodical "change something, see what happens to ECU and engine operation".

The voltage check is just in case. The engine will run quite happily without alternator charging for quite a long time.

I believe its too early to think about replacing parts but sometimes there's little option - but the trick would be to pick which part!

When you disconnect the wires from fuel pump have a look to see how clean the pins in plug and socket are.
 

SJI

New member
#45
Ok good stuff, I'm not looking at replacing the pump just yet, completely agree would rather find the problem for sure first!! I have disconnected the pump and was getting fault code 26 and 11 when that was disconnected, which are Governor Servo, but I dont have a fault code 26 in my list of codes, but either way it is picking up a fault which isnt there when it is connected, so wondering if this may be not the one.

As for voltage check am getting 12.4 volts on idle and 13.85 when running.

Just started looking at the accelerator position assembly, and what I have on section 13-54 of the manual is a complete;y different loom and pedal assembly from what I can see, so come back inside and on computer to due and look through the manual again!
 
Top