Need help from the T2 FB guys, not running very well
07-17-09, 06:10 PM
#26
Fail - unless I am now the proud owner of 3 bad ecu's the ecu is not the problem. I'm thoroughly baffled on this one, it runs just fine on the N351 NA ecu, but won't run for crap on any of the 3 N374 ecu's I have. It has a strong smell of gas as it tries to start, and once it does start it doesn't run well at all, same symptons as above.
Guess at this point i've gotta pull the engine harness out and start tracing wires.....
Guess at this point i've gotta pull the engine harness out and start tracing wires.....
07-17-09, 06:33 PM
#27
Weird. Only thing I can think of is something that is part of the TII circuit, but not part of the NA setup (i.e. knock sensor). I know you weren't showing any codes, but it might be worth checking the wiring at the knock sensor and checking the sensor itself if you can. I would also go through and probe all of the voltages at the ECU and compare to what they should be. This test should be with a warm engine. If you can't get it warm with the TII ECU, warm up with the NA one and then switch over.
I know with the codes, it will usually only throw one if something is disconnected or shorted. A sensor could just be way out of range and not show a code.
It still sounds ECU/sensor/harness issue to me. Just need to track down the exact problem.
I know with the codes, it will usually only throw one if something is disconnected or shorted. A sensor could just be way out of range and not show a code.
It still sounds ECU/sensor/harness issue to me. Just need to track down the exact problem.
07-17-09, 09:39 PM
#28
Thanks Kent, I agree, it has to be something that isn't part of the NA circuit, of which there really are only a few things, I guess at this point the most likely candidate would be that knock sensor. Unfortunately I'll have little if any time this weekend to work on it, guess it'll be next weeks project.
07-20-09, 07:43 PM
#29
Put a few hours in tonight and still nothing. Put the wiring diagrams side by side and crossed off anything that was the same, only a few things left, only two of which seem to make any sense, the boost duty solenoid and the knock sensor. Tested the boost duty solenoid and the resistance was spot on, although it does click like crazy with the key on and engine off, not sure if thats an issue, but its been doing that basically forever.
As for the knock sensor, the "test" method mentioned in the manual is a joke. "Bang on engine, nothing should happen, ground test connector and bang on engine again and timing should retard" Nothing happened either time, and the timing is spot on in any case....
I guess at this point the problem must be in the wiring harness itself. Oye....
As for the knock sensor, the "test" method mentioned in the manual is a joke. "Bang on engine, nothing should happen, ground test connector and bang on engine again and timing should retard" Nothing happened either time, and the timing is spot on in any case....
I guess at this point the problem must be in the wiring harness itself. Oye....
07-20-09, 10:02 PM
#32
Sean: if the timing doesn't retard in either situation for the knock sensor test, then isn't the knock sensor bad (or disconnected)? The ECU should throw a code if it was disconnected, though.
Have you tested all voltages at the ECU? Page F2-77 (section 4b) of FSM shows the correct voltages. It looks like a lot to test, but it goes pretty fast if you are organized. I would use Excel or something and make a list of the pins and a spot to write the voltages. Also have a diagram (FSM page) of the ECU connector printed out so that you know which pin is which. You can measure each pin and have someone else write down the values. After you are done, go through and compare measured with what it should be. This can point to something disconnected (such as something that doesn't produce a code) or if there is some problem with the harness.
Good luck.
Have you tested all voltages at the ECU? Page F2-77 (section 4b) of FSM shows the correct voltages. It looks like a lot to test, but it goes pretty fast if you are organized. I would use Excel or something and make a list of the pins and a spot to write the voltages. Also have a diagram (FSM page) of the ECU connector printed out so that you know which pin is which. You can measure each pin and have someone else write down the values. After you are done, go through and compare measured with what it should be. This can point to something disconnected (such as something that doesn't produce a code) or if there is some problem with the harness.
Good luck.
07-21-09, 12:10 PM
#33
Yeah, if the timing doesn't retard they say the sensor is bad, I'll have to redo that test though to make sure as I was banging on the alternator bracket and not very hard at that (didn't want to bend it) The test says to strike the tab for an engine crane to attach to which currently is not on the engine. I will put one on and retry it.
Have not tested the voltages at the ecu yet, kind of dreading that, but you gave some good tips to keep organized about it. I'll try to work on that later in the week when I have a chunk of time to commit to it.
Thanks for the support everyone, I probably would have lost it already
Have not tested the voltages at the ecu yet, kind of dreading that, but you gave some good tips to keep organized about it. I'll try to work on that later in the week when I have a chunk of time to commit to it.
Thanks for the support everyone, I probably would have lost it already
07-27-09, 05:53 PM
#35
Ok finally some results, I think I've tracked down the problem, now I just need help on how to fix it. Basically this is what happened, I was checking voltages at the ecu, and when I got to 2B (air flow meter) and stuck the probe in (the other probe was grounded) the car smoothed right out!
The voltage was just a hair under 2v, so it was right on spec with the FSM. I checked several times just to make sure, and everytime I check that pin it smooths right out and runs like it should. I even revved it a few times just to make sure and it runs great.
Now the problem is:
1) if the NA and T2 ECU both share this pin why does it run fine with the NA ecu and not the T2, and
2) What does this mean and how the hell do I fix it?
I checked the afm previously and it checked out fine, and since it does the same thing with 3 t2 ecu's I can't believe its that, is it something as simple as that wire being burned up/shorted in the harness somewhere?
Glad I'm making progress now I just need to finish it off lol
The voltage was just a hair under 2v, so it was right on spec with the FSM. I checked several times just to make sure, and everytime I check that pin it smooths right out and runs like it should. I even revved it a few times just to make sure and it runs great.
Now the problem is:
1) if the NA and T2 ECU both share this pin why does it run fine with the NA ecu and not the T2, and
2) What does this mean and how the hell do I fix it?
I checked the afm previously and it checked out fine, and since it does the same thing with 3 t2 ecu's I can't believe its that, is it something as simple as that wire being burned up/shorted in the harness somewhere?
Glad I'm making progress now I just need to finish it off lol
07-27-09, 06:03 PM
#36
Must be a loose pin in the connector. The meter should have an extremely high input impedance (resistance). This is so that when you measure voltage, you are not adding another load to the circuit and changing the results. So, the electrical part of the meter shouldn't change it. It is probably the physical pushing the probe in that is making connection. To test the theory, you could push a plastic pin or something in there and see if it behaves the same or not.
It must have just been when you tested the NA ECU that this wire moved a bit and was making a better connection.
It must have just been when you tested the NA ECU that this wire moved a bit and was making a better connection.
07-27-09, 08:08 PM
#37
That would be the most logical assumption, but it doesn't seem to be the case. I tried wedging it in there and no luck, the only thing that makes it smooth out is if the other probe of the multimeter is grounded.
Also, as much as I didn't want to start cutting wires just to test I ran a new wire between the MAF and pin 2b to see if the wire is shorting somewhere in the harness and no luck, grounded 2b using the multimeter with this setup and again it smoothed out.
A few weeks ago I did test the MAF, but I'm thinking I should go back again and test.
Also, as much as I didn't want to start cutting wires just to test I ran a new wire between the MAF and pin 2b to see if the wire is shorting somewhere in the harness and no luck, grounded 2b using the multimeter with this setup and again it smoothed out.
A few weeks ago I did test the MAF, but I'm thinking I should go back again and test.
07-27-09, 08:53 PM
#38
Sean: I'm just looking at the S5 NA section right now (should be the same as T2). 2b looks like it should be 4v with engine off and 2.5-3.5v at idle. What were your readings in both cases?
Go ahead and test the AFM. It should only take a few minutes to test. It looks like 2b (corresponds to Vs on the AFM) is the flow output from the AFM (voltage changes with flow).
If it was E2 from the AFM, then I would say you were missing a ground somewhere. Even though alost no current should flow through the multimeter, there is a little..so it could complete a missing ground. However, it doesn't seem to make sense for Vs.
Double check that it is really 2b that you are testing and use wiring colors to confirm.
Go ahead and test the AFM. It should only take a few minutes to test. It looks like 2b (corresponds to Vs on the AFM) is the flow output from the AFM (voltage changes with flow).
If it was E2 from the AFM, then I would say you were missing a ground somewhere. Even though alost no current should flow through the multimeter, there is a little..so it could complete a missing ground. However, it doesn't seem to make sense for Vs.
Double check that it is really 2b that you are testing and use wiring colors to confirm.
07-28-09, 07:30 AM
#39
I'll double check what I wrote down tonight when I get home, but I believe with the key on engine off it was about 3.8 volts so a bit low and with it running it was like 1.96 volts, so again a bit low, but I also got slightly lower than spec readings on several other sensors (maybe theres a bigger issue here?)
And the wire in question here is green with no tracer, its the middle plug (the smallest of the 3 on the ecu) bottom right pin (looking at it from the wire side of things when its plugged in, if that makes sense...). Likewise if you look at the AFM with the engine side of things on the left and the air filter on the right its the first wire on the left (engine side) also green no tracer. If I check the voltage at the green wire there it does the same thing, as soon as the meter is grounded it smooths out, and the voltage is the same as it is at the ecu.
And the wire in question here is green with no tracer, its the middle plug (the smallest of the 3 on the ecu) bottom right pin (looking at it from the wire side of things when its plugged in, if that makes sense...). Likewise if you look at the AFM with the engine side of things on the left and the air filter on the right its the first wire on the left (engine side) also green no tracer. If I check the voltage at the green wire there it does the same thing, as soon as the meter is grounded it smooths out, and the voltage is the same as it is at the ecu.
07-28-09, 08:17 AM
#40
Here is what I would do:
1. test the AFM (resistances, etc.)
2. Check voltages related to AFM (2I, 2B, 2K, 3D, 3C)
3. Both E2 pins at AFM (2nd from the right, and middle) should be grounded. Check continuity with chassis ground (key off).
4. 3D and 3C tie into the pair of 'E2' at the AFM. Test these for continuity with ground (key off).
5. 2B appears to be a shielded wire the shield splits off at some point (probably near the ECU). This shield also connects to the shielded wires of the O2 sensor and the knock sensor (pins 2C, 2M) and is grounded somewhere. The shield is there to prevent electrical interference on these wires. Trace this shielded wire and make sure it isn't frayed and making electrical contact with any of the pins. The shield should only connect to ground and nothing else. Trace it also to its grounding point. Clean the ground and make sure it is secure. If the strands are exposed somewhere, check the voltage with the key on (should be 0v) and then test continuity with ground (key off). Since things work okay with the NA ECU, especially check 2M (knock sensor pin) when checking out the shield wire.
Check that stuff and see what you can find out. If that doesn't lead anywhere, post up all of your measured voltages at the ECU. Maybe something will stand out. In some cases, low voltages may be due to a weak battery or poor connections at the fusible links. If the supply voltage at the ECU is low, it can bring down the other voltages too.
1. test the AFM (resistances, etc.)
2. Check voltages related to AFM (2I, 2B, 2K, 3D, 3C)
3. Both E2 pins at AFM (2nd from the right, and middle) should be grounded. Check continuity with chassis ground (key off).
4. 3D and 3C tie into the pair of 'E2' at the AFM. Test these for continuity with ground (key off).
5. 2B appears to be a shielded wire the shield splits off at some point (probably near the ECU). This shield also connects to the shielded wires of the O2 sensor and the knock sensor (pins 2C, 2M) and is grounded somewhere. The shield is there to prevent electrical interference on these wires. Trace this shielded wire and make sure it isn't frayed and making electrical contact with any of the pins. The shield should only connect to ground and nothing else. Trace it also to its grounding point. Clean the ground and make sure it is secure. If the strands are exposed somewhere, check the voltage with the key on (should be 0v) and then test continuity with ground (key off). Since things work okay with the NA ECU, especially check 2M (knock sensor pin) when checking out the shield wire.
Check that stuff and see what you can find out. If that doesn't lead anywhere, post up all of your measured voltages at the ECU. Maybe something will stand out. In some cases, low voltages may be due to a weak battery or poor connections at the fusible links. If the supply voltage at the ECU is low, it can bring down the other voltages too.
08-01-09, 02:45 PM
#41
Ok, been tinkering with this over the last few hours and here's what I've found so far
With the N374 ecu the voltages for the AFM related pins are:
2I - 5 volts
2B idle - 2.01
2B key on - 3.69
2K 1.19
3C - 0
3D - 0
So 2B and 2K are below spec, and the two grounds seem ok, although I'm not entirely sure how I check continuity with the grounds on my meter...
Started checking some of the other voltages and I can't see to get any reading on pin 2C, the O2 sensor, this could be a clue to something.
Here's my question, with much of the wiring harness, at least near the ECU, wrapped in that copper looking tape how do I take it apart without damaging anything? Do auto parts stores sell that copper shielding tape (I assume its for shielding?)
One more piece of info that might or might not mean anything: I had the NA ecu on the car to run it up to operating temp, and before putting the T2 ecu in I decided to see what 2B was doing, and once I put the meter on there it runs like crap just like the T2 ecu. So this is the exact opposite of what happens with the T2 ecu since that makes it smooth out.... again not sure what that means, but strange in any case...
With the N374 ecu the voltages for the AFM related pins are:
2I - 5 volts
2B idle - 2.01
2B key on - 3.69
2K 1.19
3C - 0
3D - 0
So 2B and 2K are below spec, and the two grounds seem ok, although I'm not entirely sure how I check continuity with the grounds on my meter...
Started checking some of the other voltages and I can't see to get any reading on pin 2C, the O2 sensor, this could be a clue to something.
Here's my question, with much of the wiring harness, at least near the ECU, wrapped in that copper looking tape how do I take it apart without damaging anything? Do auto parts stores sell that copper shielding tape (I assume its for shielding?)
One more piece of info that might or might not mean anything: I had the NA ecu on the car to run it up to operating temp, and before putting the T2 ecu in I decided to see what 2B was doing, and once I put the meter on there it runs like crap just like the T2 ecu. So this is the exact opposite of what happens with the T2 ecu since that makes it smooth out.... again not sure what that means, but strange in any case...
08-01-09, 03:48 PM
#42
Sean: Have you tried to check the resistances of the AFM itself?
To check grounds, set to the lowest resistance range (20 Ohm or whatever) and stick one end in the ECU pin (3C or 3D) and connect the other meter wire to chassis ground. The resistance should be very low (maybe 0.2 Ohms or something).
The other way is most meters have a continuity setting. If you set to that and do as above, the meter will beep if there is continuity. This setting is often the same as the diode testing setting.
For the shielded wire, usually the auto parts stores don't seem to stock it. Easiest way to get some is at an audio shop (by the foot) or you can simply get some cheap shield audio connectors and cut off the ends.
To check grounds, set to the lowest resistance range (20 Ohm or whatever) and stick one end in the ECU pin (3C or 3D) and connect the other meter wire to chassis ground. The resistance should be very low (maybe 0.2 Ohms or something).
The other way is most meters have a continuity setting. If you set to that and do as above, the meter will beep if there is continuity. This setting is often the same as the diode testing setting.
For the shielded wire, usually the auto parts stores don't seem to stock it. Easiest way to get some is at an audio shop (by the foot) or you can simply get some cheap shield audio connectors and cut off the ends.
08-01-09, 06:35 PM
#43
Ok tested continuity (thanks for telling me now, its very simple isn't it lol) at the two grounds both at the AFM and ECU and it is present. Also checked resistance on the AFM itself, everything seems within spec, here are the results:
E2-VS Closed - 323
E2-VS open 65
E2-VC open/closed 280
E2-THA (its around 70 degrees today) - 1736
Also I did the above tests using both E2 pins to make sure the results were the same, and they are.
Attached is a pic of the shielding tape I'm referring to, I don't actually mean the shielded wires themselves, this like a copper tape of some kind that goes over the whole harness, a ground is actually soldered to it as well.
What do you think about not getting a reading at 2c, the o2 sensor pin?
E2-VS Closed - 323
E2-VS open 65
E2-VC open/closed 280
E2-THA (its around 70 degrees today) - 1736
Also I did the above tests using both E2 pins to make sure the results were the same, and they are.
Attached is a pic of the shielding tape I'm referring to, I don't actually mean the shielded wires themselves, this like a copper tape of some kind that goes over the whole harness, a ground is actually soldered to it as well.
What do you think about not getting a reading at 2c, the o2 sensor pin?
08-01-09, 07:17 PM
#44
Ok, here are some voltage results, didn't really bother with any of plug 1, and only a few on plug 3 as most of them aren't really applicable to this problem (headlights, ac/ps, etc etc)
2A off - 0, idle - .84
2B ign on - 3.69, idle - 1.9
2C - 0 in all cases
2E - .8V at ~140 degrees , I'll have to recheck with the engine cold
2F - ign on idle pos .75, ign on wot 3.9 idle running .78
2G - ign on idle pos .65, ign on wot 3.25 idle running .39
2H - not able to correctly test, will have to attempt again
2I - ign on 5, off 0
2J - didn't test, will check later
2K - 1.26
2L - .87
2M - ~0 all situations
2N didn't test, will check later
2O - idle 13, half throttle ~1.6
2P - idle 1.2, above 4k rpm 13
3E - .2 all situations
3K - ign on 11.8 idle .23
3Q - idle 10.5
3R - idle 2.47
Seems like a lot of the voltages are lower than they should be, yet the ecu is seeing 12v with key on and almost 14 with the engine running at pin 1A not sure what that's about, Not sure why I can't get a reading on the O2 sensor guess thats my best lead at this point?
2A off - 0, idle - .84
2B ign on - 3.69, idle - 1.9
2C - 0 in all cases
2E - .8V at ~140 degrees , I'll have to recheck with the engine cold
2F - ign on idle pos .75, ign on wot 3.9 idle running .78
2G - ign on idle pos .65, ign on wot 3.25 idle running .39
2H - not able to correctly test, will have to attempt again
2I - ign on 5, off 0
2J - didn't test, will check later
2K - 1.26
2L - .87
2M - ~0 all situations
2N didn't test, will check later
2O - idle 13, half throttle ~1.6
2P - idle 1.2, above 4k rpm 13
3E - .2 all situations
3K - ign on 11.8 idle .23
3Q - idle 10.5
3R - idle 2.47
Seems like a lot of the voltages are lower than they should be, yet the ecu is seeing 12v with key on and almost 14 with the engine running at pin 1A not sure what that's about, Not sure why I can't get a reading on the O2 sensor guess thats my best lead at this point?
08-01-09, 10:54 PM
#45
Sean: Can you check the voltage at the AFM (key on, engine off, engine doesn't have to be warm). I'm interested in what the BR/W wire reads. This should read the same as 2I (5v). It also powers the TPS and OMP. See..the TPS should range between 1v and this voltage (5v) and you are a bit low. If this is off, it will also make your AFM measurements to read low.
Not sure where you could get that copper tape. Does it look like there are shields from the individual wires soldered to this? If not, then maybe the inner wires are normal and the copper is the shield for all of them. If the copper is broken or not grounded, then the shielding will not work.
For the O2 sensors, they can be a little hit/miss. Normally the mixture is on the rich side for idle, so the voltage should be more like 0.5v-1v. The FSM just says below 1v at idle, which doesn't help much as the O2 sensor is always between 0v and 1v.
Not sure where you could get that copper tape. Does it look like there are shields from the individual wires soldered to this? If not, then maybe the inner wires are normal and the copper is the shield for all of them. If the copper is broken or not grounded, then the shielding will not work.
For the O2 sensors, they can be a little hit/miss. Normally the mixture is on the rich side for idle, so the voltage should be more like 0.5v-1v. The FSM just says below 1v at idle, which doesn't help much as the O2 sensor is always between 0v and 1v.
08-02-09, 09:26 AM
#46
Sure Kent, I'll check that voltage later on today. As for the shielding, there is shielding on the individual wires within the harness (I have part of another harness that was cut so I took it apart to look) and then the whole harness is wrapped in that copper tape, and as far as I can tell the only thing soldered to that copper tape is a ground wire, the one that bolts to the floor right by the ecu.
As for the O2 sensor they do seem to be hit or miss, a friend was over the other night and we went through and were testing sensors, was hard to get a reading on that one. Still though it seems like I should have some reading at the ECU from it, I'll double check again today
As for the O2 sensor they do seem to be hit or miss, a friend was over the other night and we went through and were testing sensors, was hard to get a reading on that one. Still though it seems like I should have some reading at the ECU from it, I'll double check again today
08-02-09, 12:50 PM
#47
Just checked and it reads 5 volts just like at the ECU... Also just for kicks used some jumper cables between the engine and frame just to make sure the grounds I have in place solid, and no change, so everything seems ok there.... Guess the next step is a new harness?? I can't imagine it would go just like that but who knows, I guess it could...
08-02-09, 08:27 PM
#48
Well not sure if it was the smartest thing to do, but I got fed up with the endless testing that seemed to be going nowhere so I pulled the damn UIM off again and yanked the wiring harness out. Unwrapped it to see if anything obvious was wrong and of course there wasnt. The only thing I did notice is that it was soaked in oil toward the middle of the harness, not sure if that would have caused any issues or not. I'm thinking not.
So now I've got a few options, put this harness back together and reinstall, modify the spare NA harness I have and install that, or buy another T2 harness since I have it all apart and use that. I'm leaning toward modding the NA harness I have because I can route it a bit neater than the JDM harness was (its one of those things that always bugs me when i open the hood) but I'm not sure.
I'm not convinced the wiring harness would go just like that, but I'm running out of possibilities... What do you guys think?
So now I've got a few options, put this harness back together and reinstall, modify the spare NA harness I have and install that, or buy another T2 harness since I have it all apart and use that. I'm leaning toward modding the NA harness I have because I can route it a bit neater than the JDM harness was (its one of those things that always bugs me when i open the hood) but I'm not sure.
I'm not convinced the wiring harness would go just like that, but I'm running out of possibilities... What do you guys think?
08-02-09, 09:32 PM
#49
Hmm..it is a strange deal. It doesn't seem likely that the harness would just go 'bad', but it wouldn't take much to cause problems. Some corrosion on a connector, partial short, broken connection..who knows??
At least if you went with the modded NA harness or a new T2 harness, then you can see if the problem lies in the harness or somewhere else. I would lean towards modding the NA harness since you already have one.
At least if you went with the modded NA harness or a new T2 harness, then you can see if the problem lies in the harness or somewhere else. I would lean towards modding the NA harness since you already have one.
08-03-09, 09:57 AM
#50
Yea its a strange problem for sure, I'm leaning toward the modded NA harness as well. To be honest I've never been very sasisfied with the way that JDM T2 harness fit, since it comes out on teh left side near the oil filer I ran it along the firewall and then in a hole near where the A/C evaporator would be. Looked kind cheesy. With the NA harness I'll be able to run it more like the US FC's and GSL-SE's where it goes it all the way on teh right side of the firewall. While I'm at it I'll clean up some other wiring as well, its a good excuse to do all this stuff and keep me motivated lol.
Worst case scenerio is that it still has the same problem when I'm done and I'll know its not the wiring.... I'll keep you guys updated.
Worst case scenerio is that it still has the same problem when I'm done and I'll know its not the wiring.... I'll keep you guys updated.