|
|
|
Hi Brickboarders,
I have a new 240 85 wagon automatic (LH 2.2 I guess). It has an acceleration problem when cold. It starts perfectlly everytime and idles great but as soon as I give it gas, it bogs and occasionally pops through the air cleaner until it warms up.
I replaced the following:
AMM (Air Mass meter)
Accordion intake hose from AMM to TB
Cleaned TB
Replaced and readjusted TPS
Cleaned flame trap
Checked all hoses for vacuum leaks
Replaced ECU inside of car
Checked operation of both fuel pumps (both are running)
Replaced fuel pressure regulator
Then it dawned on me this might have to do with the coolant temp sensor, duh!
I disconnected it and lo and behold the thing accelerates and runs perfectly when it is cold. No bog, no hesitation. So I replaced the coolant temp sensor thinking I finally found the probelem, reconnected it and the problem is still there!!!!! WTF???
So it only accelerates when it is cold with the coolant temp sensor disconnected.
Anybody got any ideas on what is going on?
Thanks...
|
|
-
|
|
That would point to some of that crap insulation falling off and shorting out the wire. Real issue for both your 85 and my 86 years. The wires on that connector in Inga was so bad that I had to chop them out and replace them.
Bet if you look around there is a bare spot touching chassis some place. You can break out the Ohmmeter to be sure, put a lead on chassis (Engine Block) and that wire and see if you have a low resistance.
Be ready, if that is bad you may find others in the same bundle of wire.
Good Luck,
Paul
PS: Did not know that would cause acceleration issues, good find!
|
|
-
|
|
|
Thanks to all of you that have replied. It looks like a consensus on the wiring being shorted to ground. I'll try to diagnose this in the next few days and I promise I'll get back to the thread to confirm or deny suspicions.
Many thanks again and to Jarrod as well for keeping such a great resource alive.
|
|
-
|
|
|
I'm having to think on this one too!
The coolant sensor work on a resistance value to the ECU.
The more resistance the colder the ECU thinks the engine temperature is.
A disconnected wire means a open circuit = A lot of resistance means no flow of signal to and from the sensor.
Interpreted inside the ECU as a engine very cold or on the newer cars an engine fault code and possible light.
The statement will read "engine coolant sensor" in the a translator's manual.
Like in other languages the "translations" loses the concept of complete thought and a conversation. The reason for your "WTF" statement.
You have a problem of a wire to the ECU from the sensor is shorting to a ground, the wires or chassis. A short circuit is closed circuit with "0" resistance. It should have approx. 350 ohms.
The "0" sent to the Ecu and it thinks a hot to very hot engine and leans the fuel.
If other things are working correctly like the 0-2 sensor, the battle begins. The ECU has a conflict of its program and at about 15 to 28 times a second samples the readings and tries to adjust mixture, rpm and timing etc.
This self tuning creates symptoms of the other things we want to think could cause this erratic behavior.
The best thing is to unplug the ECU connector. Find the pin to the coolant sensor and trace it to find the bad/bare spot. Then splice around it. It could be it the connector on the sensor or the two wires (to each other) going home to the ECU.
Happy Hunting and hopefully this helps & Happy Motoring later.
|
|
-
|
|
|
The popping sound would indicate a very lean condition on a cold start which in turn indicates that the ECU is not compensating for the cold coolant temp. Since you have replaced the ECT without any results, and disconnecting the wire allows you to run, I would suggest checking the length of the wire for a bare place where it may be grounding to the chassis. If you have a spare ECU, you can swap it out and see if it has failed also. My bet would be in the wiring insulation.
Regards,
--
Will Dallas, www.willdallas.us, www.willdallas.org, www.willdallas.com, www.dallasprecision.com 86 245 DL 222K miles, 93 940 260K miles, 88 765 GLE 152K miles
|
|
-
|
|
|
Thanks for the reply.
I've read the Fuel Injection manual on K-Jet.org for the LH 2.2 and it says the ECT (Coolant Temp Sensor) when disconnected simulates a cold start. Resistance increases as temp increases.
If I understand the wiring diagram, it looks like one wire of the ECT goes to ground and one goes to pin #2 on the ECU.
Looks like a pretty simple trace operation to check for ground on pin #2 when the plug is connected and no ground when disconnected. See any problems with this logic?
Thanks for your suggestion. Very helpful indeed.
|
|
-
|
|
|
Not sure if you have a typo, but the resistance decreases at temp increases on the sensor. Therefore if you have a short to ground in the wiring, the ECU will read that the engine is hot, and lean the mixture. As you noticed, disconnecting the sensor, and therefore having a high resistance, would simulate a cold engine signal from the sensor.
--
84 242Ti IPD bars&springs, 89 745 16v M46 IPD bars, 89 744 16v M46 IPD bars, 90 745 AW70, 91 245SE AW70 IPD bars, 93 245 CLassic M47
|
|
-
|
|
|
Oops. Sorry that was a typo. You are quite right. Resistance decreases as temp increases.
Anywho, last night I tested the coolant temp sensor on a cold engine and it read 2240 (within specs according to the Green book factory manual). I reconnected the plug and went to pin #2 on the ECU and it also read 2240. If there was a short in the circuitry, I would not have gotten the correct reading at pin #2. So the theory about a bare wire unfortunately is not the problem. I even wiggled the wires all the way from the ECT to the ECU and no fluctuations in the readings. I also checked ground at pins 11 and 25 (direct connections to the grounds on the manifold) and all was well. On the surface, all seems to be well in the coolant temp sensor department. Yet still the car won't accelerate when cold unless it is disconnected.
So now what? If the ECU is getting the correct signal from the ECT could another component be sending erroneous info to the ECU causing it to run lean and pop during cold running? Perhaps the AMM has a wiring issue. Like I said, I don't think it's in the ECU itself because I swapped it out with a good working one.
I suppose I will have to run through all of the diagnostic tests one by one as outlined in the GReen book and see if I find anything out of whack.
Anyone got any other ideas?
Thanks, Tony
|
|
-
|
|
|
You are sure putting a twist in this. I thought you said you took the wire on and off and it ran different. By chance did you check the sensor after it is hot for the 350 ohms?
My guess is that you are getting better at troubleshooting the harness!
Now that you are so good at tracing to the ECU. I would make sure that the throttle switch is for sure sending a signal to the ECU.
There needs to be a signal to ground pin. Either closed or open just as the throttle is cracked opened.
This tells the ECU what you want it to do and when.
Just thinking of what to rule in!
The plugs, wires cap and especially the rotor button are good. Right? Vacuum tee's and lines. I threw this in cause you have to reach right pass these to get to that connector.
If these check out as working every time I would "lean" towards an overheated Air Mass Meter or a air leak between the AMM, the throttle or intake.
I'm still leaning here?
Inspect the bellows tube carefully. Especially the port where the flame trap connects on. Both these hoses and intake side of the flame trap have to be good and tight.
In fact I would pull the bellows and the AMM. Look inside it with a bright light and look for the really fine wire stretched though loops in the center and make sure its intact.
Flex the bellows and I like using a compact fluorescent light inside it. Holes or cracks in the convolutions.
There is a lean condition being presented from what you have said. Excess air getting in that you got to locate or the AMM isn't reading air coming through it correctly.
I hate to pick on the electronics but the AMM are more prone than the ECU. This is due to failures caused by heat from the air cleaner thermostats failures or dirt and oil smut vapors.
Bosch knows there can be a problem with this or they would have not built in to the ECU program a 1 or 2 second burn off of the heater wire in the AMM. Every time the ECU records a rev over 3000 rpm it is programed to burn the wire to clean off impurities. It does this burn after the key is turn off. Its pretty hard to drive the car and not go over the 3000 rpm. So it is possible that it has burned it own wire?
This is as far as I can think this unless you come across something more twisted. Keep us posted!
|
|
-
|
|
|
Hey Machineman, thanks for the suggestions.
I guess my post was a little confusing. To reiterate: I did take the wire on and off (i.e., unplug the ECT and plug it back in) and yes it runs differently. Unplugged when cold, the car runs perfectly. Plugged back in and the car stalls and pops. So I switched off the ignition and 1) unplugged the ECT and tested the ohms to make double sure it was within specs and it was and 2) plugged it back in, went to the ECU, disconnected the harness and tested pin#2 and ground. This reading also produced the same ohms as the ECT disconnected so it means that the wiring to the ECU is sound. Several Brickboarders suggested that based upon my symptoms that there was a short in the wiring. I confirmed that this was not the case. SO is this clear as mud now?
Anyway, back to your suggestions. My initial post states that I swapped out the AMM and the ECU and if I didn't mention it, I also swapped out and readjusted the TPS. You suggestion about tracing the TPS and AMM circuits are good.
P.S. I did not check the ECT 350 ohms when hot. I figured the problem is cold running and the ohms are correct for that. Also this is a replacement ECT as well and both read exactly the same ohms.
I also replaced the corrugated hose going from the AMM to the TB. No cracks or anything and all fittings are tight.
Yup it's something twisted or it's me missing something really obvious. I love challenges though.
Thanks again everyone for your replies.
|
|
-
|
|
|
I read the other post and it appears we are all down to the wires!
It must be something really twisted and according to Paul Driver he is betting on the wires in the harness. He makes a good case since he had the experience.
The wires are twisted in the harness and I think they are stranded like US!
AS far as throwing things on the wall. Here one that may be a little wet from my noodle!
I think you have change everything that we can think of for you to change.
Paul has got the best final solution. Go find a good harness likes he suggest.
Lay that puppy out on the engine and plug it up right there on top the engine to all the components.
That may be where I'm all wet. Will they reach? I have no experience in this!
Paul can chime back in on this question! He's been there, done that!
I was thinking it might be quick and less dirty attack! That way you don't go to the effort of banging your knuckles to find out.
Looking for one may give you insight to what Paul knows and a more overall feeling of how these things are failing.
I'll need to keep watching, this DRAMA!
|
|
-
|
|
|
OK, if we look at what is happenning here. When you have the sensor connected, the car runs badly (too rich or too lean, would be nice to know) and the ECU "sees" 2250 ohms. When you disconnect the sensor, the car runs better (leaner or richer?) and the ECU sees infinite ohms.
I think that the ECU sends a signal through the wire and uses that to set the pulse width to the injectors (along with other factors). A longer pulse width equals more fuel, richer mixture.
One would assume that with low ohms the mixture is "normal" and with 2250 ohms it is "richer" (to promote better cold running) then infinite would give "very rich" mixture.
So could you have a lean mixture most of the time? What readings do you get from the O2 sensor? Might there be an air leak?
If you have the resources, it would be useful to be able to read the exhaust gas when cold and connect and disconnect the sensor.
--
84 242Ti IPD bars&springs, 89 745 16v M46 IPD bars, 89 744 16v M46 IPD bars, 90 745 AW70, 91 245SE AW70 IPD bars, 93 245 CLassic M47
|
|
-
|
|
|
It amazes me that so many of us out there actually get involved in these little puzzles.
Your reply is accurate. The 2250 ohms at cold start signals the ECU to provide more fuel. As the temp increases the ohms go to 350 and the ECU sends a signal to lean out the mixture.
As I have stated in the prior posts, the ECU seems to be getting the 2250 ohm reading at pin#2 (no wiring problems at least in the ECT circuitry) but it is still not sending a signal to the injectors to richen up the fuel mixture. I'm sure the problem is lean running because there is an occasional pop at the air box (a common indicator of a lean condition as I've read).
Regarding the other post from machine man to swap out the harness, I have done that on another car and it can take a good deal of time to pull one and then replace this one. This harness is not brittle. I have seen bad harnesses in which the insulation cracks when bent. This one is surprisingly flexible with no bare wires even down by the alternator where they seem to fray and split first. Of course, this does not mean the harness is good but I think it would be more productive for me to run through the step by step wiring tests as outlined in the factory manual before I replace the harness.
The electrical diagnosis shouldn't take more than an hour to check every component in the system. This weekend is reserved for that. If nothing else it will give me a more in depth understanding of what is going on.
I'll post my findings Monday. Thanks again for eveyones continuing interest.
|
|
-
|
|
|
Just throwing stuff at the wall,
Have you replaced or repaired the Engine Harness by chance? I know there are lots of postings about strange harness related issues in the Pre 87 or 88 cars. 85 and 86 are often mentioned having problems with crumbling wiring.
Wondering if issues in the harness where it goes through the 3 and 4 gap in the intake manifold or in the smaller one were it goes under the engine could cause this sort of problem?
Now to call my own wall toss idea into question. Stumbled into a genuine Volvo replacement harness in an 85 down at the PUP and gleefully installed both into my 86. The four wires on the big connector were crumbled and as we say in Texas had lots of necked(naked)copper stands exposed. Wiring going to some of the under the hood lighting and all of the washer motor connections were a mess. Those issues along with that lots of BB postings on 86's with bad harness, lead me to believe the whole harness was going bad. A couple of months after transplant surgery, decided to do a postmortem on the two old greasy stiff harness that came out (before tossing). Sliced off the old stiff black tubing at the intake manifold bend and several other places. Found the only issues were the wires from the gray connector on the firewall. The large harness and the rest of the small harness wires did not have a single wire damaged or even close to having a problem. The smaller harness may have been replaced by the dealer, found a cryptic reference about replacing distributer signal harness on a dealer service invoice in the file. The previous owner (my Mom) does not remember what that was about. Guess at some point before 88 better wires were installed in at least some of the harness used.
Might be worth crossing it off your problem check list if there are insulation issues in your main harness.
Regards,
Paul
|
|
-
|
|
|
BrickBoarders,
Several times I have read threads on this forum only to find that they trail off with no resolution. Not this time!
I spent about 2 hours tracing each wire from each fuel injection component in the engine compartment back to the plug that goes into the ECU inside the car. I was just about to give up and replace the entire wiring harness hoping that I had missed something and that the shotgun approach might work. I had only the TPS wiring left to check.
As I looked at the wiring diagram for the TPS to the ECU I started wondering why the hell the 3 wires on the TPS didn't match the colors for the wiring diagram. All of the other wire colors matched perfectly. WTF???!!!
So then in the back of my mind I remember reading a post on this website that said the plug for the TPS and the plug for the IAC are identical and sometimes if someone isn't careful they can switch them. In fact I believe there was a Volvo service bulletin warning of this possibilty becasue it was so common.
Anyway, I disconnected the plug at the IAC and sure enough the wire colors matched the colors in the wiring diagram that should go to the TPS switch!!!
Whoppee! Someone switched the plugs. I reconnected the plugs to the right components, started the car and vrooom the idle shoots up to about 5000 rpm and then drops to around 1000 rpm and then just keeps surging back and forth between 1000 and 5000 rpm. Now what?
Luckily, I had encountered this symptom before in another car and was pretty sure what to do next. I ran off to the Pick and Pull and grabbed a new 511 ECU, brought it home, plugged it in and voila!, the car runs like a dream! No hesitation when cold, no surging, perfect idle, no issues whatsoever.
So why did I need another ECU? I assume that the TPS and IAC switched plugs fried the internals of the ECU and sure enough the new ECU proved this point.
The only drag was that I had had another good ECU spare in the garage but during the initial phases of debugging before I discovered the switched plugs, I swapped the good one in and the screwed up circuitry fried that one instantly so I had to go buy one that hadn't been exposed to the mixed up wiring.
Hope this proves helpful to others out there.
Thanks for everyone's input.
|
|
-
|
|
|
Great Job!
I fried an ECU once during a re-wire project by swapping the two plugs. Sorry that I didn't think of it.
Regards,
--
Will Dallas, www.willdallas.us, www.willdallas.org, www.willdallas.com, www.dallasprecision.com 86 245 DL 222K miles, 93 940 260K miles, 88 765 GLE 152K miles
|
|
|
|
|
