View Full Version : 92 civic engine cranks but it won't start!
civicdx92
Jun 25, 2006, 04:54 PM
Hello,
I have a 92 civic dx manual with 188k miles
My son drove it to School had it towed home.
The engine cranks over but it won’t start.
I’m not getting fuel over to the fuel rail.
I have no pressure.
I got no problem codes.
Check engine light never came on.
I Checked and replaced fuel pump relay no fix.
I Checked and replaced fuel pump no fix.
I checked and replaced ICM along with Coil no fix.
I replaced all these parts because the car is over 13 years ald and I've never really had any problems with it.
I plan on keeping this car for another 10 years or so Hondas are great cars.
What I’m I missing here guys?
Thanks,
TxGreaseMonkey
Jun 25, 2006, 07:34 PM
I suspect the voltage going from your ECM to the MAP sensor is the problem. Check for 5v. Going to the MAP sensor's reference wire (+) and ground. It's probably a fraction of a volt, indicating the ECM is failing. If the signal wire from the MAP sensor is real low, you won't get spark and the fuel system will shut down. This is the KEY sensor on Hondas and a major design flaw, in my opinion. They should have designed a more fault tolerant electrical system. The real test is to turn the ignition switch OFF, disconnect the D harness to the computer, turn the ignition switch ON, and measure the voltage between computer pins D19(+) and D21(-). If there is less than approximately 5 v. then substitute a known good ECM. If you have approximately 5 v. between D19 and D21, then repair an open between the ECM and YEL/GRN reference wire going to the MAP sensor. Odds are your computer has failed and it's time to replace it. MAP sensor failure is another possibility, as well. Both are easy to check out.
In return for the help, let me know the ultimate solution.
civicdx92
Jun 26, 2006, 11:43 PM
I get only about 300mvdc at the MAP sensor connection.
I'm having a problem IDing pin D19 and D21 at the ECM My book is real generic and basic.
I'm not sure if I mention that I get no codes but the check engine light stays on all the time and never goes off.
Thanks,
TxGreaseMonkey
Jun 27, 2006, 06:33 AM
The odds are your MAP sensor test confirmed your ECM is bad. By the way, the check engine light staying on and no codes being thrown are characteristics of this problem. Checking the voltage at the ECM only confirms that you do not have an open between the ECM and the MAP sensor. Honda wants about $850 for a new ECM. I suggest going to O'Reilly Auto Parts and buying an A1 Cardone remanufactured ECM, with a lifetime warranty, for around $260. I'm certain this will solve your problem.
ECMs are nothing more than "power transistors," whose job is to supply the proper voltage to a host of sensors, under constantly varying conditions. Some websites bogusly claim the odds of a Honda ECM going bad are virtually zero. Don't believe it for a minute. When you remove your ECM, disconnecting the negative battery terminal first, look at the heat discoloration on the metal covers. You will observe a spiral blue pattern. ECMs don't have fan motors to cool them and, to make matters worse, are covered by carpeting. Heat, vibration, and shorts reduce their lives. You did well getting 188,000 miles out of your ECM. Once you replace it, your car will fire right up, run like a top, and the check engine light will go out. Most Honda technicians are in the dark when it comes to diagnosing the situation you encountered. They don't understand the vital significance of MAP sensors in Hondas. They are the single most important transducer in these cars. If they don't get the proper voltage, your ignition and fuel systems shut down. You will have just saved yourself a lot of money and learned some valuable information about trouble-shooting your car. It's all in the diagnosis, not in "replacing and playing." Too many main relays, fuel pumps, ICMs, and coils are being unnecessarily replaced for a "won't start situation." Even the Honda Shop Manual did not include this test for this situation. Spread the word and help others.
Please keep me posted with what happens.
TxGreaseMonkey
Jul 5, 2006, 08:20 PM
So what did you end up doing?
civicdx92
Jul 5, 2006, 09:12 PM
So what did you end up doing?
Well I ended up purchasing a used ECM from my local junkyard $125 from a 95 civic.
I would like to thank you for all you great input, I would not had been able to figuire it out with out your help.
THanks, :D
TxGreaseMonkey
Jul 6, 2006, 06:46 AM
Your posting makes my day. Thanks. I was really interested in your situation and I'm glad you were able to fix it so reasonably. You're a good dad and I'm sure your son appreciates your effort.
kean
Jul 31, 2006, 03:26 AM
Good day txgreasemonkey,
I have the same problem as civicdx92.
I only get more or less 2V on D19 and D21 of the ECM(tested that while the ECM was unbolted from the body of the car, I don't know if there will be a grounding issue). I checked the map sensor pin connector and it gave me the same reading (ECM was unbloted from the body).
I suspect my ECM is bad also, but sometimes it does work!
After several times of turning the ignition key back and forth sometimes the CEL goes out and the engine will start OK.
So I isolated any problem with distributor.
I had my mitsuba relay resoldered, and I can hear it click.
Is my ECM bad? I would want to change it but the OEM is quite expensive and I doubt the reliability of a used ECM because it might perform just like mine?
Please help
Thank you in advance.
TxGreaseMonkey
Jul 31, 2006, 07:37 AM
kean, I'm sure your ECM is bad. Try and buy a remanufactured ECM with a lifetime warranty. You can buy a Cardone ECM from O'Reilly Auto Parts, with a lifetime warranty, for $293.99 in the U.S. Other outfits sell Cardone ECMs but only with a 1 year warranty. Here's O'Reilly's offering:
CARDONE INDUSTRIES - Remanufactured Electronic Control Unit
Part No: 72-2048
1 Per Vehicle
Limited lifetime warranty
w/OEM #37820
P06
A00, A01
ENG CTRL MODULE
w/MAN TRANS
EXC CALIF
Our Price: $293.99
Each
[Core: $70 ]
kean
Jul 31, 2006, 09:39 PM
kean, I'm sure your ECM is bad. Try and buy a remanufactured ECM with a lifetime warranty. You can buy a Cardone ECM from O'Reilly Auto Parts, with a lifetime warranty, for $293.99 in the U.S. Other outfits sell Cardone ECMs but only with a 1 year warranty. Here's O'Reilly's offering:
CARDONE INDUSTRIES - Remanufactured Electronic Control Unit
Part No: 72-2048
1 Per Vehicle
Limited lifetime warranty
w/OEM #37820
P06
A00, A01
ENG CTRL MODULE
w/MAN TRANS
EXC CALIF
Our Price: $293.99
Each
[Core: $70 ]
hello txgreasemonkey,
sorr for not being convinced but I looked for a same model car and we switched our P27 ECM, his ECM just worked fine with me, my ECM was really bad. For the moment I will just buy a used ECM to be able use the car. But I don't think it will stay long since it will be the same age as my ECM and the probability of failing is high. That includes humid, hot, and cold weather, and more places are flooeded during rainy days.
I plan to keep the car for long since it is my 1st car.
looking for branches of Cordone Industries, I found one in mississauga ontario canada where most of my ralatives live. I hope they have a stock
I will just ask them to buy me one and give it to me on their next visit to us this November.
thank you very much for your help
zortmo
Aug 8, 2006, 06:54 PM
Txgreasemonkey,
I also have similar prob. I show no voltage between 19 & 21 at the ecu but I show .12 at map sensor. Is that possible or am I doing something wrong? I have 2 ecu's, one is supposed to be known good. They both measure the same voltages. I'm pretty sure I have fuel pressure as I hear it buzz for a couple of seconds at key on. Just no spark...
Thanks,
Mike
TxGreaseMonkey
Aug 8, 2006, 07:32 PM
Your description is an impossibility. You must have at least .12 v. at the ECM, if you record .12 v. at the MAP sensor. Are you turning the ignition ON, when you record your readings at the ECM? No matter, your readings are SO low that both ECMs are bad, if you are running the tests properly. Is your DMM set for DCV?
I think I just figured out what you are doing. You are testing for voltage on the D19 and D21 "female" connector. That's why you are reading zero volts. You need to test the voltage on the ECM's "male" pins.
Do you hear your fuel pump run, when your ignition is turned to ON? Does your Check Engine Light stay on?
Where did you go?
zortmo
Aug 9, 2006, 04:50 AM
You are correct... I was checking female end (DCV). The fuel pump runs for a couple of seconds after turning the key on and the CEL comes on and then goes off about the same time as the fuel pump stops buzzing. I have to go to work... I will test again this afternoon. Thanks for your help.
kean
Aug 9, 2006, 05:43 AM
You already have a known good ecu then why don't you just plug it to ecu harness and check if there will be approximately 5V on MAP sensor pin?
Or just start the engine.
Having a solid CEL is one sign of a bad ecu, and you can't retrieve any code.
In your case having a click in the main ralay and fuel pump goes off after 2 secs as well as the CEL, isolate the problem with a bad ecu.
Once you fixed the problem, you may try to use your old ecu and check if it works.
TxGreaseMonkey
Aug 9, 2006, 07:27 AM
At this point, I don't think your problem is going to be your ECM. Just for the heck of it, though, take a MAP sensor connector reading on the right most wire and let me know what you get. Make sure you get a really good ground on the black test lead. Press the lead into the G101 bolt, where the ECM grounds.
I'd focus more on your distributor; e.g. igniter and coil.
zortmo
Aug 9, 2006, 02:40 PM
Ok, I have 5 vdc leaving the ecu to map and I have the same voltage at map sensor. I didn't have a good ground in previous test.
Here's where I am... I have battery voltage at both coil leads/ICM (key on). I have 9.7 at the other lead on top right of ICM... Battery voltage (12.5) at the lead on the side of ICM. I have to wait until some gets home to measure voltages while cranking. I read somewhere I should test at coil +... connected and disconnected whil cranking. Is that correct?
If I didn't tell you before, I have a solid CEL when 2 pin connector is shorted @ ECU. No error codes. If I remove the jumper from 2 pin connector, CEL goes off like normal after a couple of seconds after key on... fuel pump buzzes during the couple of seconds.
Thanks much for the help.
Mike
TxGreaseMonkey
Aug 9, 2006, 02:50 PM
Based on the age of your Civic, I'd replace the igniter, coil, rotor, and distributor cap. I'd recommend this, even if your car started. You are at the stage where you don't want things to go bad on you one at a time. Go for "like new" performance and put some new parts in it. Advance Auto has the cheapest ICM and RockAuto.com has the best price on coils. O'Reilly Auto Parts sells Wagner rotors and distributor caps with brass contacts and a lifetime warranty. Hondas are notorious for problems with ICMs, coils, and main relays. Just my opinion, though.
Let's go back to the MAP sensor connector for a minute. With the ignition ON, what voltage do you get between the YEL/GRN reference wire+(right wire) and the GRN/BLU neutral wire-(center wire)? How about between the WHT+ signal wire (left wire) and the GRN/BLU neutral wire-(center wire)?
zortmo
Aug 9, 2006, 03:26 PM
My wire colors are different than you describe... yel/red -- grn/wht -- what
Between yel/red and center 5.0 vdc
Between white and center 3.5 vdc
Mike
TxGreaseMonkey
Aug 9, 2006, 03:52 PM
The voltage between WHT+ (D17) and center- (D21) should be 5 volts. Here's what you need to do. Disconnect negative. Battery terminal and remove the "D Harness" from the ECM. Then, reconnect the negative. Battery terminal and turn the ignition ON. Measure the voltage between male pins D17+ and D21- on the ECM. Tell me what voltage you observe.
zortmo
Aug 9, 2006, 04:02 PM
3.54 vdc @ ECU, between d17+ and d21- (When I measure it starts at 3.9 and quickly drops to 3,54)
Just to add... same voltage on both of the ecu's
TxGreaseMonkey
Aug 9, 2006, 04:09 PM
I believe both ECMs are bad. You should've read 5 volts! To me, this is why your Civic won't spark and ties in with the Check Engine Light staying on, when the service check connector terminals were jumped.
Remember, the MAP sensor controls timing (similar to the old vacuum advance on distributors) and air/fuel mixture, depending upon manifold pressure changes. Like any transducer, it converts energy from one form to another. Here, it converts throttle body vacuum pressure changes into fluctuating electrical current signals to the ECM. If this sensor does not receive the proper voltage from the ECM (basically, a giant power transistor), you will experience catastrophic ignition and fuel system failure.
zortmo
Aug 9, 2006, 04:30 PM
Is it odd that both ecm's measure the same voltages? Have you ever tried to repair the ecu? Do you know which chips go bad?
TxGreaseMonkey
Aug 9, 2006, 04:46 PM
That's way over my head. I've been told that, over time, the electrolytic capacitors start to leak on the 1989-94 ECMs. Heat is the big culprit. Do you see blue swirls on the covers? If you open them up, you may see several microprocessors that are "crazed" from heat. Vibration and heat can also affect these "giant circuit boards," just like Honda's main relays, causing cold solder joints. Buy a remanufactured Cardone ECM from O'Reilly Auto Parts for around $255 with a lifetime warranty. Other outfits only give you a 1 year warranty. Another alternative is to buy a used one on EBay for around $110, freight included. Some might come with a 1 year warranty. New ECMs from Honda are over $850, plus tax, and only have a 1 year warranty.
Where did you get the second ECM? Yes, it's very interesting. Both could have been damaged in the same way; e.g. not disconnecting the negative. Battery terminal when working with them. It's hard to tell.
Honda Civics will go 350,000 miles, if you take care of them. I keep my cars for a long, long time. If you like the car and would like to keep it for a long time, then I'd put a new ECM in it.
TxGreaseMonkey
Aug 9, 2006, 04:53 PM
You can perform other tests, but I don't think anything else will get it to spark. Tell me about how your problem began. Was it sudden?
zortmo
Aug 9, 2006, 05:12 PM
Yea it was sudden... going down the road about 45 and just died. I checked for spark and fuel first thing. I had no spark and I smelled fuel at the exhaust after cranking the engine over and heard the fuel pump as usual, so I figured I had fuel pressure. I opened the distributor and saw a small amount of oil inside. I figured it was enough to short out the igniter or ICM or whatever the heck it is. I bought a remanufactured distributor and it still had no spark. I returned the distributor and searched the net and found someone saying if I had a solid CEL on shorted service connector I had bad ecu. I also bench tested the main relay... it's fine. So I bought an ecu off Ebay for $90. I have an email to the seller to see if it can be returned. I probably get to eat it.
TxGreaseMonkey
Aug 9, 2006, 05:18 PM
Sudden death situations are often caused by computer failure. People think these ECMs (particularly Honda) are bullet proof, but they are not. Makes me think even more that its your computer.
Here are the likely culprits, when you don't get spark: ECM, igniter, coil, rotor, distributor cap, wires, main relay, and MAP sensor. ECMs and igniters are the usual culprits--both solid state. They either work or they don't.
zortmo
Aug 9, 2006, 05:26 PM
I really appreciate all your help. I'm also a believer in Honda. This is the first time one has laid down on me. I'm thinking about an buying an Insight. Maybe now is a good time.
Just thinking out loud here... do you think any other sensors / transducers being bad could cause this condition? (I understand there are other sensors that effect the spark... )
TxGreaseMonkey
Aug 9, 2006, 05:33 PM
If your car is in otherwise good shape, I wouldn't get rid of it. The MAP sensor is the "Key" sensor on Hondas and is the only one capable of knocking out spark and fuel. The same can be said for the ECM.
Somebody else will come along and get your car for a steal, put in a remanufactured ECM, and drive it for years. Trust me, I've seen other people get frustrated with otherwise fine cars and sell them to an individual who knows what he is doing for peanuts on the dollar. Good way to make a living, if you wanted to. You are on the verge of a great personal victory--don't stop now.
TxGreaseMonkey
Aug 9, 2006, 05:39 PM
How many miles are on your Civic?
zortmo
Aug 9, 2006, 05:42 PM
Ok,OK... I'll have to save for a couple of weeks for the parts. I'll post back after the next round. Thanks again
150,000 mi give or take... run hard and put up wet...
TxGreaseMonkey
Aug 9, 2006, 05:50 PM
Hang tough and never, never, give up!
TxGreaseMonkey
Aug 9, 2006, 08:43 PM
zortmo, here is what it would cost to replace the crucial ignition parts and get a new lease on life:
Ignition Coil (RockAuto.com), Part No. 30510PT2006, $36.79
Main Relay (RockAuto.com), Part No. RY422, $38.89
Igniter, Advance Auto, Part No. 115070, $69.88
ECM, O'Reilly, Part No. 722048, $256.99*
Total $402.55
* Have O'Reilly match Advance Auto's price and get a lifetime warranty for the same price.
kean
Aug 9, 2006, 09:30 PM
You may try to test both ecu to a same model car/engine and see if it works so you could isolate the problem with it.
I thought several times trading my honda to a corolla bacause toyota is known to be very reliable, but when I had the chance drive the corolla, I realized that I get more pleasure driving my civic, damn fast for a SOHC but economical.
I hope you've seen some souped and very well dressed civic around your place, I'm sure you'll think twice before getting rid of it. The EF,EG, and EK civics are good looking and I think even after 10years it will still be.
The good thing about having a problem like this is you learn about your car.
And if you encounter the same problem in the future, you already know how to solve it. Save money on labor cost.
Just change the crucial parts as txgreasemonkey had stated and you will feel your car perform like it was new. Anyway those parts are part of the car's service maintenance, don't be sad to spend few bucks, you will benefit from it in the long run.
TxGreaseMonkey
Aug 10, 2006, 10:27 AM
zortmo, try this test:
. Turn ignition switch OFF
. Reconnect the 3P connector to the MAP sensor
. Turn ignition switch ON
. Measure voltage between D17+ male terminal and D21- male terminal on the ECM
. Record the voltage and let me know the reading.
zortmo
Aug 16, 2006, 06:55 PM
Fixed it... finally. The first distributor I bought was bad. I got another one and it fired right up. I was able to rule out the ecu by using a logic probe to verify the signal between ecu and ignitor. It had to be in the distributor... I bought one from Oreilly with all new oem parts and a lifetime warranty for 175.00.
Thanks for the help and the inspiration to stay after it. Great forum.
Mike
TxGreaseMonkey
Aug 16, 2006, 07:34 PM
Good job. That's got to make you feel really great!
Doug777
Jun 17, 2008, 01:06 PM
I was having the same problem with my 92 civic not sparking. I finally figured out from this forum how to test the ECM by map sensor voltage and it tested bad. So I got a used one and it started up just fine and worked for a day. I ran a engine code test and got straight blinking... nothing I could discern different from long - short blinks. They all seemed the same to me and just kept blinking(Is there something wrong with that?). Now today I go to start it up after work and it won't start again. So I checked the engine code again and got a bad ECM signal.
So my question is, what is it that is killing the ECM? Is there a short somewhere that is frying it? If so, how do I diagnose it?
I've noticed for a while that every time I try to get out of my car I get shocked when I touch my door, which was annoying at least, but got me wondering if there was something wrong. Next thought is that the door locks randomly quit working and I don't remember being shocked before that. Could that be the culprit?
TxGreaseMonkey
Jun 17, 2008, 04:17 PM
Doug777, it's very important to disconnect the negative battery cable, before working on the ECM. You also should ground yourself first, so you don't short out the new ECM.
This link may help:
https://www.askmehelpdesk.com/cars-trucks/faq-how-troubleshoot-repair-maintain-hondas-46563-3.html#post235038
Doug777
Jun 17, 2008, 07:04 PM
Doug777, it's very important to disconnect the negative battery cable, before working on the ECM. You also should ground yourself first, so you don't short out the new ECM.
This link may help:
https://www.askmehelpdesk.com/cars-trucks/faq-how-troubleshoot-repair-maintain-hondas-46563-3.html#post235038
Yup... I read through all that before and did it all step by step. The negative cable was disconnected when I did the swap and I reconnected it afterward. I was also grounded to the car body when I did all the work. It ran and I drove it about 2 miles, put it to sleep for the night, and drove it another two miles to work in the morning. When I went to start it again when I was done working it wouldn't start and that's when I did the engine code check and got the bad signal.
Now I'm just trying to figure out what is causing the ECM to go bad? There has to be something screwing it up for it to be going bad after 15 minutes of driving.
TxGreaseMonkey
Jun 17, 2008, 07:28 PM
Doug777, was the replacement ECM an exact part no. match? When I replaced my ECM two years ago, I bought an ECM with a lifetime warranty from O'Reilly Auto Parts.
Try applying dielectric grease to the ECM's connectors.
Doug777
Jun 18, 2008, 09:38 PM
Yeah... the first row of numbers was the same. I figured that was all that mattered. It came out of a working 92 civic 1.5liter manual tranny. It worked perfectly for a bit before crapping out on me.
As far as the dielectric grease goes, can I get that at autozone or do I have to go somewhere special? I can try it but I'm still pretty sure that something electrical has to be going wrong as well.
TxGreaseMonkey
Jun 19, 2008, 08:31 AM
I'm concerned the replacement ECM is not an exact match and that is the heart of your problem. Dielectric grease is available at almost any auto parts store.
Doug777
Jun 20, 2008, 12:18 PM
So here's my plan for now. I'll probably end up buying a new/reman ECM, but first I'm going to try to fix several of the electrical issues that I already know are there.
One, the cruise control tries to turn on then doesn't stay on.
Two, the door locks quit working a few weeks ago
Three, there is something loose in the wiring for the dash lights. They didn't work at first, then they did for a day, then they didn't work any more after that.
Four, the horn doesn't work.
The fact that I get shocked every time I get out of my car makes me wonder why, since that's the only car I get into that I get shocked from... and it makes me think something electrical is goofed.
Now I'm not super knowledgeable about car computers, but I do know that with desktop computers one part can at times cause other parts to go bad. Maybe you could shed a little light for me about how the car computers interact with a short to ground and stuff like that... and also something about that shocking problem, if introduced to a microchip will fry it. I am not going to buy a new ECM without getting some of this fixed to begin with.
TxGreaseMonkey
Jun 20, 2008, 05:53 PM
Doug777, you need to eliminate stray current from damaging your ECM. That was part of what the recommendation to use dielectric grease was about. Try isolating what circuit the stray current is flowing through and fix the problem. The link below may give you some ideas on how to proceed:
https://www.askmehelpdesk.com/cars-trucks/faq-how-troubleshoot-repair-maintain-hondas-46563-3.html#post241804
Even static electricity is a threat to ECMs, which are full of microchips and other sensitive components.
xxfile
Nov 10, 2008, 10:44 AM
Hi, I have the same problem with my 92 civic. I was giong down the highway and the car suddely lose throttle. I towed the car back home, I check for spark, no spark. I replaced the distributor, and now I have spark, but it doesn't start. I checked the MFR, it works, when I turn the key to on, I can hear the fuel pump. So what does the problem seem to me? Could I have a bad map sensor? Or could I have a bad ECM?
TxGreaseMonkey
Nov 10, 2008, 11:04 AM
xxfile,
. Check for a broken timing belt.
. Check each cylinder's compression.
. Clean main ECM ground on the thermostat housing.
. Perform K-Test on ECM.
. Test all under-dash and under-hood fuses with a test light or multimeter.
xxfile
Nov 10, 2008, 12:26 PM
So I know timming belt is fine since I can rotate the crank-pully with a break bar and the camshaft will rotate
How do I check cylinder compression and K-test?
Where do I find the main ECM ground on the thermostat hosing (if it's in the hayens manual, Ii can read it)?
I'll test the fuses after work today and come back w/ more questions. Thank you. ;)
xxfile
Nov 10, 2008, 12:29 PM
Well, I just realize you have a whole how to posts. I'll read them first before I ask more questions.
xxfile
Nov 11, 2008, 07:11 PM
xxfile,
. Check for a broken timing belt.
. Check each cylinder's compression.
. Clean main ECM ground on the thermostat housing.
. Perform K-Test on ECM.
. Test all under-dash and under-hood fuses with a test light or multimeter.
Hi txgreasemonkey,
So I did what you said.
-Timming belt is good,
-I don't know how to check cylinder compression with the tools I have, but when I take one sparkplug out and check for spark, I try to stuff the hole with paper towel and it shoots the paper towel right ouw, I don't know if that means I have compression
-I don't know which one is the main ground exactly, but I looked for all the grounds and clean them, they seem good.
-I have 5V between the yel/grn and the blue wire (left pin and middle pin)
-I didn't find any broken fuse, I especially check for fuse 24, because it controls the MFR and ECM, but that one is good.
Does this leads to off timming when I did the installation? And how do I check for timming?
TxGreaseMonkey
Nov 11, 2008, 07:17 PM
xxfile, I'm not sure what else to suggest. Could you have blown the head gasket, for instance? Maybe someone else will chime in with additional ideas.
xxfile
Nov 11, 2008, 07:38 PM
Do you think it can be timming issue? Because when I crank, if I hold the key long enough, I would hear back fire (like that bomm noise when you have unburn fuel in your exhaust)
xxfile
Nov 11, 2008, 07:42 PM
And if I have spark, then we can basicall rule out bad MAP sensor right?
TxGreaseMonkey
Nov 11, 2008, 08:03 PM
Yes, MAP Sensors are very reliable (and very expensive). I don't think it's the MAP Sensor at all.
xxfile
Nov 11, 2008, 10:10 PM
So how do I check timming? Because I thought I the distributor can only go in two directions, and one directions is 180 degreess off. So given that timming belt isn't broken, the timming shouldn't be off right?
xxfile
Nov 12, 2008, 01:41 AM
So if anyone can help, please do so, here's a list of what I have done so far:
So... Here's what I have done so far.
Check spark: I took the spark plug out and then ground it, when I crank, I can see that it has spark. This also tells me that map sensor is good.
Check compression: I didn't use a gage, but I stuff paper towel into the top of the cylinder when I check spark, and when I crank the car, it shoots right out, so I assume I do have compression and it's probably good?
Check ECU: I did the K-test, which is, taking out the map sensor and check for 5 volts between ground and it's power (yel/grn and blue). I did have 5 volts.
Check for exhaust: I once ran into a problem when I was in an auto shop that a car had a bad cat-back and it stuffed up the exhaust and hence the car won't start. I took out the two bults between the down pipe and the cat-back and try to crank the car, no luck, and besides that, I smell gas, so I'm assuming it's spraying gas into the cylinder?
Check voltage on fuel injector: I checked the voltage between the fuel injectors, I got 4 volts when the key is on. I'm not sure whether it's right, but given that when I check for exhaust, and I can smell gas after cranking the car, I assume the fuel injectors are good?
So at this point I guess it' leads to timming? So does anyone know how I should check timming? I mean... is there a way to check if and totally off timming and the car won't start?
xxfile
Nov 12, 2008, 06:26 PM
Yo txgreasemonkey, question for you. I just tested the ECU, like I jump the ECU test terminals (the one with two wires), and the CEL staied on, no blinks, so I did the map sensor test again (last time I only test the yel/grn and grn wire) so I still have 5 volts between yel/grn and grn wire, but I don't have any voltage between the what and grn wire, so does that tell me my ECM is dead?
TxGreaseMonkey
Nov 12, 2008, 06:38 PM
Xx, your ECM is likely good, if you performed the K-Test properly and it indicates 5 volts. Be sure to run the test, as I described it. I'm not sure you did it properly. The red wire of your DMM goes to the reference wire on the MAP Sensor connector and the black wire goes to ground.
xxfile
Nov 12, 2008, 06:52 PM
So I put red terminal on to the middle (green) wire, and red on yel/grn and white? Like I have to check both wires right?
TxGreaseMonkey
Nov 12, 2008, 06:58 PM
The K-Test: Remove the MAP Sensor connector and turn the ignition switch to ON (not start). Using a multimeter, check for 5 volts going between the MAP Sensor connector's reference wire (+) and ground. As you look at the connector, this is the socket on the right. Really press the black test lead into a cleaned main ECM ground on the thermostat housing. If the voltage is low, it's probably indicating ECM failure. Most failed ECMs will record a fraction of a volt. To me, the K-Test is simple, elegant, and accurate.
xxfile
Nov 12, 2008, 07:07 PM
Tgm, I just went down and tried the test real quitck so here's what I did:
Voltage between yel/grn(+) and grn(-): 5volts
Voltage between white(+) and grn(-_: less than 1 volt
Voltage between yel/grn(-) and grn(-): - voltage (I can't see negative how much, but it's negative)
Voltage between grn(+) and white(-): no voltage
I tried the same thing using battary ground, and I have the same result. I jumped the terminal for CEL codes (the two terminal connector with the grn/wht and brown wire) and the CEL stays on
xxfile
Nov 12, 2008, 07:11 PM
Where is the thermastate housing?
TxGreaseMonkey
Nov 12, 2008, 07:12 PM
Use any engine ground.
xxfile
Nov 12, 2008, 07:18 PM
All right, let me try again
xxfile
Nov 12, 2008, 07:24 PM
So yeah, it's the yel/grn wire, the terminal on the right, I press the black lead of the multimer into it, and check it against ground (using red terminal), I have a negative voltage. Does that indicate a bad ECM? Because when I try to jump for the CEL code, the CEL stays on. But why am I still getting spark?
TxGreaseMonkey
Nov 12, 2008, 07:26 PM
You hooked the leads up backwards.
xxfile
Nov 12, 2008, 07:34 PM
No, I have the multimeter with me here, it's correct...
povkham
Nov 17, 2008, 12:18 AM
Where is e c m locate
TxGreaseMonkey
Nov 17, 2008, 07:51 AM
povkham, this should help:
https://www.askmehelpdesk.com/cars-trucks/faq-how-troubleshoot-repair-maintain-hondas-46563-3.html#post235038
gerrythrash
Aug 17, 2009, 06:10 PM
Hello txgreasemonkey...
I did a Google search on the problem I'm having with my Honda Civic and ran across this thread. I hope you can offer some insight. My 95 Civic has 326,000 miles on it. I know that's a lot, but I'm out of work right now, and can't afford a new car... so I need to try and keep this one going, especially to help me with in my job search. Anyway... it cranks... but won't start. I had it towed to a repair place... and they told me the distributor failed. So I later put in a new distributor... and new spark plugs, plug wires and did a tune up. When I try to start it... it still cranks but won't start. It sounds like it wants to start... but something isn't happening. I read this thread about the ECM... so I'm wondering if that could be it. The check engine light is not on... and when I just turn the ignition switch on... the engine light goes out after a couple of seconds. Do you think I should test the ECM? What else should I test for?
Thanks for any advice you can give.
Gerry
TxGreaseMonkey
Aug 17, 2009, 06:31 PM
. Test all under dash and under hood fuses with a test light or mult-imeter:
https://www.askmehelpdesk.com/cars-trucks/faq-how-troubleshoot-repair-maintain-hondas-selected-other-vehicles-46563-3.html#post252145
Be very thorough in testing EVERY fuse, since it will save a lot of time and money.
. Ensure you can hear the fuel pump run, when the ignition switch is turned to ON (Position II).
. Clean the main ECM ground (brass connector), located on the thermostat housing.
gerrythrash, since the CEL comes on and goes off after 2 seconds, when the ignition switch is turned to ON (Position II), the ECM is likely okay. If the problem persists, after performing the steps above, I would focus on the distributor, even though you've had a new one installed. We have had many problems with new after market distributors for Hondas. After market Ignition Control Modules and coils are fine, but the problem is with the distributor housing--likely with the internal sensors (CKP, CYL, and TDC). I would take the distributor back and try and get another one. I only recommend genuine Honda distributor housings.
gerrythrash
Aug 17, 2009, 08:21 PM
Thank you so much for the information. I'll check the fuses and see what I come up with. It's a brand new distributor I put in... so I'll be bummed if that's the problem, since I ordered it through an online auto parts seller, and would have to go through the trouble of packing it and sending it back. I'll let you know what happens.
gerrythrash
Aug 17, 2009, 08:33 PM
Thank you so much for the information. I'll check the fuses and see what I come up with. It's a brand new distributor I put in... so I'll be bummed if that's the problem, since I ordered it through an online auto parts seller, and would have to go through the trouble of packing it and sending it back. I'll let you know what happens. One thing I don't understand, however. You suggest checking all the fuses. So if a fuse that has nothing to do with the ignition system (cooling fan fuse, for exampe) goes, that can cause this crank/no start problem?
TxGreaseMonkey
Aug 17, 2009, 08:51 PM
I'm quite sure the problem is with the distributor, if everything else checks out okay. On my 1993 Civic, I purchased just the distributor housing from Honda and installed an aftermarket ICM and coil. That was almost 8 years ago.
Regarding the fuse question, the answer is no. Since it's so easy to check everything with a test light or multi-meter, why not test everything? The prime fuse for your car starting is the ACG (S) (ALT) fuse, in the under dash fuse/relay box.
I'm virtually certain the problem is with your new distributor. Be sure to keep me posted.
gerrythrash
Aug 17, 2009, 10:02 PM
Thanks again. I'll let you know what I come up with.
gerrythrash
Aug 18, 2009, 10:04 AM
Hi again TxGreaseMonkey...
So here's the thing. I bought a testing light and tested the fuses. All the ones in the under-hood box are fine. But... in the fuse panel under the dash, NONE of them work. Not one. And I know that ACG fuse is there. But here's the weird thing. The headlight fuses are there also, and if I turn the lights on, they still work. So I'm wondering if the main relay or some other relay could be the culprit.
Your thoughts?
TxGreaseMonkey
Aug 18, 2009, 10:56 AM
Sounds like your ignition switch is bad. Check it for wear, pitting, and melting:
https://www.askmehelpdesk.com/cars-trucks/faq-how-troubleshoot-repair-maintain-hondas-selected-other-vehicles-46563-2.html#post224652
The flow is positive battery terminal, under hood fuse box, ignition switch, under dash fuse/relay box. I think that says it all. From the fuse check, I now think your ignition switch is the culprit. Should be an easy fix.
This is why I keep harping on checking fuses with a test light or multi-meter, where you are testing for power through the fuse. A simple visual inspection of the fuses would not have uncovered the problem.
Good job!
gerrythrash
Aug 18, 2009, 11:11 AM
Hmmm... that's interesting that you say that. This past December, my car was stolen, and when I got it back, the ignition switch was totally screwed up... Icouldn't get the key to turn it. So I had it towed to a Honda dealer, and paid them $500 to put in a new ignition switch assembly. It worked fine for about three weeks, then when I went to start it, it cranked but wouldn't start. Also, incidentally, during this three weeks, I noticed the SRS light would stay on. So I'm wondering if all of that is related. So this is a relatively new ignition switch.
I'll check it out. Thanks again.
TxGreaseMonkey
Aug 18, 2009, 11:42 AM
Go back and check the 50A (IG) fuse in the under hood fuse box. IG means it goes to the ignition switch. Turn the ignition switch to ON, and touch the test light to the screws on each end of the fuse. It should also have a window for you to examine.
Clean the main ground. Ensure your test light was clamped to a good ground, when you checked the fuses in the under dash fuse box.
gerrythrash
Aug 18, 2009, 11:49 AM
Hi again... so check this out. I just went down to check the ignition switch. I completely took out the lower cover, and all the panels. Just for fun, I retested the fuses on the under-dash fuse box (thinking I could get a better angle at the fuses). So... all of the fuses are fine... except Nos. 7, 9, 11 and 13. Fuses 7 and 9 are the headlight low beams, 11 is for the taillight, instrument panel and license plate light... and 13 is the starter signal. I replaced them with spare fuses, and they don't work. I then put one of those fuses in a slot (matching the amps) that had a fuse that worked, and the replacement ones also work. So what do you make of that?
gerrythrash
Aug 18, 2009, 11:53 AM
I also checked the 50A ignition fuse under the hood... it's fine.
TxGreaseMonkey
Aug 18, 2009, 12:02 PM
Not sure I understand that last sentence.
On older cars, and especially trucks, it's not uncommon for fuses to be good but they don't work because of corrosion. Just remove and clean them with abrasive cloth. Apply a light coating of dielectric grease or spray with WD-40. The key fuses to focus on are #13 (Starter Signal Fuse) and the ACG (ALT) (S). Fuse 13 controls power to main relay Terminal 6, the second relay within the main relay. Did you depress the clutch when you tested Fuse 13? The ACG (ALT) (S) fuse control power to the first relay (Terminal 5) within the main relay.
gerrythrash
Aug 18, 2009, 12:17 PM
I didn't depress the clutch when I tested #13. Should I?
TxGreaseMonkey
Aug 18, 2009, 12:17 PM
Yes.
TxGreaseMonkey
Aug 18, 2009, 12:27 PM
Can you confirm that the ACG (ALT) (S) fuse is fine?
gerrythrash
Aug 18, 2009, 12:40 PM
OK... I just tested the ACG (ALT) S fuse... it's fine. I took the starter signal fuse #13 out, cleaned it, put it back in, depressed the clutch... and nothing. I also put a spare 7.5A fuse in that slot, and tested it... nothing.
Weird, huh?
TxGreaseMonkey
Aug 18, 2009, 12:48 PM
I can only conclude there's a problem with the ignition switch, since fuses 7, 9, 11 and 13 are not getting power.
What do you think?
gerrythrash
Aug 18, 2009, 01:00 PM
It's very possible... even though this one is still relatively new. I'll do some more diagnosing and if I can't come up with anything, I might just bite the bullet and have it towed to the shop. And if it costs unreasonably much to fix, I'll probably just junk it and ride the bus until I can save to get another car. I guess I can't complain... I did get over 320,000 miles out of it.
But thanks again for all of your help. I really appreciate it.
TxGreaseMonkey
Aug 18, 2009, 01:06 PM
New ignition switches usually cost around $65, maybe less if you shop for a deal online. I just had end links go bad after 18 months--fortunately, they had a lifetime warranty and I exchanged them. So stuff does go bad surprisingly early, sometimes.
Don't give up.. . victory is very close!
TxGreaseMonkey
Aug 18, 2009, 01:20 PM
I just had a thought. With so many fuses not receiving power, disconnect the negative battery cable and examine the ignition switche's connector that goes into the under dash fuse box. In fact, examine all connectors going to the ignition switch. There may be corrosion/contact issue. Check things out and spray WD-40 on the connectors. It may just be a connector issue.
gerrythrash
Aug 19, 2009, 12:13 PM
Cool. I'll try that... and let you know what happens.
TxGreaseMonkey
Aug 19, 2009, 12:15 PM
Great.
gerrythrash
Aug 19, 2009, 01:19 PM
Hi there... OK I checked the connection to the fuse panel and I didn't find any obvious signs of corrosion or anything... but here's what I DID find.
Several years ago, I had a car alarm put on. It later stopped working so I just disconnected the little box that the wires go to so I could start the car. Apparently, part of the installation includes clipping a thin yellow wire to the BLK/YEL wire coming from the ignition switch. This yellow wire is clipped to the BLK/YEL wire with a little plastic clip, which has a little blade that cuts into the wire from the ignition switch (my guess would be so some of that current will go to the alarm system). I took that clip off, and the BLK/YEL is partially cut through. So... I'm thinking my next move should be to cut this wire, strip the two ends and tie them together and wrap it with some electrical tape.
So... what gets me is this... when I had the ignition assembly replaced back in January... would the dealer just put on a new cylinder housing assembly, and left me with the same ignition switch/wires?
What do you think?
ID
TxGreaseMonkey
Aug 19, 2009, 01:37 PM
Completely removing alarm systems is always a hassle. I know your problem only involves the ignition switch. Why else would so many fuses not be getting power? Since I'm not there to look at things, just use your common sense and go over the ignition switch and harness very carefully.
If splicing those two wires solves the problem, use heat-shrink tubing, not electrical tape.
It appears that's what the dealer did.
TxGreaseMonkey
Aug 19, 2009, 01:59 PM
With the negative battery cable removed, turn the ignition switch to ON (Position II) and run a simple continuity test on each end of the harness. That should tell you a lot.
gerrythrash
Aug 19, 2009, 02:38 PM
OK... I was able to remove the alarm system wiring (it was a Y harness between the 7P connector from the ignition switch and the main harness). I connected the 7P to the main harness, and tested everything... still no power to the fuses. I'll try the continuity test, and if necessary... just get a new ignition switch.
So... let's say that a new ignition switch doesn't solve this. Does the focus shift back to the distributor?
gerrythrash
Aug 19, 2009, 02:51 PM
Also... the other thing I'm beginning to notice is that when I try to start it, the cranking is getting slower and slower. That's probably a batter issue. Right?
PublicEmily
Aug 19, 2009, 03:09 PM
All right I'm going to hop in here quick. I'm a novice but an ingenious learner. Cranking no starting. 2 second CEL, hear the fuel, hear the click. I need to buy something to read the volts I'm guessing. Is there a place with a picture of the wires I need to check/ After I check all the fuses I guess I'll have more answers.
(I can't type a question mark for some reason... )
gerrythrash
Aug 23, 2009, 04:30 PM
Good news! I my car running! After I put the new distributor on... it still wouldn't start. So I then tested the computor, main relay, map sensor... nothing. I also noticed that it would crank slower and slower. So... that, and knowing that the car has been sitting for almost 7 months... I took the battery in to be charged. I brought it back, put it in, and after a couple more start attempts... it started! It was a bit rough at first... but now it runs like a top. I'm so happy to have transportation again.
Thanks again for all of your help!
TxGreaseMonkey
Aug 23, 2009, 05:36 PM
Good job! I'm so glad you got it running--the alternative was not pretty. I apologize for not seeing your postings on August 19th.
Did you figure out what was preventing so many fuses from getting power?
gerrythrash
Aug 23, 2009, 06:08 PM
Good news! I my car running! After I put the new distributor on... it still wouldn't start. So I then tested the computor, main relay, map sensor... nothing. I also noticed that it would crank slower and slower. So... that, and knowing that the car has been sitting for almost 7 months... I took the battery in to be charged. I brought it back, put it in, and after a couple more start attempts... it started! It was a bit rough at first... but now it runs like a top. I'm so happy to have transportation again.
Thanks again for all of your help!
gerrythrash
Aug 23, 2009, 06:12 PM
Whoops... didn't mean to post that again. I still don't know what was going on with the fuses... but they're all working now... so who knows.
MattXFX
Aug 12, 2010, 12:48 PM
Hey,
We are having the same problem, check engine light won't go out, the fuel pump won't turn on and we aren't getting a spark at the plugs. We have checked everything from the distributer to the ECM, eventually replacing it. The replaced ECM started the car once in about 2 seconds and made the engine light go away. So we put everything back together and when we were about to drive out to get some connectors for the sound system, we got a check engine light again and now it won't start.
This was a gradual problem, it started with the car not starting every time you turn the ignition over, you had to wait for the fuel pump to kick in. We have fixed the main relay, and know that it works, and we know that fuel pump works.
We are wondering if we are just frying the ECM's somehow.
MattXFX
Aug 12, 2010, 12:49 PM
We also have no idea where the map sensor is on the engine
tomwiller
Sep 20, 2010, 01:33 AM
If Your Engine Cranks but Does Not Start Follow this Troubleshooting Guide
# Step 1 - Anytime you have a problem with electronically controlled components such as an engine, transmission, ABS brake, or SRS (supplemental restraint system, Air Bag) inspect all fuses using a test light and check the under hood power distribution center and under dash fuse panels. If all fuses test OK continue to the next step.
# Step 2 - To check for problems with electronically controlled components such as an engine, transmission, ABS brake, or SRS (supplemental restraint system, Air Bag) and the fuses test OK a trouble code scan is needed to identify any system trouble. Use a simple scanner tool to retrieve trouble codes and see if they relate to the specific problem, like a crank angle sensor failure code. If the trouble code present does not pertain to the immediate problem like an EVAP code ignore it until a later time, after the car is running. The reason we repair non-related codes after the engine is running is because sometime false codes can be triggered by the engine not running. Once the engine is running again the code present might cycle and turn itself off. You might say "if the engine doesn't run shouldn't it have a trouble code?" Sometimes conditions occur that will not be detected by the computer, example: if the fuel pump fails the computer cannot detect the failure, so the engine doesn't start and the computer thinks everything is OK with no codes. If no trouble codes are present proceed to the next step.
# Step 3 - The spark plugs in your engine are used to ignite the compressed fuel air mixture. If the condition of the spark plugs are fouled by excessive fuel or carbon the engine will not start, backfire or run rough. Remove all spark plugs to inspect their condition. Please use this spark plug condition reference guide to see how the spark plugs are operating.
# Step 4 - Determine if the engine has compression, this can be done a number of ways but the most complete method is to perform a compression check. Remove the spark plugs and perform a compression test on one cylinder. If one cylinder has compression then the remaining cylinders usually will be close to the same. Crank the engine over about 5 seconds, normal compression readings should be between 125 psi and 160 psi on each cylinder. If no or little compression exists additional tests will be needed. The most common reason for an engine to lose compression is a timing belt or timing chain failure. If low or no compression exists remove the oil fill cap and observe camshaft rotation when the engine is cranked over. If no rotation exists the timing belt or chain has failed. If your engine has a timing belt and you cannot see the camshaft easily remove the upper bolts to the timing cover and gain visual access to the belt, recheck cam rotation by cranking the engine over. Sometimes a timing belt or chain can jump causing the camshaft to lose correlation with the crankshaft and therefore causing low compression. The best test for this condition is to remove the timing belt/chain cover and inspect timing marks. If the compression is OK proceed to next step.
# Step 5 - Test the ignition system output, ignition systems can vary in configuration but operate on the same principal. Ignition systems can consist of a coil, pick up coil, crank angle sensor, cam angle sensor, spark plugs, spark plug wires, distributor cap, ignition rotor and a distributor and any variations of these components. An ignition coil is a voltage stepper coil that transforms a low voltage (12 volts) signal into tens of thousands of volts needed to jump the gap of the spark plug. This coil is activated by an ignition module triggered by using the camshaft/crankshaft angle sensor; timing is adjusted by the ECM (computer).
If this doesn't work try to consult from Certified Auto Techinicians/Engineer. :D
kieshmudva
Aug 4, 2014, 02:41 PM
Been through the same thing for the past 4 days trying to figure out how to fix this problem. When the car is not starting you need to check the spark if it looks weak and you actually try bypassing the relay and force the fuel pump to run and it still won't start It's the IGNITION SWITCH also a very common problem with these cars. One of the symptoms is an irregular shift. Of course we didn't realize it was shifting as bad as it was until we by-passed the ignition switch and the car ran great shifted great. New switch is about $200 I don't have that much pulled one from a junk yard for $10 hopefully it will last awhile...
Anybody gets this post and wants more info feel free to email me
[email protected]