Car doc I need some help

[justen]

Interesting that I stumbled on this. I had the same exact issue, only it's on a old honda motorcycle. Came out my coils were not working right and the cdi (computer) was not functioning correctly.

I got mad, replaced the cdi.. But while there. I replaced the coils also for they are 32 years old. Heating and starting issues gone. So maybe look at that area old dog. With my cdi hot and coils hot. Was getting a weak spark. One or the other.. Or both was the issue, If you have gas coming out the exhaust. For some reason either spark is not burning gas. Or a something is off saying to dump more fuel in then is needed. Could be a easy as a miscalculation from something like air to fuel ratio.

[olddog]

I measure all the spark plug wires. They run between 6000 and 8000 ohms. Depending on manufacturer that is not unreasonable.

So I pulled the distributor cap and measured the resistance from the coil male lug on top, to the button on the bottom that the rotor rides on. Zero ohms. Then I started measuring the plug wire lugs. To my surprise I was getting 23 meg ohms on about half of them, and 2-3 meg ohms on the rest. It seems there is a layer of carbon built up on them on the inside of the cap. It is a dark hard surface. If I scratched around with the lead and pressed real hard I could get it down to ~1000 ohms. I sanded and polished one lug and it came down to 0 ohms.

Ok I found the problem. I ordered a new cap and rotor. Put them on and guess what. It still does it. !@# (*# @$$!@!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

I have check rotor to cap distance. Rotor tip is about the mid point of where the spark plug lugs are half milled off. Maybe a little lower than the mid point. No idea if that is correct or not. I just doubt that temp could cause this alignment to change enough to make the problem only happen when hot. Also the problem is getting worse with time and no change with a new cap and rotor.

I am getting a hot spark to the distributor and intermittent spark to the plugs. The only thing I haven't changed is the spark plug wires. Wires are not arcing, by visual check at night, but not when symptom occurs. Also you can hear the spark tool crack pretty good. I have move all the wires around with a plastic stick while symptom is happening, and I cannot see nor hear any arcing in the daylight.

Well I have the unplanned pleasure of redoing my septic system. So I'm broke and don't have much time to work on it any more right now. THIS SUCKS!!!

[thcardoc]

Originally Posted by olddog

I measure all the spark plug wires. They run between 6000 and 8000 ohms. Depending on manufacturer that is not unreasonable.

Then you get a car that is difficult to diagnose and you suddenly find out that measuring the resistance of the plug wires failed to give you an answer to the problem. This turned out to be one of the tests we were taught to do that was simply wasted effort. The sad part is they still push this and it fails to help nine out of ten times. Yes it can find a wire that has gone open, but having a substitute wire handy to use with the spark checker would have provided an answer faster.

Ok I found the problem. I ordered a new cap and rotor. Put them on and guess what. It still does it. !@# (*# @$$!@!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

I have check rotor to cap distance. Rotor tip is about the mid point of where the spark plug lugs are half milled off. Maybe a little lower than the mid point. No idea if that is correct or not. I just doubt that temp could cause this alignment to change enough to make the problem only happen when hot. Also the problem is getting worse with time and no change with a new cap and rotor.

There is another alignment that you must check and it's where the rotor is pointing at the time that the coil fires. If the distributor body is loose from the shaft the distributor cap may have moved in relation to the shaft. Your spark occurs at the correct time on the crank, but the rotor is inbetween cap posts instead of pointing right at one. This creates a very high spark demand (Kv for kilo volts) in just jumping that gap and there is little to no potential left to actually fire the plug. If you tried the old spark plug stuck in a wire on a valve cover, it would likely fire great all of the time. But since the spark checker takes 25Kv it fires randomly or not at all, as I described before, you diagnose as no spark.

I am getting a hot spark to the distributor and intermittent spark to the plugs. The only thing I haven't changed is the spark plug wires. Wires are not arcing, by visual check at night, but not when symptom occurs. Also you can hear the spark tool crack pretty good. I have move all the wires around with a plastic stick while symptom is happening, and I cannot see nor hear any arcing in the daylight.

Well I have the unplanned pleasure of redoing my septic system. So I'm broke and don't have much time to work on it any more right now. THIS SUCKS!!!

You either have bad wires inspite of everything that appears to suggest otherwise or your rotor isn't pointing exactly at the cap terminal when the coil fires. You have good spark in confirmed, but poor or no spark back out to the plugs. Roll the engine over by hand to ten degrees before TDC #1 and pop the cap off to see where the rotor is pointing. Grab a known good wire to substitute and see if it makes a difference.

[olddog]

I have to level out some dirt and get it ready to seed this week end, but I hope to get a little done on the Cobra as well. I'll check the rotor phase angle as you suggested. I wish I had some spare wires or at least another car I could rob one from, but I don't. I never did Ohm the coil wire. I would like to check it and then check it and the plug wires while they are hot. More out of curiosity than actual troubleshooting.

I'm going to replace the wires and wire looms as soon as I can get the money together.

Here is a link to the wire looms I'm interested in. Anyone have any experience or opinions good or bad, I would like to here it.
Buy Mr. Gasket Universal Spark Plug Wire Divider Bracket Set 6023 at Advance Auto Parts

[olddog]

Got to spend some time this week end on the Cobra.

The rotor is dead on #1 @ 10 degrees advance. At 50 degrees advance it is almost past the post. I expected better, but it should work.

Well I got some really nice wire looms and replaced the wires this morning. I really like the looks. No more over the valve cover sloppy wires. They are routed around the heads and then along the valve covers. Very nice.

90+ F today. Ran it up to operating temp. Let it sit for 40 minutes. Did the same thing, as it has been doing. Put the spark test tool on it. I have a hot spark now, but it still wouldn't run. Didn't have a timing light with me, so I couldn't confirm the spark is at the correct time.

I could smell fuel and the few times it would fire it blew back (rich) smoke. So Every stinking part in the ignition system has been replaced now. I'll still try to confirm the spark is on time, however I am starting to think it is fuel. In fact, I am looking at the ECU with suspicion, now.

At this point, I have been proven wrong at every turn. Confidence destroyed. Self esteem very low. Advise would be appreciated.

[thcardoc]

Originally Posted by olddog

Got to spend some time this week end on the Cobra.

The rotor is dead on #1 @ 10 degrees advance. At 50 degrees advance it is almost past the post. I expected better, but it should work.

That sounds good, I won't draw this out too far but the lower compression at 50 BTC allows for a lower spark demand voltage so a little wider gap at that point is OK.

Well I got some really nice wire looms and replaced the wires this morning. I really like the looks. No more over the valve cover sloppy wires. They are routed around the heads and then along the valve covers. Very nice.

90+ F today. Ran it up to operating temp. Let it sit for 40 minutes. Did the same thing, as it has been doing. Put the spark test tool on it. I have a hot spark now, but it still wouldn't run. Didn't have a timing light with me, so I couldn't confirm the spark is at the correct time.

OK, you good hot spark during "the event". You're at one of the points in auto repair diagnostics that can be very frustrating and many people would try and say that you ripped them off or didn't know what you were doing and got "the diagnosis" wrong. The simple facts are you confirmed an issue, corrected it and now you have confirmed that you made a difference with a needed repair. Did you fix "THE PROBLEM"? No, not yet. But had you not dealt with this problem you would still of had it looming in the background ready to show up and throw you some symptoms of it's existence whenever it felt like doing so.

The measure of a mechanic/technician isn't how good they are when things are easy, it's how they handle the nightmares that get thrown at us.

I could smell fuel and the few times it would fire it blew back (rich) smoke. So Every stinking part in the ignition system has been replaced now. I'll still try to confirm the spark is on time, however I am starting to think it is fuel. In fact, I am looking at the ECU with suspicion, now.

At this point, I have been proven wrong at every turn. Confidence destroyed. Self esteem very low. Advise would be appreciated.

Yea BTDT and had to teach myself a lot of routines or tricks to work my way back out of these situations.

Rule #1 at each event, you need to test correctly and completely while forgetting everything else that has been done. That takes some discipline but it pays off with efficiency in the long run job in and job out.

You keep saying that you smell fuel when it acts up. The questions that raise are:

Is it getting too much fuel?

How could you find out? Fuel pressure gage, volt meter to measure injector on time, voltmeter to confirm inputs such as the coolant temperature sensor and the intake air temperature sensor. What other inputs are needed for the computer to calculate the injector on-time during that hot re-start? (RPM, TPS, MAP/MAF?)

What would too much fuel do to the ignition system, specifically the spark plugs. Are they being fouled which would ground the spark around the gap of the plug? How could you tell? When it doesn't start, does it try to occasionally fire? Does holding your foot to the floor make a difference in the vehicles attempt to restart? (WOT throttle plus a crank command causes most system to go into a clear flood mode which can shut the injectors off completely, or else still pulse them but at a very short on-time.)

The fact that the car runs well until you get it into a hot soak actually makes it possible for you to gather known good information and practice retrieving what you need to know so that you are more efficient during an event.

The only part of all of this that you should even consider close to being a dissapointment is that what you are learning may never be put to use again as the odds of you having to experience a failure like this are extremely remote. But the reward of having done it will last forever.

So right now, outline your game plan for testing. You have now confirmed good spark where before it was clearly questionable. Next?

[olddog]

As I see it at the moment, the problem is some electronic item braking down with temp. There is the ECU, MAF, ECT, ACT, and Barametric temp.

Engine temp nor air temp, in my mind should not read be able to read so far from actual to shift either the air fuel ratio or the timing to cause the symptom. When this all started I did meter those sensors with the engine hot and cold and they were dead on, but the symptom was not happening when I took the measurement. My industrial experience with resistance temp devices is they fail open or shorted and that is easy to detect as a failure and ignore. I'm tempted to rule these out.

The barometric pressure did leave a code in the continuous memory this last time. It is a suspect. Any idea if the ECU detects a bad signal and ignores it? It would seem to me that this sensor would affect drive-ability, but not to the point of the symptoms I get. How do you test this bugger?

The MAF is a GM (Corvette) with a signal converter used to calibrate it to look just like a Ford MAF. I have the voltage verses air mass chart. If the signal converter is breaking down or the MAF itself it may give a strange reading. Question: if the TPS is on idle and the MAF gives a large reading will the ECU dump the fuel based on the MAF reading or will the ECU look at TPS at idle and the rpm and realize the MAF reading is bogus and therefore not dump too much fuel? If the ECU realizes the MAF is out of range wouldn't it trip a code? Most of the time it will start and idle, maybe a slight miss, but once you rev it, all he!! breaks loose. Lately it dies and will not restart until it cools down. Rule it as a possible suspect?

The ECU is mounted in the passenger compartment, above the passengers legs in the foot box. The foot box is near the headers and not a lot of air space between the top of the foot box and the hood. It does see heat. Transistors that fire the injectors could very well cause my symptom.

My thought at this point is to heat components with a hot air gun, when the engine is cool and there is no symptom. Then see if I can cause the symptom. Work my way through the parts one at a time.

[thcardoc]

Originally Posted by olddog

As I see it at the moment, the problem is some electronic item braking down with temp. There is the ECU, MAF, ECT, ACT, and Barametric temp.

I agree that is quite likely.

Engine temp nor air temp, in my mind should not read be able to read so far from actual to shift either the air fuel ratio or the timing to cause the symptom.

At times it appears that you are describing the engine flooding. Those sensors when the engine is being cranked cause the computer to add additional on time to the injection pulse for a cold (colder) engine. The way to think of the strategy is they cause the computer to add fuel to the injection pulse they cannot cause the computer to take fuel away from it.

When this all started I did meter those sensors with the engine hot and cold and they were dead on, but the symptom was not happening when I took the measurement. My industrial experience with resistance temp devices is they fail open or shorted and that is easy to detect as a failure and ignore. I'm tempted to rule these out.

Don't rule them out without testing and proving that they are working correctly, DURING AN EVENT!

Back probe, or tag wire them and get used to measuring their voltages under different operating conditions so that when the next event occurs you can efficiently prove exactly what they are doing. This is where the discipline I often refer to being needed to solve these kinds of problems really becomes evident.

The barometric pressure did leave a code in the continuous memory this last time. It is a suspect. Any idea if the ECU detects a bad signal and ignores it? It would seem to me that this sensor would affect drive-ability, but not to the point of the symptoms I get. How do you test this bugger?

First you do need to know what the atmospheric pressure is and what voltage that sensor reports at that pressure. Now you can compare the MAP sensor, KOEO (Key On Engine Off) to the baro sensor voltage once you know how closely they relate to each other. This is why scan data can be so nice, but there are times when failiures occur (a trouble code sets) that the microprocessor uses substituted data in it's calculation and scan data often shows the substituted data instead of the actual sensor value.

The MAF is a GM (Corvette) with a signal converter used to calibrate it to look just like a Ford MAF. I have the voltage verses air mass chart. If the signal converter is breaking down or the MAF itself it may give a strange reading.

This is one of the most troubling aspects of the system you have in your car. There is so much room for trouble to occur and no clear references that could be looked up. I would approach this with a four channel scope and preset the math channels to recreate the conversion that the sensor signal is ultimately supposed to mimic. That could take an hour or two all by itself and that could be time spent to find out, you didn't find anything wrong.

Question: if the TPS is on idle and the MAF gives a large reading will the ECU dump the fuel based on the MAF reading or will the ECU look at TPS at idle and the rpm and realize the MAF reading is bogus and therefore not dump too much fuel?

Normal strategy for a MAF sensor equipped system is for the computer to look at the TPS more for change and rate of change than it does for any given value.

Imagine half throttle voltage reading as being 2.5v, the computer actually wouldn't do anything different if half throttle voltage was 1.5v to 3.5v. But have that voltage change from 1.5 to 3.5 and the computer will have a signiificant reaction to that change.

If the ECU realizes the MAF is out of range wouldn't it trip a code?

If it recognizes a failure, yes it should and newer systems are much more prolific with their testing, and especially cross testing against other inputs or said another way performance testing.

Most of the time it will start and idle, maybe a slight miss, but once you rev it, all he!! breaks loose. Lately it dies and will not restart until it cools down. Rule it as a possible suspect?

The car has a problem, nothing is positively ruled in or out yet until you make the measurements that actually prove a failure. Remember one of the rules, forget you ever saw this before and test from that perspective. There's that need for discipline again.

The ECU is mounted in the passenger compartment, above the passengers legs in the foot box. The foot box is near the headers and not a lot of air space between the top of the foot box and the hood. It does see heat. Transistors that fire the injectors could very well cause my symptom.

My thought at this point is to heat components with a hot air gun, when the engine is cool and there is no symptom. Then see if I can cause the symptom. Work my way through the parts one at a time.

The problem is you could heat up something that actually isn't getting as hot in real use and cause a false symptom that you could positively fix but have that hav nothing to do with the original failure. Could it work for you this time? Yes it could, but given enough chances it woud not only not work, it would mislead you and have you replacing a part that isn't bad.

Practice, plus Patience, plus Discipline and you'll earn the answer to this.