[mechanic58]

I have a 97 Corolla CE that has developed a random crank but no start issue. Whenever this occurs the temp gauge will peg to HOT as soon as I turn the key on. Also, sometimes when this condition is present, whenever I turn the key off, some of the warning lights and the CEL will be dimly lit for a few seconds even after the key is off. I have done some troubleshooting, including checking grounds, but I have not figured this out just yet. I do know that when this happens there is no spark and no fuel. When it clears up everything works normally and there is no codes. I figure this issue is probably ECM related. Hoping for some insightful feedback, any help is much appreciated.

 

[Pete Suhman]

My first thought was grounds, but it could also be on the positive side, at a connection point like a fuse box or connector or something like that.

 

[mechanic58]

My first thought was grounds, but it could also be on the positive side, at a connection point like a fuse box or connector or something like that.

I did have the engine out of this car recently. This car spent about 6 years sitting in a junkyard with a bad rod knock. I pulled the engine out and overhauled it. After I put it back in the car, brand new, I felt like it wasn't making the power it should have. Sometimes the power was even intermittently variable. Then one day it started skipping, randomly. Each time it would do it the check engine light would flash with each skip. But it still wouldn't log any codes. Then it started just cutting off for no reason but would always re-start. Eventually it got to where it wouldn't start at all, except on rare occasions. All of this has unfolded inside of about 500 miles of driving since I put the new engine back in the car.

 

[mechanic58]

ALL of the ignition parts are brand new, from the dist to the spark plugs. I even dropped $100 on a set of OEM plug wires. Has a new CPS also. The problem is either wire harness-related or the ECM. Has to be. I have another OBDII Corolla that I have been swapping parts back and forth with in my troubleshooting efforts, haven't found any failed components yet. I can't swap the ECM's because the other car is a 5-speed, this car has the 3-speed automatic.

 

[mechanic58]

Something else I found that was odd, when this issue is active, if I disconnect the cooling fan switch on the t-stat housing, the cooling fan will come on and run immediately when I turn the key on. I do not believe this is a normal indication.

 

[94RollaDad]

Something else I found that was odd, when this issue is active, if I disconnect the cooling fan switch on the t-stat housing, the cooling fan will come on and run immediately when I turn the key on. I do not believe this is a normal indication.

It's by design so a failed switch doesn't cause overheating.

 

[94RollaDad]

Might just have to whip out the multimeter and take some resistance readings on the sensors when this occurs to trace the error either to the sensor or to the wiring or ECM... If you don't have the FSM I recommend getting it for the troubleshooting steps here.

 

[DrZ]

Something else I found that was odd, when this issue is active, if I disconnect the cooling fan switch on the t-stat housing, the cooling fan will come on and run immediately when I turn the key on. I do not believe this is a normal indication.

Wrongo! Why did you believe this to not be normal? You need a wiring diagram and repair manual to properly know how things are supposed to work. The fan switch is normally closed (circuit). When it heats up it opens the circuit and this turns on a relay. The whole circuit is complicated, so you can't just guess at how you think it should work. Try not to assume you know how things work.

I doubt it's the engine computer because the temperature gauge in the cluster is not controlled or wired into the computer in any way.

The temperature gauge has 2 solenoids in it. Each tries to pull the needle in different directions. When the needle goes high immediately that tells me that one of the solenoids isn't getting grounded properly. One is grounded on the back of the engine block and I suspect this could be loose. The wiring diagram says intake manifold, so it might be there, but I think the wiring diagram is wrong. This is also the ground point for some other sensors, so it might explain some of your problems. This ground is hard to get to and inspect. It's near the starter motor. You may need to crawl under to inspect or remove some parts above.

 

[JacRyann]

There's a loose wire somewhere. Also sounds like fan-circuit may be connected improperly. Majority of circuits in his car are ground-switched, like modern cars. Might have power-wire going to fan-switch when it should be ground. Unplugging fan switch might be disconnecting that short-circuit and allow power to go where it's needed.

As mentioned, use official FSM wiring diagram and trace all circuits one at a time and verify that are connected properly. Should measure correct voltages and grounds at proper times.

 

[mechanic58]

Wrongo! Why did you believe this to not be normal? You need a wiring diagram and repair manual to properly know how things are supposed to work. The fan switch is normally closed (circuit). When it heats up it opens the circuit and this turns on a relay. The whole circuit is complicated, so you can't just guess at how you think it should work. Try not to assume you know how things work.

I doubt it's the engine computer because the temperature gauge in the cluster is not controlled or wired into the computer in any way.

The temperature gauge has 2 solenoids in it. Each tries to pull the needle in different directions. When the needle goes high immediately that tells me that one of the solenoids isn't getting grounded properly. One is grounded on the back of the engine block and I suspect this could be loose. The wiring diagram says intake manifold, so it might be there, but I think the wiring diagram is wrong. This is also the ground point for some other sensors, so it might explain some of your problems. This ground is hard to get to and inspect. It's near the starter motor. You may need to crawl under to inspect or remove some parts above.

Why? Because it is NOT the industry standard. I have never run across a cooling fan circuit that functions in this way. It IS a good idea, but this is the first automobile I have worked on that worked like this. I have been a mechanic for 35 years.

Thanks for the note about that ground on the back of the block. I'll have to investigate that. My lift is tied up at the moment, hopefully I'll be able to get this car up on it one day this weekend to check that out. As mentioned before, I did have the engine out of this car recently and so it is entirely possible I either left this ground lug fastener loose or failed to connect it all together. We'll see.

Thanks!

 

[mechanic58]

Might just have to whip out the multimeter and take some resistance readings on the sensors when this occurs to trace the error either to the sensor or to the wiring or ECM... If you don't have the FSM I recommend getting it for the troubleshooting steps here.

If I had to buy the FSM for every vehicle I traffic in, I'd a been broke decades ago. I usually do just fine without it. That's what these forums are for.

I have already ruled out all of the instruments on the engine as the source of the problem. I think DrZ is onto something with that ground on the back of the engine under the intake. That's one that I have not verified yet and is highly suspect.

 

[94RollaDad]

Yeah I think @DrZ hit on something important: The temp gauge and the sensor it receives its readings from are completely independent from engine operations and the engine computer. The fact that it displays an erratic reading in concert with no-start means two completely independent systems fail together, so you'll have to look at a common denominator - ground straps being an excellent place to start.

 

[DrZ]

Since it will crank I assume the big ground straps are fine. That's why I suggested checking that small ground on the back of the engine block that is used by some of the sensors and gauges.

It wouldn't hurt to check the other grounds anyway because they are easy to get to. There is one on top of the transmission under the air filter area. There is one that splits off that wire and goes to the fender/strut area near the air fliter. And there is another by the timing belt/pulley area coming off the last intake manifold bolt to the right strut area. This one has a connector which could be disconnected.

 

[JacRyann]

Thing you have to be careful is if ground for starter isn't firm and good, starting-current will go through ECU and fry it...

 

[mechanic58]

Welp, this has been fixed the problem was the ground on the main engine harness that is located on the back of the engine block. I had failed to connect it. Once I connected it the problem was solved. I appreciate everyone’s input to help me get this fixed.

 

[JacRyann]

Woohoo!!! Good job tracking that down!