'97 B4000 "turns over" but sometimes won't start

[cobrajocky]

I have a 1997 Mazda B4000 V6 (same as a Ranger 4.0 liter) PU 2WD, only 80k miles, had the CPS replaced 20k ago. For no reason at all, on occasion with no predictability or reason, from cold only, the motor will crank and crank and crank but will not start for hours or the whole day or two. I'll come back later and bang it starts. Once or twice I've heard it backfire through the intake manifold. It has never done this hot, never two days in a row and it only has been doing this about a year, well after replacing the CPS. It may only happen once a week, or again after 2 weeks, or once a month! Sometimes it'll happen two or three days in a row! Nuts!?

We checked for spark when it does this and no spark, but there is a minor smell of gas. So I've replaced the distrib coil. I was going to replace the Ignition Module, but the Mazda dealer parts guys laughed and said they virtually have no history of these failing, none even in stock, even from the warehouse! The Mazda techs said they can't think of a specific reason and can't diagnose it unless it's dead ... wanted to have me leave it there, even though I told them it may not happen for a week or two or more! It's my only transportation, so can't do that.

I am at a loss that the local Mazda dealer is not any help.

 

[kemo]

CEL by chance? (check engine light)

 

[cobrajocky]

All the dash board lights are on when you turn the key to start! That doesn't tell you anything. The CEL is only an indication when a motor is running or stalls after running.

 

[Mirage]

You may want to hook up a code scanner anyway. Sometimes a code is thrown, say when you couldn't get it started, then a day or so later it starts just fine, and resets the error code. But in most cases, the last error code is still saved and can be retrieved. It may just point you in the right direction. It couldn't hurt.

 

[kemo]

If there is a code stored, that's somewhere to start. if not, time to hunt for where continuity ends since it sounds like an electrical problem if that's the case. But, your engine won't fire if it has no idea where the piston is in cylinder 1, hence why I ask if you had a CEL PRIOR to your truck not starting. Throwing parts at the truck won't fix it... well, it will if you get lucky. Thats how one guy I work with at my other job fixes things. He has no idea what he's doing, so he moves parts around until he THINKS it is fixed. Dono how he hasn't been fired yet...

 

[cobrajocky]

Logic always wins.

I agree with you both, that a Code Scanner might show something, but I don't have one and I don't want to buy one since I am planning to sell the truck this fall (moving) and buying a nice new Toyota Tundra.

Since this problem is erratic as all hell, I can't borrow one, as it might not happen for days or weeks.

I'll watch for a CEL before I crank it, but I still doubt that it will tell me anything specific since the CEL is a highly generic alert light anyways, requiring a scanner to pull a code out, which still might be a foggy answer.

I doubt that it is a fuel problem, since I've checked pressure which is fine and while it won't start I do get a fuel odor, once even got a backfire in the intake manifold (I believe) when it suddenly "got spark".

On one other Ford / Mazda truck forum, some have pointed to a flaky "Relay Box"; others to the CPS (crankshaft positioning sensor shaft). But I had a bad CPS (actually broken) before and it ran better than it does now after having a new CPS put it! And I've read in post that after 1994 or 95 Ford V6 motors, the CPS really does nothing since the newer SS ignition system used does not reference the CPS.

I only have about 82k miles on this '97 truck, and I'll be damned if I spend more money to fix it than it's worth.

 

[kemo]

These codes really aren't "generic". In fact, they'll point you where to start looking.

And to be correct, it's not CPS, it's CKP.

Here is an example of a CKP code and what it causes...
http://www.obd-codes.com/p0016

The engine may crank but not start

Further reading that you NEED to understand...

http://www.freeautomechanic.com/ckp-crank-position-sensor.html

Related Symptoms

* A No Start / Intermittent Start condition – Can be caused by a faulty crankshaft position sensor due to loose connections, bad grounds, high resistance in the circuit, or opens in the circuit.

Just because you recently replaced it doesn't necessarily mean that it is out of the question as to being your problem.

 

[cobrajocky]

These codes really aren't "generic". In fact, they'll point you where to start looking.

And to be correct, it's not CPS, it's CKP.

Here is an example of a CKP code and what it causes...
http://www.obd-codes.com/p0016

Further reading that you NEED to understand...

http://www.freeautomechanic.com/ckp-crank-position-sensor.html

Just because you recently replaced it doesn't necessarily mean that it is out of the question as to being your problem.

All I know is the Mechanic showed me the online Mazda parts page and it did say CPS for the Crankshaft Positioning Sensor parts. Leave it to Ford for pervert a P in to a K in their parts listing.

I've been suspicious of the CPS, but why would the B4000 run fine if not great with it broken as previously, and now be intermittent with a new one that may well be flaky?? Makes no sense at all.

 

[kemo]

As I previously quoted before...

Can be caused by a faulty crankshaft position sensor due to loose connections, bad grounds, high resistance in the circuit, or opens in the circuit.

The 3 in bold make perfect sense for what would cause an intermittent issue.

Get a volt meter, backprobe your ckp sensor and try to catch the minimum voltage and maximum voltage while it is being cranked over. You'll see anywhere between 0 to 5 volts. Good meters have a min/max feature, you'll get your info that way. You can also unplug the connector and check the resistance of the CKP... although you have to probe the right terminals.

CRANK POSITION SENSOR CHECKS

Whether a crankshaft position sensor is the magnetic type or a hall effect switch, most problems can be traced to faults in the wiring harness. A disruption of the sensor supply voltage, ground or return circuits can cause a loss of the all-important timing signal resulting in an engine that cranks but won't start.

Also, on some vehicles, damage to the notched sensor ring on the crankshaft, harmonic balancer or flywheel can cause an erratic crankshaft sensor signal.

When troubleshooting a suspected crankshaft position sensor problem, you should follow the diagnostic flow charts in the vehicle manufacturer's service literature to isolate the faulty component when a fault code is present, otherwise there is no way to know if a no-spark starting problem is a bad ignition module, coil, computer, wiring fault or ignition switch.

Magnetic sensors can be checked by unplugging the electrical connector and checking resistance between the appropriate terminals. On a GM 2.3L Quad 4, for example, the sensor should read between 500 and 900 ohms. Always refer to the vehicle manufacturer's test specifications when testing these sensors. Obviously, if you see a zero resistance reading (shorted) or an infinite (open) reading, the sensor has failed and needs to be replaced. If viewed on an oscilloscope, a magnetic crank sensor will produce a waveform similar to that below:
Crankshaft Position Sensor waveforms

A good magnetic crank position sensor should produce an alternating current when the engine is cranked, so a voltage output check while cranking is another test that can be performed. With the sensor connected, read the output voltage across the appropriate terminals while cranking the engine. If you see at least 20 mV on the AC scale, the sensor is good, meaning the fault is probably in the module, coil, wiring or computer.

Hall effect crankshaft position sensors typically have three terminals; one for current feed, one for ground and one for the output signal. The sensor must have voltage and ground to produce a signal, so check these terminals first with an analog voltmeter. Sensor output can be checked by disconnecting the coil and cranking the engine to see if the sensor produces a voltage signal. The voltmeter needle should jump each time a shutter blade passes through the Hall effect switch. If observed on an oscilloscope, you should see a square wave form (see above) that changes in frequency.

If your diagnosis reveals a faulty crank sensor, the only option is to replace it. With Hall effect sensors, the sensor must be properly aligned with the interrupter ring to generate a clean signal. Any rubbing or interference could cause idle problems as well as sensor damage. Magnetic crankshaft position sensors must be installed with the proper air gap, which is usually within .050 in. of the reluctor wheel on the crankshaft.

 

[cobrajocky]

Sounds like good solid advice

Kemo,

This sounds like a good solid clue, but it may be over my head for doing the testing. But I can take this thought to the mechanic that put in the CKS / CPS last year and have his test it.

I found on another forum (this problem does come up a lot on both Mazda and Fords with V6's - same motor, well same truck) a suggestion that it might be the inhibitor switch (neutral safety switch). Something I can only check when the truck won't start. But I would think that this switch would also inhibit the engine from even turning over, right?

 

[kemo]

Kemo,

Something I can only check when the truck won't start. But I would think that this switch would also inhibit the engine from even turning over, right?

Correct.