Cranks but won't start

Im spitballing here,
Did the engine run before the TB service?
Is the gasoline fresh?
Are you able to verify TB alignment?
Are the spark plugs new?
I’ve never tried with a coil pack, but it used to be a simple check to pull the spark plug wire off and while cranking, hold the boot close to the block to see spark. Not elegant, but effective. I would think it is possible to do the same with a coil pack, although a little more involved due to the harness and mounting.

If I remember correctly the factory service manual has a thorough flowchart for troubleshooting. Go down it step by step, even if you’ve already done it.
Engine ran good prior to TB service, but 12 years ago. I checked alignment marks multiple times during and again after replacement. Fresh gas. Lines clean. New plugs. But I'm second guessing everything now.

I've been doing everything by myself so I need to get a helper to check spark, injectors, etc. I need to be 100% sure about spark and fuel to move on. I still think it may related to sitting for so long.
 
Apparently misfire sensors have two connectors exactly like ignition modules. Kinda strange.
Yes, they look pretty much identical and I think they might even be be plug swappable because they both have the same 6 pin / 8 pin plug arrangements and the big and small plugs appear to be arranged the same on both modules. They might be plug swappable; but, definitely not functionally interchangeable. It is almost like they decided to save money on designing a new case and just re used the ignitor module case design.

Keep this up. Have enough of these 'that didn't work out' events and pretty soon you will qualify for the Tech Expert badge :) .
 
Engine ran good prior to TB service, but 12 years ago. I checked alignment marks multiple times during and again after replacement. Fresh gas. Lines clean. New plugs. But I'm second guessing everything now.

I've been doing everything by myself so I need to get a helper to check spark, injectors, etc. I need to be 100% sure about spark and fuel to move on. I still think it may related to sitting for so long.
I’m a bit rusty diagnosing but unless an engine is seized or a completely destroyed, it should try to fire. Even if one cylinder is able to build pressure, gets fuel and spark, it would / should sputter.
I’ve seen an engine missing the back 2 (of 8) connecting rods run. One of the connecting rods was sitting where the last quarter of the single cam shaft was!
Keep with it. Take your time and follow the steps. It is going to be something simple.
 
Igniter didn't fix it. I'm running out of ideas.

Well that is disappointing because the fault analysis was pointing to that as a high probability (although no the only) candidate. Have you had a chance to try the noid light test to see if the injectors are firing?

Noid lights are a convenience. If you have a multimeter you can sort of do the same test by back probing one of the injector connectors with fine wires and connecting the multimeter across the plug so that you are measuring the voltage across the injector. If the injectors are firing you will see the multimeter voltage switch back and forth between zero and something. The '91 has an external injector resistor so I am guessing the injector voltage might actually be around 2 - 4 volts. This is a bit of a kludgy test because you are not always certain that the back probe is making contact and sometimes they will pull out during the test. The noid lights make it much, much easier. Also, at a cranking speed of about 150 RPM the injectors will be firing about 2 times per second. Most less expensive digital multimeters have slow A to Ds so all you may see is constantly changing stuff on the display when the test is working. If the display remains fixed then the injectors are not firing or one of the back probes has disconnected. Noid lights are so much easier!

The spark test during cranking as proposed by @Miner should theoretically work. However, as noted I think it will be a bugger to execute. However, as I recall the coil harness enters from the cam belt side. If you completely unbolt the harness on the front head and unplug it from the other coils you might be able to lift the harness up enough so that you could remove the #6 coil and spark plug while still connected to do the spark test. Failure of the spark test would point you to
- the ignitor (two bad modules seems unlikely)
- wiring issues
- ECU
- crank / cam sensor

If the noid lights are not firing and the spark test shows no spark, those are two semi independent systems with the following common elements
- the ECU
- crank / cam sensor
- wiring issues affecting those particular elements

In your first post you mentioned wet spark plugs so I assumed that the fuel pump is priming when you switch the ignition key to run (before you try to start the engine). That is a fundamental check for the ECU (and the main EFI relay). If the fuel pump is not priming then you have problems further 'up stream'. When you mentioned wet plugs, did you check all the plugs or just one? If all the plugs were wet that suggests that all the injectors are firing, or at least going through the initial wall wetting cycle.

During all of this, is the check engine light illuminating on power up and then going out? That is also a basic self test indicator for the ECU. Have you used the service check connector to scan for any stored error codes in the ECU? Error codes 4, 54, 9 and 59 cover the four individual crank / cylinder position sensors. I assume that the check firmware is a pretty basic open / short circuit check and should be able to flag an error very quickly if one exists. If you did the tach test and you are reading RPM and the ECU has no CRK/CYL sensor error codes then balance of probabilities says the ECU and the sensor might be OK.

Circling back to the 'wet spark plug thing'. Since the car was in storage for 12 years, what did you do with the injectors? Have they been replaced or serviced (and confirmed working)? If the injectors are new or have been confirmed as functional then the next comment is not relevant. I had inferred from the wet plug comment that all plugs were wet; but, if you only checked one is it possible that the other injectors are severely plugged up and the engine is not getting enough fuel to start up? It would be a rare set of circumstances that engine would not even cough even if the injectors were plugged up.
 
Keep at it- you'll figure it out and we are here to help!

In situations like this, I like to start at the beginning and assume nothing. What is known: The engine will turn over but not fire. Even a drippy injector or bad plug/coil in a cylinder or two will still result in a start or at least a sputter. To me, this means the starter circuit is fine, but the engine isn't getting any fuel, any spark or both.

FUEL

Step 1 is the ECU. If the ECU isn't working, the engine will not run because it will not get fuel or spark. When you turn the ignition to run, you should see the check engine light illuminate for about 2 seconds and then turn off. This is the ECU check routine. If the light never comes on, the ECU is not working. It could be something as simple as a loose plug/shorted wire or bad as a fried ECU. The ECU troubleshoot list starts on page 11-26. Assuming you get the light, go to step 2.

Step 2 is the fuel pump. When you turn the ignition to run, you should hear the fuel pump run behind the firewall for a few seconds and stop. It's a humming noise. Turn off the climate fan to help hear it. If you don't hear the pump, check the following:
  • Fuel pump wires behind the driver's seat that go through the metal plate (could be chewed)
  • Fuel pump relay- also behind driver's seat. Test is on page 11-101.
  • Fuel pump- test is in page 11-102. If you get more than 11.2 oz of fuel for the 10 second test, then you know that fuel is at least making it through the rails and the regulator.
Step 3 is the main relay. The test is on page 11-103. An easier way is to find a friend with a known good relay to swap. If the car starts, there's your problem! I can send you my spare from the Zero for testing if you want.

Step 4 is the fuel pump resistor. A bad resistor shouldn't prevent a start, but the engine will not run. It will start, sputter and die. But it's still worth checking. Test is on page 11-106. Also look for frayed or cut wires where they attach to the resistor- this is a common problem on the NSX.

Step 5 is the fuel injector resistor. If this unit is faulty or the wires are damaged, then the injectors will not fire. Look for damage to any of the wires and then do the tests on the bottom of page 11-90 to 91 and 11-94.

Step 6 is the fuel injectors. Do the resistance test on the top of page 11-90 for each injector. Though, if the engine isn't starting at all, I doubt all 6 injectors are bad. Still, assume nothing and test them all. Check the wires for damage.

Step 7 is the fuel injector signal wires from the ECU. Do a continuity check for each injector to make sure there is a clear signal to each injector. Again, a failure on one or two of these wires would not prevent a start and it's unlikely all 6 wires are chewed, but assume nothing and check them. The manual (page 11-91) requires the use of the ancient and unavailable ECU tester, but you can use a multimeter and just check pin-to-pin.

If all of these tests check out, you've confirmed 100% that the engine is getting fuel inside the cylinders and you can move on to spark.

SPARK
Step 8 is the ignition signal from the ECU. In Step 1 you confirmed the ECU is functioning. Thus, we know it will send an injector pulse signal out. Now, you need to check if the signals are making it to the ignitor. Do a continuity test (page 11-57) to check each wire.

Step 9 is the ignitor ground wires. If the ignitor ground is bad, it will not fire to any of the coils. The test is on 11-57 and there are two black wires. Make sure they are not chewed and securely bolted to the ground point (G103 shown on page 23-92) on the intake manifold.

Step 10 is the ignitor itself and the output wires to the coils. The test for the unit is on page 23-93. I know you already tested this, but test it again. Assume nothing.

Step 11 is the coils. At this point (assuming the ignitor isn't bad), you've established that there is a clear signal and power getting to each of the coils, so you know there is the correct voltage at each coil harness plug. Now you need to check the coils and plugs. Again, it's highly unlikely that all 6 are bad, but test them (page 23-95). Check the plug color and carefully measure the gap.


If you have gone through all 11 steps and it all checks out, then the engine should run, even if it runs poorly. Given that you just went through a TB/WP job, my suspicion is that you forgot to plug something back in or attach a ground somewhere. Running through this progression should catch the culprit. Generally, you should make sure the main harness engine grounds are attached as well- there's one under the coolant bottle from the transmission, and depending on year, one on the upper firewall coming off the front cylinder head or one/two attaching to the right side body next to the alternator/fuel lines.
 
I can't thank you guys enough for the time and expertise you're giving.

Where I work we sell repairable vehicles that we buy at auction. I'm a Buyer but help oversee the shop. Our mechanics spend a lot of their time diagnosing odd problems since most have collision damage. A tried and true test for fuel is using starting fluid to eliminate lack of fluid as the problem. I should be able to get enough in the intake to test it at lunch today. It's usually very reliable.

After that I'll work through Honcho's steps.
 
Yes, they look pretty much identical and I think they might even be be plug swappable because they both have the same 6 pin / 8 pin plug arrangements and the big and small plugs appear to be arranged the same on both modules. They might be plug swappable; but, definitely not functionally interchangeable. It is almost like they decided to save money on designing a new case and just re used the ignitor module case design.

Keep this up. Have enough of these 'that didn't work out' events and pretty soon you will qualify for the Tech Expert badge :) .
yea your right ,my last post i was looking at the fuel pump resistor , had to move the snowblower to get a look at the misfire detector ,sorry for inaccurate post , but happy we dont have any snow
 
SOLVED. And I feel embarrassed and lucky.

I tested for spark through a plug and it worked so I decided to take the valve covers off and double check the cam gear positions. Somehow I managed to line up the front intake gear on the 2 with the arrow instead of the 1 with the line. I still haven't found what those extra marks are used for but after setting it to the correct spot I did a compression test to verify then fired up the motor. Sounds good now.
 
SOLVED. And I feel embarrassed and lucky.

I tested for spark through a plug and it worked so I decided to take the valve covers off and double check the cam gear positions. Somehow I managed to line up the front intake gear on the 2 with the arrow instead of the 1 with the line. I still haven't found what those extra marks are used for but after setting it to the correct spot I did a compression test to verify then fired up the motor. Sounds good now.
@Miner 's 'not elegant; but, effective' check dispels the no spark problem. Since you had spark and fuel I am surprised that you didn't get the occasional 'pftt' out the tail pipes.

Don't feel bad. 15 years ago I was assembling a vintage Volvo engine and the SM said #1 firing was with the timing marks on the crank and cam gear opposite one another. I took 'opposite' to mean lined up across from one another - they way most timing marks are registered. Nope! They meant on opposite sides of the timing gears from one another (180 deg of rotation). Nothing except for the occasional pfft out the exhaust was happening. Finally did the inelegant spark plug test with the valve cover off and discovered that I was firing TDC on #1 with the exhaust and intake valves open. Not wanting to remove the cam gear I pulled the distributor and clocked it 180 deg. More than one way to fix a problem.

Good that it is fixed.
 
@Miner 's 'not elegant; but, effective' check dispels the no spark problem. Since you had spark and fuel I am surprised that you didn't get the occasional 'pftt' out the tail pipes.

Don't feel bad. 15 years ago I was assembling a vintage Volvo engine and the SM said #1 firing was with the timing marks on the crank and cam gear opposite one another. I took 'opposite' to mean lined up across from one another - they way most timing marks are registered. Nope! They meant on opposite sides of the timing gears from one another (180 deg of rotation). Nothing except for the occasional pfft out the exhaust was happening. Finally did the inelegant spark plug test with the valve cover off and discovered that I was firing TDC on #1 with the exhaust and intake valves open. Not wanting to remove the cam gear I pulled the distributor and clocked it 180 deg. More than one way to fix a problem.

Good that it is fixed.
i think that if you rotated the crank 360 degrees the camshaft would routate 180 degrees and the valve timing would be spot on. But the distributor would still have been 180 degrees out.

Gosh, I miss distributors; sorta.
 
i think that if you rotated the crank 360 degrees the camshaft would routate 180 degrees and the valve timing would be spot on. But the distributor would still have been 180 degrees out.

Gosh, I miss distributors; sorta.
Oops

You are correct. What I did was clock the distributor wires 180 deg (switch 1-4 and 2-3) to get it to work. Removing the distributor, rotating the crank 360 deg then re inserting the distributor would have been the correct procedure. However, that vintage of Volvo engine had the oil pump driven off the bottom end of the distributor and getting the distributor shaft back into the oil pump drive if there was the slightest bit of mis alignment was misery. I opted for the easy to do non OEM spark plug wire arrangement.
 
@e9coupe If you found that your front intake cam was aligned with #2 TDC marking when the crank and other 3 x cams were at #1 TDC, you were very lucky.
Probably you got away without bending the valves.

You may want double checking the TB tension as most likely, it was not adjusted properly if the initial alignment was wrong.

Also, may want carrying out the leak down check.

Kaz
 
@e9coupe If you found that your front intake cam was aligned with #2 TDC marking when the crank and other 3 x cams were at #1 TDC, you were very lucky.
Probably you got away without bending the valves.

You may want double checking the TB tension as most likely, it was not adjusted properly if the initial alignment was wrong.

Also, may want carrying out the leak down check.

Kaz
Compression is 170-177 across all cylinders and that was after sitting for 12 years. Engine runs good. Not sure why the TB tension would be questioned because the gear was positioned wrong, unless you're suggesting since I made one mistake I've made more. I do question how accurate the TB tension is using the tensioner spring process without being able to physically see the TB slack (lower cover on). Since it's designed to do it that way I didn't take the lower cover off and have to trust it. I considered taking the cover off just for piece of mind.
 
If I am reading this correctly, it ran properly before the TB service, correct?
If so, it has to be something that was disturbed; assuming all mechanical components are within specs.

Is there a security system?
 
Last edited:
I'm having a hard time following the tenses of what was and what currently is.

There is a sweet spot of low/no tension in the valve train at the blue mark on the harmonic balancer. This special valve train position allows the timing belt to be tensioned without influence.

If one of your cams was off the mark, then you need to pull the harmonic balancer, pull covers, re-index everything, verify everything with a straight edge (parallax on this car a big deal), and re-tension the timing belt.

I'm with Kaz: you are probably OK. However...

Borescopes are cheap and I would inspect the front bank for witness marks. You're gonna have to remove all the spark plugs, so a lot more to gain than to lose.
 
I'm having a hard time following the tenses of what was and what currently is.

There is a sweet spot of low/no tension in the valve train at the blue mark on the harmonic balancer. This special valve train position allows the timing belt to be tensioned without influence.

If one of your cams was off the mark, then you need to pull the harmonic balancer, pull covers, re-index everything, verify everything with a straight edge (parallax on this car a big deal), and re-tension the timing belt.

I'm with Kaz: you are probably OK. However...

Borescopes are cheap and I would inspect the front bank for witness marks. You're gonna have to remove all the spark plugs, so a lot more to gain than to lose.
The latest... TB was realigned on the front intake gear to the correct 1 mark. All marks line up perfectly with "white" crank mark lined up with pointer on cover. My pulley doesn't have any colors left on the marks but the TDC mark is easy to figure out.

Compression is 170-177 on all cylinders.

I did the belt tension (loosen bolt) at TDC. I now know this is incorrect. Which mark is the blue line? The manual shows a mark but it looks like it's by itself. Is that the one?
 
The latest... TB was realigned on the front intake gear to the correct 1 mark. All marks line up perfectly with "white" crank mark lined up with pointer on cover. My pulley doesn't have any colors left on the marks but the TDC mark is easy to figure out.

Compression is 170-177 on all cylinders.

I did the belt tension (loosen bolt) at TDC. I now know this is incorrect. Which mark is the blue line? The manual shows a mark but it looks like it's by itself. Is that the one?
I'm a little worried about that compression reading. A "healthy" C30 near sea level will usually show above 200 psi. Did you do the test on a cold engine? That might explain it. Does the car have any history of FI or nitrous? Though, if it was a valve-contact situation, you would only be seeing compression issues on the front head, since the rear head cams were presumably aligned properly. I'm trying to visualize where the front intake valves would be in the combustion cycle with a 15-tooth offset. For peace of mind, I'd at least do a leakdown test and maybe borescope the front head pistons to look for witness marks on top of the pistons and check the rims of the intake valves so see if the black nitride coating has been chipped/scraped. The Honda valvetrain is fairly stout, so you might have gotten lucky! ;)

I've alluded to it on many threads, but the belt tension process is not nearly as simple as the manual makes it look. You really have to check the tension by flexing the belt between the rear EXH cam gear and the tensioner (you're looking for about 10mm of deflection). I had to do the 9-teeth and then snap the tensioner process several times before it tensioned properly (you can do it as many times as you want- they key is to hit the right tension). The friction of the tensioner bolt on the pulley and the tensioner pivot point are variable and can cause different belt tension results each time (they resist the force of the spring).

The blue mark on the pulley is for advancing the belt 9 teeth before you loosen the tensioner bolt for the second time. Technically, this is supposed to remove all the slack from the belt. I just mark the gear:

20210910_214811.jpg
 
Yeah, when I put the belt on I followed the tensioning procedure but ended up tightening based on what seemed right, which ended up like you explained. The belt seemed too loose by just using the tensioner. I'll take the covers off and start over. Bah.

I have the car on the ground in my other garage for interior and paint work. I may take a break from the motor and come back to it with a clear head.

Compression was done dry. 12 years dry. My friend has a leak down tester so I'll do that too. I have a bore scope for my phone but haven't used it. Not sure if it's any good.
 
Last edited:
R&R'ing the NSX TB scares the heck out of me, so I spend a crazy amount of time verifying my verifications of all the index marks. A small metal scale straight edge is extremely helpful in this situation. And then I do it all over again while talking myself through it.*

On the bright side, I've seen very loose and mis-indexed TBs that seem to work just fine.

The compression doesn't concern me at this point as it is all even.

I *would* do a vacuum check once everything is up and running. And, borescopes are $20 at Harbor Freight, so get one. You should also check the intake VVS for loose screws too.

Getting an assortment of mirrors is also quite helpful and one should have them to verify the valve cover seals.

I really do admire your confidence with this install, but the NSX TB is rather difficult and non-intuitive. It is difficult, but less so than a F355 or Audi S8.

*I probably talk to myself when working on cars more than I'd like to consider, but it does keep me on track. This might just be me to have gotten old.
 
Last edited:
BTW, would an OBD1 show a misfire if a valve was damaged? I didn't see a misfire code on the list in the manual and I haven't worked on an OBD1 car in quite a while. I figure it wouldn't show a specific cylinder but maybe a misfire or an 02 sensor out of range.
 
Here's a video I did showing a too loose belt after the standard tensioning procedure. You can see that there is proper tension at the cam gears and water pump, but the belt between the rear EXH and tensioner is way too loose. If I had been doing the job with the engine in and cover on, I would never have caught it and just assumed that the belt was A-ok based on doing the service manual procedure exactly. This is why I strongly suspect that the "one tooth off" we typically see on many NSXs when they are opened up is the result of dealer techs doing exactly the same thing. They follow the steps, line up the blue mark, release the tensioner, torque it and assume it's fine.

Don't get me wrong, the NSX will run fine with this loose of a belt, but if you run it hard, it likely will slip a tooth or two. I know it's a royal PITA, but you're smart to double check it. It has to run with this tension for 7 years...

View attachment 20210910_220916.mp4
 
BTW, would an OBD1 show a misfire if a valve was damaged? I didn't see a misfire code on the list in the manual and I haven't worked on an OBD1 car in quite a while. I figure it wouldn't show a specific cylinder but maybe a misfire or an 02 sensor out of range.
OBD-I has no misfire detection circuit. The coils are "dumb" in that they just receive the signal and fire the pulse.
 
Also, the front intake cam gear is where the Crank Angle Sensor pick up is located.

Still very strange that the engine didn't sputter when being cranked.

This will be an interesting case study.
 
Also, the front intake cam gear is where the Crank Angle Sensor pick up is located.

Still very strange that the engine didn't sputter when being cranked.

This will be an interesting case study.
It's been a few years since I looked, but I'm pretty sure the teeth are evenly spaced on the rear side of the gear wheel. So I don't think the crank sensor would realize anything was wrong. I can't remember if there is a "key" tooth (or gap) that tells the sensor it's at #1 TDC.
 
Back
Top