NB2 Won't Start After Timing Belt
 

Hello everyone, my NB2 has a fairly recent engine rebuild on it. It probably has 5k miles on it, and ever since the rebuild the timing belt has made a fair bit of noise, so I took it off and set it again ensuring I do not make it too tight.

My problem is the car won't start. I have checked the timing (I am almost certain it is correct), made sure all of the electronic connections are there, and my CAS is gaped correctly. The car ran absolutely fine before the timing belt was removed, which is why I cannot think of a reason why it won't start. When I do try to start it, it will just crank uneventfully. If I put the gas all the way down the car will almost try to start. It will start to smell like gas if I hold the gas down and crank it enough times. The car has 120,000 miles on it.

Could it be fuel or spark? Everything was working fine before I reset the timing belt. Or could I really have messed up the timing belt alignment again?

Thanks!
Nick

You have done one of the following things:

• Messed up the installation of the cam gears on the camshafts
• Messed up the timing belt alignment
• Reversed the orientation of the 4-tooth trigger wheel behind the crankshaft damper

Set the motor to #1 TDC with the cam marks pointed at the I/E behind the cam gears, remove the valve cover and front timing covers, and take a photo of the camshafts and cam marks so we can check your work

Thank you sir, I will be back with pictures. First thing I'll check is the trigger wheel, as that is the easiest thing to correct. If that doesn't fix it, I'll take some pictures of the top cam alignment, and the bottom.

I got a dollar on #2

Well you may just earn a dollar, because I just checked the trigger wheel and it is fine. I had to stop because of rain, but I'll post pictures here with cams after it lets off.

My favorite technique. (warning involves math)

I remember talking to that guy on M.net back when he made that video, and pointing out to him that his belt tensioning method is basically wrong (never a mention of 1-5/6 crank rotations or the correct tensioning mark on the case). He eventually acknowledged it and said he'd either edit the video or create a new one specifically on belt tensioning, but apparently he never bothered, so this video lives on with errors in it.

Two other gripes: I also wished he had scrubbed off the old, unnecessary, and extra yellow marks on all the sprockets, which just adds confusion, and that he had gone ahead and lined up the crank sprocket at TDC mark (just for OCD reasons).

It is too bad though, because the picture quality is really good; it could have just been so much better overall without errors.

I push the belt over the pulleys, lock the tensioner all the way back, loosen it to tension, let the spring spring, lock it down, the rotate through 720* twice. Use a cam-lock or whatever to make sure the top end doesn't rotate. Works fine for me. Worst case, if you're having trouble, figure out which direction you're out of time, and over/under advance said pulleys. Do it the exact same way, now your method works, because you compensate for it.


To me, the BEST way to mark the timing, is a plumb-bob. Rotate the engine to TDC, then use the plumb-bob to mark 12 o'clock on the cam gears. After your rotations of the engine, set the crank mark to TDC, then check your marks up top. Great for engines with poor visibility/access for the timing marks.

The whole point of the factory spring is to provide the proper tension for the belt. The spring is supposed to be calibrated/intended for setting that tension when the cams are off their lobes and the valve springs are offering the least resistance. This condition occurs at 1-5/6 rotations after #1 TDC (they even gave us a nice little mark there for our convenience); anywhere else, and you may or may not be getting the intended tension from that spring.

I really don't know why various alternate tensioning methods keep come up again and again, as the factory method works just fine, and the procedure is clearly specified in the FSM. That being said, the tolerance range for belt tension is probably sufficiently large enough to accommodate alternate techniques as well. Although, it does seem that all the threads with reports of (over and under) tensioned belts occurred when they were not using the factory method.

That spring assumes you use a factory belt, which I don't do. One factory belt failure after 10 hours in a race motor was enough for me

How do you tension them?

That's true enough, although for a brand new belt, my sense it it wouldn't make all that much difference. So, like aidandj said, what is your method then? Is it a modified "by feel" sort of thing, or is it something even a monkey/M.netter could do repeatedly?

I add to the spring a fair amount of pressure from my thumb. To what tension, I don't know.

Only the spring would give quite different tensions depending on how freely it can move (lubrication etc).
The "hurt" level of a thumb is easier to repeat IMHO.

With a kevlar belt:
Effects of loose belt, we can all guess.
Effect of too tight, ?

If one mechanically locks the cams in place, doesn't the whole valve spring tension issue disappear?

No, in theory, I don't see that being the same thing at all. If the cam gears are frozen, how is that allowing the tensioner to pull slack from the left (idler pulley) side of the belt (the crank would have to rotate backwards then), or from across the top between the cam pulleys (impossible)? Again, in practice, it may work fine for you, but it's only because you have a "feel" for it and it ends up being within the spec range. It's not a "repeatable-for-any-man" routine like the FSM method.

I tension mine to about 1/2 the deflection of the factory spec/factory procedure for checking deflection. (tighter) I run heavy double valve springs, so more load on the belt to spin the cams, more likely to jump a tooth. I used to run a lot more tension, but I think it's unnecessary and don't do that any more.

All I do is set the engine at top dead center, lock the cams, Remove old belt and idler/tensioner, install new tensioner and idler, put on the belt, release the tensioner, then lock it down.

Start the belt on the crank, take it up to the exhaust cam, then intake, then tensioner. Then loosen the tensioner bolt, let the spring do its job, then tighten the bolt down. Very repeatable.


Spin through 720 twice to confirm timing.

:dealwithit:

I use that video only for the math. I use Keith's Miata Performance Projects book and a Mazda shop manual for the procedure.

Hey everyone, I appologize for the wait. Here is what the timing belt looks like right now. This is the exact position of the belt that caused the car to not run. Or, at least I think it caused it to not run. Does it look ok? I'll use that video's tooth counting method to check.
https://www.miataturbo.net/attachmen...ine=1447878540


https://www.miataturbo.net/attachmen...ine=1447878540

Isn't there also supposed to be a tick mark on the edge of the intake pulley at the 12:00 position where the 19 belt tooth count ends?

Post a picture of looking at the head from the front. I want to see the front cam lobes on both cams.

^ I would have said the exhaust, but I agree in principle.

Thank you everyone. Looking at it now, maybe the exhaust cam is a tooth too low.
https://www.miataturbo.net/attachmen...ine=1447880188


Top of cylinder 1


https://www.miataturbo.net/attachmen...ine=1447880188


Exhaust side


https://www.miataturbo.net/attachmen...ine=1447880188


Intake side


https://www.miataturbo.net/attachmen...ine=1447880188


Top view

Cam lobes are correct. I don't see anything wrong with that.

What about clearance on the crank sensor? Did it get moved? Is it too far away?

any missing grounds? are the plugs wet after cranking?

Hmm.. No, there are no missing grounds. There weren't any before the timing belt was removed. This is why I am so confused. The car ran fine before I adjusted the timing belt, and now that I have pretty much been confirmed that it is not a timing issue, I have no idea what it could be.

Is the crank trigger wheel on backwards? (Sav's original #3)
Should be convex side toward the radiator.

Nope, the crank trigger wheel wasn't removed from before. I honestly have no idea. I wonder if maybe exhaust cam is off a tooth, or maybe I had the crank sensor too close.

My impression is that being off one tooth won't prevent it from starting. It may stumble and idle poorly, or it may have no top end (depending on which way you're off), but it will start.

I agree that the most likely cause is electronics. Do you have a stock ECU, or aftermarket? If aftermarket, many of those have cam/crank pulse debugging tools. If it's a stock ECU, you might be able to hook up an OBD2 scanner and see if it outputs an RPM value while cranking (not sure if the stock ECU does or not). If you get an RPM signal, you know the cam/crank pulses are good.

You say you think you smell gas -- do you actually? If so, then your cam/crank signals are good (ECU won't inject fuel without a sync) and that would suggest it's probably a spark problem.

Stupid question, but you did get the plug wires back on the right cylinders, right? :)

On the "missing ground" topic, the main ECU harness is grounded through a bolt at the front of the intake manifold. The harnesses are somewhat in the way for the timing covers, so it's not unusual to take that bolt out to get more flexibility in the harness, and if you forgot to put it back on, or missed one of the ring terminals that goes under it...

--Ian

OK, now replace your cam angle sensor :party:

Thank goodness I have a spare engine sitting around. I'll salvage a cam sensor from it after I put my engine back together and see if that fixes it. :party:

Ian, thank you for a very intelligent and insightful answer. I love it when people explain things so thoroughly and in a "point A leads to B, which causes C" manner. I will test the spark after everything is buttoned back up. :bowdown:

Subscribed.... Facing same issues.

Cam Angle Sensor didn't work
 

It still will just crank and crank without ever starting.
-timing is correct
-not crank sensor
-not cam angle sensor

Known working sensors or just random ones believed to have worked at some time in the past?

Connectors or ECU?

Stupid stuff like a non-clicked connector can spoil a lot.

The car literally ran absolutely fine before I removed the timing belt because it was too tight. I put the belt back on, and it won't start. I'll check the connectors, but I am pretty sure they are all clicked. The ECU hasn't been touched in a long time. All of the grounds are good too.

Are you on the stock ECU by chance? Curious if a simple log would tell us what the computer sees. Check fuses?

Yep, stock ECU unfortunately. I'll check the fuses in a second.

99% of the time, if you took it apart and put it back together and it doesn't work, it's something you put back wrong. :)

Have you pulled the plugs? Is it injecting fuel? Is it sparking?

--Ian

Output test mode would be so sexy right now

I made a Youtube video showing what the car does when I crank it. I apologize for not being the smoothest talker in the world, and the lack of turbo :D

Have you pulled a plug and checked to see if you actually have spark?

I notice one screw is replaced on the cover, but is there any chance you opened up the never-to-be-touched voodoo VVT innards (behind 4 hex bolts) on the end of the intake cam?

You need to charge your battery before you do any more cranking or you're going to kill it.

Might want to check my thread for some troubleshooting. I know its not what you want to hear, but perhaps a very slow redoing of timing belt job to ensure it still lines up fine?

Were you sure to check the trigger wheel orientation?

As mentioned, have you verified spark by sticking a screwdriver in the plug wire and grounding while someone cranks?

Lastly as mentioned, def charge the battery or jump it when you try cranking again

So I pulled the coilpacks today on my NB2, and while cranking the car I could barely see any spark. We tested both coilpacks and both of them barely put out any spark. I checked the grounds and fuses, and both are all good. So it sounds like it is the coilpacks? The sparkplugs look like the car was running rich, which is also why I believe it is the ignition system. If it is not the ignition packs, any ideas what else it could be?

Thank you!

Seems quite odd that both would fail simultaneously though. No offense, but is there any chance they are not hooked up to the right cylinders?

No offense taken, but how would one hook them up wrong?

Go back to any pics you have, or others of a stock/put together setup...aka how it should be. Could be loose plug wire at either the pack or going into the cylinder... Could be u might have put together in the wrong way

For it to have been running perfectly, and all of a sudden both packs die is very unlikely

Also, u sure you grounding properly when u check? And lastly make sure plug gaps are within limits and plug wires are all on nice and snug. And make sure coil packs are tightened down all the way (don't over do it) along w plugs

I'll add to that...you shouldn't see a lightning bolt when testing FYI, so diagnosis of "barely" any spark might just be sunlight etc..

Just trying to help however I can

Since I don't have a pic of your current stock nb2 "cop" and plug wire arrangement, it was just a wild speculation that it is theoretically possible to configure your pairs in a 1&2, 3&4 arrangement instead of the correct 1&4, 2&3 arrangement. This theory would probably also need to include non-stock plug wires for it to even be possible though.

As has been mentioned before, the nature of your type of symptoms, where everything worked perfectly before it was disturbed, but not after, is classic for the oversight of some simple bungle in reassembly. You can look straight at it a hundred times, and it all looks good, but then once you finally see it, you're embarrassed at how stupid your mistake was, and often, how easy it is to then fix. We've all been there, we've all done it. Hopefully your fix is just that close.

Thank you TonyV and good2go for the advice.

Let me recap what I have done so far:
So so far I have
-Made 100% sure the timing is right
-quaddruple-bazillion time's double checked the grounds, connections, and sparkplug leads
-made sure the crank sensor is gaped correctly
-bought a new camshaft position sensor
-bought all new coilpacks
-bought a new battery

And the car still won't idle. When I crank it, it sounds like it wants to start, but it never does. If I pump the gas when cranking it, it will try to "catch" but will die without ever really idling. Cranking it in succession makes the car feel like it want's so start, but it will just sputter and die if I ever get it to reach 800+ RPMS. Does anyone have any ideas? I almost think it could be the fuel pump.

New thought: could this be an anti-theft immobilizer malfunction issue?

I have never heard of that before.

Every NB2 has got one. It's part of the whole chipped key fob deal. There are numerous accounts of them failing too.

http://www.ebay.com/itm/2001-Mazda-M...=Model%3AMiata

also, at The Parts Group:

https://www.miataturbo.net/attachmen...ine=1449624740
ECU w/ Keys,Immobilizer,Trunk Lock,Halo & Remote
Came From A 2004 Mazdaspeed.
$250

I am looking it up, and the symptoms certainly look right, but I cannot find a solution.

Have you verified fuel pressure?

Too many times I've heard, read, and done it myself where its running and then not...gotta be something we're overlooking

OP...beating the dead horse ONE more time, when you say you've triple checked timing...you're visually confirming the CRANK is at TDC per the mark you can only see taking the Crank pulley off...AND the timing marks are lined up on gears vs metal plate behind gears?

TDC on Crank per arrow, gear on left lined up w E and gear on right with I? All THREE confirmed at same time?
(It'll be last I mention...promise just that most of the time THATS the culprit)

And....how about a shot w the valve cover off while everything is lined up...in my issue turns out one of the camshafts was out of alignment...when all is lined up the intake (left) and exhaust(right) should be at 9oclock and 3oclock respectively

***When you pulled the pulley and timing wheel off...how did you reinstall the timing wheel? Tons of reads about people installing backwards and the car starts first try once corrected (won't notice poor fitment during reassembly... Just a matter if markings/orientation)

Let's confirm all this and leave timing out of it...ditto w air, leaving spark and fuel

Spark....screw driver in plug wire-resting on Engine Hook, someone cranking should see a good spark (do in garage or at night so very easy to see)

Fuel...pull intake tube and open throttle body butterfly after some cranking...or pull a spark plug or 2...smell fuel??

Let's be detailed about how your confirming all this, and supporting w pics would be huge as alot of members on here can spot something out of whack easily

On the previous page are detailed pictures of the current timing marks. I have done that job twice before, and every time it has ran fine.

As for the pulley, I did not remove the trigger wheel. It was facing the right way before I took the crank pulley off.

Spark: Spark is good. I thought it might be the issue because when we pulled the coil packs out, we could see spark, but it was faint. The car has pretty new ignition lines and spark plugs, and I just installed new coil packs.

Fuel: I can certainly smell it when it is cranked, but I have not tested it.

I'll offer another WAG that I've heard of before --> you didn't possibly stuff a rag into the air ducting (to block debris entering during repairs) and then forget to remove it, did you?

I don't think so lol. Again, I appreciate any ideas you have, but that is funny you ask because I'm sure someone has done that before. I have an Autoexe cold air intake, so I cannot even really reach the filter to block it.

I have been looking up the Immobilizer issue, and people are saying that the car will at least start for a little while before the immobilizer will kill the engine. Mine doesn't even start for a little while. Also, the immobilizer cuts spark, which isn't right because I have that.

Your exhaust cam is a tooth off in this picture.

Also, the crank keyway looks like crap/damaged. And you didn't have the bottom pulley exactly centered on the mark either. I would change that keyway and gear piece with new parts that aren't damaged.

Fix cam timing first. Take CLEAR PICTURES and post them up when you do that.

Like this, this is my motor timed correctly. Note I put the camera in a straight ahead view so you can actually see if they are lined up.

Attachment 183831

Attachment 183832

Apologies...with my own troubleshooting didn't remember previous pics. Seems like most of the above I posted has been ruled out. Seems maybe a tooth off but don't think that'd make a no start condition...

Agree doesn't sound like a immobilizer issue..

Hmmm, time to brainstorm... Will stay tuned and post anything I can think up