Replacing the timing belt and related items has been on my list of things to do and I finally got around to it this past weekend. While doing the timing belt, it made sense to also install a new water pump, replace a leaking front crank seal, replace the cam shaft seals, and put on a new valve cover gasket.
For this and many of the other typical tasks I’m doing, I use at least two sources of information:
- Mazda Miata MX-5 Performance Projects by Keith Tanner
- 1999-2005 MX5 Workshop Manual from Mazda
Tanner’s book is very good and is great for getting the lay of the land before doing the job. And the Workshop Manual from Mazda is a critical resource though it is designed for the professional and not the enthusiast.
Another book that would be useful for most folks (highly recommended!) is Mazda MX-5 Miata 1.6 Enthusiast Workshop Manual or Mazda MX-5 Miata 1.8 Enthusiast Workshop Manual by Rod Grainger. Unfortunately, the 1.8 manual is currently out of print on Amazon but may be available on the publishers web site. I’ve not used these resources because even though a 1999 MX-5 is a 1.8 because the 1.8 manual doesn’t cover the 1999+.
Two supplemental resources for any Mazda work, especially for Spec Miata, is the invaluable Spec Miata web site forums and the helpful Miata.net garage. The Miata.net garage has helpful instructions on most common tasks (e.g. timing belt change). And the Spec Miata forums cover many issues specific to race car prep.
In doing my timing belt change, I started off by removing the radiator. This isn’t strictly necessary, but it sure is nice to have the room. And radiator removal is one of the easiest jobs you can do on a Miata. In fact, draining the radiator fully of water/coolant probably takes longer then removing the radiator fan and related items. So I highly recommend doing this.
Once the radiator is safely out, the fun begins. Dis-assembly goes fairly quickly if you’ve already removed the A/C and power steering components. I started with:
- Air intake
- The valve cover gasket
As every component is removed, the parts go into separate labeled plastic bags. It is even better if you take photographs both before and after removal in case you need to refer back to what goes where later on. As this job progresses, there are a few brackets and cables to disconnect. Photographs are very helpful for quick reference if you get stuck the many hours later during re-assembly.
With the valve cover off, I moved on to removing the covers for the timing belt and pulleys to get access to the water pump.
With all the work to do related to a timing belt change, the water pump is itching to be replaced too. And it is a pretty simple task compared to the overall job.
Replacing the water pump was a real breeze. Unfortunately, it was not a sign of more good things to come.
My leaky crank seal was next on the agenda and didn’t look to be such a daunting task. But it was mildly annoying due to the composition of the crank seal.
Tanner’s book suggests using a seal puller, but this did not work at all. The seal on my car was rubber over a metal ring. The seal puller simply scraped the rubber off the ring and left me with a bigger problem to tackle. On the other hand, the Mazda manual suggests a screwdriver cover by a cloth. This turned out to be much more on target. With some measure of bad language and dark thoughts, I was able to finally get the seal out and the new one in.
At this stage, I went into a time warp and forgot to take pictures until I had the new timing belt on and was part way to putting the whole thing back together. Everything seemed to be going very nicely.
Alas, it was not to be. Upon firing up the car, well, there was no fire. Despite my careful approach to putting on the timing belt, everything was not aligned properly.
I am fairly certain the crank shaft started at top dead center but moved on me when I was fiddling with the cam shaft belt alignment. So off everything came (sadly) and I was back to getting the belt put on properly.
The second time around seemed to go better and I was meticulous about checking everything once, twice, and even three times. After it was all reassembled for a second time (or was it third?), the engine fired up straight away and sounded great. I ran it for a while checking carefully for leaks and other issues. Everything looked good, so I took it out for a short drive.
On my short test drive, the engine ran smoother then it had in the past and things looked good. I took it back to the garage and again checked for leaks or anything unusual. No leaks were observed, so then I went out on a longer run into the mountains.
On my mountain drive, I pushed the car HARD. No issues were apparent for the first 10 miles. Then suddenly, I got some engine misfires. The tach was seriously bouncing and the spark plug firing was cutting out. I eased off and drove home another 5-6 miles without any further issues.
Upon arrival back home, I downloaded codes from the car using my ODB-II reader. Pending codes were reported as P0339 (crankshaft position sensor intermittent) and P0302 (cylinder 2 misfire). The Miata workshop manual suggests a P0339 is caused by loose wiring harnesses, so I centered my debugging in this area.
I rechecked all the wiring harnesses and didn’t find any issues. However, I did find a loose metal bracket that was dangling around. It seemed like this could potentially interfere with the sensor readings since it appears to be based on signal generation as metal passed through a magnetic field. I quickly removed the offended bracket since it was a remnant from my removed power steering pump.
Again, I took the car out for a long run and was hopeful. It was not to be. Again, I encountered misfires but later in my drive.
At this point, I have re-checked the cable connections but don’t have any strong leads. Since I was pushing the car very hard in steep terrain, I suspect there may be some heat related issues. But for the moment, I’ve got many other tasks to work on and am going to wait before running through the full set of diagnostic procedures suggested by Mazda.
All-in-all, this set of tasks (seals, belts, gaskets, etc) was more time-consuming then challenging. Of course, maybe I’ll change my tune if I can’t find the solution to P0339.

















