Head gasket replacement and injector troubleshooting

The time finally came to start putting things right and keeping fingers crossed that the warped head was ineed where the issue lie, and not, say, a cracked block for example.

While we were in there, I wanted to eliminate, hopefully permanently, the head hardware from the mix. A friend suggested a bolt may have stretched, and since the Mazda hardware (bolts) costs about the same as a set of ARP studs, the studs were a no-brainer.


Since I had the luxury of having my brother, an engineer at Maserati in town for the holidays (we’ll ignore the fact that he does suspension & vehicle dynamics…) I enlisted his help to get the car squared away. Here he is with the big torque wrench helping to get the head nuts tightened down on the studs. We did it in 3 different steps to 55 ft/lb (using the ARP Ultra Lube), and did the last step twice just to verify that they were correct.




Didn’t take too many detail pictures of getting the head back on, but it’s pretty straight forward. Put the HLAs back in, in the order they were removed, put the cam in and index it with the Cam Angle Sensor, put the caps on in the correct order, and torque the bolts down in the prescribed pattern, to 120 in/lb. A healthy dab of assembly lube was used on any of the rotating and reciprocating bits for the first start.


After that, it was a simple matter of making sure the seals were pressed in correctly, then bolting the gears back on (SUPER easy using the Flyin’ Miata cam tool, easiest timing belt job I’ve ever done), the timing belt, belt covers, accessory pulleys, and the intake & exhaust manifold. Note the removed Thermostat water neck / housing. More on that later.


After getting everything buttoned back up, it was time to spin the engine with the ignition & injector bank disconnected to build up some oil pressure. And here’s where I’m a bonehead: I did this without pulling the pugs, which meant that my tiny, 6 lb ETX-9 battery was fighting compression, instead of the engine spinning freely. Once we finally got oil pressure to read on the gauge, the battery was dead. Like, capital D dead. I’ve had experience doing deep discharge cycles on these batteries before, and it never works out well. We left it on the Battery Tender for a few hours to see if it could be brought back from the dead, but it wouldn’t ever charge fully and could barely crank the car. Fortunately, my local O’NAPAdvancedAutoParts had an ETX-9 battery on the shelf so we took that home and plopped IT on the charger to try again. After the family Christmas Eve dinner, I decided, because I’m stubborn, to give it 1 last shot for the night. So we hooked the battery up and attempted to fire it up again. It would crank fine, but alas, it wouldn’t fire. So we covered the basics. What do ICE engines need? Air, Spark and Fuel. We had air (even tried playing with the throttle to see if it would touch off, but no joy). We tested the spark and were getting power to the plugs. But when we pulled a plug, it was bone dry. The ECU was having the fuel pump prime, so we should have pressure, but it didn’t appear that the injectors were firing. A cursory check didn’t reveal anything wrong, but the 30A injector fuse didn’t look… right. The filament was all there and accounted for, but it was a different shade of color compared to the unused, spare 30A fuse right next to it. At this point I figured why not try and see if the problem went with the fuse, and on the next crank, she started to cough to life. Gave everything a moment and hit the start button again, and lo and behold, it was a Christmas Miracle (though, a ton of man hours, effort and troubleshooting probably helped): the Miata roared to life! High fives were shared all around and we went ahead and brought up Shadow Dash on my phone so we could keep an eye on things as it warmed up to operating temperature. This was important because, as you may remember from THIS POST, the fan wasn’t coming out quite as soon as I’d like. The working theory was because the coolant neck simply wasn’t getting enough hot water to get an accurate temperature reading to the fan switch. As I pointed out above, the temp sensor is now directly in the head:


Now that it’s reading temperature from the front of the head, and coolant is going all the way through the motor courtesy of the re-route (instead of just circulating around the #1 and #2 cylinders, as it does with the stock setup) the temperature as read by the ECU (at the hot side / back of the head) at which the fan kicked on dropped nearly 15 degrees from where it was when the fan switch was in the old Thermostat housing. I’m calling that a big win, and that mod had the exact effect I intended. It’s nice for SOMETHING to finally go right with this motor.

Merry Christmas & Happy New Year, y’all!


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s