,
Fast forward, again, to three weeks ago. The Frankenmiata, after running a staggering 10 endurance races at nearly double its stock output, finally needed a new engine. Being LeMons, we had to find a replacement engine for as close to free as possible. We tried begging, but apparently our first engine build had sucked up the world’s supply of free Miata engines.
 |
| Textbook early Miata crank damage. The woodruff keys in these early 1.6 Miata engines were too short and could, with time or high revs, walk off the nose of the crank until there wasn’t enough engagement. The overloaded keyway would give way, letting the timing belt pulley rotate. SInce the engines are non-interference, nothing bad really happens, as long as you consider a trashed head, timing belt and front main seal to be nothing bad. |
Finally, we found a well-maintained 146,000 mile engine for $50. Why so cheap? It was a victim of the dreaded 1.6 Miata crank wobble, and although mechanically perfect in almost every other way, it needed a new crankshaft. A week before the Buttonwillow Arse-Freeze-Aplalooza, we swapped our old crankshaft into this new engine, fired it up, and called it good, without so much as even checking if the head bolts were properly torqued.
The new engine was, indeed, quite good. It made the same power as our old one, but without the old engine’s quart-of-oil-per-tank-of-gas thirst.Everything seemed fine until the first pit stop.
That garden sprayer still travels with me to every race, and topping off the coolant with it is a standard part of every Eyesore Racing pit stop. The procedure is simple. Just pump it up to 20 or 25 psi before the stop, then plug it into the radiator hose when the car arrives and open the valve. The 20-psi water in the bottle flows into the 12-psi cooling system, blowing the relief valve on the radiator cap.
The Frankenmiata’s cooling system is completely concealed behind coroplast ductwork, so we can’t actually see when water starts coming out the cap. Instead it flows into the also-concealed overflow bottle, then out the bottle’s spill hose, which sticks out the side of the car. In a normal pit stop, when we open the bottle, there’s a 2-3 second delay between opening the valve and water pissing out the overflow bottle’s overflow hose.
This time, it took a lot longer. There was a long, pregnant, 20-30 second wait while the trusty garden sprayer pumped three quarts of water into the cooling system. This did not seem good.
But maybe it was just a bubble, we rationalized. Maybe we hadn’t properly filled the cooling system when installing the engine. Maybe that three quarts of air was in there the whole time…
Not so much. The next stop took another 3 quarts. Still in denial, we assumed we had a coolant leak. An engine swap gives you plenty of opportunity to disturb crusty old radiator hoses, some of which are bound to protest by starting to leak.
At the end of the first day’s racing (the Buttonwillow race is really around 14 hours, since they don’t allow night racing), we pressurized the cooling system and went leak hunting. We found nothing.
Next, we pulled the spark plugs, figuring that if it was a head gasket, as we were starting to fear, one of the plugs would be white with the evidence. All the plugs looked the same, though. There was no evidence of any head gasket issue. Finally, someone had the good sense to shine a flashlight down the spark plug holes and discovered an ever-deepening lake on top of the #2 piston.
It was just then that we remembered how the car was running on 3 cylinders when we pulled it off the trailer. And how it had sounded just a little off when we started it up at the end of our last pit stop. Of course, we’d been starting the car with a wet cylinder.
But now what? First: triage. With water pissing into the cylinder, the first thing to do is stop the pissing. Removing the radiator cap drops the cooling system to atmospheric pressure, giving the water no more reason to rush into the cylinder. Next, we cranked the engine with the fuel pump relay disconnected and the #2 spark plug out, giving the water somewhere to go so the engine didn’t hydro-lock. Finally, we put the plug back in and fired it up, warming up the cylinder enough to dry out all the water.
If the head gasket problem was serious, this should have sent combustion pressure into the cooling system. With the cap off, water should have erupted from the cap any time we revved the engine. This didn’t happen.
Somehow, once again, I had one of those mysteriously-benign coolant-only blown head gaskets. No oil was getting into the coolant. No water was getting into the oil. No combustion pressure was getting into the radiator. We just had a leak in a particularly inconvenient place.
See the eroded gasket around the small coolant hole? That erosion is the only evidence of the 3-quart-per-tank coolant leak that almost cost us the LeMons championship. Obviously, we only found this the week after the race.
Our course of action from here was clear. We had a spare head gasket and time to change it, but odds were good the head was severely warped so there was a pretty good chance the new gasket would leak too. The odds of us screwing something up during that head gasket swap were pretty high, so it seemed too risky. Instaed, we re-torqued the head bolts to the top of the torque tolerance (loosening each bolt 1/8 turn before re-torquing it to spec), finding several bolts WAY out of spec. And then we crossed our fingers.
For all of day one, the car had been leaking at a constant rate, and that rate seemed sustainable. Even three quarts down, the car never ran hot. If the leak continuted at this rate, and we kept topping it off at each stop, we shouldn’t have to slow down at all.
So that’s what we did. And 8 hours later, we blazed across the finish line, 3 quarts low on coolant, 39.6 seconds behind the leading E30 BMW of POS Racing. Second place was still good enough to clinch the 2010 24 Hours of LeMons championship, though, which means this garden sprayer, which I found in my rented garage 10 years ago, just won its second championship.
