FrankenMiata: Part I – Freebuilding the Engine

Frankenmiata Mazda B6 Miata engine by the dumpster

FrankenMiata: Part I – Freebuilding the engine

By Dave Coleman

There is a moment of magic at the end of every engine build. It's that mysterious and suspenseful instant when a cold lump of metal comes to life for the first time. One second its just a pile of all the sharp, oily bits you’ve been lacerating yourself with for weeks on end, and the next second, if you’re lucky, it’s vibrating, belching smoke, perhaps even purring. Suddenly it generates its own heat, moves under its own power; it can take you places, power out corners, do burnounts, drive to Philadelphia; it's instant mechanical freedom.

If the elaborate mechanical and financial ballet of buying new pistons, rings, bearings and gasket sets; paying for machine work, taking blocks here and heads there, observing torque specs and generally doing things right is the internal combustion equivalent of giving birth, then building an engine entirely out the broken parts others have thrown away; digging through piles of old bearings to find the least worn; looking for the used gaskets that still have a little squish left in them, and using sandpaper where others would use a machine shop is surely the mechanical analogue of reanimating the dead. If the first lap around the block in your new engine brings a tear to your eye, then imagine the giddy, uncontrollable, Frankensteinian joy of the first dyno pull with the unholy creation you resurrected from the grave.

This is a tale of that joy. A tale of building a Miata engine for $96.46.

The engine used in this project was freebuilt from no less than four junk engines. Freebuilding is like rebuilding, but without all the stuff that costs money. Ignore all your instincts to think long term, and focus only on what is absolutely necessary to make the engine perform the task at hand — surviving 14 hours of abuse before exploding. Labor cost, it should be noted, is not considered in LeMons accounting, a fact that can lead to countless lost weekends and the absolute destruction of any social life. You have been warned. 


Inspection and Teardown

The first step of any freebuild is figuring out why the donor engine was free. If the engine you're starting from wasn't free, you're doing it wrong.  The previous owner's cover story is a start, but should never be considered definitive. If they really knew that much, you wouldn't have this free engine in the first place.

Frankenmiata cylinder bore 
Those vertical scuff marks aren't exactly a great sign, but they aren't the end of the world either. It's quite normal for the highly-loaded piston skirts to wear the sides of the bore a bit. Most importantly, none of those scratches were deep enough to catch a fingernail. Neither was the ridge at the top 

Our first engine was a dead powerplant from a rare automatic-transmission equipped Miata. The automatics had an even lower performance version of the B6 engine, offering only 9.0:1 compression vs. the manual car’s 9.5:1, and using camshafts with 1 mm less valve lift. Those low-compression pistons are perfect for turbocharging. The low-lift cams, though, have a date with the scrap metal bin.

Our first engine's previous owner was Mark DiBella, who had never seen the engine running properly (he bought the car for the body). The owner before him, the story goes, was a half-assed ex-mechanic who “fixed” the engine only months before it started running on 3 cylinders. Doesn't sound good…

Before disassembly, a leakdown test was performed and cylinder #4 was found to have 100% leakdown. After teardown, the culprit was found to be exhaust valves that had large notches burned out of them.

 Frankenmiata Mazda B6 Miata engine bad bearings
 On the one hand, bearings that look like this are a bad sign. Clearly, a lot of crap got into the oil at one point and did tremendous damage. On the other hand, the car still ran well enough to burn an exhaust valve. Bad bearings did not kill this engine. LeMons demands this kind of robustness.

The rest of the head was deemed “good enough” through a series of simple, free inspection techniques. For example, we removed the valve springs and retainers (with a hammer) and pulled all the valves out about an inch, letting us check the valve guides. Rocking the valves side-to-side reassured us that valve guide wear was not excessive, and sliding the valves up and down by hand reassured us that there were no bent shafts, galling, or debris wedged in any valve guides.

We checked the block using the highly scientific look-at-it-and-poke-it-with-your-finger technique, confirming that the bores were in surprisingly good shape. When looking at a bore, shine a light on the cylinder walls and make sure the original crosshatched hone marks are still visible everywhere. Bores can wear unevenly and this will be revealed by smooth spots with no crosshatching. Also, make sure there are few, if any, vertical scratches in the walls. These scratches can happen when debris gets caught between the piston and bore, or when a cold engine is driven too aggressively, heating and expanding the piston before the bore has had a chance to expand a corresponding amount. Once the gouges are there, they provide a leak path past the rings, lowering effective compression.

Frankenmiata Mazda B6 small-nose damaged crank 
Early 1.6-L Miata engines are notorious for their easily-damaged crank keyways. The dreaded “crank wobble” is caused by sloppy mechanics bungling the woodruff key installation when swapping timing belts. It is relatively rare to find an early, small-nose crank with a keyway that doesn't look like the one above. This was our first crank, which we threw in the dumpster. Using a crank like this will lead to incorrect cam timing at best. At worst, if you also bungle the installation, the nose of the crank can snap off. Later cranks had a larger diameter nose that was more monkey-tolerant. Our final crank had a rare, undamaged small nose. 
Frankenmiata Mazda B6 Miata big-nose crank 

When poking a bore with your finger, drag your nail up the top half-inch of the bore. Peak cylinder pressures happen near the top of the bore, and those pressures push the rings harder against the cylinder walls. Typically, it’s the top inch or so of the cylinder wall that wears the most. The rings never touch the top quarter-inch or so of cylinder wall, though, so there should be no wear at all there. If the bore is worn excessively, there will be enough of a step between the worn and unworn parts of the wall to catch your fingernail. Don’t be mislead by carbon deposits that tend to build up on the top of the bore, though. Clean the top of the bore with some brake cleaner and a rag before doing this test.

When you remove the oil pan, check it for debris. Chunks of metal, blobs of silicone, lost pocket change. Any of this stuff is bad news. Also check the screen on the oil pump pickup. Our pickup had several little metal grenades, some blobs of silicone goo, and even one long, curly steel shaving stuck in it. This was a warning of things to come… (If they aren’t numbered already, be sure to mark the main and rod bearing caps as you take them off so they go back in the same position. Swapping caps can result in less-than-round bearings.)

You can learn a lot about a bearing just by looking at it. If there are gouges, that’s obviously bad. Most bearings are made of multiple thin layers and worn spots will show up almost like tree rings as the various substrates are exposed. Because of the way the bearings are loaded when the engine is running, Its normal to see wear on the top and bottom of the rod bearings, but not the sides, and only on the bottom of the main bearings.

If there are no gouges in the bearings you may be able to re-use them. If some are slightly gouged, you may be able to use some light emery cloth to basically de-burr the gouges. If they’re severely worn and gouged, you should really start looking for better ones. Our bearings were severely gouged.

Making matters worse, our crankshaft, a notorious small-nose crank, had a damaged nose with an enlarged keyway that would at best screw up cam timing and at worst cause the nose of the crank to snap off halfway through our race.

So, to review, our engine needed only manual transmission camshafts, two exhaust valves that weren't burned, a full set of un-gouged bearings, and a decent crankshaft. Other than that it was great. Money and a machine shop could have fixed everything that was wrong with our engine, but that costs money. Instead, we rummaged for better bits.

Leave a Reply

Your email address will not be published. Required fields are marked *