1997 Mazda MX-5 Miata Long Term Road Test

  • Full Review
  • Pricing & Specs
  • Road Tests
  • Comparison
  • Long-Term (10)

1997 Mazda MX-5 Miata: Assembly By Keegan Engineering, Pt. I

August 13, 2012

1997 Mazda MX-5 Miata: Assembly By Keegan Engineering, Pt. I

It's been a while since you've heard about Project Miata's long-rod BP engine, but that doesn't mean nothing has happened. After a few unrelated delays, it has been coming together quite nicely at the experienced hands of Mike Keegan of Keegan Engineering.

Some might call it obsessive. I'm referring to the discipline required to be a professional engine builder. You might think that the majority of time spent in building an engine is, well, building the engine. 


1997 Mazda MX-5 Miata: Assembly By Keegan Engineering, Pt. I

The best engine builders spend most of their time making measurements. This was hammered home when I watched Mike Keegan of Keegan Engineering setting to work on bringing Project Miata's 1.9-liter long-rod VVT engine into reality. His assembly process entails assembling and disassembling the engine several times in order to obtain the measurements he uses to set the final configuration. Patience, he has it.

Mike is too modest to have a motto, but if he did, I bet it would go something like this: Assume Nothing, Except That Every Dimension On Every Part Is Wrong Until I Confirm That It Is Right Or Make It So. 

Measuring Valve Travel

1997 Mazda MX-5 Miata: Assembly By Keegan Engineering, Pt. I

If you have big cams, big valves, high-compression pistons or some combination, the valves and pistons might have a territory dispute. The valve overlap period is when this is most likely to occur -- as the piston rises toward top dead center, the exhaust valves are closing and the intake valves are opening.

One of the many checks Mike makes prior to assembly of every engine is a valve travel measurement. He starts off with the bottom end bored, assembled and torqued to spec to ensure that the pistons' position will be representative of reality. For this process, the head has no cams, lifters or retainers but the valve seats are cut to depth and includes the guides and valves to be used in the engine -- in our case Supertech +1 mm valves

Then he puts a dummy head gasket on the head and torques the head down. 

1997 Mazda MX-5 Miata: Assembly By Keegan Engineering, Pt. I

A dial indicator is set on the head and situated onto the exhaust valve stem to measure valve travel. A degree wheel is set up on the crankshaft nose. Mike sets the piston to 10 degrees before top dead center (BTDC), then 5 degrees BTDC, then zero degrees BTDC. At each position he extends the exhaust valve down until it touches the piston and measures the associated valve travel. The process is then repeated for the intake valves at zero, 5 and 10 degrees after top dead center (ATDC).

By examining the travel of the valves during the overlap period in this manner, insight is gleaned beyond simply piston-to-valve clearance or whether the pistons might touch the valves in operation. He compiles this valve travel data and plugs it into his software to reveal exactly how much he can deck the head, whether the pistons' valve reliefs need adjustment and how much, even how aggressive the VVT mapping can be. 

Setting Up The Block

1997 Mazda MX-5 Miata: Assembly By Keegan Engineering, Pt. I

Our JE pistons are 1mm larger than stock, requiring the blocks' bore to opened to 84mm. Keegan has found that the BP will accomodate 85mm bores without worry, though he cautions that 85.5mm requires the block be sonic checked to ensure there's adequate meat. 

Piston rings seal on the cylinder walls best if those cylinders are straight, round and of the correct surface finish. However, the act of torquing the head down to the block significantly distorts the cylinder bores. If it had nice round bores prior to installing the head, it won't once the head is torqued .

This is why good engine builders use a torque plate during honing of the bores. The torque plate is a slab of metal that is bolted to the block's deck prior to honing its bores.  

However, if the torque plate is too stiff, it won't induce the same distortion as does the head. This is why Keegan first performs an analysis of the stock head to determine its inherent stiffness, then creates a torque plate that emulates this stiffness. For extreme builds he employs the hot hone process, whereby a hot liquid is circulated through the block and torque plate during honing. This simulates -- and accounts for a lot of --  the heat-induced bore distortion of a fully-warmed, running engine. 

During honing, Keegan allows no more than 0.0005" out-of-straightness from the top to the bottom of each bore, and less than 0.00025" out-of-round. He employs a plateau hone for the bores. If you were to look at the finished plateau honed surface under a microscope, you'd see that the peaks of the surface have been knocked off, but tiny channels have been behind. These channels are critical to retain oil for proper ring lubrication -- you don't want your cylinder bores to be polished smooth, else the rings would quickly weld themselves to the bores due to lack of oil. Doh.

1997 Mazda MX-5 Miata: Assembly By Keegan Engineering, Pt. I

Tight squish relies on very stiff rods. Our Mil.Spec billet 4340 H-beam connecting rods are made specifically for long-rod BP engines to withstand much higher loads than stock. Despite their added length relative to stock, the Mil.Spec rods weigh 516 grams. Stock BP rods are about 550 grams, though this number varies from widely from rod to rod. 

Deck height, or the gap between the topmost portion of the piston crown and the surface of the head, will be set to approximately 0.030" for this engine. The smaller this gap, the better for "squish." 

Squish results in high velocity, radially-inward gas motion when the piston reaches top dead center, and is a great way to speed the combustion process and stave off knock. Knock resistance (and control) will be crucial to support our goal for this engine -- to maximize output and powerband with 91 octane.

Next up: short block assembly.

Jason Kavanagh, Engineering Editor

1997 Mazda MX-5 Miata: Assembly By Keegan Engineering, Pt. I

To set up the valve travel measurement, the prepped block first gets a degree wheel installed on the crank.

1997 Mazda MX-5 Miata: Assembly By Keegan Engineering, Pt. I

Then a pre-compressed mockup head gasket of nominal thickness in put in place. Once the valve travel is known, a thinner or thicket head gasket than nominal may be used in the final assembly.

1997 Mazda MX-5 Miata: Assembly By Keegan Engineering, Pt. I

Mike lowers Project Miata's ported head over the head studs.


The dial indicator is fixtured to pick up the exhaust valve stem.


Keegan Engineering - www.keeganengineering.com

Mil.Spec - www.mil-spec.org

Post a Comment

You must be signed in to post a comment.

Research Models