Click on a thumbnail to see the full-size image
Miata Madness: Getting Into The Flow With Mild Port Work
By Dana Johnson
At my shop, Import Machine Service in Framingham, MA, we do a lot of flow testing and head porting, and are continually working to improve induction systems. Our work, however, is limited to aluminum only. Because I do work for many racers, I get many requests to do "performance" enhancements.
Over the years, Mazda Miatas have developed quite a following. These little roadsters are now very popular on racetracks and streets across the country. With classes from Spec Miata (a road racing class with stock, equally prepared cars and strict rules) to CSP Autocross and SCCA (Sports Car Club of America) road racing classes, the venues to run these cars are virtually limitless.
In 1990, the Miata emerged, along with a multi-million dollar ad campaign, as a very desirable, "fun" car for the motoring public. The early cars came with a 1.6L 16-valve twin-cam engine. The 1.6L was bumped up to a 1.8L for the ’94-’97 editions, which were hydraulic bucket-style heads. Then in 1999 Mazda went to a different style head with shim-type buckets and raised intake ports for improved flow.
One of these Miata heads with shim buckets and raised intake ports recently found its way into my shop. Its owner wanted to add performance to the stock Miata head. Normally, these stock classes do not allow for head modifications, but it is my philosophy to do what the customer asks. So off to clean the head I go.
First I strip all the parts off of it and thoroughly clean the whole head and its components. Next, a pre-flow test is needed to establish a baseline from which to compare. In order to be able to test any intake port, a radiused entry is needed to introduce the air into the port in a consistent manner. I had a test plate CNC’d from a 1/2˝ aluminum plate to match the intake gasket, and it fit perfectly because I used the intake gasket as a template for making the plate.
The head was assembled with stock valves, soft test springs and the radiused entry plate was attached. Then I clamped the head to my JKM Custom Flow 400 flow bench. As a rule, I test at .050˝ increments up to .500˝ lift, even if it will never see that much lift. If the flow continues to rise at higher and higher lifts it shows that the head has an efficient port.
I was surprisingly pleased with the results for a stock head flow test. It was tested at 28˝ of H2O and measured 36.7 cfm (cubic feet per minute) at .050˝ lift and 205.2 at .500˝ lift. For the finished product, the stock valves were to be replaced with stock size pieces from SI Industries. So, out of curiosity, I swapped the valves and re-ran the test. I was not surprised to see a 2-5 cfm increase all around due to the backside shape of the stainless steel valve.
Running the stock induction system is pretty straightforward stuff, so doing the prep for the port work was not rocket science. The intake and exhaust surfaces get Dykem applied to the outside to make it easier to see the lines that are about to be scribed. The gasket for both intake and exhaust are used to scribe lines all the way around.
The port work for these stock systems involves matching the transition points and working the port floors and ‘shortside’ radius. The roof of the port is reworked to make it as smooth as possible until it meets the bowl. The bowl should have as smooth a radius as possible without making it too large, blending it to the valve seat. Another area to pay close attention to is the port divider. I thin it and sharpen it quite a bit, and I am very careful when radiusing the top and bottom to blend to the roof and floor of the port.
Doing the exhaust port work is very similar to the intake side. However, be careful that you do not make the exits too large. Make sure you match the exhaust port to the pipe where they meet. Also, keep the floor of the port as high as possible because it is more advantageous for maintaining a high flow velocity. Remove only enough material to get the shape and smooth radius that you desire.
After shaping the ports, the head is set up on my Serdi machine for a multi-angle valve job. I’ll also unshroud and smooth the chambers a little and then resurface the head. When all of these steps are completed it goes back to the flow bench to test the results.
After seeing the results, I’m more than happy with the work that I’ve done. The intake went from 36.7 cfm at .050˝ lift to 51.9 cfm. And at .500˝ lift it went from 205.2 cfm to 242.8 cfm. On the exhaust side, the results went from 35.3 cfm to 53.0 cfm at .050˝ lift, and from 173.7 cfm to 226.1 cfm at .500˝ lift.
In a small engine like this one, low-lift and mid-lift flow is very important for torque and throttle response, and from the results we have seen this has definitely been achieved. If the flow bench is any indication of how big the differences are, it should make a big improvement on this racer’s stopwatch, too.
Dana Johnson owns Import Machine Services in Framingham, MA. He has been rebuilding VW and Porsche engines since the 1960s.