Click on a thumbnail to see the full-size image
OBD II Shop
By Norm Brandes
Speaking From Personal Experience, OBD II Can Be A Headache For Engine Builders
But before you have to pay for your mistakes, learn from the ones I made
The main purpose of previous columns has been to discuss the effects of OBD II on engine rebuilding. But one question many of our readers keep asking is "how real is this stuff?" Does OBD II really make that much difference in the way late model engines should be rebuilt?
Take my word for it, it’s real. OBD II is something every rebuilder has to consider when rebuilding late model engines because of the way tolerances can affect emissions and misfire detection.
There’s a sign in my office that reads, "Experience is what you get when you didn’t get what you wanted." In other words, we learn from our mistakes. What you don’t know can often get you into trouble, so learn from my experience and save yourself some headaches. Over the years, we’ve worked on a lot of different engines in our shop. We’ve done stock engines, performance engines and a lot of cutting edge projects including performance development work for some vehicle manufacturers. We’ve done Viper V10 performance packages as well as GM LS1 engine programs.
A number of years ago, I bought and restored a 1965 Superstock Dodge A990 with a 426 hemi. The car runs consistent 11.20s in the quarter mile. We restored the engine using all original parts, and by paying careful attention to tolerances we ended up with better driveability than stock.
One project we did recently for General Motors was a street performance 1968 Firebird. The project car was called the "Historian." The basic idea was to build a streetable exhibition drag car using a RAM Air III 455 engine. Some Pontiac engineers bet me the car would never meet emissions. I took their challenge and said it would meet 1975 standards. It turned out we were both wrong. With the converters on the engine it was actually able to meet 1990 emission levels – and still run 11.60s in the quarter mile!
Shortly after we finished that car, we did another project for GM. We were asked to build a performance stroker motor for a 1998 Pontiac Trans Am project car called the "Goldrush." We installed a stroker crank in the LS1 engine to bring the displacement up to 382 cubic inches. The engine’s power output increased from a stock 300 hp at the rear wheels on the dyno to 425 hp.
With the converters on, the engine would still pass an I/M 240 emissions test, so there was no sacrifice in emissions or drivability. By carefully matching the heads, camshaft and compression, we were able to create an engine that was very drivable on the street, delivered gobs of torque with no significant increase in emissions.
But, we also learned something from this project that we hadn’t anticipated. To our surprise, the MIL light would come on unexpectedly and log a P0300 "random misfire" code. Yet the engine was running smoothly with no detectable ignition misfire.
The OBD II system on this vehicle monitors changes in the speed of the crankshaft to detect misfire. Anything that causes even the slightest variation in the acceleration of the crankshaft between power strokes can mislead the OBD II system into logging what it thinks is a misfire. As the "misfires" add up for the various cylinders, it eventually reaches a threshold that triggers the MIL light and logs a random misfire code.
On most engines, a random misfire is caused by a lean fuel condition — usually an air leak that allows unmetered air to enter the intake manifold. But on this engine, there were no air leaks and fuel trim was right on the money. So what was causing the misfire code?
As it turned out, we discovered that a stack up of manufacturing tolerances in the stroker crank and rods was creating enough of a variation in piston protrusion height to affect compression. This, in turn, caused a slight enough difference in power, cylinder-to-cylinder, to affect crank speed and mislead the OBD II system into thinking there was a misfire problem occurring.
General Motors says that there should be no more than a .010˝ variation in deck height, otherwise it may affect the misfire detection capabilities of the OBD II system. What I found was that if the piston protrusion height varied more than .008˝ it would cause a problem.
On the next LS1 motor we built, we kept the piston protrusion height to within .0015˝ on every cylinder, and equalized combustion chamber volumes to within half a cubic centimeter. This eliminated the false random misfire code problem.
When you’re building an OBD II compliant motor, you need to check stroke length, rod length and piston height. By mixing and matching parts, you can equalize piston protrusion to minimize the risk of a false misfire code. You can also compensate for variations in piston protrusion height by adjusting the volume of the combustion chambers. One way to do that is to recess the valves slightly in the head to alter the displacement of the combustion chamber.
If you don’t know how sensitive the OBD II system is to variations in combustion chamber volume, piston protrusion and compression, you can end up building problem motors that turn on the OBD II light — and you won’t know why it’s happening.
What we learned on the Goldrush project, we applied to another project car we did for Chevrolet. The goal here was to build a small displacement, high winding engine that was capable of producing 400 hp, yet would be streetable and be OBD II compliant. We decided to build a 302 V8 with LS6 heads and intake manifold based on the heritage of the 1969 Camaro Z28.
In my next column, I’ll detail everything we did to the engine. But basically I opted for a mechanical roller cam to handle the higher rpms, upped the compression and modified the heads to improve low end torque.
When you’re building a street performance engine, you want to keep port velocity high to improve torque at low rpms. Tall narrow ports and intake runners are better than large ones because they keep the airflow velocity high. You also have to pay careful attention to the profile of the intake ports and the shape of the combustion chambers so you get good airflow past the intake valves.
As for the camshaft, I like a high lift, short duration cam with lots of ramp speed to open the valves quickly. You also have to pick the right lobe centers and separation to get good low-end performance and emissions. Advancing the lobe centers helps low rpm torque on the street.
The finished 302 met all of our performance goals, including being OBD II compliant. The motor put out over 400 horsepower and could wind safely to 7,500 rpm.
Yet, it could idle at 850 rpm with plenty of intake vacuum and pull away from a stoplight without hesitating or having to feather the throttle. Best of all, this car can really light up the tires when you punch it!
Norm Brandes owns and operates Westech Automotive, Inc., a machine shop and vehicle repair service business located in Silver Lake, WI. You may e-mail Norm at email@example.com.