Danke Sehr, Herr Doktor Ohm

Danke Sehr, Herr Doktor Ohm!

www.unsage.com

My 3rd race of 2004 had been at Pacific Raceways at the end of May. I hadn’t placed well. It was clear I needed to find ways to make the car faster. There weren't a lot of obvious options left. However, a project that I had considered for a very long time was to modify my engine management system to increase the RPM-limit. This was a big limitation of my car – I often found myself running out of RPMs mid-corner. Bouncing off the engine rev limit isn't great for the mechanicals, as it works by reducing / removing gasoline flow to the engine. This causes it to combust very lean, i.e. hot. More importantly, suddenly being robbed of go-juice mid-corner can significantly slow lap times (duh). Before I had begun racing, I had queried the race steward at the time whether this modification would be Production class legal, and received an affirmative response. The steward told me that I wanted to do was the same as modifying the air / fuel mixture on a modern, CPU controlled car at high RPMs.

My engine management system was Bosch L-Jetronic. This is a mid-60s designed analog circuit system. The circuit board has small capacitors, resistors etc… which control the fuel injection. There is no software, firmware or CPU anywhere on the board. In theory, it's a very simple system. In practice, it's very complicated and difficult to troubleshoot, much less modify. Cars in my class with pure mechanical systems, such as carburetors, are able to tune their cars by modifying the carburetor. Cars in my class with modern fuel injection can simply install a new ROM chip that provides new fuel mapping or a greater redline – whatever is necessary.

Although I was only intending on racing one more year with this engine and drive train, before upgrading to the more modern Bosch Motronic controlled 6-cylinder motor, I decided to dive in. My previous race had really rubbed in the fact that I needed more speed. If I was going to remain competitive for the championship, I had to find a few more tenths per lap. A higher redline meant more flexibility in what gear I chose. I’d be able to choose a lower gear for some corners. I was in the lead for G-P class, but I wasn't going to remain there if I couldn't keep up.

First, I did some online research. I found a very blurry picture of a two-part PCB circuit board labeled as Bosch L-Jetronic. The minimal accompanying documentation indicated that one of the resistors controlled the fuel injector signal responsible for the RPM limit. Theoretically, shorting that resistor would make the RPM limit go away.

I chatted Craig McLuckie, a co-worker who had helped me with my freeze plug issue two years previously by pointing out that the co-efficient of thermal expansion of brass is much higher than iron, thereby saving me enormous frustration. Craig is a multi-talented fellow with a degree in Electrical Engineering. He knows this stuff inside and out. I explained what I was doing.

"You don't want to short the resistor," he told me. "That's not a good idea – it can fry something downstream.”

“OK, so what do I do?”

“You can just wire a second resistor in parallel to decrease the circuit resistance.”

I think I said something sharp, like, "Whazzahuh? English please.”

Craig gave me two key pieces of information. First, how to read the little color bands around the resistors. This would tell me how much resistance the circuit had. Second, he explained the equation Rtot = R1xR2 / R1+R2. This essentially says that the Total Resistance (Rtot) decreases as you pile resistors on the circuit. Electricity does not take the path of least resistance. He suggested calculating out the resistance of the circuit in question and adding a resistor that was a few times greater. This would only lower the resistance slightly – but should be enough to tell whether it had worked or not without burning up anything important on the circuit.

I stopped by Radio Shack and picked up some copper alligator clips and a big pack of resistors.

This was going to be easy! This was awesome!

I took the ECU out of the car, and with some vigorous bending and prying, I was able to remove in the circuit board from the aluminum case. Immediately, I saw a problem. There was only one board, not two as my research had indicated. The circuit board that I had looked nothing like the picture and it was crowded with resistors. It looked like there were about 400 total on the board. I had absolutely no idea where to start. The numbered resistor in my picture corresponded with two open holes on my fuel injection board. Not useful.

This was hard! This sucked!

I offered to buy Craig a cup of coffee if he could spare the time to look over the circuit board with me. He peered at it suspiciously. "Wow," he said, "that's pretty complicated." Not a good start. "Uhhhm… this probably won't be easy." Thanks, but… yeah. "Well…" Craig pointed out that a likely candidate was going to be one of the 50-odd raised resistors on the circuit board. The raised resistors are designed to be easily swapped in and out during the manufacturing process. There was also a good chance that it was a resistor of the same magnitude as the one in the picture, i.e. about 100k ohms.

Of course, this was also assuming that the rev-limit was still controlled by a resistor! It was likely that, along with the new circuit board, Bosch had changed the way in which the rev limit was controlled. They probably did it just to piss me off and waste my time. Teutonic sadists.

That evening, I tried my first raised resistor. First, I calculated the resistance via the color bands. In my Radio Shack multipack, I found a resistor that was about 2 times as large, as this would only marginally reduce the circuit's resistance, and hopefully not fry anything. Second, I used the alligator clips to wire it in-line. Third, I plugged the ECU back into the car and rev'd it up. It hit 6200 RPMs and stopped, as always. I repeated the process twice more. I had now tried all 3 of the raised resistors that were of the same magnitude as the one labeled "RPMs" in the crappy, blurry photo I had. So much for easy possibilities.

I put the project down until the weekend. That weekend, I decided to try a few more resistors. The first circuit I tried was a very odd one. It already had two resistors in parallel. I calculated the resistance at 172k ohms, threw a resistor on the circuit, reassembled and rev'd the motor – right past 6200 rpm. Hot diggity! That wasn't so hard. I immediately wondered whether I could find a resistor that would allow me to boost my engine redline, but keep the rev limit. That was optimum, as it would keep me from accidentally overreving my motor. The second resistor I tried was 1M ohms. It allowed me to rev to about 6400 RPMs. This proved that my theory was possible. The next one I tried was 470k ohms. When I rev'd the motor it stopped at 7000 RPMs – exactly what I was aiming for.

I whipped out a soldering iron, soldered the new resistor into place, reassembled the ECU case and went to start the car… and nothing. Chugachuga – nada. What the heck? Disassembled, the car would start fine. With the case assembled, it wouldn't start. With some troubleshooting, I identified that the problem was having the ECU grounded. For some reason, with the ECU grounded now, the car wouldn't start. If I grounded the ECU while the car was running, it continued to run just fine. Weird. Since the car was a race car, I just degrounded the ECU on the premise that it only had to last the rest of the season.

My fourth race of the year, back at Portland International Raceway, was coming up soon. I couldn't wait for the chance to see whether my times would be improved by the new modification.