Hello people. I too have been reading this tread for about a year and just now decided to get my butt in gear.
Though your circuits work fine, the engineer in me chuckles and cringes at the same time. Just a technicality, but the opposing diodes are called a "clip" not a "clamp". Personally I used a Zener diode in reverse bias with an optoisolator because it is adjustable and gives a much cleaner signal than a regular diode in forward bias. Also, when your diodes fail --or when one of those wires shorts out -- you'll be getting the full 50 volt spikes into your laptop. Or your injectors might be shorted to ground when you're trying to impress the girl in the convertable. So be careful.
I am impressed with the Java program, too. I don't know anything about the language, but this project will give me a good excuse to learn it. Some crafty iPhone or G-phone owner could turn this into an App, then have a bluetooth module send the two signals as auido to the phone wirelessly! Now that would be cool.
And a little advice on finding the Vss: it's got to be somewhere on the back of your speedometer. Probably not feeding into the ECU, though, as is the case on my '91 Corolla.
I was worried since the only diodes I could find at Rad Shack said 50v, but they clamp it to .7V.
You probably should be worried about that 50V because one of the diodes is reverse-biased and injectors create huge spikes in excess of 50V. Usually a diode is rated by its maximum reverse voltage before breakdown.
I'll draw a schematic sometime, but it is just as simple as the setup you have already: just replace the two diodes with one zener reverse biased (backwards) then use a few resisters to divide the voltage to whatever you want. I guess you don't really need a cleaner signal, though, as long as there is only one rising and one falling edge per injector cycle.
As for 50v into a laptop, I guess I don't know if it would be bad or not. I'm going to try to avoid it because I'm a big pussy.
I'm going to try wiring this thing up on someone elses laptop tonight ... wish me luck! I'll take pictures of the mess I make.
So I got it all hooked up last night. I recind what I said about the Vss not being on the ECU; it is there on my toyota and it made things really easy to wire up.
I got really good signals into both left and right channels. There is a "drive to work" image of the signals attached below. Injectors on top (using the zener) and Vss in the botom of the image (using the opposing diodes the OP used).
What I could not figure out were the "injThreshold" and "vssThreshold". What is meant by "value above the noise"? And why are some of the values negative? If you look at my image from CoolEdit, none of the samples are negative on an 8-bit recording.
The gage said I was always using .0055 gallons of gas at every instant, whether the car was running or not. The miles were working, I think, but they were not quite right; I'll have to mess with the fudge factor.
Hello, EvilToothpaste! Nice to hear I am not the only one left researching this project I am actually using this idea as my course project at university (big thanks to skewbe for the idea/java realisation). The difference is that I am coding this in C++ (I really don't like java.. no reason, I just plain don't like it), currently on linux, but if everything works as planned, it will be easy to port to windows or a mac.
EvilToothpaste, seems to me that you know electronics (which I don't). Would be really nice if you drew a schematic for your diode/resistor setup Also, since nobody answered my previous question, would you be so kind as to share your recorded sound file with me? Even better if you could also provide the information (like how much fuel you spent on that trip and what distance you covered), so I could work on my software before I hook a laptop to my car. My laptop doesn't have a line-in, and neither does my mother's. So I need to buy an external sound card, but, well, I am broke right now