M54. Let the foolishness begin.

[matt]

Nevermind, I'll just reply to your email.

[2002maniac]

What size are the M54 injectors? Maybe you could move one step up there?

[hoveringuy]

From what I've found they are 22lb, 235cc injectors, but I don't think it's an injector issue. It's possible that I have the fuel pressure too low, the e46 is supposed to be 3.5bar and I have it at a tad over 3.0 bar to match OBD1 specs. I don't know if the returnless fuel rail requires more pressure or the e46 injectors??

Hmmm... that's a good question. Do I need to have 3.5 bar for those injector's rated capacity or are they all measured at the same pressure???

I will reconnect my pressure gauge tomorrow, check the vacuum and probably boost the pressure by 5 lbs.

BTW, I installed the final version of my Hall to VR conversion circuit today. Works titties!

[matt]

OBDI is 3.5 bar.

Do you have a vacuum reference connected to your FPR?

[hoveringuy]

OBDI is 3.5 bar.

Do you have a vacuum reference connected to your FPR?

I have it set to 52psi which is 3.5 bar. (I thought it was 3.0bar) I have a vacuum reference line and I'll check it again tonight.

[5mall5nail5]

Steve loving your circuit - so simple and concise. Good work on that!

[hoveringuy]

VANOS controller

I've been working on my standalone VANOS controller and have made some decent progress over the last several weeks.

My first milestone was getting the board to recognize TDC off the 60-2 wheel and blink an LED at TDC. That works great.

Then I got the LCD to display the RPM. Check.

Then I added the cam input and added some code to display the cam angle based on how many teeth have passed on the wheel. In doing that, I got some insight into why the cam signal wheel has 180 degree shapes on it. Since the cam spins at 1/2 speed of the crank, the cam sends a change in the signal once per crank revolution. Convenient!

Finally, I hooked the whole thing up to my car today and the car immediately stalled...

Connected a 1k Ohm resistor to keep the crank signal from being attenuated and now I can read the RPM and cam angle in real time off the (running!) engine.

However, I'm reading 240 degrees for the cam. It's jittering back and forth a few degrees so maybe they're not reading exactly 180 apart. I'll need to average between the two...

If someone has a motor on a stand I'd sure like to know when the cam wheel passes the sensor compared to the crank wheel.

Otherwise I'll configure the O-scope to read the Crank and Cam inputs simultaneously and I can figure it out that way.

[hoveringuy]

Love the O-silly scope!

I changed my software to average two cam signals and got the jitter out of the reading. It's rock steady at 250 degrees now.

That reading still seems kinda odd so I hooked the O-scope up to crank and cam signals simultaneously. The first picture clearly shows that the cam changes its state at right around 250 degrees of crank rotation. The cam wheel goes from the big part to small.

My LCD readout is dead on. Awesome!

The second picture shows why I was getting jitter when I read the pulses individually. The wheel isn't exactly in 180 sections. It's more like 174/184. At least that's how it's read. You can clearly see that one pulse is wider than the other. The LCD was constantly changing between 256 and 244 so that's why I couldn't read it.

Averaging is key otherwise the rapid changes would throw the Derivative part of my PID controller into chaos.

Also, VANOS retarded it reads 250. Advanced it reads 208. Do the math and you get 21 degrees of advance at the cam. Perfect!

[CorvallisBMW]

:bow:

You, sir, are an inspiration!!!

[SpecM]

Yay for you! that's so cool...

[Jean]

In no time you'll develop your own ECU ;)

[bender]

anymore details on your vanos controller? sounds really interesting. dual vanos right? so i'm guessing pwm circuit with some lookup tables?

very impressive project!

[hoveringuy]

anymore details on your vanos controller? sounds really interesting. dual vanos right? so i'm guessing pwm circuit with some lookup tables?

very impressive project!

Correct. I'm using a Microchip dsPIC30F4011 which was designed for motor control. It runs at 40Mips and has 1k of EEPROM as well as a raft of cool built-in features.

Now that I can read the cam position, I will need a lookup table to say where I want the cam position. A PID controller will drive the onboard PWM generator to control the cam.

I will definately be able to display command position and actual position of both cams, I'd love to be able to change the set points on the fly.

Think of this.... what if the e36 VANOS could be modified to expand it's range of motion to more than 12.5 degrees?

One thing I will need to take into consideration is that I can't completely usurp cam control from the ECU. It is still switching tables based on whether or not whether or not it has VANOS engaged. So, I need to give the ECU a "vote" or force it to use a single table somehow.

[e30polak]

Correct. I'm using a Microchip dsPIC30F4011 which was designed for motor control. It runs at 40Mips and has 1k of EEPROM as well as a raft of cool built-in features.

Now that I can read the cam position, I will need a lookup table to say where I want the cam position. A PID controller will drive the onboard PWM generator to control the cam.

I will definately be able to display command position and actual position of both cams, I'd love to be able to change the set points on the fly.

Think of this.... what if the e36 VANOS could be modified to expand it's range of motion to more than 12.5 degrees?

One thing I will need to take into consideration is that I can't completely usurp cam control from the ECU. It is still switching tables based on whether or not whether or not it has VANOS engaged. So, I need to give the ECU a "vote" or force it to use a single table somehow.

Dude you are a fucking genious! More info on this Vanos controller!

[Drew325is]

More awesome work Steve!!! Hmmm...next Hoveringguy project a cam position gauge???