333pg333

We recently had something similar setup for our car and have seen some gains on the track from it. Pretty much outlined by what Iridium said. More linear and easier to drive. Equals more comfortable for the driver therefore faster. I think it's pretty much boost by throttle position but don't have all the details. Was only done last minute when we switched intakes and gasket. The engineer who wasn't there on the following abbreviated trackday looked at the data later and confirmed that we were quicker in certain parts of the track. That was on a short twisty layout. Will be interesting to see how it goes on the larger layout.

Wonder if you can switch over to a stock turbo r&p Shawn. Would that help?

User 52121

...subscribing.

One day, when I have more time, I will look into doing something similar. I've had some ideas on this for the last ~4 years or so and just never had the time to implement.

refresh951

Really loving this project. All the parts are ordered. Finished basic block diagram and started on the code. Tachometer module is ready for testing thanks to some code I found here. Planning on testing that tomorrow night.

Going to use a different MAP sensor with a larger range (75 psi) and that also adjusts for local atmospheric pressure:

http://www.digikey.com/product-detai...00DP-ND/608501

Looking at 2 modes, Open and Closed. Then 4 different presets for each mode:

- Street Mild - 18 psi (4th) - gear based
- Street Aggressive - 22 psi (4th) - gear based, rpm based (ramp up to 22 psi after peak torque)
- Drag Race - 24 psi (4th) - gear based, rpm based (ramp up to 24 psi after peak torque)
- Track - 21 psi (4th) - gear based, rpm based (ramp up to 21 psi after peak torque), tps based

Tom M'Guinn

That tachometer code would need to be adjusted to produce the correct rpms on our motors. If you want to calculate rpms I'm sure I have a small sketch for our motors if helpful, but I'd argue that's unnecessary processing for your purpose. For gear identification, you just need the ratio of the speedo to tach sensor pulses -- converting to rpms and mph may make it easier to think about, but will just eat up processing time unless you plan to display it for human consumption... Rpm-based boost is the same -- the processor can be programmed based on the pulse rate without converting to a human-friendly rpm number... I suspect the arduino will have ample extra processing power, so maybe no harm done doing the extra math, but it always seems better to code it as efficiently as possible...

In addition to a handful of canned boost curves, you could also add ***** (pots) to adjust from the cockpit -- maybe one per gear, along with adjustable spool up rate(s) so you can dial it in on the track while you're there... I've always thought about using the now-useless fader dial for that...

FRporscheman

Subscribing.

I work with Arduinos so this is cool to see!

refresh951

Making progress. Still waiting on my solenoid to arrive. Should have the an open loop version running in a few days. Shift light is set to 3K rmp for test purposes.

URG8RB8

Amazingly fast results as always! Very exciting!

refresh951

Going to add logging ability. This will allow me to log speed, rpm, gear, and boost as a function of time. With the logs I will be able to see if the clutch is slipping and exactly where it occurs (thanks Tom for the idea). Also, I can run simulated 1/8 or 1/4 mile runs and know 60', 1000', and trap speeds. I will be able to look at those runs in great detail.

I addition I want to add knock protection to the boost controller...

Tom M'Guinn

Very cool. You're mission creep is getting into the other project I've had on the back burner... I have a half-done safety-box project that monitors knock counts, AFR, boost, engine temp, and a few other things I'm forgetting, and dumps boost if things get ugly (e.g., overboosts, AFR run lean on WOT, knock counts go through roof, etc.). Mine is an all or nothing thing that just opens the wastegate if bad conditions exist, but you could probably dial back boost if the clutch slips. If you have ABS, you could also dial back boost if the rear tires spin faster than the fronts... You gotta love that little Arduino...

refresh951

Interesting about the ABS, I have an 86 so no ABS. I was thinking I can use some math based on weight, boost, and acceleration (rate of change of speed) to determine if tires are slipping. Yes, I am loving the Arduino. Thanks for all the help.

Tom M'Guinn

If you get a pulse from the front ABS sensor and compare it to the rear speedo sensor, you can detect when the wheels are spinning... If no front sensor, you might check out some of the Arduino accelerometer shields...

URG8RB8

You two are amazing, great work!

refresh951

Yes, a bit of mission creep

URG8RB8

Neat and tidy work space, I love it!

mtnman82

Shawn, why the little RTC breakout board - doesn't the data logger board have a RTC on it?