AIM Lambda Install

John Denman · Posted: Mon Feb 02, 2009 9:11 pm Post subject:

That installation is more then clean, its a work of art and needs to be in a museum. Cool

I was wonedring how they held up to lead, thanks for the answer Joe.

Now if only AIM can come up with a Detonation Sensor. MoTec has one, I wonder what it would take to fit it to a Mychron.
_________________
John Denman
Producer for RTMP
http://www.kartweb.com

Daryl Choby · Joined: 08 Dec 2006 Posts: 221

Chris Livengood · Posted: Tue Feb 03, 2009 11:23 am Post subject:

We mounted ours directly in the header trying to avoid contamination. I think my off throttle data was more reliable than what Joe wrote of.

Though since my PC crashed (imagine that) I don't have any data to show.

My setup was a Bosch sensor sent directly the PI system and also to a small LED read out.
_________________
http://www.Chrislivengood.net
http://www.Work-Racing.com
http://www.OurZeal.com
http://www.Karting101.com
"Auto racing, helping white guys get laid since 1887!!!"

Jordon Musser · Posted: Wed Feb 04, 2009 1:47 pm Post subject:

FYI,
I wrote a program for predicting gear ratios.. and for that I needed to know when I was off/on throttle also. so I looped through the long/lat acceleration data and developed an algorithm that guessed while I was at WOT or not.

even better, just add a throttle position pot, and whenever it is not at 100% have the data ignore the Lambda sensor..

Charles Kaneb · Posted: Wed Feb 04, 2009 2:15 pm Post subject:

Even better than that, use a switch at the end of the throttle travel to do the same thing.
_________________
"If you're still in full control, you're not going fast enough" - Fred Frame

Jordon Musser · Posted: Wed Feb 04, 2009 2:48 pm Post subject:

Yea, but if you are going to add a channel you might as well have something that will help in other areas too.

joseph hollinger · Posted: Wed Feb 04, 2009 3:00 pm Post subject:

I don't think its really necessary. You're really only interested in the sections of the Lambda curve where the corresponding RPM trace is increasing. Everything else is trivially rejected. If you wanted to be super paranoid, you could add some redundancy by looking at the EGT trace as well. Falling EGT means that the Lambda data is invalid.

I think the data is already there. If you could decipher .drk files (a process I'm working on) it would be pretty easy to get the information you need to optimize Lambda.

John Mulvihill · Posted: Thu Feb 05, 2009 4:05 pm Post subject:

Hi Joe,

Do you have a 'target number' on the Lambda that you shoot for?

Thanks,
John

Jordon Musser · Posted: Thu Feb 05, 2009 4:09 pm Post subject:

Joe- you can already export to excel and do that... just eliminate non acceleration data (with the long accel, or calc derivitive off time and RPM)

joseph hollinger · Posted: Thu Feb 05, 2009 4:21 pm Post subject:

joseph hollinger · Posted: Thu Feb 05, 2009 4:22 pm Post subject:

Jordon Musser · Posted: Thu Feb 05, 2009 4:32 pm Post subject:

I love excel Wink

But you can read it straight into matlab/mathcad etc if you prefer something more powerful.

I imported it straight into a VB program I wrote that just parsed the file and then I could work my magic on it.

joseph hollinger · Posted: Thu Feb 05, 2009 4:35 pm Post subject:

Jordon Musser · Posted: Thu Feb 05, 2009 5:09 pm Post subject:

No student discounts here!

I wish. I was getting solidworks with a student discount several years after I graduated through the old FSAE team.. but i didnt this year.

like I said.. I actually do most of my file manipulation in visual C++ or visual basic. (express, which is free)

I actually am not a huge mathcad fan.

BTW, I am an engineer by trade. Now that I am farther up the totem pole, I have complete access to all the engineering tools I need on my workstation laptop that I bring home on a daily basis.

Cesar Rull · Posted: Sun Oct 21, 2012 1:28 am Post subject:

can someone post a picture of where they installed their wideband sensor on their shifter's pipe?