Engine Management Forum Use this area for discussing Haltech, Wolf 3D, Power FC, AEM and any other aftermarket ECU upgrades. Help/Questions/Tuning

ECU Design

Thread Tools
 
Search this Thread
 
Old 11-21-06, 10:20 AM
  #1  
Full Member

Thread Starter
 
rx7__newbie's Avatar
 
Join Date: Nov 2006
Location: USA
Posts: 125
Likes: 0
Received 0 Likes on 0 Posts
Talking ECU Design

I am an Electronic Engineer with embedded software design experience and hating the fact that we are having to depend on companies from Australia and new zealand to support our hobbies.

I want to design an ecu that would work on all the 2nd and 3rd rx7s and possibly the rx8's. I would like to know what all of you guys like about haltech and microtech and the PFC and possibly incorporate all the likes and remove the dislikes in all these ECU and make a product that is American made and American's supporting Americans. How would like to help me with this project and the start of my company to help produce something special for you guys.

Thanks and God Bless.
Old 11-21-06, 11:41 AM
  #2  
Lives on the Forum

 
rynberg's Avatar
 
Join Date: Dec 2001
Location: San Lorenzo, California
Posts: 14,716
Likes: 0
Received 8 Likes on 8 Posts
Good luck with the RX-8...it's an encrypted 32-bit design that controls ALL of the aspects of the car....big companies have had zero luck with a standalone.....

Build a PFC with the benefits of the Datalogit built-in, and no problems with the A/C, sell it for the same price and you might have something.
Old 11-21-06, 12:21 PM
  #3  
~17 MPG

iTrader: (2)
 
scotty305's Avatar
 
Join Date: Jan 2004
Location: Bend, OR
Posts: 3,293
Received 226 Likes on 152 Posts
Here are a few of the features I would want:

-Plug-and-play installation, using the OEM wiring harness (similar to the Apex'i PFC)

-Closed-loop operation, checking 0-5V wide-band O2 output, and adjusting fuel trim to acheive desired AFR's (I think certain versions of Megasquirt or Motec have this)

-Ability to use different MAP sensors, such as 3-bar sensors for people running increased boost.

-Closed-loop wastegate control that works better than the PFC

-Ignition retard and/or cut during fault conditions, with variable thresholds: knock sensor, IAT, ECT (coolant temp), AFR, boost level.


-I'm sure people will want it to work with the 20B, although I don't personally need this.


Personally, I was planning to implement a Megasquirt to work with the FD, for starters. There seems to be a large community of people writing code.


-s-
Old 11-21-06, 01:33 PM
  #4  
Full Member

 
wardaj's Avatar
 
Join Date: Jan 2002
Location: Oshkosh, WI
Posts: 54
Likes: 0
Received 0 Likes on 0 Posts
If you decide to go ahead with this PM me. I might be interested in sharing some knowledge. I work for a company called MotoTron and I write software for engine managment systems. I am currently testing some software on my FD. We have hardware allready designed for other applications so I didn't have to mess with that.


When I was in school I did the FSAE thing and we built an engine managment system from scratch. It was pretty intense. Anyway, let me know. I can say now that the megasquirt would be a good starting point as you would allready have the time domain->angle domain software and could just write the application code from that, plus hardware.
Old 11-22-06, 05:07 PM
  #5  
~17 MPG

iTrader: (2)
 
scotty305's Avatar
 
Join Date: Jan 2004
Location: Bend, OR
Posts: 3,293
Received 226 Likes on 152 Posts
wardaj, which FSAE team were you on? How did the custom ECU turn out?

-s-
Old 11-23-06, 08:44 AM
  #6  
Full Member

 
wardaj's Avatar
 
Join Date: Jan 2002
Location: Oshkosh, WI
Posts: 54
Likes: 0
Received 0 Likes on 0 Posts
Oklahoma State. It was our best year, 10th overall. Doesn't sound good until one consideres our engineering dept. is into aerospace and pretty much doesn't care about FSAE.

Our ECU turned out OK. Nothing to fancy. Just the essentials to run an F4i motor. It was a modular design and used CAN to communicate between modules. We had a spark control, fuel control, CF card datalogger, telemetry, and dash. The schematic was incorrect on the ignition part and the bad trace was on an inner board layer so it rendered our CANbus useless. This meant we had to reburn the processor to load new timing maps and we couln't get any data about the engine state( temps and such). We did an AFR tune so we just ran the stock timing map anyway since we were using octane fuel. And we used T0-220 package IGBTs for the injector and coil driver circuits. They kept shearing off from vibration. Other than these issues it was running very well. I think we were in the top 5 in NA power that year. I redesigned it all to fix the mechanical problems and moved it all to one processor but never had a chance to test it. Most of our team graduated and I had to take the lead powertrain position.
Old 11-30-06, 12:06 PM
  #7  
~17 MPG

iTrader: (2)
 
scotty305's Avatar
 
Join Date: Jan 2004
Location: Bend, OR
Posts: 3,293
Received 226 Likes on 152 Posts
wardaj, sounds like a really fun project. On my FSAE team, they're afraid to let the electronics guys control of the engine; I've worked on the pneumatic shifter controls and dashboard instrumentation. How many students worked on the custom ECU project, and how much time did it take?


rx7__newbie , you might be interested to read this: https://www.rx7club.com/3rd-generation-specific-1993-2002-16/apexi-power-fcs-coming-end-600946/


-s-
Old 11-30-06, 07:55 PM
  #8  
Full Member

 
wardaj's Avatar
 
Join Date: Jan 2002
Location: Oshkosh, WI
Posts: 54
Likes: 0
Received 0 Likes on 0 Posts
There were three of us. I have never been more stressed in my entire life! It was cool to show up with one of the few truly custom boxes.
Related Topics
Thread
Thread Starter
Forum
Replies
Last Post



Quick Reply: ECU Design



All times are GMT -5. The time now is 10:11 PM.