Propeller EFI with spark control and Idle and Boost and more. Running 4500RPM 3

tripptripp Posts: 52
edited 2012-02-16 - 11:04:21 in Propeller 1
This project is not new, it started a while·ago when I and my friend we decided to
make an EFI system.

Initially the system was based on a PIC18F452.
The system·was functional but had some problems that were difficult to find, and·the PIC·had a·very limited RAM.

But when I found the propeller It felt quite obvious to upgrade to it.

The propeller is perfect for an engine management system, with true multitasking, it will be much easier to succeed with
a project·like this.

Here is a picture of the most recent prototype:

4.JPG



Now we have a prototype that works very well, but lacks a few features that we're working on right now.
The system is similar to other aftermarket engine management systems.

The ECU will have support for 4,6,8 cylinder engines to begin with.

The ECU controls both ignition and fuel.
16x16 matrices for Ignition, VE, Lambda. with interpolation.
Measuring water temp, air temp, TPS, MAP,·EGT x 2, Battery voltage, RPM, Lambda, etc..
Manages 60-2 Trigger and HALL Trigger.
Have·a data log function.
Control idle with both the PID controller and ignition timing controller.·
Boost control.
Waste Fire / normal fire.

We also·plan to add some unique features smile.gif

The software is written in C # and works well, but is far from finished.

4.jpg

I have also made simulator engine that really has simplified the development of the system.

simu.jpg

Recently, we've run a motor in a·engine test bed·with very successful results.

1.jpg

Now it seems that all the basic functionality works as intended.

Here is a short video that proves that is works: smile.gif

http://www.youtube.com/watch?v=OGMC5JC5ye8



Post Edited (tripp) : 7/27/2010 10:51:05 PM GMT

Comments

  • smbakersmbaker Posts: 162
    edited 2010-06-16 - 22:32:04
    Now that is cool! Nice job!
  • Timothy D. SwieterTimothy D. Swieter Posts: 1,613
    edited 2010-06-16 - 22:44:43
    I bet it is exciting see all your hardwork and learning experiences paying off!

    I noticed the different fonts on your PCB, what program did you use to do that? I use EAGLE and wish I could change fonts, but off the bat I didn't see how and have always expected the standard font.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Timothy D. Swieter, P.E.
    www.brilldea.com - Prop Blade, LED Painter, RGB LEDs, 3.0" 16:9 LCD Composite video display, eProto for SunSPOT, PropNET, PolkaDOT-51
    www.tdswieter.com
  • BradCBradC Posts: 2,601
    edited 2010-06-17 - 00:45:03
    Is that motor on the test bed a B230?!?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "I mean, if I went around sayin' I was an emperor just because some moistened bint had lobbed a scimitar at me they'd put me away!"
  • pmrobertpmrobert Posts: 579
    edited 2010-06-17 - 00:47:32
    Very, very nice! I'm ecstatic to see that a Prop based ECU is under development!
  • JavalinJavalin Posts: 892
    edited 2010-06-17 - 21:16:16
    Wow - good work. Always been tempted to do a prop ECU - as you say the multitasking is ideal! You look like you have a good test bed there!

    J
  • Cole LoganCole Logan Posts: 196
    edited 2010-06-18 - 03:47:49
    So do you have any plans to sell this or is it just a fun project.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    1986 Chevy EL Camino·· No prop yet

    1984 Suzukie GS1100GK No prop yet
  • tripptripp Posts: 52
    edited 2010-07-27 - 22:47:47
    Hi all.

    Now we made some progress with the Propeller based EFI system.

    Yesterday we ran the old Volvo engine pretty hard. smile.gif

    The ECU is really robust, running over one week without rebooting.

    Everything works as it should so far.

    Now we just need to add some functions like lambdacontrol and·transient fuel.

    The software for the PCB is Labcenters ARES.

    And its a Volvo B230FT.

    Here is a video: http://www.youtube.com/watch?v=MDudc4YUUUc

    VolvoB230Ft.jpg
  • tripptripp Posts: 52
    edited 2010-12-04 - 15:24:42
    More progress:

    More boost :) 275kPa and that resulted in 276Hp and 450Nm :)

    Added and tested festures:
    Wideband lambda closed loop control.
    Transient fuel from TPS and MAP.
    WasteSpark control.
    Warmup fuel.
    Injector opening time (to get correct VE table).
    Support for 6cyl engines.

    Soon i will mount the ECU in 2 cars, one Nissan Skyline RB30 turbo, and one BMW 325 M50 turbo.
    And in this winter i will try to upgrade my old Volvo 940Turbo to a 450Nm 276Hp :)
  • Cole LoganCole Logan Posts: 196
    edited 2010-12-04 - 18:04:31
    Do you have any plans on selling it or in some way making it available for others.
  • markaericmarkaeric Posts: 282
    edited 2010-12-04 - 20:21:20
    Very slick, tripp!

    Suggestion: If you have any free COGs left, it would be cool to add video out (probably composite) to see real time parameters while driving without the need for a laptop. I haven't heard of any other aftermarket EMS with that capabilitiy. That would negate the need for stand alone gauges, and increase your products value factor if you do plan on selling it... Just an idea :)
  • $WMc%$WMc% Posts: 1,884
    edited 2010-12-04 - 21:34:45
    tripp wrote: »
    More progress:

    More boost :) 275kPa and that resulted in 276Hp and 450Nm :)

    Added and tested festures:
    Wideband lambda closed loop control.
    Transient fuel from TPS and MAP.
    WasteSpark control.
    Warmup fuel.
    Injector opening time (to get correct VE table).
    Support for 6cyl engines.

    Soon i will mount the ECU in 2 cars, one Nissan Skyline RB30 turbo, and one BMW 325 M50 turbo.
    And in this winter i will try to upgrade my old Volvo 940Turbo to a 450Nm 276Hp :)
    '
    Could You elaborate on "Wideband lambda closed loop control" and "Injector opening time"????
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2010-12-04 - 23:50:32
    I realize that this is intended for normal gasoline and automotive use. But I am wondering if an automobile engine adapted to run 'bio-methane' for stationary use at fixed RPM would benefit from the Propeller EFI.

    I've a friend in Thailand that is powering ventillation/cooling fans on a pig farm in Thailand from a Nissan engine that is less than 2 liters. Pig waste makes the 'bio-methane".

    "Bio-methane" is low pressure (about 2 psi supply) and a mix of up to 50% CO2 and the carb is replaced with a rather simple device. So it is generally a bit more difficult to manage ignition. Spark advance can be 40-50 degrees. In stationary operation at fixed RPM, it is often suggested that mechanical and vacumn advance mechanisms should be disconnected. But I've doubts that such is wise.
  • Bob Lawrence (VE1RLL)Bob Lawrence (VE1RLL) Posts: 1,578
    edited 2010-12-05 - 08:11:49
    @ tripp


    Wow! that's a beautiful design. I used to be a mechanic in the Canadian military and remember back when the first vehicle electronic systems came out. It brings back great memories. Good luck with the rest of the project and I look forward to the updates.
  • AJMAJM Posts: 171
    edited 2010-12-05 - 14:38:50
    Tripp,

    I love this project! 40 PSI? I take it that engine must be a diesel.

    If you need any testing on V8 engines, please let me know.
  • BradCBradC Posts: 2,601
    edited 2010-12-06 - 21:38:43
    AJM wrote: »
    Tripp,

    I love this project! 40 PSI? I take it that engine must be a diesel.

    The B230FT is a 2.3L 4 Cyl petrol engine. SOHC 8 valve alloy head. Volvo's mainstay from about 1975 through to the mid 90's.
  • Paul MPaul M Posts: 95
    edited 2010-12-07 - 10:37:02
    Congratulations on a fantastic project!!

    I vividly remember the early Volvos (especially the Amazons) as a teenager on holiday in Ostergotland.

    Keep it up!

    Paul
  • AJMAJM Posts: 171
    edited 2010-12-07 - 11:23:19
    BradC wrote: »
    The B230FT is a 2.3L 4 Cyl petrol engine. SOHC 8 valve alloy head. Volvo's mainstay from about 1975 through to the mid 90's.

    If this is a gasoline engine, 275kPa (39.9 psi) is very high. I have not seen many gasoline engines running such a high boost, although it is possible. It makes the project even more impressive.
  • tripptripp Posts: 52
    edited 2010-12-07 - 12:54:42
    Thanks for all comments!

    Regarding the boostpressure its 275kPa absolute, that is 25 PSI gauge.

    markaeric: nice idea, but for now we are out of cogs.

    Cole Logan: yes i hope it can be sold :)
  • FreeEMSFreeEMS Posts: 1
    edited 2012-01-31 - 15:48:05
    Hello Tripp! Nice looking design! May I ask what the connector is? And which supplier you got it from? How many pins is it? I need some connectors for a project coming up soon, and liked the look of it, so thought I'd ask. I had a google for the number, or what I think the number is, on the top, and failed. Thanks in advance, Fred.
  • DavidZemonDavidZemon Posts: 2,825
    edited 2012-02-16 - 08:37:46
    tripp - hope you're still around! haven't seen a post from you in over a year now.

    Anyway, I'm looking to do my own EFI system and currently looking at different analog-digital converters. Found what looks like the perfect chip, but want to make sure 8-bits is enough precision. It seems like it would be, but what do you think? What did you use for your sensors?
  • CircuitsoftCircuitsoft Posts: 1,020
    edited 2012-02-16 - 08:49:18
    SwimDude, what is the fuel flow of the largest injectors you're planning to use? How much fuel flow do you need at idle? Divide, multiply by 20 or so rounding up to a power of 2, and you'll have roughly the range you need.
  • DavidZemonDavidZemon Posts: 2,825
    edited 2012-02-16 - 10:06:29
    I haven't gotten near that far yet.
    Also, what do the fuel injectors have to do with the ADC? As I understand it (which could be wrong of course) the injectors are a digital output that will be set or cleared for X clock cycles right? I'm trying to figure out what kind of precision I'll need from things like: oil temp, MAF, O2 sensor, TPS, etc
  • CircuitsoftCircuitsoft Posts: 1,020
    edited 2012-02-16 - 11:04:21
    Oil temp, 8 bits is probably fine. TPS, 8 bits is also probably plenty, but you might use a little more there. O2 sensor is a long subject. MAF/MAP are directly proportional to fuel injector time, so that's what you want to match up.

    O2 could actually be done with a 1-bit input with the right cutoff voltage. I suspect even a wideband could be driven from Sigma-Delta mode, but that leaves heaters to worry about, which can be rather difficult to deal with.
Sign In or Register to comment.