Speeduino is now on Github Sponsors (Rather than Patreon): https://github.com/sponsors/noisymime
Help with building your Speeduino, installing it, getting it to run etc.
User avatar
By gpineau
#46783
Got another Kawasaki Concours that is ready for EFI retrofit. Learned a lot from the first one so this one should not be as difficult.
The previous one is running well but I still have the issue of going lean and stalling out when going quickly to WOT. Changed fueling to AlphaN but have not had a chance to tune since cold weather has come. So I will tackle it again next spring.

Meanwhile this winter I have cleaned up the garage and I am ready for another EFI project.
Today I started by making what I consider to be the most critical part of the entire speeduino implementation. THE FRIGGING CRANK POSITION SENSOR. I salvage the mounting bracket from an old Kawasaki VR sensor and mounted and potted it with my favorite HALL SENSOR. ATS627LSG. They are idiot proof and I are simple to implement. This took about half a day to finish. The original took about 3 weeks to settle on what was best for this application.

https://youtu.be/ijxOpge07cw
User avatar
By gpineau
#46803
On this project I will be using the NO2C board that was over from my first experiment in EFI on my Honda magna. I'm using the same crank 36-1 wheel and hall sensor as I used last time.

The only new thing about this vs the last project is that I will be using a different set of throttle bodies and I will attempt to control the secondary throttle plates with speeduino. I have the throttle bodies off of a Kawasaki Z1400 which is larger than needed but the spacing is closer to what I need and it will require less modification to make them fit. I'm going to need some guidance on how to implement control of the secondary plates into the firmware.

I actually have two choices for the throttle bodies. The upper one requires being separated in the center and spread by 13mm. This makes it difficult to use the secondary throttle plates because they are on the same shaft. (would require some invention) The second choice is the TB on the bottom of the photo. The spacing is very close to that of the head and I could probably connect them using a flexible hose/coupling.
Attachments
20201130_110105.jpg
20201130_110105.jpg (3.22 MiB) Viewed 1922 times
User avatar
By gpineau
#47091
This is a one-off development just for my motorcycle which will have a throttle body with secondary throttle plates. The reason I need them, or think I need them is because of my other motorcycle AFR going too lean if opening the throttle too quickly. So in theory by delaying the airflow into the throttle body until the RPM can catch up with the TPS it may solve my problem. These secondary plates also crack open the primary plates when they are fully opened. This provides air at idle and a fast idle during warm up.

Currently working on firmware routines that control and calibrate the Secondary throttle plates.
When code wakes up and starts the initialize routine, it will calibrate the secondary plates by finding the full open and full closed positions. It is repeated 4 times then the values are saved.

In the main loop, a routine is called periodically to service the Secondary Throttle plates (STP). This routine uses RPM and TPS to look up in a table the target position of where the secondary throttle plates should be. It then compares this target position to the actual position and initiates a seek in the direction of the target position. Upon initiating the seek the routine will exit. This is repeated until the position reaches the target. At the moment I have it slowed down so I can see what is going on during debug.

A while until I am done but I think this is going to work.
https://youtu.be/ia2SJ3GiNH4
By dazq
#47094
Looking good so far :-)

As a thought ..
If you are using a second MCU to control the flaps (as per your tests you did) then you could have it read the data for rpm ,TPS,rpm.dot and TPS.dot from serial3.
This would mean you wouldn't need to make any changes to the speedy firmware itself ..
User avatar
By gpineau
#47097
dazq,
My prototype was on an arduino UNO for debug, learning and practice.
I thought about using a stand alone arduino to control the plates but I don't know how responsive it would be using the serial port to get the information I need. I need TPS, RPM TEMPERATURE, and Warm Up enrichment to do a complete job.

I like being able to make changes and tweak things using the Tuner-studio interface.
Also I am cheating on the construction of the table. Since I wont be using fueltable2 I am using tuner-studio to modify it so I can easily make changes while tuning.

Sure all that could be replicated on a dedicated ECU but I like the shortcut and not having to invent something for this one off project. Maybe in time I will get around to it.

For now I have created a couple subroutines that are called from the mainline code and so far not having to modify Speeduino code other than a couple calls to my code.
By LPG2CV
#47099
Just a thought, but perhaps use the 2nd ignition table, in the 202012 FW. Then, if necessary, you can still also blend fuel tables. One using TPS and the other MAP if necessary. That may help with the stumble off of idle.

However, as you have increased the size of the throttle bodies, adding extra fuel to cure the lean out, it may make pulling away a little scary with all that power. So then the dampener can come into effect .

If you could cure the lean out by blending fuel tables, then, by using the switching feature in the 2nd ignition table, you could also have modes of dampening. :)
User avatar
By gpineau
#47105
Good Idea. I will try the second spark table instead of the second fuel table.

I had to put a filter on the analog A6. There was lots of noise spikes coming from the sensor of the secondary plates. It will sit there and chatter.

I removed the delays and placed a single call to my routine in the main loop();
Very responsive now. Follows the changes in RPM/ TPS very quickly. I'd say about a quarter second behind .
By LPG2CV
#47106
What would be good, is if you could eventually, add the 2nd throttle plate value to the log data.

Edit

Though it may actually come out as the 2nd table advance or fuel value if thats logged
By JHolland
#47117
gpineau wrote:
Tue Dec 15, 2020 6:12 am
Currently working on firmware routines that control and calibrate the Secondary throttle plates.
When code wakes up and starts the initialize routine, it will calibrate the secondary plates by finding the full open and full closed positions. It is repeated 4 times then the values are saved.

In the main loop, a routine is called periodically to service the Secondary Throttle plates (STP). This routine uses RPM and TPS to look up in a table the target position of where the secondary throttle plates should be. It then compares this target position to the actual position and initiates a seek in the direction of the target position. Upon initiating the seek the routine will exit. This is repeated until the position reaches the target. At the moment I have it slowed down so I can see what is going on during debug.

A while until I am done but I think this is going to work.
Have you done any basic characterisation? is this a DC-motor driven assembly?
  • 1
  • 2
  • 3
  • 4
  • 5
  • 7

Is it VR or Hall? How do you know it's not workin[…]

ULN2803 output random cutout

I wouldn't be surprised if this issue is fixed n[…]

Hello friends. I assembled my SCG-ECU 1.0 board. I[…]

So i ordered 5pcs of bl49sp to test on. I was abou[…]

Still can't find what you're looking for?