Fuel Injector Retrofit Project

Description of your first forum.
admin
Site Admin
Posts: 44
Joined: Mon Jul 02, 2018 1:02 am

Re: Fuel Injector Retrofit Project

Post by admin » Thu Aug 09, 2018 7:44 pm

In the back of my mind is how this translates to Kawasaki C10. 4 cylinders, wasted spark same as the Magna. There is the larger displacement, and horse power which probably means larger injectors . Need to do all that math again but is should not be too different.

Food for thought.....The real challenge is going to be the plumbing. Plenum, throttle body, injector location, Where what had how? Could just use the current air box and add another hole for the throttle body and plug all the other input ports. Going to take some thinking here.

admin
Site Admin
Posts: 44
Joined: Mon Jul 02, 2018 1:02 am

Re: Fuel Injector Retrofit Project

Post by admin » Fri Aug 10, 2018 9:27 pm

Playing with the ecu on the bench with another Arduino hooked up simulating the crank on a motorcycle cycling between 500 and 5000 RPM in 100 RPM increments.

The only thing not responding now is the temperature gauges
A few more days on the bench the downstairs to the garage for some real fun.

https://youtu.be/3xpene2kRPw

admin
Site Admin
Posts: 44
Joined: Mon Jul 02, 2018 1:02 am

Re: Fuel Injector Retrofit Project

Post by admin » Wed Aug 15, 2018 4:38 am

I left my office/lab and dragged everything down to the garage to start shoehorning the NO2C into my Honda magna 750. After getting everything set back up after the move I noticed that the RPM gauge is reading exactly half of what it should. I have everything set up as 4 cylinder, distributor, and sending it a square wave to VR1 via another arduino running the simulator program.

Supposed to be putting out 1000 RPM signal but the Tunerstudio gauge reads 500.

Its got to be the setup. I just looked a a video I made earlier when I first fired it up in the garage and it was working fine. I remember looking at the frequency earlier when it stopped working and it was 16Hz... A friend (the developer) said it should be 33Hz. So I think the problem is in me screwing up what I told the stimulator to do.

Running extra grounds from the chassis to all the sensors to avoid any noise issues. Also going to be using twisted pair CAT5 cable to the sensors also to avoid noise issues.

I am going to be using the coil signal for the RPM signal and since it is 0/12 volts I will be using an optical isolator to step it down to a 5 volt signal.

https://youtu.be/uf99iZ0j3Mo

admin
Site Admin
Posts: 44
Joined: Mon Jul 02, 2018 1:02 am

Re: Fuel Injector Retrofit Project

Post by admin » Fri Aug 17, 2018 4:34 am

The harness is nearly finished. I still need to add a fuel pump relay and some connectors for the RPM.

For RPM detection I have added an opto-isolator to the circuit board. I need to connect and test it to make sure it is converting the 12 volt coil signal to a 5 volt digital signal.

Getting dangerously close to trying to start the motor.....

admin
Site Admin
Posts: 44
Joined: Mon Jul 02, 2018 1:02 am

Re: Fuel Injector Retrofit Project

Post by admin » Fri Aug 17, 2018 7:22 pm

Finished the harness today and got a chance to test the sensors.
Still need to put in a relay for the fuel pump and I'm ready to give it a live test. Probably a few more days. I am anxious to hear some kind of noise coming from the mufflers.

https://youtu.be/aTILof9I4oI

admin
Site Admin
Posts: 44
Joined: Mon Jul 02, 2018 1:02 am

Re: Fuel Injector Retrofit Project

Post by admin » Wed Aug 22, 2018 5:39 pm

This is the end of chapter one. It has been a years worth of discussion, thinking, reading and studying. Designing and re-designing manifolds, fuel tanks and pressure pumps. Experimenting and measuring different injectors and finding the right throttle body for this size motor.

Lot more work to be done. Fine tuning for best performance and fuel mixture.
Its near impossible to tune with a narrow band sensor. They work ok after you have already tuned it and just need to keep it close but for tuning from scratch you must have a wide band controller. I ordered a wide band O2 controller and a Wide band Bosch O2 sensor. Trouble is they take about 3 weeks to arrive.

THE WORLDS FIRST 1987 FUEL-INJECTED HONDA MAGNA. https://youtu.be/_oLNUT8rP28

Connie C10 is next what to do for a manifold????
Attachments
20180822_112919.jpg
20180822_112919.jpg (7.07 MiB) Viewed 2511 times
20180822_112936.jpg
20180822_112936.jpg (7.3 MiB) Viewed 2511 times
20180822_112941.jpg
20180822_112941.jpg (7.15 MiB) Viewed 2511 times

admin
Site Admin
Posts: 44
Joined: Mon Jul 02, 2018 1:02 am

Re: Fuel Injector Retrofit Project

Post by admin » Sat Aug 25, 2018 8:22 pm

Ordered the Sigma Lambda Controller O2 controller and a Bosch 4.9 O2 sensor and a backup Spartan 2 OEM all for $130. Decided to mount on top of the muffler between the muffler mount and the shock.

Found a bung nut to weld to the muffler at O'riely's auto parts for $4.99.

Discovered that the RPM data is 1/2 half what it should be. I added a pair of diodes to combine the spark signal so it counts all 4 pulses instead of 2.
Attachments
blocking diodes.JPG
blocking diodes.JPG (35.21 KiB) Viewed 2490 times
20180825_132133.jpg
20180825_132133.jpg (7.07 MiB) Viewed 2490 times
20180825_132033.jpg
20180825_132033.jpg (6.85 MiB) Viewed 2490 times

admin
Site Admin
Posts: 44
Joined: Mon Jul 02, 2018 1:02 am

Re: Fuel Injector Retrofit Project

Post by admin » Mon Aug 27, 2018 7:39 pm

This is a narrow band O2 sensor. I have ordered a wide band sensor and a controller but will continue to test with this one until they arrive..

There is little exhaust in the outer reaches of the muffler. And the sensor only reaches in about 1/2 inch. I can see the sensor switching from rich to lean. I think I will move the sensor further up stream where the pipe leaves the exhaust manifold and enters the muffler pipe. That will put the sensor directly in the flow of the exhaust.

Getting some better results on the RPM input. I tapped into the tachometer signal and ran it through a blocking diode and then through a resistor with a cap filter. The signal is much cleaner and no missing or extra pulses.

Still running rough at idle. I believe it is running too rich. It seem to run very well just as it is running our of fuel.

https://youtu.be/aPw-NtfGg3o
Attachments
20180827_133653[1].jpg
20180827_133653[1].jpg (7.17 MiB) Viewed 2465 times
20180827_124841.jpg
20180827_124841.jpg (9.01 MiB) Viewed 2465 times
20180827_124700.jpg
20180827_124700.jpg (8.08 MiB) Viewed 2465 times
20180827_124650.jpg
20180827_124650.jpg (12.18 MiB) Viewed 2465 times

admin
Site Admin
Posts: 44
Joined: Mon Jul 02, 2018 1:02 am

Re: Fuel Injector Retrofit Project

Post by admin » Tue Aug 28, 2018 9:49 pm

Could not condition the tach signal any better so I cheated. Since the tuner studio is displaying half the real RPM I changed the interrupt handler to double what ever it ended up calculating.

I guess the firmware is expecting double the number of pulses that the tach signal is producing but when I try to add them together, rather that two pulses I end up with one long pulse at the same frequency. So now my conditioned tach signal is producing one pulse per rev, I think it perfectly honest to modify the firmware to account for the missing pulses from the other cylinders.

Running better now. Idles at 1500 RPM and has quit surging. I am going to call that good for now and proceed to tuning the other parameters.
Attachments
20180828_153128.jpg
20180828_153128.jpg (13.39 MiB) Viewed 2455 times
20180828_152444.jpg
20180828_152444.jpg (9.57 MiB) Viewed 2455 times
20180827_124841.jpg
20180827_124841.jpg (9.01 MiB) Viewed 2455 times

admin
Site Admin
Posts: 44
Joined: Mon Jul 02, 2018 1:02 am

Re: Fuel Injector Retrofit Project

Post by admin » Thu Sep 06, 2018 3:41 am

Making little progress. it starts, idles and throttles up and down. Not a lot but it is progress.

Signals coming from the neg terminal of the coil are just too noisy to use. It looks good on the scope but at time I see a burst of several injector pulses that indicates the firmware is seeing extra tach pulses. So I tapped into one of the crank sensors and I am now getting a reliable pulse once per rev. Problem is the firmware is expecting two pulses on a 4 cylinder motor. The real solution for this is to add another toothed wheel and sensor to the crank but... that is a big messy job that I don't look forward to. I got around the problem temporarily by telling the Tunerstudio that the bike only has two cylinders. But that is just a band aid. Can't do any reliable tuning without at least two crank pulses per rev.
Still waiting for my wideband sensor and controller. I found out today that is wont ship until next week. Arrggg.

https://youtu.be/WY8xhAa_ZQA
Attachments
20180905_121309.jpg
20180905_121309.jpg (7 MiB) Viewed 2437 times
20180905_120817.jpg
20180905_120817.jpg (8.14 MiB) Viewed 2437 times
20180905_120505.jpg
20180905_120505.jpg (7.83 MiB) Viewed 2437 times
20180904_165421.jpg
20180904_165421.jpg (8.68 MiB) Viewed 2437 times
twocranksignals.JPG
twocranksignals.JPG (65.21 KiB) Viewed 2437 times

Post Reply