On the MotoGP bikes they use a load cell in-line with the shifter linkage. On the upshifts they kill the ignition, on the downshifts the ECU blips the throttle (all fly by wire). I posted a picture below from a google image search.

For the noise, you might try resistor spark plugs. They are sometimes used to reduce interference, maybe it will help (assuming your software solution doesn't do the trick).

https://www.ngksparkplugs.com/about-ngk/faqs/spark-plug-faqs/when-should-i-use-a-resistor-spark-plug

 

Unfortunately this project is on hold for a while as I crashed my sidecar and have suffered a badly broken leg.

Once / If I get this fully working I will of course update this thread with the results.

 

Hello David, I hope you are recovered and your sidecar repaired. My english is very poor and I have no idea of electronica but some mechanics if that is and I have found a problem in the base of your scheme. The ecu cuts the mass only when it wants ignition, ie the coil is always with 12v and more less 9 amp always less when the ECU gives the signal of ignition that cuts the mass. If you cut the positive the spark plug will spark.

Greetings.

 

Sorry to hear about the crash. Hope you get well soon. I built up a simulator so I can work on the electronics without the need of a bike.

Here you can see the logger connected up..

 

Yep I know that feeling brother, bones mend faster with a bit of Ti, and sports physio will get you mobile a lot lot quicker. You do the old "you're not cutting my leathers" thing or did you realise the ambos were probably right.

 

As my bone was sticking out of my leg at the time, I allowed them to cut the shit out of my £80 leathers.

 

Hi David, I hope you have recovered well. Did you come back to the tracks, and have chance for this again?

Regarding to your false trigger problem, it is your circuit picking up EMI noise from the electrical system of the car (it is very noisy, many noise sources: spark, coil, alternator, injector...). And especially the wire from your circuit going to the switch is acting as an antenna when the switch is open, so it picks up noise...
470uF//10uF is too much for filtering an input like this. Usually it is less than 1uF, you can try 1uF // 10nF place them close to arduino pin, and add series resistor (22 ohm to 100 ohm) on the switch wire.
Filter at power supply lines and other lines are required as well. Try adding big capacitor (470uF) // small capacitor (100nF) at 12V input of the circuit; add 100nF close to 5V pin of arduino; add 100nF on "A" and "B".

For other's concern about fuel is still injected into cylinder when ignition is cut off, is it actually a problem? I see on wikipedia saying "other product cuts ignition and/or injection", should we cut injection as well? I think a mosfet will also work to cut the injection signal?

Also I have one more question: as before cutting off the ignition, the transmission is loading as we are riding, it will be locked and even if we apply force on the shifting rod it will not move. So how do you setup your proximity switch? It should be tune carefully as we want it to detect when the shifting rod reaches the stop position (as it is locked) and the force increase enough, only then we should cut off the engine and allow it to move and gear will jump. That's why aftermarket products usually uses inline loadcell or pressure sensor.

 

Hi,

Recovery has taken much longer than expected. My external fixator was removed around 2 months ago. I am looking to ride again in March.

I did have a little play around with this circuit on my brothers sidecar this season, but testing was very limited. As you mentioned, it was very difficult the get the proximity switch working correctly. For this reason I have decided to purchase a second hand pressure sensor, it looks very similar to the one posted by MrSoftware above.

I am a little worried about the size of the capacitor I have used. It is actually lower than the 470uF I posted above, I think it is 220uF (Still sounds huge to me). I have also put a 10uF directly next to the arduino GPIO pin.

I presume that when the switch on the sensor is closed, the capacitors will be discharged to GND via the microswitch in the sensor. I have no idea how much current this will use, and whether or not the sensor I have bought can handle it, also if it takes time for the voltage to drain to a logic low level. I don't see this on the circuit I have built but I also don't have a scope to test it properly.

 

I have tested this circuit in the garage, using a pressure sensor in the gear linkage and creating artifical load on the gearbox by running the engine with the brakes on. It seems to be working really well!

I added a LCD screen and rotary encoder in a seperate box, which can be plugged into this circuit in order to change the kill time etc. This is far easier than having to change the code on the arduino by using a laptop. I have covered the circuit in hot glue from a glue gun, hopefully this will make it robust enough when it is out on circuit.

I had to add a 40ms delay, after the sensor was tripped, before cutting the ignition. My original concerns over using a relay as it was too slow seem to be unfounded, however I am happy I went with a MOSFET as it should be more reliable and not have a limited number of cycles like a relay.

I'm sure we will need to change the settings once we get to test this on track but it seems good so far. I will keep the thread updated.

 

Hi David
Hope you have recovered fully from your accident.

I am very keen and interested in doing this for my kawasaki zx7r. But i am clueless as how to go about it.
I have no experience with electronics but i can follow instructions and possible read a schematic.

Please if you wouldn't mind sharing your final diagram or layout of exactly how you have done this would be really great.

Thanks

 

Hi,

First of all you need to identify the primary feed to the coils. It will be a wire that comes from the ECU that feeds all 4 coils with power. Next you want to cut this wire before it goes to the coils, strip the ends of the wires and hold them together. Start the bike and get someone to gently rev it, you should be able to pull the wires apart for a split second and join them back together again, you should hear the engine cut out for a short period. If it keeps running then you are good. Do it a few times and check for any fault codes. On the CBR 600 RR no fault codes appeared.

If you are able to get the engine to cut out for a very short period, but it carries on running and there are no fault codes, then you can use my circuit.

 

Thanks David

I have odered the adruino uno starter kit when it arrives i will give it a go.

My bike is a carb bike so should work fine with out any fault codes.

I am new to electronics and programming etc. Have found online tutorials will fiddle around first before trying to set everything up for the bike

 

How did your project ever turn out? I assume you healed up but maybe gave up the sport. Time for an update.

 

How did your project ever turn out? I assume you healed up but maybe gave up the sport. Time for an update.

Hi Joe,

I had the external fixator removed in September 2017, recovery went well and I raced my sidecar again in 2018. I also took up cycling to get a bit fitter. Unfortunately in December 2018 I was hit by a car while cycling and broke my hip, same leg. It took about 12 weeks to fully recover from the broken hip and I have been cycling again this year and also racing my sidecar again.

As for the arduino quickshifter, it has been working flawlessly on my brothers sidecar in conjunction with a Dynojet pressure sensor. I have also made another project, which is a gear indicator and rpm gauge on some neopixel strips.

Here is an onboard video from my brothers sidecar, with the arduino quickshifter fitted. Not sure if you will be able to see it or not: https://www.facebook.com/chris.scho...054615643/?type=2&video_source=user_video_tabnot: https://www.facebook.com/chris.scho...054615643/?type=2&video_source=user_video_tab

You can see the rpm and gear indicator in action in this video:

Regards,
David

 

Nice videos at a glance. I will check them out. Sorry to hear about the second break but good to hear you are on the mend. Good work on the auto.

 

I added a LCD screen and rotary encoder in a seperate box, which can be plugged into this circuit in order to change the kill time etc. This is far easier than having to change the code on the arduino by using a laptop.

 

I added a LCD screen and rotary encoder in a seperate box, which can be plugged into this circuit in order to change the kill time etc. This is far easier than having to change the code on the arduino by using a laptop.

Oops that didn’t work out as I thought. Apologies.
Quietly I’ve been building up a copy of this unit for my GPZ sprinter. It’s progressing well considering my limited electronic knowledge and my non existent arduino knowledge. I had been pondering if a display was possible. Any chance of a diagram and the script.

 

The rotary encoder and LCD stopped working on the first version of this, and it needs quite a few wires coming out of the box. I will try to knock up a diagram and see if I can find the code.

I have made a few more quickshifter boxes since this one and made a new version, this is using an ESP32 instead of an arduino, and an Android app is used to update the settings.