• There has been a recent cluster of spammers accessing BARFer accounts and posting spam. To safeguard your account, please consider changing your password. It would be even better to take the additional step of enabling 2 Factor Authentication (2FA) on your BARF account. Read more here.

DIY quickshifter project

A

alpalwal

New member
Joined
Oct 8, 2008
Location
Alamo
Moto(s)
04 R1
I've been doing track days the last couple years and I've been wanting a quickshifter but couldn't justify $250 for the Dynojet sensor plus another $250 for a PC3 I don't otherwise need. I don't race (for now) and so I figured a DIY quickshifter could do most of what a Dynojet can while saving a bunch of money.

Here's what I have. The sensor is a pressure sensitive resistor in a custom aluminum housing I've made. There's no extra play in the shift linkage since the sensor doesn't have any travel like a standard button. It goes right in the shift linkage like the Dynojet and is compatible with their 2 piece shift rods.

The electronics are pretty simple. It's an Arduino mini that reads the pressure sensor and then trips a relay when it gets above a certain threshold. I've wired it inline with an ignition wire coming off the engine cutoff switch. All the delays are programmable and I'm hoping to have a "don't quickshift below X rpm" setting added soon.

I finally had a chance to test it out this weekend and it worked great. I'm working on getting the circuitry cleaned up now and just fine tuning the last couple settings.

Right now this is just a push sensor but if this goes well I'll make a pull one as well.

Once I'm finished prototyping does this look like something you guys would be interested in? I'd love to hear any feedback you have on the project.

tfxitNf.jpg

8lTYXYP.jpg
 
Could be fun. I've never had a quick-shifter. It'd be cool to have effortlessly smooth shifts. Some bikes just don't shift easily, no matter the technique. This would help.
 
That looks really cool. Does the pressure sensor have an analog output into your controller or is it digital on/off? It would be nice if it's analog, so you can differentiate normal vibrations and some minor "preloading" from the actual shift point. In that case, the refresh rate of your a2d conversion is going to be pretty critical since you have to get the timing down to the millisecond on something like this.

I'd also check into the force limit of that particular sensor, and realize that every bike's linkage is going to be different and thus a different force. If you exceed that force, it will ruin your zero calibration and possibly the sensor itself. You'll probably want to program in some calibration routine with some LEDs to get it just right.

I like the relay, but they typically fail off, so there is a chance that if the relay fails you will go for a shift and the ignition won't cut out. I'm not sure if this means the bike just won't shift, or it will crunch your transmission. It might be worth looking into a solid state alternative with an override, so worst case the engine dies and you have to disable the quick shifter. Or put an a transistor sensing circuit and do an auto bypass with an led or audible alarm. I couldn't help you with the design of that though.

Nice work overall! If you got it working then that's a sweet little solution!
 
Thanks!

The sensor is analog and so I've been able to tune out the risk of it cutting out from road vibrations. It's still a bit sensitive to preloading but that's just more fine tuning.

The refresh rate is working fantastic. At one point I had it down so low that the ignition was cutting out before the shift happened. I actually needed to put in a ~15ms delay to get the timing right.

I'll keep an eye on the relay. It's wired to be normally closed so in concept if it's broken it should just not open the circuit any more. That would crunch the transmission which is obviously bad but it's better than having the bike be completely dead.
Some sort of secondary check system on that circuit is a good idea. The relay shouldn't die but if it does I want to know about it!
 
alpalwal said:
Thanks!

The sensor is analog and so I've been able to tune out the risk of it cutting out from road vibrations. It's still a bit sensitive to preloading but that's just more fine tuning.

The refresh rate is working fantastic. At one point I had it down so low that the ignition was cutting out before the shift happened. I actually needed to put in a ~15ms delay to get the timing right.

I'll keep an eye on the relay. It's wired to be normally closed so in concept if it's broken it should just not open the circuit any more. That would crunch the transmission which is obviously bad but it's better than having the bike be completely dead.
Some sort of secondary check system on that circuit is a good idea. The relay shouldn't die but if it does I want to know about it!
Click to expand...

Oh great idea. Sounds like you've thought this through! What's your background if you don't mind? I'm assuming this type of thing relates to your job and/or school.
 
kuksul08 said:
Oh great idea. Sounds like you've thought this through! What's your background if you don't mind? I'm assuming this type of thing relates to your job and/or school.
Click to expand...

Actually not at all. I work in IT and went to school for business. I just enjoy making things end working on projects like this. I've made some stuff before - http://arcatch22.com but pretty much all my learning has just been reading online and trial and error.
 
Last edited:
lol, and we thought we were slick running a coil's ground wire to the horn button for a momentary kill switch. Preload the shifter, tap the button, and boom, next gear.
You crazy kids and your electronics...:laughing
That's pretty bad ass, though.
 
You must log in or register to reply here.
Back
Top