Adding a little Raspberry Pi to my 82'Heritage Special.

[BBS360]

Could splice up a short length of wire with a male and female bullet connector at each end. Connect it inline on the orange or white points wire. Tap off it directly, or with an inductive coil if there is enough flux on the 12V line to sense, and feed that into the diode side of an optocoupler. Feed the other side into the microcontroller. When the points are closed it will conduct to ground and trigger the optocoupler. When the points open, current doesn't flow, and the optocoupler is not triggered.

Not sure whether it would interfere much with the function of the ignition coils. I want to say it wouldn't be an issue since the condenser already soaks up the charge from the flux that feeds back into the primary ignition coil. As long as the sensing circuit itself is protected against damage by the counter-emf everything should work all right.

 

[rugbywarrior89]

The coils operate with a voltage between 11.5 and 14.3V so they should be able to compensate for the optcoupler I would think as long as it doesn't draw too much power. I doubt it would considering signals going to arduino/pi are on the order of 20 mA. Your coils probably wouldn't even know it was there.

 

[rugbywarrior89]

Here is a few interesting threads about Using the bluettoth RSSI to estimate proximity and run script based on the proximity of your phone. This could be used to only turn on the ignition when you're within a certain proximity. This is not ideal as the proximity detection is very rough and a class 2 adapter is still going to consume more power than a class 3 but as of now, it would be my only option.

http://www.raspberrypi.org/forums/viewtopic.php?f=37&t=47466

http://nuigroup.com/forums/viewthread/2474/#14582

 

[rugbywarrior89]

I was just perusing around online and came across a hall effect sensor. I was think for those doing the speedo/tach, couldn't you use a hall effect sensor to calculate speed instead of a gps module? You could probably even use it to calculate a rudimentary tach as well.

For the tach, you could have the program assume that you are in a certain gear at a certain speed. You know all the gear ratios so you could then calculate the rpm of the motors. Wouldn't be terrible accurate during acceleration outside the normal driving range (i.e. driving like you stole it) but it would be good enough for a monitor and kill two birds with one stone in the process, right?

 

[SuccubusJuice16]

Sounds like you could use a hall sensor. I think that's what a lot of newer computer controlled engines on bikes use for cam and crank positioning. FSAE guys use these a lot for independent wheel sensoring.
basically the same unit you can find on a lot of cars with ABS sensors on the hubs. would be a cheap part to pick up at the junkyard, and maybe you could utilize the rear sprocket somehow if you didnt want to stick one into the motor.

Edit- Also i wanted to add i think adding the pi is an awesome idea!

 

[rugbywarrior89]

Brings up a good point. Perhaps you could stick a hall sensor inside the engine somehow. Can any mechanic Gurus in here think of a spot you could mount one of these things to read the camshaft? Or any moving part of the engine for that matter. Maybe putting the sensor inside a tappet cover and putting a magnet on one of the valve guides might do the trick. I rpm = 2 open/close of the valve.

Edit: Just brainstorming ways to avoid messing with the coils.

 

[BBS360]

That'd work to get a best-guess RPM.

I have a few hall sensors lying around. They'd be fine if you used it to monitor front or rear wheel rpm, far away from any stray magnetic fields.
Trying to put a hall sensor near a points contact breaker would probably pick up a lot of noise.
On a non-points ignition timing unit that already uses hall sensors the addition of one more should be easy.
I've been looking further into tapping into the negative side of the coils. Seems like a more universal solution to get RPM across bikes with different timing units as they all have coils.

 

[BBS360]

It'd be possible to put a magnet and sensor inside the motor. I'd be cautious performing any permanent modifications to route wires inside.

 

[Norman1950]

I am liking this idea. I bought a PI to maybe use as an airride controller. But haven't worked on it for a while. But instead of using RFID could possibly use a magnetic reed switch to turn the bike to "on". But you do lose the function of perimeter sensing.

 

[rugbywarrior89]

Thougth of a reed switch ignition but then anybody with a magnet could just drive away. The advantage to that is that nobody would think to steal a motorcycle using a magnet. However, when your in a bar showing off to your friends that you power your ride with a magnet, you never know who else might be listening in.

 

[rugbywarrior89]

It'd be possible to put a magnet and sensor inside the motor. I'd be cautious performing any permanent modifications to route wires inside.

Thats why I was thinking tappet covers and valve. It's accessible and if you break your tappet, you can easily replace it. XSChris apparently has a pile of them sitting in the corner of his basement. I see him selling them on here all the time. Since the hex part of the cover is hollow, you could drill 3 tiny holes mount the sensor in the hollow part, and glue a tiny (rare earth?) magnet on top of the valve.

 

[SuccubusJuice16]

don't the stock RPM cables attach to the top of the valve cover? maybe could modify that little wheel that spins into some sort of sensor for RPM since its already in place.

 

[rugbywarrior89]

What displays were you guys looking at for outdoor lighting and weather proofing?

 

[BBS360]

Thats why I was thinking tappet covers and valve. It's accessible and if you break your tappet, you can easily replace it. XSChris apparently has a pile of them sitting in the corner of his basement. I see him selling them on here all the time. Since the hex part of the cover is hollow, you could drill 3 tiny holes mount the sensor in the hollow part, and glue a tiny (rare earth?) magnet on top of the valve.

There ya go. Might be an idea to thread a hole in the cover. Mount the hall senor on a hollow threaded rod with the wires going through it. You'd be able to ensure the hall sensor is positioned optimally by rotating the rod, locking it in position with a but.

 

[BBS360]

There are displays that can connect to arduino through the I2C serial protocol. Just two wires for power and two for data. They can be as simple as block LCD characters or full color displays. Cheap ones can be had for under $10.
Hadn't thought too much about weather proofing. Just a simple commercial project box at first.

don't the stock RPM cables attach to the top of the valve cover? maybe could modify that little wheel that spins into some sort of sensor for RPM since its already in place.

Yeah. Not a lot of clearance inside but there might be a way to connect some kind of mechanism in-line with the tach cable.

 

[levdir]

I've been mulling over weatherproofing options and I'm probably going to end up designing a housing to get printed and then just sealing it with epoxy and silicone. The screen I bought is a touch screen but I don't intend to use it that way so I'm not too concerned about sealing it away behind glass or whatever.

 

[BBS360]

As my unit is meant to replace the stock flasher I've been thinking about finding a tubular enclosure that will fit in the flasher's rubber damper. No additional hardware necessary and vibration will be minimized.

 

[rugbywarrior89]

I've been mulling over weatherproofing options and I'm probably going to end up designing a housing to get printed and then just sealing it with epoxy and silicone.

I saw that pi case that fits like a 3 in display, You could get that. cut out a tiny piece of plexiglass over the screen and silicone it all together. then you would have both you pi and screen waterproofed.

 

[rugbywarrior89]

Okay, I think I have nailed down my scope. It seems to be a wee bit bigger after taking everyone's ideas into consideration and deciding how to go about things. I've considered just using an arduino microcontroller as it has a much lower power consumption but I just don't know... It looks scarier, and I have already done so much research on the pi.lol

 

[rugbywarrior89]

Sorry there haven't been much update to this thread. I recently just closed on my new near 200 year old house and have been quite busy with renovating, painting, and moving. Aside, from practice with python code here and there, I have not really managed to do a whole lot with the project yet. However, I have a sneaking suspicion that my wife bought me the pi for Christmas, and my that I mean she ordered it using my Amazon account and saw it on my History . She got me the B+ model which is somewhat unfortunate considering the whopping power consumption. But I could expand the scope yet again in order to add an LCD display which opens up the door for speedo, tach, GPS, and media player (nothing like watching Lord of the Rings while cruising down I-76). Kidding of course. Maybe I can add a solar panel with a automatic transfer switch to keep the pi powered when I go somewhere and shut the bike off.

But First thing First: Program the lights!