Installing a Sigma BC800 onto a Ninja 250

Posted September 2, 2003

I decided this weekend that I would like to have a clock on my Ninja 250, and after a bit of deliberation, decided that getting a bicycle computer would accomplish that goal, as well as giving me a redundant speed display and additional trip odometer (for things like reminding me to lubricate the chain).

I settled on the Sigma BC800 after reading numerous accounts of other motorcyclists choosing this model. I picked up my BC800 for $18 at REI (also available numerous locations online, and local bikeshops in your area). I think typical prices are $15-20 in the US.

The Sigma BC800 is a fairly small unit, about 1.25 inches on a side, and half an inch thick. There are smaller bike computers out there, and there are much larger. I think the feature set of the BC800 is fairly complete, although I'd prefer to see a temperature readout as well. I didn't try to find a computer with that capability because of the overwhelming support the BC800 seems to enjoy in the motorcycling community. I only have two problems with the BC800: 1. when it wakes up from sensor input (ie, when the bike starts moving), it always displays trip distance. I want it to display whatever setting it was left on. 2. I wish the 12/24 hour clock display format wasn't tied to whether you have KPH or MPH selected. I want MPH and 24 hour time.

The current-model BC800 has an excellent feature for motorcycle mounting: its mounting bracket has a removable handlebar plate, which was designed such that the plate could be taken off, rotated 90 degrees, and reinstalled. However, it's also perfect for making your own bracket out of thin aluminum or steel stock.

I elected to make a bracket for it from 1/16" thick aluminum stock, purchased at the local Lowe's for less than $3. This material is definitely strong enough for the small distances and weights involved. Making the bracket mostly involved filing down the edges of my 3/4" wide stock to 11/16" (the width of the Sigma handlebar plate), and rounding the upper corners to fit into the recession on the computer bracket. I traced the mounting plate's hole pattern onto the aluminum, and drilled the holes using a 9/64" drill bit in a small drill press.

Once I had chosen the mounting location (covering the unused lower right section of the tachometer), it was a simple matter to bend the bracket to the right angle using some scrap wood (to protect the aluminum) in a vice. Mounting the bracket to the instrument cluster cover was accomplished using some small stainless 4-40 screws and nylon locknuts I already had for a different project.

With the mounting location chosen and the bracket made, that left extending the sensor wire, locating a suitable magnet, and figuring out how to light the face of the computer. Extending the sensor wire was as simple as splicing some extra wire in the middle of the cable. I recommend, if you do this, that you locate the spliced section near the computer, so that the wire hanging out in the weather is all one piece.

For a suitable magnet, I had read a recommendation of using a voice coil magnet from a pair of cheap headphones. I had a pair of Sony throwaway headphones (the type that goes over your head, with the 1" speakers), and liberated the coil magnet from one side with a bit of ripping and tearing. Indeed, it was quite strong for its size, and very flat -- ideal for placement on the inner radius of the front brake disc. The magnet was affixed with epoxy to the brake disc. The sensor was likewise affixed to the fork leg with epoxy, since there's no practical way to use zip ties at that location. Remember to have some tape handy to hold these things in place once the epoxy has been applied. Even quick-setting epoxy takes several minutes to set up, and the item being glued needs to be held steady for a good bond.

That left the question of lighting. The solution I came up with was to wire up some yellow LEDs I had left over from a different project, and epoxy the completed assemblies to the sides of the computer bracket. As you can see in the picture to the left, it's not actually a very good solution, as the LEDs I used are too focused, resulting in light by which it's difficult to read the display. I may try sanding down the lenses of the LEDs to diffuse the light, or I may create a more inflexible mounting system, and use different LEDs. The LEDs are wired with a 500 ohm resistor inline (overdriving them slightly, at about 25 mA), and are spliced into the speedometer light leads, so they're on whenever the instrument lights are on.

The resulting setup (except for the LEDs) looks pretty good. The mounting location was unused space on the dashboard, so it's not concealing any instruments you might want to see. Because the bracket is fairly tidy, and the wires are kept out of the way, it all looks fairly professional. The computer is easy to see during the day, and is marginally readable in the dark.

A quick and highly inaccurate comparison between the BC800, a Garmin GPS-12Map and the stock speedometer revealed that the BC800 updated faster than the GPS, and typically read a little bit higher. The stock speedometer read a little bit higher than the BC800, but I haven't had a chance to test on the freeway yet, which is where the stock speedometer gets really inaccurate. I expect that the BC800 will certainly be accurate enough, particularly considering that what I really wanted was a clock.

Update, October 31, 2003

I successfully installed a better lighting system on the BC800 a month or two ago, but have finally managed to get some pictures taken with the new system. It consists of an aluminum arm that sticks out over the face of the display, with three small red LEDs aimed down onto the face. In order to weatherproof the light, I added an aluminum cover, and filled the back-side of the LEDs with epoxy. It's not as pretty as it could possibly be, but it's functional and effective.

The speed sensor also fell off this week, fortunately when the bike was parked. It would have been disastrous if the epoxy had failed while under way. I have reglued the sensor in place using JB Weld, although that also appears to be epoxy, but with pigment in it. Only time will tell. I wish there was some more secure way to fasten the sensor to the fork leg, but I haven't thought of one yet.

Created by Ian Johnston. Questions? Please mail me.