Monday, May 6, 2013

Parts Are Arriving


Prototype parts are starting to show up. The first items are the cranks. Currently the line up include 4 x SRAM Rival GXP 53/39 172.5mm, 1 x S950 50/34 170mm. I’ll likely be picking up a few more in the coming weeks. I want to get a shorter 165mm arm unit as I want to explore the unwanted force rejection characteristics of a shorter arm.


I’ve also received a used Quarq S975 (50/34 172.5mm). However, in putting it on my bike I had to lower my front derailleur hanger. That was fine, however, I find that for my cadence and the local roads have me switching between small ring and big ring a lot. With a full size crank I was either in the small ring or the large ring more consistently (53 for flats, 39 for climbs). I might look for a 52/36 110bcd chain ring setup, but then I have to send it in for calibration. The S950 (or maybe the existing crank arm) will be instrumented for a double-power meter. This will likely be the unit I send to Ray over at

I picked up a used Trek Madone 5.2 frame, but it was setup for Shimano cranks. I therefore had to remove the bearings. They aren’t “pressed” in tight, so the non drive side came out easily but the drive side did not want to. An 11/16th socket made a good bearing removal tool as you can see below.


Sadly it came with a 130mm seat cap, which comes up 10 – 20mm shy of what I need to match my setup on the GT GTR bike seen above. I need to source a 175mm cap version now. Hopefully a Trek dealer in area or near Toronto has one.

Programming is going slowly. I haven’t looked at it in a week, so I’ve been stuck at 80% complete. The App Timer for measuring cadence with a magnet sensor is the only thing I haven’t sorted. I was really caught up with the force rejection and thermal testing and I wanted it sorted as best as I could. I’m still not 100% satisfied but I am generally settled on the strain gauge and temperature sensor design.

As for the gyro / battery debate I’m just making a call and sticking with it until I get out a minimum viable product for people to test. I am thinking on designing the circuit board to support both though. That way switching down the road is more about a new rapid prototyped enclosure than a major redesign.

I’m also in talks with a couple of companies to supply me with programmers for the prototypes. These boards I am designing use a two wire interface for reprogramming the nRF51422 chips. In order to field update firmware these would have to be included with the prototypes (or built into them which I do not wish to do).

What I’ve heard most is that people would like their devices to update more to get more accurate data. ANT+ is every 0.25 seconds but it will only update the page information that is transmitted once a rotation. This is where the gyro can come in. It can give much faster, and therefore more accurate information. Like I said I am likely going to include the gyro, but not enable it initially. I’m aiming for coin cell but will include a charging circuit for Lithium that can be disabled. These are low cost to front load and doesn’t take up much physical space.

I’m taking three days off the day job to recover and work on the circuit board design.

No comments:

Post a Comment