MiniPOV with Accelerometer

[Jorge Pinto aka Casainho]

Hello :-)

Just one idea. I found this whiling browsing Make blog:

http://www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1223449559

I would prefer to not have to put a magnet on bicycle... if there was an way to do the sincrinization with another kind of sensor, would be great :-) -- but price is always a question ;-)

[Jorge Pinto aka Casainho]

Sáb, 2008-10-25 às 06:04 -0700, kvas_off escreveu:
> > I would prefer to not have to put a magnet on bicycle... if there was an way to do the sincrinization with another kind of sensor, would be great :-) -- but price is always a question ;-)
>

> Jorge, what's the problem with magnet on bicycle? Any cyclocomputer
> uses it (but, to tell the truth, on a spoke; we need yet another
> magnet to use it vice versa).

Yes, but when we buy it it comes already with a magnet and with a
support for the magnet - on Bicycle LED POV, we need to find one magnet
on the market and attatch it on a DIY way... it complex the thing.

> And, yes, MEMS accelerometers are pricey a little bit in comparison
> with Hall effect sensor. But they're really cool.

First I would like to know if a sensor like that could tecnhicaly
substitue the sensor hall effect on this system, after we can think on
advantages and disadvantages.

> P.S. Didn't you send me PCBs? :)

I had a busy week, I just could send them yesterday at the end of the
day :-)

[Jorge Pinto aka Casainho]

Sáb, 2008-10-25 às 07:05 -0700, kvas_off escreveu:
>
> Jorge Pinto aka Casainho:

>
> > > Jorge, what's the problem with magnet on bicycle? Any cyclocomputer
> > > uses it (but, to tell the truth, on a spoke; we need yet another
> > > magnet to use it vice versa).
> >
> > Yes, but when we buy it it comes already with a magnet and with a
> > support for the magnet - on Bicycle LED POV, we need to find one magnet
> > on the market and attatch it on a DIY way... it complex the thing.
>

> Well... well well... :) Piece of rubber and nylon cable ties should
> help. It isn't more complicated then mounting LEDPOV board itself on
> spokes, shielding, hermetisation, etc.

>
> > First I would like to know if a sensor like that could tecnhicaly
> > substitue the sensor hall effect on this system, after we can think on
> > advantages and disadvantages.
>

> If we'll know centrifugal acceleration then we will know velocity. But
> how then we define where is the top and where is the bottom?

Doing some practice tests.

Anyway, right now I am busy working on API and on it changes
(again :-) ) -- I will report soon all the changes needed on API for
software working.

> > I had a busy week, I just could send them yesterday at the end of the
> > day :-)

> So you sent them? Thank you very much! ^_^

[DZiems]

I think an accelerometer is overkill. A hall sensor, neodymium
magnet, and some hot glue to hold the magnet onto the bicycle frame,
should cost no more than $4 total, while accelerometers will cost
plenty more, as well as being less reliable and harder to code for.

On Oct 25, 9:27 am, Jorge Pinto aka Casainho <casai...@gmail.com>
wrote:

[DZiems]

Also, I don't think an accelerometer would work. The centripetal
acceleration of the wheel would likely be great enough to overwhelm
the sensor to where it would not be able to accurately detect the
force of gravity, and thus its orientation.