The main criteria for any mowbot platform must be adequate space for the
cutting mechanism. Grass must be allowed to stand up as much as possible so
it can be cut effectively. Most modern lawnmowers have twists in the blades
to create an upward airflow that helps make the grass stand up, this may
not be an option for Mowboters as it will increase current drain.
My chassis is essentially a box at the rear, that holds the batteries and
drive motors, with a flat projection at the front to carry the cutters and
front castor.
The clearance at the front is 6 inches, and 2 inches at the rear. It's made
of aluminium for strength and lightness, aluminium is also cheap and easily
bent to form a closed chassis. Aluminium angle section holds it together
and keeps it strong and straight. It's bolted together with galvanised
roofing bolts.
The castor is the largest plastic wheeled castor I could find, a 5 inch
diameter Fallshaw castor.
I chose plastic wheels and castors to keep the weight down as much as
possible.
Moving back, there is a clear region of about 6 inches where the cutters
will go. Behind that is the battery/motor box. The box to ground clearance
was dictated by the wheel size and the motor underhang as I wanted the box
completely sealed to avoid water ingress.
The wheels are Rover plastic lawnmower wheels with a diameter of 8 inches.
This is the largest plastic wheel I could find and seems to be a reasonable
comprimise between speed and torque.
Let's say your motor turns at 40RPM, an 8 inch wheel covers 25 inches per
revolution, a 6 inch wheel covers 19. By changing wheel size you can change
speeds, but you pay for it in reduced torque. In rough terms, you can push
30% less weight with 8 inch wheels over 6 inch wheels, 50% less with 12
inch wheels.
Robotic platforms are a constant tradeoff between weight and power. You
need bigger batteries to go further, but you need bigger motors to haul
bigger batteries, bigger motors require bigger batteries, ad infinitum.
The batteries go in between the motors and slightly forward to balance the
robot and stop it tipping backwards while going up hill, this puts most of
the weight on the drive wheels which are 2 inches wide, the castor is freer
to turn this way.
The motors as I mentioned in a previous post are from electric windows and
were very cheap. I had axles lathed to connect the wheels to the motor
output shafts, the shafts go through sealed bearings to carry the weight. I
could use flange bearings but at the time I didn't know about them. A
flange bearing has flanges on the outer edge to bolt to a flat surface.
Each motor is controlled by a DPDT relay and a FET. The DPDT relay changes
the polarity to the motor, and the FET drives the ground input so I can
turn the motor on and off. If I needed it I could PWM the FET to control
speed. There are 2 other FETs to switch the relay coils. The relays are set
up to provide positive motion when off, so I only need to energize them for
reverse. I could have used an H-bridge but it seemed excessive for such a
simple platform.
Disregarding the aluminium casting I had made, the total cost for the
platform was about $120AUD. Quite cheap for what is a relatively
professional chassis.
The motor battery cost me about $80AUD. I chose a sonnenschein 9.5AH
battery. It is a sealed lead acid deep discharge battery. You cannot get
away with a trickle charge battery, trickle charge batteries are designed
to give up their life to run a system in a once only event. I have a
separate 2AH battery for electronics.
I have chosen to use dual cutters, this is for 2 reasons, one I won't
mention because it is to do with sensors, the main reason is that it allows
a wide cut path without making the robot too long. The cutter region of my
robot is only 6 x 12, if I used one cutter, it would be 12 x 12. The
cutters will be attached to a platform that I can height adjust.
The rear box dimensions were determined like this:
width is the width of the motor battery plus the motors, length is the
length of the motor and electronics batteries, height is the ground cutter
height minus rear ground clearance. The box didn't need to be as wide as
the cut path, I used angle brackets from a hardware store to hold a bump
ring out at the right distance. (see my web page).
The current aluminium shell was bent on a corner thingo by someone who owed
me a favour. I bent my first shell by hand using pieces of wood I'd cut to
the right length, sandwiching the aluminium between them and bending and
hammering.
It seems to drive over 5 inch grass fine. Going uphill is harder, but when
the cutters are installed the grass will only be 2 inches by the time the
rear wheels hit it.
Everything mechanical, except the axles and flange bearings can be bought
at the average hardware store.
I'll put together some photos on the web site in the next few days.
Dave Everett
Mowbot Project Website
http://www.idx.com.au/~deverett/mowbot/index.htm