Today turned out to be a good day. Despite daytime temperatures in the mid to high 90’s (roughly 20 degrees f above my personal cranky threshold), I made some significant progress. It sure seemed like most of what I did was just sitting in the hot sun, and all I got done was to install a second cross-brace, this time a flat bar across the aft motor mounting holes. This was part of setting up rails to hold the battery box, which I plopped on the bot. I had been planning to mount the batteries between the motors and slightly behind. This would have placed the batteries a lot lower as well. I was worried about the center of gravity being low enough to prevent roll-overs on inclines. As it turns out, it’s a good thing this plan didn’t work out. I ended up placing the batteries pretty much right on top of the imaginary line between the centers of the wheels (where the axle would be if there were one). The Big ‘ol Switch came in from The Robot Marketplace, and I quickly realized that I had just about everything I needed for a quick test run!
I hooked up the batteries and the electronics, and went inside to find the RC transmitter.
After a few minutes making sure that all the loose wires were clear of the wheels, and a few more making some adjustments to the RC setup, I discovered why having the batteries hang off the back would have been a problem: with both sticks forward, the bot REALLY wanted to lift it’s front wheels off the ground! The CG may have been raised a little from what I had originally intended, but it was also significantly more forward than it would have been. As it stands, I am not sure if the weight of the lawnmower motor will in itself be enough to keep the front end down. I had been thinking that I would install the smaller mower batteries in with the mower, and I think I will certainly be doing that. Even if they don’t end up holding much of a charge, they will act as ballast in a more forward position. All of this may sound sort of alarmist (and I suppose it is), but the real point here is: it worked! As Dr. Frankenstein declared in the movie: “It’s Alive! It’s ALIVE!”
This thing is FAST! Of course, it is currently one lawnmower lighter than it will be when it’s finished, but I am confident that it’s eventual top speed will be too fast to effectively mow the actual lawn. That’s okay, it doesn’t have to run at full speed all of the time. The Sabertooth seems to be pretty good at modulating the PWM signals, so with the DIP switches set for “Linear” mode (switch 5, for those who are interested), the speed is pretty controllable. I will be switching into “exponential” mode for the next round of tests, and see if the decreased sensitivity near the center of the sticks’ travel helps any. Since the bot uses tank steering, I’m not actually sure if that’s going to make modulating the speed of each motor to maintain straight-line travel easier or harder, but I suppose I’ll find out soon enough.
It’s the next day. “Exponential” mode didn’t really do much, so I switched it back to “Linear”. What I did discover, though, was that the fail-safes, which were a big feature of the BR6000 did not get set up correctly. When I tested the setup by turning off the transmitter (simulating a loss of signal), the right motor stopped, like it should, but the left motor took off in full reverse. This had the net effect of sending the bot spinning in place, and VERY fast. Not so good. I repeated the binding process a couple of times, with little success. Then, looking instead at the DX5e manual, and saw that I wasn’t quite doing it right. I was turning on the DX5e while holding the spring-loaded trainer switch, which was right, but was holding it up until the binding process was complete, then turning off the transmitter with the switch still held. This isn’t right, as it happens. I tried it again, this time only keeping the switch held until the light blinked, then releasing the switch, removing the binding plug from teh BR6000, then turning off the bot. I then placed the power plug back over the BAT pins. This time, when I switched off the DX5e, the left motor went into full forward. No real improvement. What finally did work was leaving the power plug over the set of pins I had used to power the receiver while the binding plug was over the BAT pins. Weird, but I can live with it for the time being. I’m a little bit torqued, though, that I seem to be losing a channel in the deal. If anyone has any further thoughts, I’d love to hear them.