Date: Tue, 31 Jan 2006 00:50:15 -0600 (CST)
Subject: schedule off by two weeks but looking good
X-UID: 130
Content-Type: IMAGE/JPEG; name="img2579.jpg"

I slept most of Saturday so was one day behind this weekend. It was into 
Sunday morning and I had hours more wiring to do. Very tired and without 
any functional testing, I had to call off any end-to-end test outdoors.

Tonight, I powered up the onboard computers (with the current limiting 
power supplies - be safe not sorry) and connected the motors to the 
batteries. So far, everything looks good. What a difference a few days 
makes. Last week, I thought I had burned out both computer boards.

The steering motor is undervolted and a little sluggish. However, it 
should have enough torque. It is a low priority fix. Most importantly, the 
ball bearing assembly epoxied on the end of the cordless screwdriver 
works. The gearbox does not jam now. Every rotating shaft on this robot 
has wound up with ball bearings whether originally designed that way or 
not. Nothing mechanical works without bearings.

If I have time, I'll make a 12 volt bipolar battery pack using the solder 
tab battery holders to power the steering and electronics cooling fan.

Most important is that the drill motors spin up. The rear wheels are 
propped off the floor with a .30 caliber ammo box and telephone book under 
the rear of the frame. This prevents it from running away.

At 2 KHz PWM, the motors emit a loud whine, kind of like an electric car 
from a 1970s science fiction movie (seriously, it sounds just like that). 
All of the electronics and batteries twist in the frame under the applied 
torque as they are connected to the drill motor mount. I'm not sure if 
it's good to leave it like this. It looks cool, like the robot has a lot 
of power. If you've ever seen a motor in a car's engine bay twist in the 
motor mount as it revs high, this is the effect the robot has except 
everything on the outside moves as there's no hood to cover it from view.

So it has power. Looking at the wheels spin, I'm not sure what it's 
ultimate speed may be. I do think that driving it at high speed is ill 
advised until the control software can handle it. I only drove it at about 
40% duty cycle for one second. That was enough to worry me about the robot 
breaking itself.

The robot is heavy. I didn't pay much attention to weight while designing 
or building it. This shows. The front rides a little low although is still 
adequate. The lexan crash shield at the front may have real use in case of 
bottoming out on the ground.

My plan at this point is a photography session outdoors next weekend. 
Mostly, I want some glamourous looking pictures of the robot for the 
website. Some video of it moving would be nice too. Then in the week or 
two after that, the robot will be announced to a technology news portal 
site after I've added a clearer explanation and narrative about the robot.

For engineering, I can see that now a huge amount of work is required 
developing software before the robot is ready for real world operation. 
It's taken about one year (most of that being my basic electronics 
education) to build the hardware. Future development is dominated by 
software. It's too bad that you can't just buy robots like this off the 
shelf. Then there'd be all sorts of cool software written for them 
already.

As it stands, the remaining hardware work is:

1. wire/mount Sharp IR distance measuring modules and Futaba gyro
2. wire/mount steering potentiometer
3. make screw-on antenna mast (not much to do for this, mostly done)
4. cable the front wheels to steering spindle
5. wrap loose wires to prevent tangling and snags
6. tighten some screws/nuts and tape over exposed contacts for safety