When we had finished filming in 2000, SMIDSY was virtually undamaged, we
could have recharged the batteries and fought again (if we hadn't lost). For
2001 we didn't need to do anything more than replace the fake leather
covering and fill in another application form. Except that we wanted to do
even better, and the weight limit had gone up to 100kg and it would have
been silly to enter an 80kg robot, so we had the option of entering a
completly different robot.
Looking over our performance last year a few things were obvious candidates
for change:
- We needed a better weapon.
- The jaws weren't powerful enough or fast enough to lift anyone.
- We were able to push some other robots around but Panic Attack was
stronger than us.
- We wanted to make it easier to steer - driving around the barrels
three times before knocking them over was embarrassing!
We considered building a huge walker (because of the higher weight limit)
with a big spinning disk around the outside, and an axe on a turret. It
would be slow but the other robots would have to approach the disk, and we
could swivel the axe to meet them. However we thought it wouldn't be very
interesting to watch.
We reconsidered the design and decided that 4 wheels still gave better
traction than 2 and were less prone to getting stuck on things. We still
thought that an invertable design was better than a self-righter because you
can start driving much sooner after you've been flipped (and we hate the
term 'shrimech').
We tried to think of a way to prevent spikes getting stuck in our wheel
arches (which is how Panic Attack beat us) but decided it wasn't possible
and opted for a better weapon instead. The best defence is a good offence,
and all that.
Weapons
We didn't want to lose the jaws, as they are a bit of a trade mark and stop
us looking like yet another wheeled box but we had to get them working
properly - in 2000 they couldn't lift enough weight and were still too slow.
They also warped slightly from repeated ramming. Everyone said we needed a
better weapon, and we eventually lost because Panic Attack was too tough to
be smashed apart by ramming (and he wouldn't stand still long enough for us
to get a good run up, most unsporting ;-).
We looked at the different weapons that are allowed (all the really good
ones like anti-tank rockets and thermic lances are banned). Basically they
fall into 4 catagories, ramming, flippers, axes (including spikes that fire
in a straight line) and saws (including Hypnodisk).
- We had tried ramming, it works but needs good driving and an almost
stationary opponent.
- We thought another flipper would be dull, there are loads already and
almost everyone can now self-right.
- Axes are too difficult to aim, it's difficult to make a good one and
even the good ones do little damage.
That left us with saws. Dead Metal's is very impressive but less powerful
versions just jam. Even Dead Metal's saw needs quite a long time to do work,
and he needs to catch and hold his prey first. Hypnodisk was by far the most
destructive robot for the last two years. There were a number of other
spinning disks that didn't do as well (Banshee and Warpig). They did as much
damage to themselves as to their opponent, we took this to mean that this
kind of weapon needs careful thought and proper engineering but has the most
potential.
Having decided to use a spinning disk, we had to work out where to put it.
It could go right around the robot, doubling up as armour, but the bearings
would be really difficult for an invertible robot and we'd lose the jaws. We
could do what Hypnodisk does and have really small wheels with the disk on
top, but that's not invertable at all. We decided to have a small disk,
between the rear wheels. To make up for the lack of size we can make it
thicker and run it faster. Fortunately, because our design has the disk in
the middle, we can support both ends of the disk axle. This means it's held
firmly enough to be spun very fast indeed. We couldn't make a disk as thick
as we wanted so we had two disks made and mounted them with a slight gap.
This protects the drive belt but is mainly because it looks good. We
anticipate a number of robots will have had disks added since last year and
we wanted to be a bit different. To make room for the disk we made the robot
slightly wider and moved some of the electronics from the back to the front.
Better Driving
We also wanted to improve the driving characteristics. Most people don't
realise that a robot is really hard to drive - they just think that most of
the roboteers are bad drivers. In fact a robot is much harder to drive than a
radio controlled car, especially in the heat of battle. The skid steering
means that attempting to turn while driving tends to result in
a spin. When you're spinning on the spot it takes a lot of power to start a
robot spinning and once it starts spinning it doesn't stop straight away but
carries on past the heading you were aiming for. Even just pushing the stick
forward when the robot is stationary doesn't always do what you think! Often
one side will power up slightly faster than the other, or one side will slip
and the robot will turn slightly before it starts moving. One of my lego
test robots would always do a 90 degree turn (in a random direction) before
starting, it can be that bad. We wanted more pushing power and better
control over our spin rate.
I really liked the steering mechanism Killahertz uses, it very cleverly
combines car type steering (which is good for high speeds) with skid
steering for low speeds. Unfortunately it's not 4 wheel drive and pushing
power is even more important than control. We considered using hydraulic
transmission to make a 4 wheel drive version. Hydraulic transmission would
be really nice for other reasons too, it means you can use a petrol engine
without worrying about stalling or reversing and you can have automatic
gears, giving you more pushing power and a higher top speed. Maybe we'll do
that next year, but we didn't have time this year.
What we did instead was to fit bigger speed controllers. The old NCC70s,
although recommended by both 4QD and the motor supplier (Ellis) don't give
as much power as the motors will draw when stalled. The new ones will do, so
we should be able to push twice as hard; they'll also burn out our motors if
we're not careful. The new robot is slightly wider, this should make it
easier to start spinning as well as making the resultant spin a bit slower.
The Jaws
In 1999 the jaws were fairly powerful and reasonably fast but the mechanism
kept breaking and took hours to strip down. It also got in the way of
everything and weighed far too much.
In 2000 we changed it so it didn't break or get in the way, but it wasn't
nearly powerful enough. We changed the gearing so it was more powerful but
it was then really slow - we couldn't keep another robot on the jaw long
enough to lift it! (and it was still too heavy).
This year we've made our own rack and pinion system, it's much more efficent
so it's faster and more powerful. It's also smaller, lighter and doesn't get
in the way of the other bits. We've also remade the jaws to be a bit
stronger and stiffer and fitted a tool steel blade that's sharp enough to
shave with :-)
We'll just have to wait and see if we've made the right decisions.
Last modified: 2003-08-04 19:28:24 127
