Friday 18 October 2013

A2, some pictures and videos

Awhile back, I posted a few fuzzy iPod pictures of A2, my most recent cheap-hacked-rc-car-chassis-robot and promised some pictures and videos, well, in no particular order, here they are:













Someday over the rainbow I will make a proper LMR post about this guy (perhaps when I get a GPS functioning on him) but for now, here are all the recorded runs that I have of him obstical avoiding, the first 3 are videos of him, and the last 2 are from the camera mounted on his nose.


A HOG drive, take 2 pt.2

I was going to update my Going Whole Hog post on LMR, but I have been pretty darn busy whenever I have internet acess, so I think I will just throw these pictures up here with a quick discription.

I finished up the chassis that I started in the last HOG drive post:


This one was slapped together (yeah, I know, you can tell) but I tried a couple of new things on it, compared to the first version: The new gimbal design, smaller, tougher, better. Mounting the battery on the front axle, giving it a much better C of G, as well as more weight on the front wheels, which prevents them from lifting up in a tight turn. Wider wheels with better grip, and a wider front stance, partially for the same reasons as the battery, and also for turning and tracking straight, I needed more grip.

Because it was slapped together, it wasn,t very strong, and I did not like a few things about it, asmyetrical steering because of the front swingarm not being parallel to the ground, general shoddyness on the build which caused it to fall apart whenever it hit something, that kind of thing. So I built a new one, from the ground up. The only thing structurely that is the same on this one is the rear gimbal, which proved to work great.




Learning from the lessons from the first two, I implemented better wheels, the battery on the front, the improved gimbal, and a stupidly sturdy frame. The frame is made up of 3 layers of studful technic beams arranged in the pattern you can see in the last pic. The top and bottem frame layers are identical, they are fastened together with friction pins, and have a layer of overlapping bricks in the middle fastening the two layers together. The two servos are fixed in place by the shape of the frame, and two rings of zip ties each. This seems to work rather well, and I have had no problems with them. The frame also has zip ties keeping it together in various places, making it extremly sturdy, I have crashed into too much stuff already, and it hasnt shown any signs of breaking or loosening. I also replaced the fan for the ESC, as during one of the previous versions of GWH, one of the fan blades had broken off making the fan unbalanced and very noisy. I ripped the fan you see off of a broken 2000W inverter.
At the moment it is still an RC beast, but hopefully soon I will mount some sonar on the front and a picaxe or arduino on it let it rip. The way these things are made, it has two throttles, one controling the speed of the motor, and the forward/backward servo/gimbal so I will let the microcontroler have control of the forward/backward servo and the steering servo, and keep the ESC throttle on manual RC control. This will let me have a panic kill switch, and also fine tune the speed for the amount of space it has to run on. So far I have only been able to run him around my kitchen and a little on the wood floor in the house, but I stopped on the wood floor because the rubber wheel leaves little rubber streaks.
Here are two videos of it running on my kitchen floor:


 
And a video that shows the basic movements that the servos do, along with how that changes the angle at which the hemisphere touches the ground.             



Air cannon

My title is sadly lacking in pazazz, but it aptly describes what this post is all about: My (right now at least) only pneumatic projectile firing device. First, a cupple of pics:






As you can see, it was constructed from two discarded propane bottles that have had their valves unscrewed, leaving a standard 3/4 in NPT threaded female fitting. These are screwed into custom manifold made from PVC pipe and fittings that leads to an 1/4in "M" type quick connect for connection to a compressor, and to a modified 1in sprinkler valve. The valve was modified by removing the electrical selonoid, filling the hole it screwed into with epoxy, and drilling a hole in the centre of the device, threading in a 3/8 to 1/4" metal bushing ( I orriginally had just the 1/4" fitting coming off the hose, but stress on it caused it to move a bit and leak), and finally threading in a coiled hose that leads to either a blow gun or (as I have right now due to lack of parts) a simple ball valve. When the blow gun (or ball valve) is opened, preasure on one side of the diaphram in the sprinkler valve is released causing the diaphram to open and lots of air to rush by. After the valve is an adapter that goes from the 1" female output of the valve, thru a cleanout tee and to a 2" male threaded fitting. I have another large combustion launcher that uses a 2" male threaded output, so with this I can use the same barrels on both of the cannons. The pictures are with my (rather pitiful) 1.5 in barrel, but I also have a short 4" one (you can see that one in the pics too), a 10 foot 2" one, a 20 foot 1" one, and an 8 foot 1/2" one. even with the inch and a halfer you see on there, I managed to put rocks and 1/2" pvc thru a large metal can, split a 1 foot long 2x6 piece of wood with a rock and break into multipul pieces a 2.5 inch thick paving stone using a metal pipe nipple.
I also once shot an unfrozen paintball thru a piece of 1/2 in thick wonder board using that 20' long barrel I mentioned.( for those of you who do not know, wonderboard is pretty much drywall, except made from cement.) Anyway, you get the picture, pretty powerful device, and lots of fun. Two last things before I go, one, if you are going to experiment with potato cannons etc, please wear, at the bare minimum, safety glasses, and do not use sun faded/scored/cracked or un pressure tested pipe, and two, NEVER, ever point these at anything living, or that has a chance of ricocheting back into you. I know, I hate dire warnings too, but you are dealing with stuff that WILL kill you given the chance.

Thursday 17 October 2013

This T-shirt is just awesome.

I follow Letsmakerobots.com fairly closely, and this morning when I logged in, I found an interesting post.
You can buy this tee shirt:
from the site teespring.com as part of a sort of test-run by one of the LMR members. Here is a link to the LMR page that explains it a little better than I can, or if you want here is a link directly to the t-shirt buying site. In order for this to run sucessfully, he needs 30 orders by the end of the week. You can pay with Paypal :)

If you aren't convinced already, watch this video:T-shirt video

Tuesday 8 October 2013

Little Victories :)

Awhile ago I got an Adafruit ultimate GPS module that runs at 10Hz with all the goodies. Awesome! but what now? Her sample sketchs were... Complicated.... And I am, well, not the best programmer, to put it lightly. I loaded up the echo sketch, ran it at 10Hz, and it went fine :) tried the parsing sketch, fine, parsing sketch at 10Hz, not so fine. That and I didn't understand what the heck was going on. Fast forward a bit and I find that arduino 1.0.5 has a CSV parsing function built in. Awesome! sit down and figure this should be pretty easy to implement. Nope. Struggled for 2.5 hours without sucess, and then finally, I UNDERSTOOD! I had to wait till the serial buffer was almost full, if (GPS.available() > 62) not just greater than 0, and I figured out that GPS.find("$GPRMC"); The GPRMC had to be in quotes for it to work. So now I have a sketch that just needs to parse the CSV's into some floats, and I'm golden :) Here is the whole sketch as it stands. It brings the CSV strings from the GPS in over softwareserial, parses them, and sends them out over serial to my laptop running the serial moniter. It refreshes every 15 seconds right now, because I am going to glean some real life data as I walk to school today, and perhaps post that later. Right now the GPS is spitting out at 1Hz, 9600 baud, it's default setting, but this SHOULD work with heavier duty speeds too. It uses the same softwareserial pins as in the adafruit tutorial, and spits it back out over the USB port. If anyone actually reads this, feel free to use however you wish, it should just copy and paste into arduino and compile correctly. It does in arduino 1.0.5.

#include <SoftwareSerial.h>
SoftwareSerial GPS(3, 2);

void setup() 
{
  Serial.begin(9600);
  GPS.begin(9600); 
}

void loop()
{
  if (GPS.available() > 62)
  {
    GPS.find("$GPRMC");
    float tim = GPS.parseFloat();
    float lat = GPS.parseFloat();
    float lon = GPS.parseFloat();
    float spd = GPS.parseFloat();
    float hdg = GPS.parseFloat();
    Serial.print("Time: ");
    Serial.print(tim, DEC);
    Serial.print(" Lat: ");
    Serial.print(lat, DEC);
    Serial.print(" Long: ");
    Serial.print(lon, DEC);
    Serial.print(" Speed: ");
    Serial.print(spd, DEC);
    Serial.print(" Heading: ");
    Serial.println(hdg, DEC);
    delay(15000);
  }
}