Archive for the 'Progress Report' Category

Progress Report: Measuring and Propeller Arrival

Published by admin on September 28, 2011 in Progress Report. 0 Comments Tags: , , , .

Taking measurements

Got a caliper from the local hardware store and made measurements of the model parts for interfacing. Did all measurements in millimeters. Mostly because it’s a Japanese model and so the original design was probably done in metric. I’m unsure of how fine the 3D printing output will be. I’m going to have to check out the documentation again to see about their accuracy.

Parts in SketchUp

Instead of spending a lot of time hemming and hawing over which 3D package to try out, I just decided to go with Google SketchUp. Spent about an hour watching their instructional videos and then just jumped in. Most valuable lesson learned so far: Pull an object into 3D before adding anything to it. Going to have to design everything with an eye towards keeping it all simple, and all walls thicker than 2mm. Looks like translating the files into the final printing format (STL) has the potential to be kind of painful.

Propeller Board

Also, the Propeller board has arrived. Originally I was hoping to get this done with a Basic Stamp, mostly because I’m already familiar with the platform, but my plans include a feedback system for the motors now, and I know from experience that it’s going to exceed the limits of what I’ll be able to get a Stamp can do at one time. Sadly, Parallax was out of the cheaper, more interesting looking Quick Start Propeller boards. So I went for the Gadget Gangster. On the plus side, I’ll basically have every single pin of the micro available. Having built a couple of carrier boards for microprocessors in the past, I’m really looking forward to using a solution that someone else has engineered.

Theo Jansen Gakken Kits – Assembling

Published by admin on September 19, 2011 in Progress Report. 0 Comments Tags: , , , , , .

Theo Jansen Gakken Strandbeest finished

The kits are really cool. I built half of the rhinoceros kit, and assembled the entire Strandbeest. The kits go together differently. The rhino kit is more like a conventional model kit, except that it all snaps together and so there’s no need for glue. The strandbeest kit can also be assembled without glue, but it uses a clever interlocking hub system for all the parts to go together. It’s a little hard to explain how it works without showing someone. (These can be ordered on eBay, here. Alternatively, a person can order them for many more clams via Amazon, here: Strandbeest and Rhinoceros.)

Here are the kits in action. The rhinoceros is a lot easier to make walk around than the strandbeest. The strandbeest doesn’t really like to walk without the wind powering it. This is kind of too bad because it makes the thing a little bit impractical, unless you’re going somewhere windy, like a beach. The key to getting the strandbeest to walk around by pushing it is to push it low on the device. It definitely doesn’t like being pushed or pulled from the top.

It’s cool how the Theo Jansen creations all have repetitive parts. It makes them look more natural or something. They end up with a complexity that your brain can’t really sort out. Both kits are easy to split into their component parts. So there’s twelve legs in each complete set. A person could very easily modify a kit so that there are only eight legs. Also, a person could split the device in half or thirds rather easily.

I’m pretty sure that the next step is going to involve purchasing an accurate measuring device. I need to take some measurements that can be turned into 3D models. I’m also trying to figure out if I can get my idea to work with a Basic Stamp or if I should try a Parallax Propeller. It definitely looks like I need to implement some shaft encoders for the project. The problem with going with a Propeller is that I’m going to have to learn the system and (admittedly dead easy) language. It’s still time!

Work With Basic Stamp 2 and Hitec HS-55 Hobby Servos

Published by admin on September 13, 2011 in Progress Report. 0 Comments Tags: , , , .

I pulled out my Basic Stamp 2 Board of Education to help with the hacking of some Hitec HS-55 Feather Servos (these servos can be bought here on Amazon). These probably wouldn’t have been my first choice for hobby servos, because I wanted something with metal gears, but I was able confirm online that these could be hacked for continuous motion, and was able to find several different sets of instructions on how to accomplish the task, here: one, two and three.

Hacking hobby servos

At first, I decided to follow the instructions in the first one pretty much all the way, except to avoid soldering the resistors directly onto the board of the servo controller. I’m not good at soldering tiny things onto other tiny things. Especially when they can be damaged by shorts caused by errant solder. I just won’t do it. So I soldered some resistors onto the ends of the wires that connect to the internal potentiometer. Not to get too deeply into the process here, because these three other articles covered it so well, the process is:

  1. Open the servo.
  2. Pull off the gears (make note of the order first!)
  3. Unscrew the screw that holds in the potentiometer.
  4. Pull the pot out and cut off the wires (make note of which go where!)
  5. Hack the pot so that it has no stops and put back into the servo.
  6. Put the screw back in.
  7. Solder resistors to wires to replace potentiometer.
  8. Cut off a stop built into the top of the servo case.
  9. Reassemble everything, tucking the resistors back in to the servo.

So, as noted, I used resistors, and even found some in my electronics parts stash that were around the right values and had a 2% variance rating. After hacking them, I got out the Basic Stamp 2 and the Board of Education and tested them both. They worked fine, but to my dismay neither of them seemed to “center” where they should. These hobby servos (like most standard servos) are supposed to sit still at 1500 uSec. They both moved quite a bit. I also found that one was only moving in one direction, no matter what I changed the pulse width to. Of course, I’m sure this is due to my bad soldering skills and the sensitivity of the equipment!

As a result, I decided to try using an external potentiometer to replace the internal one, so that I could adjust them to center on 1500 uSec. I pulled out my potentiometer bag from my electronics stash and after quite a bit of trial and error, I settled on two 2 K ohm, linear potentiometers. I tried some larger standard pots, and also tried some with extra resistors in series and etc. Somewhere in the process I managed to fry one of the servo control boards. (Luckily I ordered an extra with the idea that hacking these might prove fatal to at least one).

The most painful lesson of the day: Don’t forget to attach all grounds when dealing with separate power supplies for the logic and motors!!!!

It seems kind of silly to attach so much gear to a servo the size of a quarter, and maybe later I’ll look for a cleaner solution, but for now these are working great. I’m able to adjust the potentiometers such that there isn’t any movement at 1500 uSec pulse width. Sadly, these two do not seem to be willing to move at the same speed, so sooner or later I’m going to need to implement a shaft encoder system of some kind. Part of me is not looking forward to that. The mad scientist part of me is delighted, however.

Of course, moving to a shaft encoder system is going to require a different microprocessor setup…

© 2009-2012 Robot Vacuum Cleaner and iRobot Roomba Reviews All Rights Reserved