I’ve been doing test runs with the new wheel casters for a few days now, and they are definitely better in some regard and not so good in others. One big advantage is the ability to go over cracks much easier and they’re more durable. However in my limited testing they’re not as good on door jams and similar obstacles.

Also, with installing the larger 3″ casters in the front foot my droid is now lifted and angled a tad.

Unfortunately, the shoulders and legs are fixed with no real easy way to adjust them and the outer feet are tipped with the rear much closer to the ground.

I’m not sure how much of a problem this really is, only time will tell. If it does become a problem some possible solutions might be

• Adjust the high of the outer feet drive system – might be tricky, and not sure I want to add any more ground clearance.
• Try and move the 3″ casters further into the center foot shell – not really an option as space and margins are already tight.
• Try a smaller wheel caster – probably the easiest thing to try, but I then loose some of the advantages of the larger 3″ wheel.

One of the issues I see right now are the ominballs in rear of the outer feet. Before I moved to the 3″ caster they didn’t touched the ground, this was intentional on my part. I could have easily removed them a long time ago, but they’ve gotten me out of a fix on more than one occasion.

With the belt drive system, the wheels are locked and hard to move when there’s no power applied, with the omni’s in place I can tip Artoo back onto them and maneuver him a little – not long distances mind you, but very handy in a pinch – So I’m hoping to keep them in place.

They are adjustable, but already on the max setting. One of the big problems is there’s always a chance the drive wheel could loose contact with the ground if the rear omni’s ride on something.

I have a big event coming up this weekend with lots of mixed terrain so the jury is still out if I’ll stick with the wheel casters, maybe I’ll try a smaller diameter wheel or even go back to the omni’s.

Related:

• Center Foot Wedges
• Caster Fix and Test
• New Wheel Casters
• Center Foot Ball Caster Maintenance

I had to add some wedges/stops to the side of the ankle to stop the foot from tipping when going in reverse. As usual rather than reinventing the wheel I copied Victor.

I made a quick paper template that I traced directly from the ankle/foot. I then cut them from some scrap 1/8″ aluminum I had sitting around.

And finished them of with a clean coat of white (including the screw heads)

I tapped the side of the ankle, and also included a wooden block inside (also drilled/tapped) just in case the ankle wall thickness wasn’t enough to hold the screw.

A couple of months ago I’d posted to the main astromech.net forum about modifying the JAG center foot to include a more pronounced skirt at the bottom. At first not everyone quite understood what I wanted to achieve.

Here’s some reference shot of the Uber R2 that may help

And here’s a screen cap from ANH with an even bigger skirt on the center foot

What set me down this path was seeing my droid next to Gerard’s (his is on the left). Note the difference in the center foot?

I think weathering helps define the foot as well, and to stops the white making the foot look more like a blank floating triangle/wedge.

This is Gerard’s raw Weaver foot

After a lot of debate most people finally got what I was trying to say, including JAG who may include the mod on the next run of feet. But I thought I’d see if I could modify my current foot in the meantime.

I created an little frame out of angled aluminum (1/16″, 1/2″x 3/4″), and the plan was to pop rivet it onto the existing 3/4″ lip around the foot.

It wasn’t perfect and I knew I wouldn’t be able to get the corners angled correctly, but was hoping to blend in the frame using bond or something.

Here’s the foot with the temporary skirt. A little too tall I think and I’m not crazy about the corners being straight as opposed to angled, so I’ve decide to see if I can find someone to lend me a welding ring to make a more accurate skirt/frame.

Well, I’ve fixed the caster problem after a bit of detective work.

A few of you wrote to ask if the 3″ casters were touching the side of the foot shell, they weren’t but clearance is tight.

To simplify troubleshooting I took the center ankle/foot assemble off Artoo and was able to duplicate the locking wheel on the workbench. It even did it with the foot shell removed.

Turns out it was as simple as some crud in the casters bearing. They’d been siting in my workshop for quite some time, so who knows what had gotten in there. And no matter how careful I was, there’s a good chance I got something in there while I was trimming the caster base plate.

The symptom was when the caster did a quick turn it would freeze and drag, slow turns were fine – but anything rapid would cause the lock.

I got some degreaser and soaked the bearing. I then blasted it with air to try and knock out whatever was causing the problem. Once I was happy the bearing was turning fine I re-applied some bearing grease using a small syringe.

And here’s a short video demonstrating the new casters in action

Only time will tell if the new wheel casters perform better in the field compared to the ominballs.

I made a bit of progress tonight getting the center foot ball casters replaced with regular wheel casters.

One of the issues with using wheel casters is finding one’s that will fit. Many builders opt to use smaller 2″ or sometimes 2.5″ casters, but I really wanted to maximize the size to help with difficult terrain. I was lucky to get some of the magical 3″ casters from the Ace Hardware store down in So Cal with the help of a friend.

At first I thought they weren’t going to fit, so I double checked with Victor to confirm that he and Mike really are using 3 inchers. He assured me that they were and sent me some current photos to prove it, along with a few suggestions and tweaks they used to get them to work.

My plan is that any mods I make will not interfere with me going back and installing the original ball casters – just in case the new one’s do not work. e.g. I’ll be using the same mounting plate but will drill some new mounting holes.

I knew I had to remove some of the lip on the front and back of the JAG center foot shell – which I did with the Dremel and a file.

I also had to remove some of the casters mounting plate to allow it to sit closer to the edge of the foot shell.

It wasn’t important to keep the mounting holes as I’d planned on drilling new one’s to match the original mounting plate used with the ball casters.

Here’s the first real test fitting before any holes were drilled

Like an idiot I forgot to take photos of the new casters bolted into place, but I did shoot this short video showing them in action

As I show in the video, ground clearance is now 1.5″ as opposed to the 1″ I had with the ball casters. But this maybe a good thing as I plan on adding an extra “skirt” at the bottom of the foot shell to mimic the center foot as seen some of the movies.

It was getting late, but I did get the foot mounted and a quick test run highlighted a problem with the rear caster not spinning fully. I’m really not sure why it’s not keeping up with the front. It spins freely when I move it by hand but when I run the droid it seems to stick to one side and drag. If I spin him enough it’ll rights itself and sometimes it does keep up.

Here’s quick shot of it sticking – I was trying to do a right turn when this happened.

I suspect it might be something to do with the fixed angle in the outer feet/legs, using wedges, which should be an easy test to see if it’s the problem.

I also need to add some wedges to the front foot to stop it tipping. With the three original ball casters located in the extreme corners of the foot this rarely happened – but now with the casters closer to the center it tips/drags very easily.

I’m definitely getting into the fun part of the project. I spent part of the evening trying to add a servo to the front holo projector. It was a little more tricky than I thought it was going to be. I’d seen photos and vaguely remember a setup on one of the tables at C4, but until tonight I didn’t understood the nuances of the angles and forces need to move something with a servo, so it was frustrating and fun at the same time.

Most people seem to use the same basic method to move pie panels, doors, holo projectors etc. And that’s to convert servo rotation into a linear movement using a push rod attached to the servo horn.

I’d bought some random mini servos from Tower Hobbies a while ago (Hi-Tec HS-55), and a push rod/linkage assembly from ebay. Problem was the servo was too small to fit the rod attachments. I managed to improvise and this is what I finally came up with

Please ignore the kinks in the rod, it really should be straight, but I’d tried to copy what I’d seen at C4 and got it totally wrong 😀

If I was to do this again I’d probably skip the pre-made assemble and make something in acrylic and parts from the local hardware store.

I also created a short video to summaries and demo the new setup.

As I explain in the video it’s not perfect, and I need to affix the servo more permantely to the dome rather than using velcro and I’m not entirely confident the linkage to the back of the HP will last very long.

The good news the Hi-Tec servo is definitely powerful enough to move the HP and will probably work for the pie panels as well.

I also need to decide if I’m going to add some code to my micro-controller to automate random movement.

See Also:

• Slip Ring Servo Test

I wanted to report that the new method of attaching the JAG legs and ankles worked. I didn’t have any problem with screws working loose while at WonderCon, unlike at C4, where I had to constantly tighten them up – and had two screws sheer on me.

Back in December of last year I’d decided to use pan head screws with a combination of fender and lock washers , instead of the recommended countersunk machine screws.

They’re do stand out a little but, but once I’d painted them up nobody really noticed them.

Related Posts:

• Finishing Battery Boxes and Feet

I’ve worked on a bunch of things the last few days, the simplest is probably getting my shoulder hubs finally installed.

I borrowed another idea from Victor and used rare earth magnets fix the hubs in place.

I started off by marking and drilling the holes. The magnets I used are 3/8″ x 1/8″ thick from MagCraft. There’s two magnets per hub, and two matching magnets inside the leg.

I’ve been doing a lot of drilling the last few days, and I’ve been extra careful to secure everything before I drill – even when I’m pressed for time or tired of doing it. You can never be too careful.

I then checked the polarity/orientation of the magnets to make sure they’d not repel another magnet. Otherwise I’d be in trouble when they’re presented to the magnets in the legs. Sounds simple but I almost got it wrong at least twice.
I roughed up the back of the magnets and epoxy’d them into place

I took two more magnets and epoxy’d them inside the leg stacked on top of small nuts.

Here’s a short video demonstrating the new magnetized hubs

I’ll try and post some more updates on new things in the next day or so.