Monthly Archives: January 2014

//January

Domeworks, part II

On the 9th of April, I posted a lot about how I was fixing up my dome for the first time.

I have, now and then put up posts of more work, but nothing has been as groundbreaking as what I’ve done lately!

A little while ago, I put up a post about cutting up the panels in the dome, well this is partly that, and partly what I’ve done with the opened panels..

First off, I’ve had a surveillance camera laying about for a while that I got from a friend of mine, Magnus Bodin, and I wanted to put this inside R2’s head, so we can actually see what he sees. An “R2-Cam”, if you will! =)

IMG_2699

So, the camera will go behind R2’s “radar eye”. For those of you not well versed in Astromech terminology, this is where it goes:

r2_eye

So, the place that looks like an eye, will literally become an eye. How imaginative!

First off, remove the surrounding blue aluminium radar eye housing, and expose the dome underneath. Please notice that I’ve added screws to the screw holes, since the dome is actually two domes inside each other, and I want to keep them as close to each other as possible when cutting through the almost 3 mm thick aluminium.

IMG_2684

I’ve tried to find the center of the circle as close as I can. In this case, it doesn’t matter if I’m not exact, as I have at least 5-6 mm on each side that I can be off, and it will still give more than enough opening inside the dome, yet still be totally invisible from the outside.

Let’s start with just using a punch to make a serious dent in this, so my drill can have something to cling on to!

IMG_2685

 

Ok … no turning back now!

One small hole, followed by a larger, followed by an even larger hole, to act as pilot hole for my hole saw.

IMG_2689

 

Yes, it is made for drilling in wood, but really, all I need is a serious enough groove in the aluminium…

IMG_2688

 

…. and a serious groove, that is!

Next up, bring out the trusty dremel and start cutting slices in a cross-wise fashion..

IMG_2690

 

No, that is not the death star laser. I had to pause here for a while, as my dremel was on the verge of burning up. White smoke emerged from within, and it did not smell good. Once all slices have been cut proper, I started with the outer layer, bent the pieces outwards and after bending back and forth, they snapped right off, as planned.

Next step was the inner layer. I used the dremel to cut along the edge, and bend the inner layer inwards back and forth until they snapped. Follow that up with a sanding disc, and you’re left with…

IMG_2692

A gaping hole! Fortunately it is in the exact right place. Funny enough, when cutting with the dremel, I took great care not to slip and accidentally score the surface of the dome. Why didn’t I take the same care when sanding the edges? my sanding disc slipped…. but not a big issue. Easy enough to sand that area and work away the rough pieces until I got a smooth surface again…

IMG_2695

 

This is just a test fit with the camera inside. Since the camera is booting up, you see the amber light surrounding the lens, but without that, it should be fairly invisible. However, since the lens is see-through, I did not want anyone to be able to see any of the LED’s on the inside of the head, or use the eye as a device to peek into the droid.

So, I needed a half-round object. Did not have anything at home that worked. I tried cutting up an old christmas ornament, but it was way too fragile (even though it was plastic) and I had to  make my own ..

IMG_2697

 

… made from duct tape. (No, I didn’t have any black duct tape at home, only the silver kind… so I spray painted the inside!

IMG_2696

Now, since the camera is light gray, I decided it was best to paint hat as well…

IMG_2698

 

A quick look at the mounting system inside, with the blocking shield in place…

IMG_2701

 

So a new look from the outside, looks like this:

IMG_2702

 

So much for the radar eye! What else can one do in a day? .. well .. time to work on getting the servos in place for the dome panels, of course!

After some trial and error I have determined that some of the hinges needs straightening up, and since they are only glued in place with E-6000, I can absolutely move them.

I have 3d printed a block of plastic with a curved surface that matches the inside curve of the dome, and I’ve super-glued them to my servos, and then used a very high bond tape (120 kg draw per meter) to stick these to the insides of the dome. At first I was a bit worried that they might not be enough, but after having to reposition them slightly, I am no longer worried. took me 5 minutes with the help of a scalpel to be able to pry one of the servos loose!

IMG_2700

 

So, servos in place … time to connect them to a small servo board…

(powered by an arduino), add a bit of test code…

IMG_2706

The end result? … well … check for yourself!

 

 

 

By | January 14th, 2014|R2-D2|2 Comments

Fixing a hobbled droid, part II

So, once the shafts were drilled, it was time to play the waiting game.

I wanted the wheel hubs to sit firmly and not spin, but also retain the option of some day remove them, in need. Not saying it has to be easy, just … well .. possible! =)

In order to make sure the grub screws stayed inside the grooves in the motor axles, I ordered a bottle of this:

treadlock

Loctite Thread Locker! High Strength! High temperature! Now they should be really sticking in there!

But wait…. if I can’t get the motor hubs off .. what happens if the screws holding the motors (that are right underneath the big hubs) comes loose, I can’t fasten them… Better loctite them down as well! (medium strength thread locker this time around).

Hubs in place, sticking really hard. I almost completely destroyed a hex screw trying to (by force alone) unscrew a grub screw once set with the high strength thread locker. Not gonna happen.

Last step for the feet: screw the motor plates together, and also add a small holder for the bottom of the white insert.IMG_2705

 

 

OH, and before you ask: “Isn’t that James Short’s design on the motor holders?” Let me clear up this once and for all. It is not. Not one bit. In fact. James will see this for the very first time after these are posted, and invent a time machine, go back in time, and make a set that looks 99% identical, make a blueprint off them, and send them to my mailbox in the past, so it looks like he might have emailed me the plans for his, and then, based on those, I’ve made these, with just a tiny, minute difference (that he even will be so cleverly evil to put into his blog post about his motor holders that he should have done (again, posted in he past through time machine so it looks like it predates mine)).

I’ve also replaced the center wheel to the same type of wheel I had previously, as it lifted up the center foot better from the ground, and should work great with these wheels.

Step three of this, is to drill upp the holes in the styrene ankles. I suspect (well .. quite strongly, in fact, since I just drilled them freehand without any pillar drill or anything. Basically eyeballed it) that the holes through the ankles of my droid is not perfectly perpendicular to the surface they are in.

This might be why my droid is veering off to the left all the time when driving!

My solution for this is simple. The bolt going through the ankle is 8mm, I’ve got a 10mm metal tube inside the hole through the ankle to protect the styrene. I just need to firmly stick the ankle to something straight, drill a new, bigger hole in its place, make sure the plastic doesn’t try to veer off in any direction (Might even mix up a batch of styrene putty and pack tight into the ankle hole and let dry first, to make sure this doesn’t happen easily.), then insert a new metal insert into this bigger hole. That is another job for another day. I can do the prep job, but drilling the holes and making new inserts on a lathe is something my friend Anton will have to help me with.

By | January 14th, 2014|R2-D2|1 Comment

Fixing a hobbled droid

The past year, I’ve taken R2 out on various outings and had lots of fun.

However, I’ve made the biggest mistake of them all.

“If it ain’t broken, don’t fix it!”

I had a working drive train with R2 using scooter motors, but I wanted to change into a better drive system that uses up less power and weighed a lot less than the old system.

I opted for using Jaycar motors, and found a wheel with a hub that was made for mounting directly on the the drive shaft. Only problem: it had another dimaneter than the drive shaft on the actual motor.

Not a big issue, I found a piece of fitting aluminium pipe and cut that off to the correct dimensions and all is well.. or so I thought.

On the first real outing, I encountered a piece of thick, sticky carpet, the worst kind of surface to drive on –ever! Half an hour into the event, He stopped moving. HE. STOPPED. MOVING! .. Panic!

So what it turned out to be, was that the small grub screw that locked the hub on to the ankle had slightly loosened, and the entire hub (with wheel) was just spinning loosely around the shaft.

So, I had to do something about this. I worked on another system, based from fellow UK R2 Builder, James Short, using two drive motors in each foot, hoping that the added torque (and better fit with better wheels) would prove better and fix the issue.

IMG_2629

 

IMG_2628

Well … it was better, but it did not fix the problem.

The grub screw still comes lose and skids around the shaft, damaging it, and makes R2 immovable. This time we had two wheels on each side so it is a lot better but not good.

This is what the shaft looks like after a short while. Note the damage on the left side. The dent is due to the grub screw slipping.

bild 2

Removing the hubs from the shafts turned out to be quite difficult. I had to oil up the shafts and start the motors running at full speed, while trying to pry and slip the wheels off.

I asked around for various solutions, and fellow R2 Builder and good friend Sigge, checked with a few people at work, and basically came back with the best option: To drill a small hole (not through) into the drive shaft, so the grub screw would have somewhere to really stick, and seal the position of the grub screw with a thread locker, such as Loctite 272, the hardest out there.

Drilling a hole in to the shaft turned a bit easier than I thought, possibly due to me just byuing hardened steel drills that were up to the challenged. I started with a 1,5 mm hole, went up to 2,5 mm on the second time, and finally settled in a 3,2 mm hole. The hole itself is about 1,5 mm deep into the shaft, so just deep enough to allow the grub screw to take hold.

bild 1

Not bad for being done freehand without a pillar drill!

I am awaiting new wheel hubs, since they old one did not get through the removal process untainted, and the thread locker. Once that has arrived (should be next week) , the wheels will be assembled again, and hopefully it will all work out as intended!

My last option, is to drill down through the hubs and shaft all the way through and insert a steel pin, but that will severely weaken the shaft… but if worst comes to worst, that is the last option! =)

So, for now, this is the situation. Next up: time to rewire the electronics in the dome and mount those servos!

By | January 5th, 2014|R2-D2|Comments Off on Fixing a hobbled droid

Building R2-D2 is Stephen Fry proof thanks to caching by WP Super Cache