At the recent Clone Wars events it became clear that the little Power Distribution Board that was supplying power to the dome servos wasn’t going to cut it. It would work for a while, but the onboard 7806 regulator would soon overheated and shut down to protect itself.

I’m not surprised, I currently have 8 servos in the dome, 4 for the pie panels, one for the front Holoprojector, one for a side panel, and two for the periscope - with more to be added. Idle, each requires around 8-10mA, but when active they can easily require an Amp or more under heavy load, which can add up fast when you have a few things going at once.

Space is tight on the power board so it’s hard to add decent heat sinks, and even if I could I wasn’t convinced that the single 7805/ 7806 (which can deliver a maximum of 1Amp) was up to powering all the servos from one IC.

One of the problems with servos is that they can sometimes stick and not return to neutral fully, and will continue to draw more power than necessary. Anyone who’s played with servos will know that if they’re even slightly misaligned or put under stress in the neutral position will cause them to hum (which means they’re drawing more than the normal 8-10mA). I’m pretty sure that’s what happens on some of my dome servos, but not all the time. It’s virtually impossible to adjust them all to be perfect, especially as things can move or shake in a fast spinning dome.

The dome is fed by a 12VDC supply coming up thru the slip ring, and servo/signals are connected to a custom board at each end of the slip ring. The 12VDC routes to the power distribution board in the dome that then re-creates the various voltages I need, from 5VDC all the way to 24VDC.

The solution to the servo power/overheating problem is to use a better DC regulator to replace the 7806, but I had a bunch of events coming up fast, was short on time and resource, and basically needed to fix the problem overnight.

So I create a simple power board with parts I had on hand, including three 7805 regulators which would power only 4 servos each, plus I added a big heat sinks on each one and a fan to help keep them cool.

Here’s the board coupled with the original servo board. It’s somewhat clunky and I will replace it at some point with a more streamlined solution and a single better regulator, but for now it worked fine to get me thru the events.

If you’re interested in makng your own servo power supply or even a fully blown Power Distrubution Board, it can be done for not much money. 7805 regulators are easy to find, most surplus stores carry them for around 50c, or at a pinch you can pick them up at RadioShack for under $2. Versions that can deliver higher loads are also available, but can be harder to find or sometimes a lot more expensive.

Google on “7805 circuit” or “7805 power supply” you will find dozens of examples of simple circuits to make your own boards. Here’s a good example.

The regulators also come in 6, 8, 9 and 12V varieties. So as I said you could make a distribution board for not much money if you’re up to the challenge.

I also have to stress again that I’m not an electronics expert. Please double check anything before trying to copy me. In this case there are definitely better was to ‘fix’ this problem.

Related:

• Electrical System Update
• Slip Ring Spin Test
• Slip Ring - Cutting the Frame
• Slip Ring Update
• Slip Rings
• Variable 12-24V Converter

I’ve been tardy posting updates this last week - been busy trying to get stuff back together for the new Clone Wars movie.

I finally got my periscope installed and functioning. Here’s a short video of it in action

There’s still some small quirks I need to iron out, but I’m pretty close to being done with it (for now). One of the big issues is that everything has to be align perfectly, and taking the dome on and off isn’t as simple as it used to be. What I’ve quickly realized is that with every gadget I add maintenance goes up exponentially.

I repainted the periscope housing after tweaking the curve at the top to match my dome. While I was at it I clean up some spot that I wasn’t happy with

Right now the dome pie panel velcro’s on top of the periscope, but it’s hard to get it to sit flush with so much spring in the velcro. It also very hard to align things perfectly and to get it to stay put. I’m probably going to use magnets instead or even use the small side screws I added. Right now they’re just for show to mimic the ROTJ periscope.

I’m not convinced that the clear lenses look correct either, it’s hard to tell but I suspect the original had a slight frosting to it.

I really need to document the wiring and micro-controller setup for this, but here’s a few overview shots of the periscope installed and there are more in the gallery.

Right now, up/down is manually controlled by the Futaba transmitter, at some point I’ll probably automate this. I also have a small 08M Picaxe micro-controller dedicated to automatically rotating the periscope when it’s fully extended. In addition I have to route 12VDC from the main dome power board back down to the periscope light circuit. The Picaxe and the small mini-servo that rotates everything runs on 5VDC and just taps off of the power for the lift servo.

There are two small micro switches, the first trigger the micro-controller to start randomly turning the periscope when it reaches the top

The second turns on/off the lights when the periscope is raised/lowered

Related:

• More Periscope Work
• Lift Mech Support Bracket
• New Periscope Lift Mech Prototype
• Periscope Port Detail
• Periscope LEDs and RoboteQ

I’m still fiddling with the periscope, I added some LEDs, a servo to make it turn and something to mount it to the lift mechanism.

But first I had to adjust the dome pie panel cut out to give more clearance. I marked off the area I wanted to cut and Dremeled in a few slots to start things off and finished it with a small hacksaw blade.

The main periscope assemble is mounted atop PVC tube to add more height.

I wanted to easily remove it for maintenance and also allow some height adjustment at the same time. This little tube is fixed to the lift mech

And the main PVC riser tube slips over and locks into place with a couple of screws

I installed a small HiTec HS-55 servo into the top of the PVC tube to turn the periscope side to side when it’s extended. Attached is a round plate the aluminum based of the periscope will screw into.

I’ve also add the LEDs to the main periscope housing. On the front is an array of 6 rectangular LEDs. These are very close to one’s used on the original ROTJ periscope

They’re glowing orange in the photo but they’re deep red in person. Power is 12V and no resistors needed as they’re in series

Here’s the block of LEDs soldered together before I installed them

I’ve wired in some bright white LEDs inside and some colored one’s on the back of the housing, I also re-appropriated one of my old PSI boards to blink a couple of LEDs to add a bit of variety.

Getting closer

Related:

• Lift Mech Support Bracket
• New Periscope Lift Mech Prototype
• Periscope Port Detail
• Periscope LEDs and RoboteQ

I replaced the original drawer slider bracket with something a little bigger. It’s cut from a piece of alu angle and makes the whole setup a lot more solid. You can see the original bracket in the background.

The slots make it easy to align on the dome ring

I still need to work on the riser below the periscope assembly and fix it to the lifter.

I’ve been busy prototyping a new periscope lift mechanism for Artoo.

Some of you may remember my earlier posts and videos demonstrating a mechanism I’d purchased from Daniel, unfortunately I gave up on it a very long time ago. In my opinion the design was flawed from the start and I never did get it to work well, but not for the lack of trying on my part.

Now that I’ve finished adding servos to the dome I was anxious to get the periscope working. I remembered seeing photos of Wayne Orr’s lift mech and thought I’d try and do something similar. Rather than fabricate something new he uses a drawer slide as the basis for the design and a small motor to hoist the periscope up on a platform.

The only catch with Wayne’s implementation is that the periscope and life form scanner mechanisms extend deep into the droids body. I don’t have this luxury as I’ve left the top ring cross bar in place to support the slip ring. So anything I come up with would have to fit within the approximate 10″ height restriction of the dome. The periscope assembly itself is around 6″ tall, so that would be the minimum requirement but if I could get more height that would be a plus as I could add an additional riser or have room for a servo to rotate the periscope once extended.

I’d shopped around and bought a selection of drawer slides, but couldn’t find anything shorter than 16 inches locally. Once I had them on the workbench I realized that it really didn’t matter because whatever extended out would need to be hacked off anyway. In the end I picked a 20 inch slider made by KV because it’s two sliding components worked in lock step, which was important to my design.

I removed the stops and hacked it down to just over 10 inches and cut a small section of the inner slider that would run up and down on the sliding bearing, which itself ran inside main track.

This is the start point with the slider on the right

And it extended all the way to the left. Notice the two part slider has moved and not just one piece.

Unfortunately I have to jump a few steps as I didn’t take many photos of my late night tinkering with the lift platform and various iterations of how I tried to move it.

I’d originally added a roller to the top (similar to Wayne’s) which would guide the rope, but in the final design it wasn’t really needed - and currently acts as the end stop.

Right now I have a fixed motor on the lift platform that pulls itself up with a piece of string attached to the top of the slider. It uses gravity to lower itself back down as the motor unwinds the string.

I’m using a small Vex motor, but basically it’s a servo that can continuously rotate, so I can plug it directly into my receiver without the need for a speed controller.

It looks like the platform is sloping in this photo, but I’m hoping it’s just the angle of the shot

Here’s a short video of it in action

I have a lot more work to do on this and not sure how well it will work inside Artoo, especially at an angle. My short list of things to work are:

• Add support to help stop the drawer slide shaking.
• Extra weight to the platform to help it lower.
• Make sure it’s all level
• Limit switches to stop the motor
• Possibly automate with the PICAXE micro-controller, rather than control it manually via the Futaba transmitter.

In closing I also wanted to mention that I’d experimenting with having a spring loaded wheel on the motor (same position as it is now) that would run along the side of the drawer slide to pull the platform along. I still like the idea and may give it another try later. It definitely had a smoother action and having a wheel on the motor made the lift a little faster - Unfortunately I couldn’t get the spring mechanism to work right.

I’ve almost finished the electronics and programming to control the new hinged pie panels. I’m using a small 18X PICAXE micro-controller coupled with a few buttons on the RF remote.

Check out the video of the panels in action.

I’m also in the process of replacing the veclro that attaches the servos to the dome with something a bit more permanent. The brackets are made from aluminum angle (1.5″x1.5″ and 1/16″ thick).

They’re attached to the top of the hinge and the bolts from the mounting plate that already fix to the dome.

As a side note, if I was doing this again I’d adjust the hinge position a tad. The pies open to the vertical position, but it would have been nice if they’d open a little further, and I figured out (too late) that the lower the hinge is on the pie panel the further it will open - but too far and the hinge will not clear the dome properly.

See Also:

• Dome Pie Panel Servo Setup
• Dome Pie Panel Hinge Experiments
• Discussion of my Pie Panel setup on Astromech.net - some really cool suggestions.

Spent the last few evenings working on the dome pie panels, and after some tweaking I think I have a setup I’m happy with.

I ended up keeping the extra plate on the inside of the pie panel. It adds some detail and I like the sound the panels makes because of the extra weight. I left the hinges in the default configuration with no tweaks apart from drilling some holes to attach the servos.

I glued the hinges in place using CA glue, which sounds like it may not work so well, but I used the same glue on the hinges on the body and they’ve held up well.

Notice how the hinges are angled slightly to follow the curve of the dome.

I could have just glue the hinges directly to the dome, but I wanted to make the panels removable, so I created a small plate that glues to the dome and then then the hinges are screwed to it.

For the front Holo projector I’d purchased some push rod assemblies to attach the servos, but they were pretty expensive for something so simple, and I thought I could do better - this is what I ended up with.

It’s made up of some very cheap and readily available parts you’ll find at any hardware store. It basically some #4-40 and #6-32 screws in varying lengths, and short length of aluminum tubing that I cut up to make some linkages. I think it cost me less than $4 to make four push rod assemblies.

The long push rod is a #6-32 screw which attaches to the hinges via a piece of the aluminum tube that’s tapped at one end for the screw.

A long #4-40 screw passes through it horizontally to fix it to the hinges.

On the servo end I tap a small section of the tube on each side, and drilled a hole in the middle that allows the long screw to freely pass thru until I lock it on the side.

The servo is currently attached to the dome with some heavy-duty Velcro, which works fine right now, but I know it will eventually fail so I’ll be making a mounting brackets next.

Here’s a short video showing the setup in action

The next step is to wire all 4 servos into a PICAXE processor and tie it all together with some software to trigger sequences from the RF remote.

See Also:

• Dome Pie Panel Control
• Dome Pie Panel Hinge Experiments
• Discussion of my Pie Panel setup on Astromech.net - some really cool suggestions.

The last couple of weeks I’ve been busy getting Artoo back together for a couple of important events. I’ve totally overhauled the electrical system (again) and I’m hoping this will be it for a while.

At the hospital visit on Saturday I ran Artoo for about 4-5 hours on the new system and the 18Ah batteries without any sign of slow down, and I continued to run him again the next day for few more hours on the same charge. I must admit that I didn’t do lots of long sprints, but I’m confident that my earlier battery problems are fixed.

With that said, here’s a summary of the electrical work and the new electrical system design.

The following schematic outlines the 3 main areas of my setup. The red area is the front charging port, flashing LEDs and the battery select/on/off switch. Yellow is the wiring harness/relays that does all the magic of switch batteries between charing mode or running the droid, and finally the blue on the right is the rear electrical panel containing the speed controllers, fuse block, battery monitor, and power distribution board. It also contains an additional relay/power jack to run the drive from 120VAC/12VDC adapter.

There’s also a PDF version that maybe easier for printing.

Here’s a photo of the battery select relay/wiring harness (yellow section of the schematic). It uses 3 automotive relays to do the battery switching for charging and to turn Artoo on and off.

I attached the wiring harness to the battery holder using a small bracket

And here’s the new batteries in place

This is the front charing port (red section on the schematic) you’ve probably seen before. The attaced board to the right is the PICAXE controller that flashes the lights when the front door opens, and the smaller board to the left just contains a 7805 5VDC regular to power the PICAXE. The wiring harness above connects to to the charging socket.

In addition to adding the extra relays to switch two sets of batteries, I replaced the large MAXI fuse block/voltage display with a much smaller ATO fuse block and a separate LED voltage meter display.

I mounted the voltage meter on the rear electrical panel, and instead of using one for each of the batteries I decided to use just one with a switch to flip between batteries. The board requires a separate 5VDC supply to operate and I got this from the power distribution board.

Here’s the new rear panel. Going clockwise, top left is the battery monitor, then the Vex Micro-controller and receiver, below that is the power distribution board, and below that the fuse block, to the left the RoboteQ AX3500 speed controller for the drive motors, and above that the Syren10 speed controller which turn the dome.

I also worked on getting the slip ring soldered up and installed.

And I made up little boards for it to plug into. 12VDC power is connected to the blue terminal block, and the control/servo cables on the 3 pin connectors.

There a very similar board in the dome, but with an additional 5VDC input from a 7805 regulator IC to power the servos.