Blog

Well, this morning I made a ton of progress on the RS-232 aspect. I replaced the chip I was using with a new one, and tried it using the STK500, and it worked great, both ways. I went to the usual haunts looking for code/hardware solutions, and found myself on AvrFreaks.net. I found a complete solution for the MAX232 setup, and it was identical to my other setup. I configured the AvrFreak code for my board, and gave it a whirl. Still no go. I switched to RS232-0 instead of RS232-1, and things started working. I got both boards working pretty well, although the topside board seems to reset a bunch for some reason.

Have a look at the serial level-shifter, after it is mounted onto the proto board (the same one as the power circuits). I also mounted heat sinks onto the 7805 and 7812 (I even used thermal grease!).

Tonight, I had some spare time, so I decided to crack open the Futaba joystick controller. The left forward/back joystick isn’t spring loaded, it has some frictionized type of setup to make it easier for keeping the plane flying level. I kept meaning to see what it would take to add a spring to that joystick.

Anyway, I carefully opened everything up, and what a mess. Nearly everything is hard-wired, and the wires are routed under parts that are firmly mounted to the board. There is no way to add springs to the joysticks without buying more parts. The connectors are stuck on to their mates pretty well too, and I have to resort to pliers to remove them. I can’t even imaging replacing the battery — I would imagine it would be easier to chuck the thing and just buy a new one when the battery dies. What a horrible design. At any rate, wires are wedged behind the power switch, and it has to be removed. The switch comes apart at the lowest level. I am able to swap the two joysticks, and I attempt re-assembling the power switch as carefully as possible, but once it is together the action doesn’t work very well (it doesn’t click to the on position), and the power meter doesn’t register. I give up before I chuck the thing into the creek.

Ok, between last night and this morning, I have both thrusters completed. This included gluing on the shafts, making the groove for the prop to sit into, as well as mounting of the prop. Yesterday AM, I welded the shaft guides.

I tried things out this morning, but either the decoding of the Futaba PWM isn’t working correctly, or there is a glitch in the PWM or motor control output circuitry. Actually, I had this problem the other day and it turned out the battery in the Futaba was running down a bit. I also found my MAX232 chips last night, so I can make another try to get RS-232 comms working again soon. There are too many circuits on the web for how to do this, and I haven’t found one which works well yet. I may have smoked the chip I have in the process. Oh well, there is always the weekend to try and make more progress. 🙂

This morning, I thought I would whip together the serial decoding. I had to roll-my-own atoi function, and along the way, I realized the pow() function was returning the wrong value. Even though I set it to use base-10, I was getting base-8 results. Very strange.

So, I coded my own pow() function as well, and the decoding for all 4 motor commands works great. I also replaced the AVR stamp board with it’s replacement, although I didn’t program it yet.

Up next — programming the new stamp board, and integrating the level-shifter circuit for the RS-232 comms on the bottom-side board.

Well, today I accomplished a bunch again. Decided to clean up a few things, and spent the morning making two power supply circuits on some proto boards. It is a much better setup than before, and reduces a lot of mess. I went as far as to put jumpers separating the two boards, and leds so the board is lit to show when they are operating.

I got RS-232 comms working pretty good on the topside board. There is some quirk that makes the board reset, but it recovers well, so we will try to somve this later. By the end of the day, sending commands works good, and receiving commands works fine as well.

Ok, I spent the entire Saturday working on the hull. I decided to trim all of the fittings shorter to save on weight, as well as make things more compact. I also decided on a single cross-bar, rather than dual cross-bars. I made the length considerably shorter, although one of the cross-bar fittings got glued a bit crooked, and the whole deal is a bit toed-in. Oh well, I guess we live with it until we get to design hull #2. I also had a thought to use reducing fittings on the tail, rather than cleanout tees. There would be an issue for how to fit stuff into the tubes, but it would have lots less drag, I would think.

Anyway, another three trips to Home Depot, and the hull is pretty much finished. Not bad. Here is a side view to get an idea for the vertical thruster idea.

I also glued up the motors, so the units are all sealed now. Have a look. I wonder what depth we can go to before they leak. I guess time will tell. 😉

Before I go to bed, I wired up the topside board so that the LCD output works, as well as the joystick port. I might add that this wiring is finalized, so I put on heat shrink and did it up right. Mighty fine.

Ok, most of the initial hull design is pretty good, but some of the fittings are beginning to become quite long, especially where the threaded parts are. There is also the vertical thruster issue — I have no idea how to mount it. I have had several thoughts of a third, smaller tube down the center, off of which the vertical thruster could be mounted.

So, back to Home Depot to fish for more ideas. I found I could put a, 1-1/2″ cross in the middle, mount it at an angle, then I can put a few street elbows, and a 45, I might have a good place to put a down-looking perspective camera.

Ok, I decided to see what Home Depot had for hull parts. After about an hour, I had a pretty good assortment of fittings, and a bunch of ideas running through my head. Firstly, I had originally wanted the main hull to be 4″ PVC, with two main tubes, parallel to each other, with two cross-members, fore and aft, like so.

Shortly into my shopping spree, I found 4″ reducing tees were $9, and 3″ reducing tees were less than $3, so the hull design quickly changed to 3″.

I also found a street cleanout fitting, where a 2″ flange came out from the 3″ part at a 45 degree angle. If I put another 45 degree, then it would be possible to mount the motors within the same tubing as the main hull, and eliminate some wiring feed-throughs.

The last major idea change I had, was to use more threaded end caps, so if something needs to be changed, at least part of the hull would be salvagable. I decided to make the nose-cones removable, as well as the motors. The initial plan was to just use them on the aft ends of the tubes to put electronics and such inside…