Blog

Yargh!

Ok, so it took me several tries to get this typed out (and the original thoughts are gone forever), so here goes. Last weekend was pretty productive. I finished up little Toby’s rocking horse, and built a giant kitty condo. Sunday was spent making some shelves, painting the beehive I built last month, and helping Sara with a quilt she made for her BFF Stephanie. I also made a second motor cowling for my rov project. The original assembled cowling needed some support on the back side to reduce shaft wobble. At some point I got email working on the server, and installed bBlog. Woohoo!

Well, as the headline reads, the H-bridge is finished and all tested out.

This morning, I wired the H-bridge up (except for the two output wires). After work, I stopped by the hobby shop and picked up two more propellors, then soldered on the output wires to the last H-bridge stage. I then proceeded to move the test motor from the first stage to the last stage, and gave it a spin. It worked, right out of the gate, so I was pretty happy about that.

Next step, we need to get the RS-232 stuff working. I was wondering earlier today of an old accoustic-coupler style modem would work underwater for any distance. That would rock, if that was the case.

Once the RS-232 stuff is working, I can start building the body. I still need to put the back cowling onto the first motor, and build the other two motor bodies.

Well, I finally got back to work on this. A short hiatis after finishing school, working on the garden, making a rain barrel, composter, and nearly finishing the rocking horse for baby, I return to make some progress.

I ordered another AVR stamp board to replace the development unit that I fried trying to get RS-232 stuff working. It has since arrived, and is still in the original packaging.

Today, I get everything working from where I was last time. I also dare to test the second H-bridge, and hook up the second bilge pump to load it. I labeled the wiring to the H-bridge, which has always been a source of confusion, so that was a big help. I added the outputs to the controller, and manage to control the thing from the joystick. There is only one pwm output currently in use, so I am not sure how to handle that, since the joystick input uses one of the needed outputs. Ultimately, I will need to put the new board in, and control the whole deal using RS-232.

At the end of the day, I have both motors working from the joystick, and they are sharing the PWM line. Not sure how to resolve that at the moment, but I will shelf that problem until later. Next I should finish building the last stage of the H-bridge module.

I decide to celebrate my success by buying another bilge pump and the last propellor, but alas, the hobby shop closes at 6, and Atwood seems to have changed models. Wal-Mart no longer sells a V-500, they now are selling the T-500, and the body style is completely different. I come home dejected, but manage to pick a couple of V-500’s up on the internet. Ebay rocks!

Ok, well I got the timing issues worked out on the LCD today, and integrated that into the future topside program as well. The symptoms were that it would skip characters, or print garbage. It turned out the issue was with the command wait function, I had to increase the delay from 9 to about 20. I also put the display code into the main loop, with placeholders for the joystick positions.

I also started compiling in the code for Futaba decoding routines that I found on the web, but they seem to be using one of the Atmel OSes — seemed to be similar to Linux, but not sure entirely. I need to figure a way to get around the task stuff, and add a way to get some time information somehow. It was also using a UART which didn’t seem to have any references, so I will need to figure a way around that too.

Well, I got the LCD display working today (kind of). I found a #define for the timing frequency that was set to an old value, so I updated it. That and two of the outputs were not working correctly. They must have gotten blown out when I was playing with the RS-232 stuff. Drat.

Anyway, I can tell it is trying to write the right thing, but it is only correct once out of every 7-8 times I reset the board. Either a character gets skipped, or it writes trash otherwise. Must be more tweaking of the timing needed.

Well, today I decided to test the H-bridge circuit. I wanted to alleviate the risk of frying the Atmel board, so I spent an hour or so rigging up a 555 timer so I could have a nice little H-bridge tester. With a .1uF cap and 1M+500K resistors for a voltage divider/RC network, I made a sweet little astable multivibrator at 7.2 Hz. 555 timers rock! I found out the 12VDC bilge pump motor didn’t turn much when feeding it 5V, so I swapped the 7805 for a 7812 regulator and I was in fat city.

First, I just jammed 12V direct to the motor to see what I could expect, then I did the same using the H-bridge circuit, with the Fwd and Rev lines tied to my trusty Bell and Howell digital IO module, and the 555 timer output for the PWM. Turns out they had the same results.