Finally working on the Pan/Tilt

[Zaibach]

lol that would be an interesting thread.

Well finally got it assembled yesterday the dual linkage at the top didn't work out even tho it looked nice, once side would bind up thankfully it was an easy fix. Just switched out the servo horn and changed out the dual linkage for one linked diagonally

001.JPG (93.54 KiB) Viewed 3436 times

Disclaimer: I actually had nothing to do with designing the front cover short of sticking the lens to it to see how a fish eye would work out. It was part of the camera enclosure I picked up on Thingiverses and I just decided to make the tilt harness compatible. Tho if I do decide the lens is going to be a permanent fixture I'll either have to make a new cover or see if I can get the STL for that cover to import and modify it.

002.JPG (106.86 KiB) Viewed 3436 times

Yes the stock ribbon cable that concerns me simply because it is a ribbon cable is far far too short. Tho I think I have found a source for longer cables.

003.JPG (84.94 KiB) Viewed 3436 times

Had to reprint another pan arm because I tried to sand it smooth where it attaches to the pan harness and managed to do so at a horrible angle..... then I remembered I am using ABS so I could have just given it an acetone wash and smoothed it right up ~.~ good times. At this point everything seems mechanically sound. Gently manipulating the servos causes the camera to move in the desired direction...... and now that the "easy part" is seemingly over its time to slam my head into a desk and cry myself to sleep writing the code to make it controllable.

[a_shorething]

That's very cool!

Yeah, the ribbon cable might be an issue. Especially since it really doesn't have a connector on the end, does it? I think it just kind of locks into place on the raspberry board.

Good luck with it.

If you want some help making a new housing let me know. I design stuff from scratch all the time for printing. Just send some measurements and what you need.

[Zaibach]

The ribbon has the standard mini locking whatever they are called, so its securely connected and I'm not worried about it coming loose. Its more that the film ribbon cable really isn't mean to twist the way I'm going to be twisting it. Of course it isn't actually going to be twisting all that much. I think most of the problem is that this cable is just way, way too short. I really just hooked up the one I had for the sake of hooking it up. Getting it all put together, fleeting sense of accomplishment before the crushing despair kicks in lol.

Once I get a longer ribbon I don't think it will be much of an issue, I hope. If it doesn't work out I have a spare HD USB camera I can switch in with a few modifications. Course I had a more interesting idea for a camera mount for that one. At least less traditional. I really need to stop thinking about the small stuff and just build the ROV or I'm going to be nitpicking until forever. But if I just built it where would the fun be, I mean half the entertainment is figuring out how to put a jigsaw puzzle together when half the pieces don't even exist yet

[Zaibach]

Well it took a few days to get my head wrapped around it, initially I was going to write the joystick controls in PyGame but I really wanted to be able to read SDL outputs from my Logitech F310 analogs. Which Pygame was more than happy to mockingly throw at me as debug code but I couldn't figure for the life of me how to actually get that value into a variable. So I eventually ragequit pygame and found a nice quiet tutorial on accessing the joystick directly. http://www.stealth-x.com/programming/dr ... python.php

Still might switch back to PyGame if it turns out it will work for the GUI/HUD

Tho I think I donked up my PWM controller :/
I've had it stop working on me twice now seems to be OK at the moment. Usually starts working again if I let it sit for a day or two. Not sure if its the PWM breakout itself or if its the power supply or a combination. Thing is when I look at either of them individually things seem to be within tolerance and sadly I don't have any readily available replacements. Its newest trick is showing up on the I2C assignment as both id 40 and 70 which I don't recall it doing previously. Not sure I would trust this in the water as it stands so I'll probably swap it out at some point but its working OKish for testing.

The Good, the basic framework now exists for controlling the Pan/Tilt and the ROV in general with an F310
The Bad, now I need all the "things" ready so I can code them into place and to track down that super fun gremlin.