This has been a couple of weeks of nothing but set backs. As you can see I was nearly finished with the body of the housing, but the main isssue was that the part was bigger than the working area of the mill. I was very careful, and moving slow, I almost finished but the part jumped and was ruined.
It can also be seen that I got the walls on the long sides of the part too thin, but I was just going to call it part of the learning curve.
None the less, I will put the mechanicals of this off until I upgrade my mill. Don't know when that will be.
There are two banana plugs on the back of the hat for communication. Inside the hat are two speakers and a microphone. They are all wired together to the banana plugs. This is a function of the hat that can not be interrupted and the easiest way to tie in. What I am thinking is run the correct wire from the hat around to my box, and then the part I want to figure out is how I am going to take the audio and modulate/demodulate and communicate back and forth with the surface.
I have had to disassemble, troubleshoot, repair and reassemble all of the radios on the page, and if you paid what they are asking, and then took a look inside you would not be happy. With the exception of the helium radios they consist of an audio amplifier, a push-to-talk button and a circuit that will switch the speaker to a microphone.
The communication wire is a 20/4 armored wire. Inside the jacket are four wires, two colors, and the two like colors are twisted together and the shield is cut back. So there has to be a dummy proof way to connect this for digital communication, and the system has to be able to fall back on analog communication if a radio is used that does not support digital communication, as well as analog communication being passed through to the surface if the box on the hat floods (additionally the radio that has digital communication capabilities must be able to switch to analog function if a digital signal is not present).
The board in the link I posted for the DECT is not something that I plan on using, I just found it interesting and I wanted to take a look and see if I could use something in a wired fashion instead of wireless.
So the first things on my list are finish the construction of the box, and figure out how to make a digital wired radio preferably with an arduino. Possibly something like this http://www.instructables.com/id/Arduino-Audio-Input/?ALLSTEPS but instead of recording the sound, pass it to the topside radio for playback.
The only stipulations that I can think of are that communications have to be realtime and reliable. They don't have to be high quality, this might end up being an improvement to existing systems.
I started cleaning up this scrap piece of delrin or uhmw (don't know which). Just ordered a end mill holder for 3/8" bits. I will be happy if I get a perfect rectangular piece as this is the first real machining I have done. The hard part is going to be the o-ring grooves, I think that there should be a gland and a face seal. If I get it right it should be water tight to 800+ fsw (~356psi), I am just hoping there is space for everything I want to put in it.
I don't think that there can be much found on the internet about making a submersible 1atm box, it is probably mostly guess and test with the big companies. I won't make 3d drawings, but I will put cad drawings in the repository after I get the box made and can get measurements.