DIY Dive Computer

[MaxAlegraD]

With the advent of highly sofisticated microchip technology such as ATMEL AVR, PIC, etc. I was wondering if someone tried to actually build a dive computer ?

All we need a pressure sensor for measuring the depth, pressure sensor for air integration, timer circuit and a case.

The programming could be made in GNU C for Atmel AVR and since most of the chips have internal flash memory and serial ports we can do all sorts of hacking.

The display could be a normal LCD inside the case with 2 or 3 buttons to control the whole gizmo.

Roughly speaking I would estimate the total cost of less than USD 100. Considering the sensors would be the most expensive items.

We can even insert some games for long lasting decompression

Any ideas ? Thoughts ?

 

[fan(t)a(s)tic]

I wouldn't want to bet my life on something like that. Do us all a favour and use a back-up for the first couple of hundred dives, dude!

I guess capability wise you're talking about more than a simple depth gauge/bottom timer! Where would you get the algorithms from? How reliable would they be? What's the max depth? (don't rely on testing, this needs static calculations!)

Again, build it, if you want to, but make sure, it'll survive long enough to keep you out of trouble.

 

[Atticus]

I've contemplated this in the past. I write firmware for a living and am relatively familiar with basic decompression theory. I even went as far as sampling some pressure sensors and drawing up some basic schematics.

A buhlman algorithm certainly wouldn't be hard to code.

To make the project worthwhile you'd need to do something more useful than straight buhlman - otherwise just go buy yourself a good dive computer and call it good.

I'm not volunteering necessarily -- I have a Nitek He that I am moderately happy with.

Drop me a line if you progress on this - I'd like to hear what you're up to.

 

[frogman]

I've contemplated this in the past. I write firmware for a living and am relatively familiar with basic decompression theory. I even went as far as sampling some pressure sensors and drawing up some basic schematics.

A buhlman algorithm certainly wouldn't be hard to code.

To make the project worthwhile you'd need to do something more useful than straight buhlman - otherwise just go buy yourself a good dive computer and call it good.

I'm not volunteering necessarily -- I have a Nitek He that I am moderately happy with.

Drop me a line if you progress on this - I'd like to hear what you're up to.

Any ideas where I can get those sensors...ones that are small?

 

[miketsp]

Any ideas where I can get those sensors...ones that are small?

Lots of manufacturers, just go into Google and type
miniature pressure sensor

Examples at
http://www.entran.com/epn.htm

Back on the original subject, I don't know if this would be a very satisfying project, after all there are many good tested products on the market. Other than to help your own learning process there isn't much application.

People build their own dive lights because they can make a significant saving. In the case of a computer this doesn't make sense.

So, why not try something more exotic and something you most probably would not, or could not buy, like an inertial navigation system that would give your 3D position relative to your entry point? With modern electronics this could be made pretty small (size of a canister light), diving is a specific application which does not require long term drift control - something like 1m error per hour would be more than enough accuracy. This would have a whole host of applications eg. wreck mapping, cave mapping.
If you need more precision you just define a reference point and take a reading at that point occasionally to obtain a drift curve for subsequent compensation of the results.

Who knows, you may even come up with a commercial product.

 

[Kim]

Lots of manufacturers, just go into Google and type
miniature pressure sensor

Examples at
http://www.entran.com/epn.htm

Back on the original subject, I don't know if this would be a very satisfying project, after all there are many good tested products on the market. Other than to help your own learning process there isn't much application.

People build their own dive lights because they can make a significant saving. In the case of a computer this doesn't make sense.

So, why not try something more exotic and something you most probably would not, or could not buy, like an inertial navigation system that would give your 3D position relative to your entry point? With modern electronics this could be made pretty small (size of a canister light), diving is a specific application which does not require long term drift control - something like 1m error per hour would be more than enough accuracy. This would have a whole host of applications eg. wreck mapping, cave mapping.
If you need more precision you just define a reference point and take a reading at that point occasionally to obtain a drift curve for subsequent compensation of the results.

Who knows, you may even come up with a commercial product.

1m error per hour sounds a little strange - what about currents? Inertial navigation will take into consideration your own mass movements relative to a stationary exterior - but how would that work if your personal mass is moving - as well as external mass?

 

[miketsp]

1m error per hour sounds a little strange - what about currents? Inertial navigation will take into consideration your own mass movements relative to a stationary exterior - but how would that work if your personal mass is moving - as well as external mass?

That's the whole beauty of an inertial navigation system - you zero the accelerometers while you are on dry land so all subsequent accelerations/movements are related to the terrestrial globe. If you stop swimming you will register the speed of the current. The only error would occur if you zeroed at a different latitude as you would then start to register the different rotational speed of the earth, but this is not relevant in a diving application.

 

[Kim]

That's the whole beauty of an inertial navigation system - you zero the accelerometers while you are on dry land so all subsequent accelerations/movements are related to the terrestrial globe. If you stop swimming you will register the speed of the current. The only error would occur if you zeroed at a different latitude as you would then start to register the different rotational speed of the earth, but this is not relevant in a diving application.

OK - I see that. Trouble is, I have a car navi that uses both inertial - and GPS - to show me my position on the display. It's the latest Toyota Japanese system and Japan is supposed to be at the front of NAVI systems. Basically it checks the satellite every couple of minutes - and in between uses the inertial sensors. It's very easy to tell at the beginning of a ride - or going through a long tunnel (lots of them in mountainous Japan) that the inertial part is very easily fooled!!

 

[miketsp]

OK - I see that. Trouble is, I have a car navi that uses both inertial - and GPS - to show me my position on the display. It's the latest Toyota Japanese system and Japan is supposed to be at the front of NAVI systems. Basically it checks the satellite every couple of minutes - and in between uses the inertial sensors. It's very easy to tell at the beginning of a ride - or going through a long tunnel (lots of them in mountainous Japan) that the inertial part is very easily fooled!!

Few things - its actually quite difficult to project an INS for a car due to the tremendous variation of G forces - bumps in the road etc. Plus they are building a mass product trying to economise cents. And they don't want it to be as big as a canister light.
An INS carried by a diver will rarely experience more than 1G in any direction. Also at some moment at the beginning of the dive the diver will know that he is motionless wrt the bottom and the device could have a reset button to indicate that this moment represents zero acceleration and velocity.
This sort of precision mechanics coupled with electronics is exactly where DIY gains the most benefit.
Nowadays with the easy availability of brushless high speed motors to drive the gyros, floating air bearings, diffraction gratings, high intensity LEDs etc, the options for the construction of sensitive accelerometers are endless, as are the use of programmable chips, microcontrollers etc to do the integration. I am pretty sure that I could put together most of the mechanics using scrap hard disk drive mechanisms as gyros for just a few US$. Mount them in gimbals with a photolithic diffraction grating to sense the movement....
Dismantle a few old watches/clocks to get the ruby bearings..
Put the gyros in the centre of the canister suspended on shock absorbers and mount the electronics on a flexible printed circuit around the inside of the tube....

This strikes me as the sort of project that would be very rewarding for anybody into DIY, electronics & programming.

 

[cornfed]

Any ideas ? Thoughts ?

Along with games I think you should program in the number for DAN so you don't have to look for it when you bend the snot out of yourself.

I seriously doubt the hardware is the difficult part of building a dive computer.