Idea = Pressure compensation with scuba dive tank

[rossrov]

Hi again Dimitris. No, that would just be the same as putting the floats on the outside, except it would create more problems.

For air controlled ballast system, I don't think there is an easier way than already discussed. You need some sort of pressure-sensing to determine depth, be it mechanical or electronic, controlling valves to let air in or out. For electronic sensing and control, Arduino would be ideal. You could sense the rate of change of pressure increase or decrease and put in a big blast of air to stop a fast descent, or a smaller amount to stop a slow descent.

All that aside, perhaps consider using logic-level FETs instead of relays. Solenoid valves likely to burn the relay contacts out.

Hope that helps, not hinders.

[dtrip]

Hi again Dimitris. No, that would just be the same as putting the floats on the outside, except it would create more problems.

For air controlled ballast system, I don't think there is an easier way than already discussed. You need some sort of pressure-sensing to determine depth, be it mechanical or electronic, controlling valves to let air in or out. For electronic sensing and control, Arduino would be ideal. You could sense the rate of change of pressure increase or decrease and put in a big blast of air to stop a fast descent, or a smaller amount to stop a slow descent.

All that aside, perhaps consider using logic-level FETs instead of relays. Solenoid valves likely to burn the relay contacts out.

Hope that helps, not hinders.

Of course it helps!

Yes I'd prefer to use transistors instead of relays. I just use the word "relay" to mean "a switch other than Arduino itself". I have not yet decided how to do it, but its not a problem.

You seem knowledgable about solenoids ? They must peak a lot of current when they switch, right ?

Update:
I have currently achieved to drive a servo, brushless motor and DC motor with a joystick, all of them plugged into the Arduino (not at once). I suppose I won't need more "micro-controlling skills" than those.

The scuba tank will delay because it is aged and the guy selling it to me prefers to take it for certification testing before sending it. Next step is to get the tank and make adaptors to [1] "pump" it up with a bicycle pump (10 bar), and [2] connect it to the solenoid.

[rossrov]

Glad to be of help! Voltage is typically the problem with solenoids. When the voltage is removed, the magnetic field collapses and creates a high voltage that can do damage, be it to relay contacts or semiconductors. This can be avoided a number of ways. The most common is to choose a transistor that has a built-in protection diode, and also to put a diode across the solenoid. Google "FET solenoid driver diagram". Also there is a bipolar driver chip designed for relays and solenoids, which can connect direct to the Arduino outputs. Can drive up to 7 solenoids/relay coils. ULN2003. Just check the current draw of your solenoids and be aware that there are AC and DC rated types. If you use a 24 volt AC coil with 24 volts DC it will overheat after a while.

Some solenoids such as those used in automotive starter motors have two coils, a high current one to "pull in" and a lower current to "hold in". You would not be using this type with valves though.

The fun really starts when you get a serial comms link of some sort remotely working the motors etc

[dtrip]

Glad to be of help! Voltage is typically the problem with solenoids. When the voltage is removed, the magnetic field collapses and creates a high voltage that can do damage, be it to relay contacts or semiconductors. This can be avoided a number of ways. The most common is to choose a transistor that has a built-in protection diode, and also to put a diode across the solenoid. Google "FET solenoid driver diagram". Also there is a bipolar driver chip designed for relays and solenoids, which can connect direct to the Arduino outputs. Can drive up to 7 solenoids/relay coils. ULN2003. Just check the current draw of your solenoids and be aware that there are AC and DC rated types. If you use a 24 volt AC coil with 24 volts DC it will overheat after a while.

Some solenoids such as those used in automotive starter motors have two coils, a high current one to "pull in" and a lower current to "hold in". You would not be using this type with valves though.

The fun really starts when you get a serial comms link of some sort remotely working the motors etc

You are indeed of help. Yes reverse voltage can be avoided with a diode. Its a similar case with the DC motor I suppose ? (although the solenoid sounds more brutal since its more rapid). I ll look into your information promptly. I dont think it will be a problem. In fact I plan to operate the solenoids without an Arduino, since they are on-off and do not require mixing of any sort.

Meanwhile, you are even more help than you think

You see, the bag idea, turns out to be not bad after all. And not just any bag. A "rubber fish" bag

In theory I can use something like this:



....and cut out two pieces of this:




glue them together (dont ask me how, its just an idea), just leave its belly open to insert the hardware
(we need an opening there, anyway)
and have a rubber fish

EDIT: I mean glue the edges only, and put the hardware inside.

Best part is that I can wiggle the tail and fins with servos from the inside.
The control will not be accurate and there will be no reverse. But, you could wiggle the joystick and it wiggles
its tail accordingly, and guide it this way, imagine that.

Its worth to give it a try I think.

I am soon buying a solenoid and try to drive it with transistors etc.

[rossrov]

Yes, I can imagine that wiggling! Will be good to see a video of the fish on the Forum . A fun way to test and develop the buoyancy system.

Motors generally do not have a diode across them, just diodes across the driver transistors. Diode across motor would mean you could not reverse it. Sometimes a capacitor and resistor in series are placed across the motor terminals.

[dtrip]

It will be a long time until I make a fish prototype, maybe even a year.
But I think I will be ready for a "plastic gas tank" buoyancy test in a couple months
Depends on how fast I can find the appropriate solenoid valve. I have email a couple vendors but no reply so far.

[dtrip]

I have also thought about another advantage of the open hull.


Since the pressure inside the hull is equal to the pressure outside,
the exit point of the thrusters' shaft could be decently sealed with some wax.

Even if it leaks, it will be teardrops of water, that will naturally drain down the hull,
and little bubbles of air escaping into the sea, which can be easily compensated
by the buoyancy system. Sure its an unwanted loss of air, but depending of the
air available in the scuba tank, it might be negligible.

What do you think ?

PS: My plan does not change, I will first do the buoyancy test before anything else.

EDIT: Oops, seems I forgot the outside wax chamber wall. Now it is corrected.

[rossrov]

Better, and probably simpler to use a normal shaft seal for the motor. Finding them in smaller sizes is a bit difficult, but they are available from 3mm up.