Buoyancy controlled submarine

Thread Starter

jackgeaston

Joined Jan 27, 2016
2
I am currently studying Aerospace engineering, and my project work is to create an autonomous underwater vehicle. I have chosen to use buoyancy propulsion as a method of propulsion. Simply by varying the buoyant force you can propel the submarine along. Being an aerospace student my knowledge in electronics is limited, but I do have some understanding.



My project requires me to create an electronic circuit which basically pumps on and off water, with time delay to control the depth. The trouble that I am having is how to control the amount of water that is inside the ballast tank. From water sensors I have used simple logic gates to control but the problem with this I understanding the difference between filling and emptying the tank. Is there any type of circuit or sensor that would be ideal to use to stop this problem?



I want to try and avoid using a microcontroller if possible.



Thanks for your time,

Jack
 

Alec_t

Joined Sep 17, 2013
14,280
Welcome to AAC!
An autonomous sub is quite complex so is almost certainly going to involve a microcontroller, unless you are prepared to cope with a shedload of logic circuits. If you need a microcontroller for e.g. navigation and sensors then you might as well use it for the buoyancy control too.
 

kubeek

Joined Sep 20, 2005
5,794
You could try measuring the pressure of the gas that is being compressed by the water. That might be easier to do than measuring the amount of water going in and out. Mind you the pressure will change with temperature, so you might need to account for that.
Also some form of last resort arrangemet to surface the sub in case something bad happens like batteries die might not be a bad idea. Something like a high pressure tank filled mostly with water and compressed air, with a valve to open it and one way valve such that it cannot get refilled by water from outside.
 

jpanhalt

Joined Jan 18, 2008
11,087
An added consideration is that as you pump water in and out, you are also pumping air out, as it is more soluble in water under pressure. Thus, after enough cycles, your pump may have to pull a vacuum to get enough water out to surface. I second kubeeks suggestion of having a high-pressure air source to ensure the ballast tank can be emptied. An alternative, but more complicated approach would be to have a non-permeable rubber membrane between the air and water in the tank.

If you have enough compressed air, you might be able to forget about pumps for the water. Just flood the tank to descend, and force water out to ascend.

John
 

Thread Starter

jackgeaston

Joined Jan 27, 2016
2
Thank you for your responses, the way in which I am attempting to go about this is to use an iterative process. By using logic to determine the state of the ballast tank. I am then thinking of having two astable 555 timer circuits, one which fills the tank for a period of time set by the timer. And the other that empty’s the ballast tank, it uses some kind of latch to keep pin 4 ‘’high’’ until logic decides that the tank is full, or empty which will then set the relevant circuit on or off.



The buoyancy controls motion uses a change in centre of gravity to initially move forward during decent. During the assent stage the pump propels water out the back creating a thrust component along with the additional buoyancy force. Newton’s laws of motion. This creates a sine wave motion forward which I will call the Y direction.

I will need to have a counting circuit, in which I have looked at flip flops, to determine the number of cycles that have been completed. Hence determining the distance travelled in the Y direction. Once this have been achieved the rudder will turn coming back on its self but with a displacement in the X direction. This will allow it to cover a field area.

As you have said I am going through a shed load of logic, I think a microcontroller is the best way to go! Unfortunately for me. Do you know of any good sites with tutorials on how to use such a device?
 

Picbuster

Joined Dec 2, 2013
1,047
Its like scuba diving keep the outside pressure constant.
If the pressure goes up empty your water tank and VV
The only thing you have to calculate is the relation between outside pressure variation and water flow to keep a steady depth.
A PID controller will do the job.
If you do not want to use MCU you should use compressed air and a pressure gauge with a valve.
 

BobaMosfet

Joined Jul 1, 2009
2,110
Yes, it's call an LED and an LDR. place LED on top, LDR on bottom, and measure the resistance off the LDR based on water level. Full tank, most resistance, empty tank, least, and you can tell by sampling which direction you're headed.
 

kubeek

Joined Sep 20, 2005
5,794
Yes, it's call an LED and an LDR. place LED on top, LDR on bottom, and measure the resistance off the LDR based on water level. Full tank, most resistance, empty tank, least, and you can tell by sampling which direction you're headed.
Sorry, come again?
 

Bernard

Joined Aug 7, 2008
5,784
Some wild thoughts: We have a flying machine that glides thru water, as we sink we glide down at maybe a 10 : 1 ratio of sink to forward motion. A pressure SW indicates 6 ft. so ballast tank is emptied to sea from stern.
by the linear actuator. With + buoyancy sub rises and sails forward on its water wings. A tilt sensor keeps angle on the bow at a fixed + or - 15 deg.by shifting water from ballast #2 to or from ballast #3. After required # of cycles & knowing speed on rising, rudder is set at 45 deg. for a calculated time to give 180 deg. After repeated cycles the low pressure SW is overridden & sub surfaces.??Sub # 1 00000.jpg
 

sailorjoe

Joined Jun 4, 2013
364
As you have said I am going through a shed load of logic, I think a microcontroller is the best way to go! Unfortunately for me. Do you know of any good sites with tutorials on how to use such a device?
Students seem to have a lot of luck by starting here: https://www.adafruit.com Click on the Learn tab.
Or here: https://www.arduino.cc/en/Guide/HomePage
Or here: https://www.raspberrypi.org
Each site is designed for their specific products, I.e., small computer boards, and they tend to cater to people, even kids, that are new to computers.
 
Top