Above is a diagram of the ballasting system I am making.
The idea is for the compressed CO2 to expel water through the bottom hole to purge the ballast.
And the pressure of the water to force air out of the air outlet and water in through the bottom.
However, due to this being a submersible that also needs to complete surface operations, the water does not have enough pressure to fill the ballast tank and force the air our of the open air outlet tube. This has been proven through testing.
I want to attach a pump to the air outlet to create a slight vacuum that will draw in water and sink the submersible. Core seems to have a few options, however I am unfamiliar with which ones will work with pumping air and work underwater.
The reason I do not want to put it on the water in/out is that I do not want to make the pump deal with the compressed CO2 that would be forcing water through it.
I would greatly appreciate any advice you would have for my selection of pump.
How big is this device and how fast does it have to fill and empty the tank. One thing to remember is that if the CO2 is expelled from the cylinder too fast it will freeze. What would be wrong with compressed air.
Cheers Bob
I must admit, fluid dynamics and pressure aren’t exactly my strong suit.
That being said, it is my understanding that you only need the pump to create a small pressure difference to draw the water in through the bottom, slowly bleeding air through the pump. Please correct me if my understanding is wrong in any way.
If that is the case, I would think something like this would work.
The inlet/outlet on that pump in particular is quite small, so it’s air transfer is rated at 1.8L/min. I am not sure how the ballast tank is, so it may be fine depending on the quantity of air inside. There is also an option from DFRobot that moves 2.2L/min
Interested to hear if this is the kind of thing you had in mind. Our pump range is fairly limited to smaller projects, but its possible they will work well for this project!
Not a very large device. I think a bit more info regarding the use would be nice.
For instance is this pump to be mounted on the vessel. If so it will have to be water proof and mounted underneath or the pipe will rotate on the water and turn upside down. The same for the gas cylinder.
Is this vessel mobile or does it just go up and down. If not mobile the pump and cylinder could be mounted on the surface and connected by flexible tubing.
A few unanswered questions before anyone can jump in with recommendations.
Cheers Bob
PS. Both those pumps linked are about half the diameter and half the length of the vessel.
There will be two of these ballasts systems making up a larger submersible, yes it is mobile and I wish for everything to be mounted inside the larger outer hull.
The whole sub is over a meter long
I am considering using a cheap small peristaltic pump, however I am unsure if this will create enough suction to draw up the water.
Hi James
Thanks. You would be surprised how much that clears things up.
I get it now. This is one of two ballast tanks for the larger vessel. Makes more sense now.
I take it you have done your sums and these 2 tanks will hold enough water ti submerge the parent vessel. That was probably a silly question. But little things like this have been overlooked before. You probably can’t finalise anyway until you know the weight of your pump(s) and gas cylinder(s).
What power do you have available to operate this pump.
Cheers Bob
Hi James,
Is there no weight to this vessel? If the vessel has some weight, and the lower valve and air outlet are open, the water coming in will push the air out.
As the vessel descends and water pressure increases, you have to add air (compressed CO2) to push water out to slow the descent and hover; if you just shut the water inlet / outlet you’ll keep descending. At 5m down you need 15L of ‘surface air’, which will be compressed to displace 10L of water (pressure is 1.5atm), at 10m you need 20L of surface air to displace 10L water (2atm) and so on.
On the ascent, you’ll also need to release air as it expands, otherwise it’ll be coming up fast!
Scuba divers do this all the time, you need just enough lead to make you slightly negatively buoyant - the difference being a flexible buoyancy vest instead of a rigid ballast tank - no need to suck out air, water pressure does the work for you.
The answer may be as simple as adding a little bit of weight to the vessel to make is just slightly negatively buoyant … and if there’s anything ‘critical’ inside the vessel, think about a redundant actuator for the air outlet.
Cheers, Dave
In theory this would be perfect, and was what I first tested. However our system (with 4mm ID tubes), did not take on water, even when pushing it deeper underwater.
Very good point about the water expanding though, I will need to take this into consideration.
The vessel has foam/lead inside it to keep neutrally bouyant, however I do need it to be able to sit on the surface, and then sink back down via ballast only.
Hi James
4mm is not very big but water should have still flowed.
Could it be that when the vessel is floating the ballast tanks are above the water line. Theoretically the water should have achieved the same level as the outside water line. This would possibly reach an equilibrium where everything balances out. If the vessel has positive buoyancy it might never sink. I don’t know.
But you would think that forcing the vessel under water would fill the tanks. The air vents have to be open. If the air cannot escape the water will not enter. If the vent tube does not come to the surface the pressure difference between the top of the vent tube and the intake tube would not be very great and with just a 4mm pipe the flow would be slow.
Just something you could try if you have not already done so. Fill the ballast tanks somehow and make sure the vessel will sink. If it still floats you are going to have to re think a bit. Like maybe add some more lead or bigger tanks.
Cheers Bob
I think it could be worth considering using a medical syringe moved by a linear actuator or motor with rack and pinion. They come in a range of sizes and would be easy to replace for the almost inevitable breakages during your making and testing.
Testing unfortunately proved this wont work, I then tried with a cheap small peristaltic pump and it worked. Looking into better, more waterproof options of pump now. I am also going to implement a bypass valve so the water forced by the Co2 does not force its way through the pump.
funny story with this, this is what I used previously, a dual servo powered rack and pinion, we could not find servos that are waterproof and would be strong enough to actuate below 1m.
Hi Michael
That would be a syringe like a grease gun. Used to have them back in the day.
In reality though it would have to evacuate the same volume of air that is in the pipe. 50mm dia X 1mm long. About 62.5cc or 62.5millilitre. Quite a large syringe but they probably exist for use by sadistic medical staff. Would put the frighteners up me anyway.
Cheers Bob
I’ve had success building a pressure chamber with a linear actuator in the past. We machined an head to fit some O rings to seal out chamber properly. Leakage was nearly negligible.
Depending on your weight/size budget, Pololu has a range of actuators that are quite strong, I can’t recall what the maximum rated pressure of our chamber was but we reached several times atmospheric with one of these actuators.A syringe-style setup with a beefy actuator should do the job.
And you can get our latest projects and tips straight away by following us on:
