How would you design this device?

[spiri]

I have a pretty specific problem that I'm trying to solve and I'm out of ideas. I can't divulge too many details, so I thought I'd use an example to describe my problem in general terms. I'm not sure if it's really possible to brainstorm using this forum, but I thought I'd give it a shot anyway, so here goes:

If you were to design a device, like a salt shaker, that could deliver a very small dose of salt (around 20 mg) at a frequency of once every 10 minutes for 16 hrs, how would you do it? It has to be EXTREMELY small, and cheap and provide precision dosing to around +/-10%. Here's the challenging part, it has to deliver the salt underwater and not allow the salt that remains in the reservoir to get wet. It would be great if it could be completely self-powered and mechanical only (electronics might take up too much space).

Any thoughts or ideas would be greatly appreciated and no ideas are too far out there.

 

[pantaz]

... design a device, like a salt shaker, that could deliver a very small dose of salt (around 20 mg) at a frequency of once every 10 minutes for 16 hrs, how would you do it? It has to be EXTREMELY small, and cheap and provide precision dosing to around +/-10%. Here's the challenging part, it has to deliver the salt underwater and not allow the salt that remains in the reservoir to get wet. It would be great if it could be completely self-powered and mechanical only (electronics might take up too much space).

Any thoughts or ideas would be greatly appreciated and no ideas are too far out there.

Define "EXTREMELY small, and cheap".

 

[mfb]

What is "extremely small"? There is an obvious lower limit of ~2cm^3 just to hold the salt, but how much more is acceptable?
What is "cheap"? 1 Euro, 10, 100, 1000?

A very simple option would be an array of small salt containers, all with a common lid slowly moving to the side, exposing one new container every 10 minutes.
With multiple rows (shifted a bit towards each other) you don't have to put 6*16 containers in a row, and you can make the device a bit more square-like.
Then you just need a small motor and a gear. I don't see much hope for self-powered systems without a power source from the environment (which is undefined right now).

 

[spiri]

Thanks for the reply. The total volume of salt is around 2.5 mL, the device shouldn't be much bigger than that. Maybe 3-3.5 mL. The cost, without the salt, has to be no more than $4 in volumes of 10 million units.

 

[mfb]

(I edited my previous post at the same time you made your second post)

10 million units, so you can mass-produce something. Hmm...

$4 rules out most active systems, unfortunately.
There are materials that slowly dissolve in water. If those are reliable and cheap enough, they could be interesting separations between container walls. But then you need some additional trick to get enough circulation for the salt to come out, because diffusion alone will be too slow. Or do you have some flow in the water?

 

[Bobbywhy]

I use an inhaled powdered medicine. It is dispenses an exactly measured dose when I need one from a roll of two (clear) tapes. Each dose happens to be visible owing to transparedt tapes. The powder is hermetically sealed until I rotate the dispenser mechanism which separates the tapes and exposes the single dose of medicine which gets inhaled.

To avoid any commercial information here on Physics Forums, this product has a six-letter name followed by "DISKUS" and the dispenser is round, purple, and plastic. It even has a counter to show number of remaining doses.

 

[thankz]

make a self dissolving tablet like alkaselser, put it on a tape roll and have the dispenser cut open the tape as it passes through.

 

[spiri]

Thanks for the ideas, very helpful. Any others?

 

[sgb27]

A few more questions first for the spec.
How many times does the device need to work for? Just one 16 hour cycle, or will it be repeatedly refilled?
Is the $4 a selling price, production cost or material cost?
What about the price/cost of the salt refills?
Does the device need to be fixed somehow, or is it ok to just float about?
If the device needs to be reused, can it be taken out of the water to be refilled?
What accuracy do you need on the "every 10 minutes"?

 

[spiri]

Great questions! I'd prefer to make it a single use device, filled before placing it underwater either by the user or pre-filled. The $4 is the total cost of manufacture (materials and production costs). The salt cost is separate and will not be an issue. The device will be fixed to something that will be underwater and won't need to be re-used. The accuracy will be +/-10% over any given 30 minute period of time.

 

[bikengr]

What is the need to keep the salt dry, if it is going to end up wet? save a lotta space by dispensing saline or pushing water through valve-isolated wet salt.

 

[litup]

You could have a CO2 cylinder powering a device that opens a valve every ten minutes that forces a rubber diaphragm open with a small hole the gets uncovered by the force which shoots the required amount out but the diaphragm closes so fast no water can get inside. Maybe two diaphragms where one would act as a shield for the salt inside. It could be arranged in such a way as it would be like a self ejecting bullet casing where a shot of CO2 would shoot any water in the outer diaphragm so the next load of salt would not get wet.

 

[Lok]

For dispensing a fixed amount with the very generous tolernace of +-10%, make plastic cylinder with gelatin pill separators, those can be made for dissolving in a fixed time. Placing them in series will dispense the salt without mechanical or electrical input.
http://en.wikipedia.org/wiki/Capsule_(pharmacy)#Manufacturing_materials