# How to control the timed release of small bubbles in a liquid?

I'd like to have electronic control of the open/close timing of a source of air at the bottom of a tank of liquid such that I have control of when a single bubble is formed.

Experimentation would be required but I expect the open period to be at most 250ms - the more scope for lower period options the better.

The process of bubble forming would be repeated over and over again - with a 50% duty-cycle the frequency of new bubble formation would be a minimum of 2Hz.

Let's say 5~10Hz would be ideal.

Multiple bubbles will need to be formed in close vicinity (~10mm) so the device would either have an appropriate form-factor, or some kind of tubing system would be put in place.

I will have close to arbitrary control over the control signals, but I expect some form of digital control per 'channel', preferably parallel and direct, i.e. 0=closed, 1=open, as opposed to some kind of interpreted data stream.

Although I understand that the dynamics of bubbles will be all over the place, in terms of control I'd expect low latency.

Question: what kind of devices am I looking at here?

Are there off the shelf components that can do this? Cost is a concern at some level but is not necessarily the overriding requirement.

• D you mean something like what is used in this clip: youtube.com/watch?t=4&v=gRSKgxgB42k? This is part of research conducted by Professor Timothy Leighton at the University of Southampton – mg4w Oct 2 '15 at 13:54
• Hook an aquarium bubbler up to a variac? – grfrazee Oct 2 '15 at 16:09
• @mg4w that looks pretty good yes. Thanks! I'll follow that up certainly. – Lamar Latrell Oct 2 '15 at 22:57
• @grfrazee I considered that, but made the assumption that a system like that would have lag and the bubbles would be too big? – Lamar Latrell Oct 2 '15 at 22:58
• No idea, just throwing it out there. – grfrazee Oct 2 '15 at 22:59

Yes they do make pneumatic valves that meet your criteria. These ones definitely exceed your criteria, but will come with a large price tag. I recommend searching vendors for "high speed pneumatic valve". http://ph.parker.com/us/12051/en/pulse-valves-miniature-high-speed-high-vacuum-dispense-valve https://www.festo.com/net/SupportPortal/Files/10047/MH234-PSI-US.PDF

It is very possible that standard direct acting pneumatic valves would meet your criteria; they will just not publish it because it is something they dont consider an important feature or dont want to be held to. You could get something like these and test to see if they meet your spec: http://www.zoro.com/parker-solenoid-air-control-valve-18-in-24vdc-b310bb549c/i/G2536274/ http://www.omega.com/pptst/V60_SERIES.html?pn=OM-AIR-V60P6DDA-XP0200

As a side note; If this is in water, you can increase bubble uniformity by adding an inorganic salt like NaCl or MgCl to increase the surface tension.

• Hello Eric, thank you very much. In hindsight "high speed pneumatic valve" is obvious, but in the initial search I didn't use that ... And yes, we had considered looking at alternatives to water for that reason, as long as it isn't expensive and/or corrosive etc. But the prices, they make use this project a no go ... Might have to home brew something. – Lamar Latrell Oct 16 '15 at 4:58
• Perhaps post another question with more details pertaining to the application and link to it in a comment. There are likely many other ways to accomplish your end goal. – ericnutsch Oct 16 '15 at 5:06
• @ericnutch The end goal I'd like to keep under wraps, but the sub-task in question is arbitrary control over the timing/release of small bubbles from the base of a tank filled with a liquid. I'm aware this is asking for "XY-problem" type criticism but it's the best I can do without giving away the game... I'll think on it more before I do that, but yeah, it looks like it'll be very task-specific and likely be developed in house. – Lamar Latrell Oct 16 '15 at 5:56
• Gotcha. One last thing you may consider is having a single valve/nozzle and traversing that nozzle through the fluid a known velocity. Now you can create bubbles for lower capital cost in a serial fashion instead of a parallel fashion. Good Luck! – ericnutsch Oct 16 '15 at 7:15
• Yep the classic spatial/temporal tradeoff :) If we have more than one bubble generator it'd need to release synchronously with the rest. One of the constraints means that if we were to implement a full or partial 'interlaced' system (e.g. offsetting the generators at double the frequency) the cams/actuators involved would have to be very fast, which means noisy and likely expensive in their own right. But still, it's an idea to be aware of yes :) – Lamar Latrell Oct 16 '15 at 8:33