Design of a Cyclone As Air Coalescing Device

<...multiply every dimension times 1.5...>

Mechanical Engineering? Pah! Nothing to it!!

I agree. That's not engineering.

Ultimately, you'll need a particle air filter, or depth filter.

I've no proof, but I don't see how you can trap 2 micron particles with a cyclone filter.

He's the boss, use his design.

Water is a liquid- particle or depth filtration won't take it out; it will slow the passage but that will eventually reach an equilibrium as well.

I do agree that using his design is best because any form of coalescing device is removing liquid from a saturated environment. If this is a typical installation & the OP having a problem with this then they must not have an air dryer. Putting a single water control device in is asking for trouble as it will typically have to be located near the compressors. This often results in additional problems as any water that hasn't been removed is joined by additional condensation as the compressed air cools travelling through the lines. This is a major reason that filtration and water control devices are typically located as close as possible to the point of use.

So if the whole design is questionable, it may be best to do it the way you're told.....

True, I gues I assumed that there were some particulates in the mix, but OP didn't say that.

So, I agree, go with the boss's design x 1.5.

No, the particles to care about are the water droplets.

We do have an air dryer, but we will dry only part of the total output of our compressors. The rest of the air should just go through a coalescing filter (or my boss' cyclone) to remove the water droplets in the air stream.

In fact, my first idea was to use a coalescing filter for all the output, store that "wet air" in a tank, feeding the dryer from that tank, and storing the "dry air" in another tank. The "wet air receiving tank" and the "dry air receiving tank" would be the two primary tanks for the whole compressed air installation of our factory.

Your idea is good, but I would suggest the first tank before the cyclone. Air reservoirs are important after a dryer when plants tend to experience cyclical draw; this helps keep the airflow through the dryer more consistent and dampens the pressure fluctuations that can otherwise occur if the dryer acts as a pinch point on the airflow. This is probably the only benefit of "octopus" networks in old plants that have been added onto over the years, as all the extra pipe gives more more capacity!

There is no advantage to having a "wet" tank after the cyclone though. You are better off to put that tank- or better yet, a couple hundred feet of oversized pipe!- upstream from it. This will allow more cooling time for the air before it hits the cyclone, and the additional surface area will increase condensation. This should also help increase the mean size of the water droplets that hit the cyclone or filter. If you can size and implement the thing to feed the entire system as you thought it won't hurt the dryer.

It's not how big. It's how small and how many. In order to remove a 1 micron particle it's velocity will need to be high. Since you have a set flow rate. Which is really not set and will depend on demand. You will need to make the cyclones small.

One other problem is that a cyclone also depends on a pressure drop. In most applications the air is drawn through them Not force into them.

You need to give the air tangentially a bit inclined towards bottom to the top so that the heavy particles are gravitated towards the bottom by the time air reaches at the bottom due to centripetal force. Hence all the heavier particles are drawn towards the centre and fresh air escapes from the periphery. The diameter of the cone, height, taper are the design parameters to be calculated based on the particle size, pressure amd flow of the air. Logically the shape would be conical with filters at the out let as no gravtitational filter( In practical) will be able to remove the finer particles.

What your boss wants is that you either copy design and fabricate/buy an 80 mm in-line moisture air separator and fit it with an air trap 1/2". You can check this out from sites of Spirax Sarco/Adca/ Gestra.

Then buy a Coalescing Air Filter 80mm and fit it after. This filter you can buy from Spirax/the Compressor supplier. Buy 2 sets and fit with 2-valves each parallel so that you can replace filter element when the "change element" indicator shows red. Check Compressor supplier's P & I Drawing.

The 80 mm size is good enough to pass your 29 m3/hr with 1.5 M/s (lowest recommended) fictional losses. You can also opt for 65 mm size but the velocity will increase to 2.5 M/s whereas 3.0 M/s is highest recommended. It is better you stick with 80mm size.

I think the OP spec'd 29 M3/MIN, not hour. A bit large for point-of-use sized solutions.

Sorry I missed that. 29 m3/min = 1740 m3/hr = Line 600 mm with velocity 1.5~2.0 M/s.

You can make/buy an in-line moisture Separator but i doubt the air coalescing filter in that size. take the air line to the air receiver and then install 80mm/100mm filters as per off shoots.

It seems that you & your boss are both right in thought, Now add a magnetic ionization bars that attract smaller particulates like a magnet at base of your cyclone chamber, think of a dis en vac inner top make up above the dust chamber.

This is some of the make up in a natural power systems. The power needed is small & the return from clean air without harmful particle.

I ask you to see some public ed awareness site renewablethermalwindpower.com it mite be use-full to you and your community.

At best, a cyclonic separator will only remove water droplets. This means that the exit gas can still be saturated with water. If you want it drier than that you will need to resort to other methods. If you cool the gas before it enters the cyclonic separator, you can condense some of the water and the exit gas will be drier than saturated.