Making a Desiccant Air Dryer with Silica Gel

[awemawson]

Having got my CNC Plasma Table up and running I've been giving a lot of consideration to the humidity of the air that I supply to it. Moisture in the air badly affects cut quality and the life of the consumables. I do have a refrigeration style air dryer, and I did resurrect it for the Plasma Table. but to be any use it needs to be switched on for long periods of time, and it draws quite a bit of electricity

(Thread here for the rebuild http://madmodder.net/index.php/topic,11815.0.html )

Now there are all sorts of ways of drying air mainly concerning condensing the moisture and draining it away, but there is another method using Silica Gel that has the useful quality that it will absorb large quantities of moisture, and better still it is re-recyclable by the simple expedient of roasting it in the domestic oven 

So my plan was to remove as much moisture as possible with a conventional air / water separator then pass the pre-dried air though a column of Silica Gel.

Ideally the column containing the Silica Gel should be transparent, as it is available in a 'self indicating' state whereby it changes colour as water is absorbed - very handy 

I hit on the idea of using a standard 10" water filter container and modifying it for use as a desiccator. These thing are modestly priced, are transparent (or at least mine is), and are designed to take mains water pressure safely.

[awemawson]

Now the input and output of this housing are 3/4" BSP female brass inserts, so connections there are no problem at all. The air needs to pass through the bulk of the Silica Gel, then up a tube leading to the output port. The tube somehow needs to have a filter or gauze on it to stop the desiccant blowing straight through to the output. I hit on the idea of using a sintered plastic compressed air exhaust silencer. They are available in all sorts of sizes, and I chose one with a 1/2" BSP male thread.

Then it was a case of making up an adaptor for the filter lid to take the 15mm copper pipe, and joining on the silencer using a standard compression 'tap connector' - it all went rather well and turned out fine.

Initially I was stumped for a bit of brass stock to make the adaptor from, as I didn't want to turn down the only rather large brass bar that I had in stock - then my eyes fell on a 1" BSP to hose barb fitting, just crying out to be modified 

I designed the adaptor as a push fit into the bore in the lid of the filter, with an 'O' ring as a retainer / air seal. In the event it turned out that the hole in the plastic lid had just the right taper for it to lock solidly - in fact so solidly I had a bit of a struggle to get it out again 

The adaptor is soldered onto the 15mm copper pipe.

[awemawson]

Now the inlet port is just a vaguely rectangular opening at the top of the filter. To prevent Silica Gel leaking back into the inlet under negative pressure situations (when turning off) I incorporated a fine brass gauze in the opening, bonding it in with high viscosity super glue

[awemawson]

Next to add the pre and post filters. The pre filter is a conventional water separator used for it's intended purpose, whereas the post filter, although again a conventional water separator, I'm using it as it has a 5 micron sieve to sweep up any fine Silica Gel dust that might 'get away'

[awemawson]

So no more ado - do a leak test. Well first a safety pressure test with it covered in a towel - shame if the poly carbonate filter housing were to explode.

All's well - no violent explosions so then the leak test. I keep a horticultural hand spray on the shelf full of soapy water, and it's amazingly good at finding even the smallest escape of compressed air - but this was sound (I use Loctite 542 thread sealant on this size plumbing and swear by it)

[awemawson]

Now as the three air assemblies of this set up are all from different manufacturers, mounting it on a wall wouldn't be easy - unless I made up some customised brackets to accommodate their differing mounting styles.

A splash of lay out fluid, a bit of die grinding, drilling, taping, bending, zinc primering and black top coating later and we have an integrated assembly ready to fix on the wall.

[awemawson]

Then, like some 'just in time' manufacturing conglomerate, the girl from My Hermes arrived carrying a 3 kg tub of self indicating Silica Gel - I reckon she'd been spying on me and was waiting until just the right time to deliver 

So - shove the crystals in - wack it on the wall and try it out 

[awemawson]

Does it work?

Well time only will tell. I did do a test cut on the Plasma Table but considerable time hopefully will have to elapse to see if the stuff changes colour ......... unless we do another test

So - shove a bit of the Silica Gel in a 1 litre pyrex beaker, drip some tap water on it and watch the results.

... well it certainly changes colour so that's a good sign 

[Spurry]

Andrew, I think you would have made a good assistant to Black Adder. "Sire, I have a cunning plan..."
Neat job, it will be interesting to see how long it takes for the crystals to change colour in your filter.
Pete

[awemawson]

Thanks for the kind words (I think !!) Pete - I've been musing over a desiccant drier for quite some time, but finding a suitable container is what stumped me.

[Brass_Machine]

I love seeing stuff like this... even if I am not sure how it is used. This would be in-line from a compressor to the plasma cutter?

Eric

[awemawson]

Yes Eric that's right. Any moisture in the air should be scooped up by the Silica Gel

[hermetic]

 Hi Andrew, the amount of satisfaction you get from a finished job like that is priceless isn't it. I am at home at the moment, having been crocked with some horrible lurgi the kids brought home from school, and can't wait to get back to my workshop, .................and now as they say in Hull, Err nerr, SNERR!
Keep up the good work!
Phil

[awemawson]

So Phil is my 'Cheshire Cat' grin getting as far as Hull then 

[hermetic]

 I can tell yer grinning, because I am too! I have one of those filter housings on my workshop rainwater system!
Phil

[awemawson]

Perhaps we should run a challenge to find other uses for them 

[krv3000]

that's a proper job

[Pete W.]

Hi there, Andrew,

Another interesting project up and running, I hope you're not neglecting the livestock!

Somewhere in my filing ('piling' ? ) system, I have a copy of the military specification for the handling and re-activation for silica gel (hasten to add - it's not classified!).

I was always happy with the process when the domestic oven was electric but, nowadays, we cook with gas.

Does anyone know if that's as effective as the electric oven?  (I'm assuming that burning domestic gas produces water vapour.)

[John Rudd]

I've worked with silica driers, regenerated with electric heater elements in pressurised vessels...works well...

[PekkaNF]

That is very inspiratonal project!

I have a small amount of Silica Gel (probably cheapest and corse, because it is ised to keep boat interior dry during winter) and I have been thinking something similar last two years. Small pack of it is on my work bench as a reminder....but you have solved the pressure vessel nicely. I have been looking for big/least partly transparent normal water separator for modification and I even had a sintered silencer for filter....but you actually made it.

Very good idea and execution.

What is the next step? Coalescing filter to produce air bearing quality air?
http://www.newwayairbearings.com/design/technical-resources/care-and-air

Pekka

[appletree]

At the power station where I worked for 40 years in the instrumentation dept, the electrical dept used an industrial martindale hot air blower (hair drier) to blast air through desiccant tubes used as moisture absorbers in very big motors (huge). This was in the times of demarcation believe it or not, if a SWA gland (steel wire armoured) required fitting to an instrument cable, an electrical fitter was called in, (its a wonder the country did not go bankrupt) oh it did?
I do not know if it was efficient or not i.e. slow cook versus forced draft either way thats what they did.

In my workshop i would use my refrigerant compressed air drier normally used for paint spraying. But yes a container containing desiccant that can be regenerated is good however i wonder about flow rates (distance velocity). There is more than one way to skin a cat, why you wold want to is a different question.

 

[awemawson]

Thanks for the various comments chaps 

Appletree - my motivation making the desiccator was not to have to run the refrigerant drier for long periods. It needs to run for at least an hour before the copper coil of tubing is cold enough, by which time I would have cut whatever I'm cutting on the plasma table, and gone in for a cup of tea 

[awemawson]

I was so pleased with the way this drier works I made a second one 

This second one takes air directly from the compressor and dries it, hence my whole air distribution system will now have properly dried air. I'm sure that there will be existing moisture in the pipework, but after a bit the newly dried air should absorb and flush it out.

Looking at the picture starting on the left is the Regulator / Water Separator taking 150 psi from the compressor and reducing it to 100 psi - the water separator is the float type that blows water down the small pipe to the outside.

This is followed by an 'oiler' that although in circuit is NOT used or filled - it happened to be integral with the regulator so stayed there.

Then comes the new desiccant filter made exactly like the previous one.

This is followed by a large volume filter / water separator, just being used as a filter, as by this stage there should be no water to separate 

Not pressured up yet as you can see from the gauges as the Loctite 542 needs another hour to go off.

[awemawson]

So the time has come for the first regeneration cycle of the self indicating silica gel crystals of my main air dryer. (This is the one that all the air from the compressor flows through)

Unscrewing the housing showed that the outer layers must be most exposed to the moist air, as tipping them into a baking tray there seemed to be far less turned blue than you'd think from looking at them in situ.

They are now in the oven at 140 deg C for a couple of hours. Meanwhile the dryer is back in service, as the 3 Kgs of silica gel that I originally bought not only filled this dryer and the one I fitted to my CNC Plasma Table, but was enough to re-fill this one while the fan oven does it's stuff driving off the water.

I'm really pleased with the performance of this dryer - it's so simple and yet seems very effective.

[DICKEYBIRD]

Very nice Andrew!  I'm always learning new things from your threads; thanks.

I'm curious what kind of pressure drop you're getting across it.  Does it get cooler as well?

If I ever do any more spray painting, I'll need to make one of those!