Building a cooling mist system

[beanbag]

Right now, the system consists of separate air and coolant lines, which converge pretty much right at the nozzle. Ideally, I'd want a system that shoots a narrow solid cone of air, with a slight stream of coolant in the middle. That way, the air has some spread to blow away chips, while the coolant is more directly aimed at the tool bit.

I made a concentric nozzle, where the air nozzle is just the usual loc line nozzle with 1/8" opening, and about 1/4" inside of this is a small nozzle for the coolant. There are separate valves for air and coolant.
The main issue is how can I determine whether the coolant is coming out as micro-sized droplets, or as a completely atomized mist? I've read that the latter is bad since it tends to fog up the shop and you end up breathing it.

Based on some reading, it seems that what makes this difference is the configuration of the junction (something about meeting at right angles and particular sizes of the air and coolant junction, as based on posts trying to replicate the Bijur spray mist system) and also something about the coolant being pressurized vs getting sucked out by the Venturi effect. The latter reason makes no sense to me because the valve is downstream from the pressurized coolant, and if you try to use that to regulate flow, then it will take all the pressure drop, and the coolant at the nozzle tip gets "sucked out" anyway.

Short version of my questions:
What's a good nozzle design to achieve the effect I want?
How can I tell if the mist is overly atomized?

Thanks

[Jimno2506]

The tank is pressurized.
The air goes into the Loc-line.
The coolant is in a tube connected to the end of the Loc-line.
The coolant nozzle is tapered at the end, it's a hex shaped to allow the air to flow around it, converging on the nozzle.

The coolant comes out from the center of the nozzle and the air carries it to the tool.

Hope that helps.

Jimno

[macona]

If your shop starts filling up with mist you know youre getting mist!

As long as you keep your pressure low ~25psi you should not get mist. The nozzle for a bijur mister has a real small hole. Something like .010". The tank the air goes into has a pressure regulator, reservoir tank, water separator and a solenoid. The regulator supplies pressure to the nozzle and to pressurize the coolant.

For the price of a NOS bijur tank on ebay you cant hardly make one.

[Jimno2506]

When I'm running the RZ at 30k rpms I blast with about 90 psi, virtually no coolant and have never gotten fog.

I could only imagine how bad the fog would be...eeesh.


Jimno

[jugs]

Whilst oil mist cooling/lubrication is ideal for metalworking, it is not suitable for humans.

It can lead to-
- irritation of the skin or dermatitis (http://www.hse.gov.uk/skin/index.htm)[1]; and
- occupational asthma (http://www.hse.gov.uk/asthma/index.htm)[2], bronchitis, irritation of the upper respiratory tract, breathing difficulties or, rarely, a more serious lung disease called extrinsic allergic alveolitis (EAA), which can cause increasingly severe breathing difficulties in recurrent episodes, following repeated exposure.

Fine particulates of oil enter the lung & cover the surface preventing oxygen from entering the bloodstream. Few face masks will prevent the most dangerous particulates (.5 -3.5um)

If the machine is totally enclosed make sure it has an effective negative air pressure filtration system.
If the machine is open, use an proper air feed mask (expensive setup but cheaper than a lung transplant), or better still dont use a mist system.

[RTPBurnsville]

I believe you are trying to build a system similar to the Trico micro-drop system. I have been very interested in also building such a system and just today brought home from work a small discarded gas cylinder to use as the coolant supply tank and some old flow control valves. From what I have read such systems pressurize the coolant, contain a pulse type timer for coolant supply metering, and separate coolant and air supply lines. I also picked up a HF air brush thinking that it could be modified in some way for a nozzle.

Clippard and Bosch are two suppliers of pneumatic controls and valves which could also be used. Bosch has a pneumatic pulse generator which looks interesting.

Robert

[S_J_H]

Right now, the system consists of separate air and coolant lines, which converge pretty much right at the nozzle. Ideally, I'd want a system that shoots a narrow solid cone of air, with a slight stream of coolant in the middle. That way, the air has some spread to blow away chips, while the coolant is more directly aimed at the tool bit.
Yeah that's exactly how the Fogbuster system works. http://www.fogbuster.com/Frame.htm
I have a fogbuster system and it works really well!
http://i109.photobucket.com/albums/n48/S_J_H/x3%20mill/test013.jpg
http://www.fogbuster.com/Frame.htm

But it's pretty expensive so many have made their own systems based on the fogbuster nozzle. I think if you do a search over at www.cnczone.com a few guys posted how they built their homebrew fogbuster systems.
Steve

[beanbag]

According to this post on a mist system based off of a patent:

http://www.cnczone.com/forums/showthread.php?t=14910&highlight=mist

the coolant opening is .04". This is a really big opening, and I think you are basically relying on the valve to prevent coolant from gushing out. It seems more like Venturi effect, and not pressurized coolant. On the other hand, a .01" opening sounds a bit more reasonable.

I also wonder about the really long nozzle (6"). Is the coolant and air supposed to be mixed that far back? If the mixture travels down a long tube, does it atomize more or condense?

[Carld]

I experimented making my own and then went to Little Machine Shop and bought one of theirs. It works just fine and it was cheap compared to others.

[Jimno2506]

I use Tricool MD-7. Here's what it says on Trico's website.

APPLICATIONS:

MD-7 is a pure synthetic based lubricant designed for use in various metal cutting operations. With the use of a Micro-Dispensing system, more effective machining productivity and saving in fluid costs can be achieved, while waste and disposal problems are reduced. MD-7 is non-toxic, non-flammable, non-corrosive, and will not become rancid.

FEATURES:

* Zero toxicity - contains no petroleum or chlorinated derivatives.
* Reduces heat and friction.
* Works in a wide-range of micro-dispensing systems.

I'm sure the guys who are using the mist system are familiar with the safety precautions necessary.

Regards,
Jimno

[beanbag]

What about Tricool TC1 or KoolMist 77? Water soluble and much cheaper. Both claim to be non toxic.

[S_J_H]

beanbag,
The fogbuster nozzle works real well. It does not "atomize" the fluid to any real extent. The fluid travels along with the air in a "semi solid stream". The nozzle can be adjusted from an almost invisible stream to a heavy soaking stream. But it does not really make a "mist" or create any sort of "foggy cloud" that hangs in the air like some mist systems.
Making an air/fluid atomizer nozzle is fairly simple. I dabbled in that area for a while making nozzles to finely atomize alcohol for water/alcohol injecting systems for supercharged/turbocharged engines. I was using high pressure pumps without air at around 300psi that worked nice with swirl designs. But air nozzles of course will atomize much finer yet( think HVLP paint guns).
Getting the nozzle to work like the Fogbuster does is probably a little trickier. I am pretty sure you can sneak a peak at their nozzle patent and just do a little testing with the design. It uses a pressurized fluid container regulated at the inlet and both air and fluid are admitted to the nozzle at around 10-15psi with equal pressure. The nozzle will work fine as low as 5psi though.
I use this system on my mill, lathe's and bandsaw as it's simple to move around. I have never found the need for a whole lot of coolant flow so there is very little "wet mess" to deal with.

I use KoolMist 77. It works very good IMHO.
Steve

[Circlip]

I'm sure the guys who are using the mist system are familiar with the safety precautions necessary.


Yes Jimno, you're possibly correct, this is however an assumption that everyone is as wise as you but doesn't warn the "Must have the latest gizmos cos Whoever has one" and after all, if it saves just one getting it wrong, it's got to be worth it??

Thanks Jugs.

Regards Ian.

[oldtiffie]

I wonder why totally enclosed high-productivity machining centres use flood coolant instead of mist coolants?

I'd have thought that copious quantities of low-speed/velocity flood coolant/tool lubricant would be better in an "open machine" and HSM environment.

Cost of replacement fluid lost in a "mist/spray" in that HSM environment should be minimal. What is the cost and any other benefit over the capital and recurrent costs of a mist system which will probably require a compressor running to feed it as well?

My first stop would be the Material Safety Data Sheet (MSDS) for the "mist" product.

I suspect that while it may not do you a lot of harm, I can't see it doing you a lot of - or any - good. Same applies to a lot of metal and electronic stuff in the shop.

It may well be one of the number of things in a shop that seems to make many machinists older and ill-er (???) before their time.

[pcarpenter]

I am no expert, but in high-volume CNC machining, the flood coolant both cools and lubricates as well as washing away swarf that would pile up inside pockets etc when milling.

As for the mist coolant issue...it's an important distinction between true mist coolant systems which make fog and something like the TRico micro drop systems which can put out tiny amounts of coolant without making a fine mist. The latter sounds ideal in a home shop.

My Bridgport came from it's commercial environmnet with a mist coolant unit hanging on the side. As I restored it and cleaned it up I vowed that I would never use the stuff. The machine looked like it had been sprayed with a mixture of swarf and wallpaper paste so that metal shards were adhered all over the column etc. As mentioned, the vaporized crap cannot be good for the lungs.

On the other hand, the micro-drop coolant/lubrication systems make a lot more sense. They are really expensive to buy (several hundred dollars) and use a special fluid that is quite expensive (reference the MD-7). I lucked out and found some really cheap and have plans to make a coolant dispenser of that type. There is a web site or three out there that talk about doing just this sort of thing. The good news is that while the coolant is expensive it goes a *long* way. Unlike a mist coolant unit that seems to blow a cloud of mist everywhere, these units can put tiny amounts right on the work. I remember reading on the Trico site that as little as an ounce per shift could be used (or something on that order). In these systems, the fluid (I noticed its about as thick as light cutting oil) is under pressure rather than using a siphon-spray type system.

Paul

[RTPBurnsville]

Paul,
The micro-drop system is what I mentioned earlier in this thread that I also have an interest in. I don't want a mist system as the shop is in my basement! Do you happen to have links to the web sites?

If I did the calculation correct the micro-drop system uses 1 oz per 8 hour shift per the documentation (2 drops per minute rate). If 1 oz = 30ml = 30cc then a drop should be 0.03125 ml.

I am not sure what type of pump/value would be best in this application as most of the off-the-shelf systems appear to use pneumatic only controls. It would be easy with some electronics and a solenoid control valve, not sure about the pneumatic only approach as that is not my area of knowledge.

I read someplace that the synthetic oils are more resistant to buildup over time compared to the vegetable based oils. No idea if that statement is true however...

Robert

[pcarpenter]

Robert-- look at the very end of the list of files here. The files named zero_fog_mister.xxx are the ones you are interested in (I think). I had no bookmarks and just found this after a bit of web searching and reading other threads elsewhere on the same topic. When I got there, this looked familiar to me. For me, this is on my "someday" list...and maybe pretty far down so I had better save off the files before they disappear:

http://metalworking.com/Dropbox/_2002_retired_files/

I read that the vegatable based micro-drop lubricant tended to polymerize, too. When I found some of the coolant cheap, the place that had it had both. I picked the synthetic stuff and lucked out. My thinking was more along the lines of it going rancid since I would not use it that fast, but I had no idea that the natural stuff would turn gummy. When I went back, the rest of it was gone.

Paul

[S_J_H]

he good news is that while the coolant is expensive it goes a *long* way. Unlike a mist coolant unit that seems to blow a cloud of mist everywhere, these units can put tiny amounts right on the work.

That is for sure! When I bought my Fogbuster system I also bought 1 gallon of the KoolMist 77 concentrate. This was quite a few years ago now. I use the system fairly often but I am quite sure 1 gallon of this coolant concentrate will last my entire lifetime :D

Steve

[oldtiffie]

Seeing as asking did no good, and as it is important to know (of and about), I went and got the Kool-mist77 Material Safety Data Sheet (MSDS).

It seems that Kool-mist77 is pretty safe to use - but there are a couple of normal hygiene requirements as is the case when working in a shop at any time on anything.

Just about every product will have a MSDS.

I make a point of getting and reading them as I can make and informed choice instead of relying on shop gossip and ignorance.

http://www.setonresourcecenter.com/msds/docs/wcd00026/wcd026cf.htm

I still prefer flood coolant as I like the better cooling. Any blockages can be cleared as we go.

[Boucher]

I have just finished building my clone of the Bjur mister. It works great. It will be a few days before I will be back to the shop and able to post pictures. The essence of the design is to be able to regulate the flow of both air and fluid independantly. The mixing tube (not nozzle) is about 6 inches long and is 0.090 id then reduces to 0.040 for the last 0.750 inch. What this does is allow the air to sweep the drops along but nor atomize them. When you are taking a cut that generates considerable heat, if you can't cool it with the air flow then the liquid will boil off and vaporize and you wind up breathing fumes anyway. In the production shops they are taking heavy cuts and use flood coolant. In the HSM environment the mister works pretty good. I used a 0.125 copper tube with a mig tip for the end piece.

[beanbag]

I tested out my system by spraying the nozzle down a clear tube, with a glass jar at the end. I didn't see any fog, nor did any other surfaces condense up with anything. So I think in terms of fogging, that will not be a problem.

In any case, the coolant I am currently using, Rustlick ws 5050, "only" has a toxicity rating on 1. I think that means that it is only slightly bad for you, it doesn't cause cancer, and you probably won't die if you breathe some of it. (MSDS claims no acute inhalation hazard). If I want to feel safer, then I might spring for the Coolmist 77.

My setup follows some aspects of the Bijur system:
The fluid is introduced when the airstream narrows down to about .1", although in a concentric fashion and not perpendicular.
The 1/8" "extension" tube on my setup is only about 1" long.
The final section only narrows down to about .063 and not .04".

The system works ok with a couple of minor problems. One is that because the nozzle necks down at the end, it creates a bit of pressure at the liquid injection point. If I don't have enough pressure on the fluid, air gets pushed into the fluid line. Secondly, this pressure is dependent on the air flow rate, so if I turn up the air, then the liquid portion decreases, or even goes to zero.

Second problem is that the coolant at any level above a minimal amount comes out kind of sputtery. I guess at least it is better than "foggy".

Third is that I think my fluid control valve is not so good. It is some cheapo needle valve from OSH hardware, and doesn't keep it's position very well, and also seems designed for larger flows. I guess I will need a better one.

[jugs]

What about Tricool TC1 or KoolMist 77? Water soluble and much cheaper. Both claim to be non toxic.

Thalidomide, asbestos, smoking etc have all been declared "safe" in the past !!!

Breathing anything other than clean air / water vapour will not do you any good, nitrogen is non toxic but in the wrong concentration it will kill you.

That said, we expose ourselves to all sorts of airborne crap every day eg, today I've welded some steel, cut wood, painted a chimney & burnt the toast :D

Flooding makes more sense to me so what are the advantages of micro-drop / oil mist systems

[John Stevenson]

That said, we expose ourselves to all sorts of airborne crap every day eg, today I've welded some steel, cut wood, painted a chimney & burnt the toast :D



You are going to get serious grief :rolleyes: :rolleyes:

[Jimno2506]

Flooding makes more sense to me so what are the advantages of micro-drop / oil mist systems

No coolant means very little mess for those of us without an enclosure. Enclosing something like my RF clone would be a PITA. Now all I need is a vacuum cleaner (Shop Vac) pointed in the general direction of the swarf. The rest gets vacuumed off the floor and put in the trash.

Cost is another issue. I bought a gallon of MD7 about three years ago and have about 1 qt left. I don't run every day but have no rancidity issues (I have a horizontal band saw so know what that's like). The air compressor is necessary anyway. Don't need a pump.

Coolant will vaporize/evaporate as well.

I really like the MD-1200.

Regards,
Jimno

[Boucher]

For my mister I used a 10" water filter housing. I installed a center draw tube with a 3/4 pvc coupling and piece of 3/4 pvc pipe. This is a tight slip fit so no glue is requirred. A air pressure regulator is used to pressurize the volume above the coolant. The air and coolant are then at the same pressure. There are both a small ball valve and needle valve in both the air and liquid line. The needle valves are used to regulate the flow and the ball valves provide a quick on/off function without changing the regulated flow balance. This has a lot more components than the Bjur mixing valve but they were things that I had in stock. The only thing that I purchased was the 1/8" copper tube and the Mig tip. This works really well for the kind of things that I do in my shop.
Previously I used WD 40 spray cans for cutting Aluminum and brushed or squirted cutting oil on steel. My worst problems in milling projects has been heat and work hardening. So far this mister has worked better than the old way. The old mister that I had was screwed up by the previous owner so I can't make a fair comparison to it. If you had to buy everything this could be built for less than $100.

[beanbag]

After some reading around, I decided to use a needle valve for a RC aircraft engine. I don't think it was designed for positive pressures, so it leaked out the knob, but a blob of silicone grease seemed to fix that. It is capable of metering out really small amounts of fluid, like one drip every 10 seconds. It only cost me $10.