What's new
What's new

Setting expectaction of coolant coalescer performance

adamm

Cast Iron
Joined
Sep 19, 2008
Location
Kingston, ON
I have been manually skimming my coolant tanks, and as you can guess, that sucks.

I bought a coalescer for one of my mills to test something more effective, and less hassle. It worked well to skim the free oil from the coolant, but it there is still a brown layer of partially emulsified tramp oil that it isn't separating. I expected the coalescer to strip the oil from this mixture, but it isn't. It passes the coolant through the coalescing media, and just mixes the brown layer back in. The brown layer disappears after skimming for a while, but when I shut off the unit, the brown layer floats back to the surface within 15 min, and covers the machine sump.

My question is whether my expectations are realistic. Do any of the coalescers strip tramp oil from well mixed layers?

The coolant is Trim E206. I'll refrain from naming the coalescer for now, because I would like to know if any of them are effective at stripping tramp oil from mixture, rather than just separating free oil from the surface.
 
Mine does, but it is just some cheap stuff I scabbed together. It sounds to me that your pump is too big and it is not giving the coolant enough time in the coalescer to properly separate the oil. I have my pump and bubbler set to run 4 times a day for 15 minuts, when I am running my mill I diable it. After the first run or two there will still be a slight film of oil on the coolant but after a couple days of sitting there is just a few spots of oil left on the coolant. I don't use any media in my coalescing tank, which is just a 1 quart plastic jar, as it is just more shit to clean and about every 2 months I dump the coalecer tank into the machine to clean it. I have a 1 micron bag filter between the mill and my coolant tank so that traps any solids to keep them from circulating through the system.

I ASSUME you are using a refractometer to keep your coolant concentrate at the correct concentration? If not then that may be your problem. I didn't and my guess was way off, like 25-30% rich, using Trim C320. One of the issues I had back then was exactly what you are describing.
 
It sounds to me that your pump is too big and it is not giving the coolant enough time in the coalescer to properly separate the oil.

The pump seems fast to me. It is about 300 gph, and seems like it flows through rather fast. The separator tank is about 4.5 gallons, and the pump is around 5 gpm, so the tank changes over in less than 1 minute.

Having said that, the oil won't separate from the emulsion when it sits motionless over the weekend. I don't think your set up without the media would strip this oil from the coolant. I think it is too mixed to gravity separate. I expected the coalescer to coalesce the oil droplets enough that the drops would get big enough to separate. I don't think that is happening.

I ASSUME you are using a refractometer to keep your coolant concentrate at the correct concentration? If not then that may be your problem. I didn't and my guess was way off, like 25-30% rich, using Trim C320. One of the issues I had back then was exactly what you are describing.

TRIM says to run it 3%-10%, I'm at about 6% for general purpose machining. Were you outside of the range completely, or just rich relative to the suggested machining type?
 
I was outside of the range, by more than 20%, probably way more. It sounds like you should try restricting your pumps flow as much as possible. I had a hard time sourceing a pump as most flow too much. I ended up with a seal-less magnetic drive pump that probably does less than 1 gpm. I have it restricted with a .16" id hose on both input and output, it is not submersible, so it flows around a pint a minute now. The "special" media is probably polyethelene, my system works fine without it. My coolant system is on a 30 taper Kitamura. I know it dumps plenty of way oil into the coolant and since I machine mostly aluminum I am pretty sure the oil is very well mixed into the coolant while running it. Granted my coolant is a full synthetic so it may not emulsify the way oil as much as yours but I know it does it enough to turn the coolant milky.
 
Part of your problem is the coolant. Soluble oil coolants are old technology. The new semi synthetic coolants are a lot better. They don't have as many emulsifiers as soluble oil coolants, so tramp oil separates much better. Normally the tramp oil separates so well a tube skimmer is all I need to clean the coolant. The machine also stays much cleaner. No more brown goop all over everything.

I had a keller coalescer that I used on soluable oil coolant and I had about the same results as you did. I eventually sold it but I think it would have worked great with a new semi synthetic coolant.
 
Part of your problem is the coolant. Soluble oil coolants are old technology. The new semi synthetic coolants are a lot better. They don't have as many emulsifiers as soluble oil coolants, so tramp oil separates much better. Normally the tramp oil separates so well a tube skimmer is all I need to clean the coolant. The machine also stays much cleaner. No more brown goop all over everything.

I had a keller coalescer that I used on soluable oil coolant and I had about the same results as you did. I eventually sold it but I think it would have worked great with a new semi synthetic coolant.

Well, that's not the answer I want, but it's the answer I need. On the plus side, I am close to the reorder point on my coolant concentrate. I guess it is time to look at switching to something else.
 
Get a wheel skimmer. So cheap and work 100%.

Unfortunately, there is no room for a wheel skimmer on any of my machines. All the tanks are inaccessible from the top when the machine is running. They can be taken out for cleaning, but then all the return coolant runs on the floor, rather than back to the tank. One of the attractions of the coalescer is that the pump sits in the sump, and I just need room for the power cable and two hoses to come out of the sump. I can make that work on all my machines.
 
Unfortunately, there is no room for a wheel skimmer on any of my machines.

Same issue, here. Which is why I am building a coelescer right now. ( literally ) The drum arrived yesterday ( Amazon ) and the pump got here Saturday. I'll cobble the bits during the week.

The pump seems fast to me. It is about 300 gph, and seems like it flows through rather fast. The separator tank is about 4.5 gallons, and the pump is around 5 gpm, so the tank changes over in less than 1 minute.

Very much so. Having used a professional unit that worked very well when I worked for others, and after speaking with others that have units themselves, I specifically looked for and bought a pump that would be much slower. I purchased a pump that is rated at 160GPH with zero head, slightly less with mild rise.

I am using a 14 Gallon drum. The plan is to put it on wheels ( small, round drum dolley ) and move it from machine to machine, as needed.

Was planning on posting a thread if it works out the way I expect.
 
IMHO you dont want a centrifugal pump. Use something like a diaphragm or peristaltic pump and you get a lot less mixing going into your separator.

I do think your expectations are also a bit high, would expect there to be some film after it sits off for a while. Just the nature of coolant, the colascer though should be doing enough to keep it in check that its not a major issue though and plumbed properly should be drawing it off the surface of the tank, you don't want a fully submerged pick up!
 
Designing and making the foat pickup is the hard part. You have to draw from the surface. If mine is an 1/8" low it doesn't get any oil. I only have a 2 quart separator with about a 2-3 minute residence time and it works well enough. A better separator tank design would help more than a better pump design in my case.
 
I been using semi synthetic for years now @ $107/pail. Probably have been ruining it too lean, as i can't seem to shake corrosion of tool holders.
Just got some old school soluble @ 75/ pail. Leaves things oily, not sticky. Think it will knock the rust prob.
IMHO, disk skimmers suck. end up pulling more water than oil.
 








 
Back
Top