vacuume coolant system

[DrStein99]

I read an article in machinist magazine which explains how to build a flood coolant system with a vacuume. The article isnt online and is kinda vauge.

Its basically a 2 hp shop-vac, attached to a drum - that has a basket inside to capture to swarf from milling.

With a check-valve and filter line comming out of the bottom, which is where the resivour of coolant resides, is a line to the coolant pump.

The article doesnt explain if the vacuume is supposed to run at the same time the pump is operating, which is what I want it to do. Ideally, I want the coolant fluid on the milling table and a vacuume hose sucking the swarf and fluid right up again; so that the mess doesnt fill the entire garage like it has in the past.

I was wondering if anyone had any advice on a low-cost pump to use. I see there are a few at harbor freight which are in my range. It must be an external pump, not a submersible one. I was wondering if at all it was possible to pump from that fluid resivour while the vacuume was running.

I dont have too much experience with pumps and dont want to buy something that has the pressure of a firehose, just simply want to pump the water-soluable oil solution up about 6 or 10 feet into my loc-line hose.

[Geof]

Originally Posted by DrStein99 ...Its basically a 2 hp shop-vac, attached to a drum - that has a basket inside to capture to swarf from milling....With a check-valve and filter line comming out of the bottom, which is where the resivour of coolant resides, is a line to the coolant pump.....The article doesnt explain if the vacuume is supposed to run at the same time the pump is operating, which is what I want it to do......I was wondering if at all it was possible to pump from that fluid resivour while the vacuume was running....

I have given some thought to a system like this but on a much larger scale; a big central coolant collection tank serving many machines with a pump delivering coolant back to all the machines via a piping system.

My conclusion was that the delivery pump would need to be inside the vacuum tank. The reasoning is that this tank is at a slight negative pressure, a good vacuum cleaner can pull a negative pressure equal to maybe 80 inches of water pressure, this is about 3psi.

This means that any pump located outside the vacuum tank has to suck stronger than this; in other words the outside pump has to be capable of sucking water up from a distance of greater than 80 inches. Few centrifugal pumps can do this, if any, and none can do it once they have lost their prime and are just filled with air. This I think would be the biggest drawback; when the outside pump stopped the fluid would be drawn back into the tank unless you hade a non-return valve. And this non return valve would be in a location where chips could get into it making it leak. And a non return valve adds drag which is the same as having to pump to a higher level.

However a pump mounted inside the tank is pushing fluid out against the 3psi inward pressure; just about any cheap centrifugal pump can do this. The pump is running fully immersed so it stays cool as it is supposed to. A normal tap water filter element can be put on the discharge of the pump so that no chips get through. This is needed because a non return valve is still necessary but this time it is in the pressure line. The filter and non return valve can both be outside the tank and it would only be necessary to have some sort of gland fitting for the hose from the pump to seal where it comes out of the tank. Actually if it came out of the top hardly any seal is needed because the vacuum cleaner will keep up with a small amount of leakage here.

[DrStein99]

The plan calls for a horizontal check valve, and filter outside of the tank. He explains there must be a check valve there to stop the vacuume from sucking fluid back from the pump.

The line also has a filter on it to trap elements from damaging the pump and clogging the lines. The author claims he uses a vane pump and says a centrifuge pump will work as well, but must be mounted higher. I have to get the article out of my car he gives the specifics on the pump (I think 10 psi and at least 300 gph pump rate).

I believe a check valve just outside the tank, and a check valve near the nozzle end will make sure that the pump is always primed, even when off - the valve near the nozzle will stop the fluid from returning back down the hose - so the next time the pump activates the line pumps immediately.

[Geof]

A vane pump will work maybe even without a check valve because these pumps are self priming. Also vane pumps can pull much better than a centrifugal so they will pump fluid out against the negative pressure in the vacuum tank.

[DrStein99]

Well, one thing is for sure - its going to be a poke in the dark for me. I dont have a bunch of pumps to test with, or any experience with pumps. I pretty much have to just buy something (that I can afford) try and hope it works.

Are ALL self priming pumps - vane pumps?

[Geof]

Originally Posted by DrStein99 Well, one thing is for sure - its going to be a poke in the dark for me. I dont have a bunch of pumps to test with, or any experience with pumps. I pretty much have to just buy something (that I can afford) try and hope it works. Are ALL self priming pumps - vane pumps?

Some gear pumps will self prime but they are not really suitable. Even a vane pump is not totally suitable because it is not very tolerant of chips which can damage the vanes.

Your least expensive route is to get a cheap submersible centrifugal pump and put it inside the vacuum tank as I discussed.

[DrStein99]

Ok I did some tests.

I have a clearwater pump and it appears to have far more pressure then I need, with 1" input/output ports. Pumping water works just great.

So for the tank, I salvaged a pool filter tank from the trash (actually got the entire assembly, so it has fittings for two hoses and a drain at the bottom theoretically is what I need).

I filled it 1/2 way with water, and hooked up the shop-vac to it. Without seeing whats going on inside, it appears to suck the water up all around the inside of the tank, causing lots of air on the pump-feed hose. I dont have a check valve yet, but dont know if that would cure the issue.

I thought about filling the bottom half with small pebble rock, or a sponge to keep the fluid from jumping around, much like a bladder works in a automotive gas tank.

[Geof]

Originally Posted by DrStein99 Ok I did some tests....I filled it 1/2 way with water, and hooked up the shop-vac to it. Without seeing whats going on inside, it appears to suck the water up all around the inside of the tank, causing lots of air on the pump-feed hose. I dont have a check valve yet, but dont know if that would cure the issue...

This is probably just the volume of air flowing through the tank; there will be a lot of air turbulence in the tank. Your shopvac is working as a vacuum cleaning; sucking a large volume of air and picking up a small volum of chips and coolant entrained in the air flow.

If your air inlet is directed down it will be impinging directly on the water in the tank and really stirring it up. Ideally I think what you need is tangential entry with the outlet to the shopvac in the center top. This will create a cyclone effect inside the tank so any water splash or spray will get flung to the periphery and not tend to get near the outlet.

The cyclone may everntually get the water moving in bulk so you would end up with a massive vortex. Your pebble idea would stop this but just a couple of plywood baffles would work just as well. Have the top of the baffle slightly below the level of the tangential entry port.

To get your tangential entry it may be just a simple as putting a 90 degree elbow on the inside of the entry port. If your are lucky this will be a pipe coupling welded to a flange and there will be threads accessible from the inside.

[DrStein99]

So if I read you right:

The input (pipe that sucks the swarf and fluid) should face directly down in the center.

The output (pipe to the vacuume) should face 90 degrees to the wall.

My current setup is very simular to that. There is a pipe in the ceter that faces down (input). There is a cage around the top of the pipe, with an open area to one 1/4 of the pipe that leads to the shopvac.
------------

I'm unclear to your idea about the plywood baffles.

[Geof]

Originally Posted by DrStein99 So if I read you right: The input (pipe that sucks the swarf and fluid) should face directly down in the center. The output (pipe to the vacuume) should face 90 degrees to the wall. My current setup is very simular to that. There is a pipe in the ceter that faces down (input). There is a cage around the top of the pipe, with an open area to one 1/4 of the pipe that leads to the shopvac. ------------ I'm unclear to your idea about the plywood baffles.

No; the reverse: The inlet should direct the flow around the inner wall of the tank. The outlet should go up out of the center of the top.

This way the incoming stream of air, liquid and particles is forced to swirl around the tank and the heavier stuff tends to cling the the sides; centrifugal force and all that stuff. The air being lightest moves to the center located outlet with no difficulty.

The plywood would be in the shape of a cross stood in the bottom of the tank. Imagine if you had a bucket full of water, you can stir it and get the whole mass of water rotating. If you blew a strong air stream just on the surface of the water eventually this will induce it to start rotating. But if the water is restrained by the plywood walls of a cross it cannot start a strong rotation.

[DrStein99]

Ok I ran some more tests. As it turns out, the tank must be ABOVE the pump in order for this to work. Once I put the tank 12" above the pump - the gravity naturally pulled the water to a pressure the pump could handle. The pump has lots of pressure to pump the fluid at least 10' high.

So this whole concept will work!

Now all I have to do is make a bag or basket to catch all the swarf / trash. I think I'm going to scout around for a 10 gallon bucket, that seems like it would be easier to work with then this odd shaped round sphere.

[DrStein99]

Ok a crutial part of this whole thing is also the size of the container. A 5 gallon paint bucket is definately too small, it sucked the water straight up.

Hunting around for an airtight 10 gallon container with a lid doesnt seem easy, I'll have to go to the scrap yard to find that.