unreliable flow on mill coolant pump

[Tony]

not the end of the world, but annoying problem from a coolant pump. am hoping someone has seen this before / has a fix?

OEM pump on a Schaublin 13. I recently cleaned the system (when I got the mill). Fresh coolant, no lumps.
It's piped vertically up the back of the machine, over the head, and down to the table/spindle. It's rigid pipe most of the way, then there's a valve I use to control coolant flow, then maybe 20" of metal flex tube that ends in a nozzle.

I keep the ball valve ~25% open as the coolant pump can deliver a lot more flow than I need.

First, when I turn the coolant on, no matter how the valve is set (unless it's closed of course), I get a blast of coolant that must be going Mach 5. I swear if it's pointed at my vise it could knock it out of alignment. water jet cutting has nothing on this.

Initial surge makes such a mess it's become second nature to point it down into the base of the machine, at the coolant return, every start-up.

Anyway, once that first blast settles down, I set the ball valve to the flow I want.

After that the flow slowly diminishes and I have to constantly open the ball valve to keep up with the cut; the 'stream' starts to miss the mark, so i open more and more to keep up with where I want the coolant to go.

After stopping the spindle/coolant to measure, clean chips, etc, the process starts all over again. fiddle with the ball valve, start the machine, initial coolant blast lifts my mill off the floor, throttle down, play with valve during entire cut.

Initially I thought I was low on coolant (ie level gets low, coolant tank can't keep up with what the pump is putting out, flow goes down), but the tank is topped up and I don't have a pool of coolant anywhere that would suggest it's not getting back to where it's supposed to be.

"My Bridgeport never treated me like this"

[Michael Edwards]

I know the problem of which you speak.
On a power hacksaw I fixed this by pumping the coolant up to a pint sized tank. Then it gravity flowed down to the nozzle.
The pump moved way more coolant than I needed, so the extra flowed out an overflow and back down to the main tank.
This works well, no big splutter when you start up. I would also install two valves in series by the nozzle. Use one for on/off and the other as a flow control.


Edit to add: My Kearney & Trecker 4H does the same as yours. I use cutting oil in it, so it would probably not be a good candidate for the gravity feed.

[RichR]

"My Bridgeport never treated me like this"

Is it over 50 years old? Maybe it has a prostate problem.

[The Artful Bodger]

I would be suspicious of the metal flex tube. Most of them have a thin wall plastic inner tube that gets kinked and twisted and causes all manner of frustration until dealt with.

[MrFluffy]

My surface grinder pump does this, but it occasionally tails off during a long grind to the point the flow isnt enough.
I have however recently found that if I kick the coolant pump and then turn it off and on again its fine so wondered if mine is spindle bearing issues, but the odd thing is it doesn't do it every use consistently. I keep meaning to pull the pump when it stops and check the impeller isnt clogged up though the suds tank is relatively clean inside.

[IanPendle]

Tony,
Is it possible that you have air trapped in the pump or line? Air will compress when the pump is switched on and then rapidly expand when the valve opens giving the powerful 'squirt'?

Just a thought, and no idea how to fix it!
Ian.

Edit: I meant the expanding air pushes any coolant ahead of it when the valve opens.

[Mark Rand]

Not so much that as if the pipe is full of air, the flow control valve won't limit the flow noticably, so the pump can move the coolant as fast as it's able to. When the column of coolant hits the valve, it's travelling at a high speed and there will be a short term splash of coolant out of the nozzle until the valve controls the velocity of the column of liquid. I get around this on the surface grinder by having the valves shut when I've finished with the grinder. Then I turn the pump on and crack the valves. That way there's more control of the initial flow.

[garyhlucas]

Nah,
The problem here is simple. I work in wastewater, solids all the time. When a valve is partially closed almost all of them produce an opening I call a Smile. Quarter moon shaped with sharp narrow corners. For a partially open valve the flow starts out fine, and then solid build in the corners reducing flow until you open the valve full again letting them pass.

The best flow control is an orifice plate in the line with just one round hole. Two or more holes won't work, a chip will staple across the openings and it is downhill all the way from there. So the best thing for a coolant nozzle is to have tips with different hole sizes to set the flow.

If you insist on a valve the thing to do is to properly size the valve. When a valve is used for flow control it must often be way smaller than the pipeline size. You want to pick one that gives you only slightly more flow than you want when fully open, such that you only will close it say 30% or so at most. Example: I had a pump with a 2 inch pipe going into a tank. With a 2" ball valve on the line I couldn't hold 20 gpm. It would drift right down to 10 gpm in an hour. I did the calculations for the valve needed using the CV of the valves (Look up CV algorithim) and a 1/2" ball valve would give me 30 gpm fully open at the pressure that the pump produced. Closing it just 20% got me to 20 gpm and it has remained there for years.

[CalM]

Have you tried just opening the control valve fully, and letting it run for a while?

Could be there is some accumulation that builds at the pump inlet. Something that "floats away" every time you shut down, then builds a blockage as the pump runs.

I had a mouse take up home in the sudspump on the Famco band saw. Nothing short of a complete tear down and wash out made a difference.

[Tony]

wow, all great suggestions, thanks, those give me some things to consider.

Except for RichR.. I'm not giving it a prostate exam.