Oil coolant pumps – which type of pump

[RJWParticipant @rjw]

Posted by Gordon W on 02/02/2017 10:10:03:

Petrol pump from old car- diaphragm type can be driven off eccentric or cam, suggest old fiat or skoda. We used one to pump a mix of hydraulic oil and white spirit for many years. Output easily adjustable with a tap.

This type are also used as a factory fitting on the Viceroy Sharpedge planer and chisel blade grinders.

John

[petro1headParticipant @petro1head]

Posted by John Reese on 29/01/2017 17:58:25:

I have used submersible pond pumps successfully for many years in water based coolants. Not sure if they are compatible with straight oil.

A lot of old machines used gear pumps. As already said, they don't tolerate particulates very well. If worn they don't prime well. If you want to go the gear pump route chjeck out carbonator pumps.

What size pond pump would you recommend?

[Martin NewboldParticipant @martinnewbold]

Wow this has become a really intereasting thread , with lots of interesting ideas. I am really impressed with some of the ideas  .I didnt realise that car motors cam in two types . I was really hoping for a 240v system but that said the speed control version with the car pup looks great. I thought no one had done this

 

M

Edited By Martin Newbold on 03/03/2017 15:01:41

[geoff adamsParticipant @geoffadams14047]

I use a pond pump on my bench top cnc mill a domestic light switch to turn it on and off a plastic storage box as the sump being using this for over 2 years with no problems

hope this helps Geoff

[Howard LewisParticipant @howardlewis46836]

If using an ex washing machine pump, the risk is that neat oil will almost certainly degrade to O ring seals. May be longer lived with soluble oil; but do check. Could be worth replacing the standard O rings with EPDM, or Silicon ones. May be relatively expensive, but you are probably only looking at a couple of seals most times, not dozens. Being centrifugal should cope with the odd bit of swarf.

Some washing machine pumps need the impellor to be kept against the housing, or leakage around the back of the impellor reduces efficiency a lot.

Gear type pumps will not tolerate swarf or particles because of the minute tooth/tooth and gear/endplate clearances. And do you really need high pressures? 4 bar will throw oil several metres. and all you want is to prject it about 100mm, which can be achieved with little pressure

(My soluble oil is gravity fed (via a syphon) from less than a metre above, through 8mm OD windscreen washer tubing, through a needle valve to reduce the flow to a steady drip. Obviously, if you want a high flow to flush away swarf, you need a continuous flow, but probably not at any pressure greater than 0.75 bar (10 psi in old money)

A diaphram type punp has been used for soluble oil on chopsaws. The diaphragm may well tolerate neat oil. The pressure is determined by the spring (That does the pumping; the eccentric merely pulls the diaphrgm down to fill the pump, compressing the spring in the process) If the valves are mica or tufnol there should be no compatability problems, but watch out for any O ring seals. As pump intended for petrol or diesel (gas oil), the seals should tolerate neat oil.

Most diaphram lift pumps only deliver 3 -5 psi. One from an in line injector pump may provide 10 psi.

The suction lift should be no greater thant six feet, (aim for minimum if possible) which should be adequate for workshop use.

Swarf in the valves will prevent it pumping, for obvious reasons.

Being intended for low viscosity oils, heavier oils may decrease flow whilst subjecting the diaphragm to higher pressures, which will decrease life. Rebuild kits used to be available many years ago, but modern pumps tend to be Kleenex "Use 'em. Throw 'em" types with the edge of the steel top rolled around the lower body.

Presumably, a SU electric pump should work O K., but a 12Volt supply needed, and they are not intended to run at full speed for long periods, rather more intermittently.

Howard

[Joseph Noci 1Participant @josephnoci1]

(The Postmaster told me my Post is to long – again!- so I had to break this post into a few…)

I am another one of those trying to make a 'fogless' coolant dispenser… Not sure why, but am nevertheless persevering..

And have spent 4 fruitless days with frustrating pump interludes.

PART-1 – the coolant applicator head – a magnetic base mount LOC-LINE head, fed with compressed air and coolant

This is the base port with the coolant inlet fitted. the second hole take an identical fitting with air pressure. The brass fitting on the right is the loc-line mounting to the applicator nozzle.

:

A burnt out JBL Tweeter was butchered for its rare-earth magnet – makes a very nice magnetic base mount.

Magnetic base in place.

The coolant inner flow tube ending at the nozzle.

Coolant tube ending in the nozzle. The nozzle is a 1mm diameter orifice in a 3mm bras tube, connected to the feed hose. Surrounding the bras tube is a air bypass nozzle, allowing the compressed air to pass by, and carry the low-flow coolant to the job.

Part 2 – the Air powered system…

[Joseph Noci 1Participant @josephnoci1]

Part 2 – The Air Powered Coolant applicator..

The Saga continues-

I started with a fully compressed air powered system. Main shop air feeds two air regulators, one set at about 3 to 5 bar, the other at around 1.5bar. The latter pressurises a coolant tank, which then pushes coolant out via a flow control needle valve, into a tube that is coaxially located inside a loc-line magnetic base assembly on the lathe carriage. The loc-line surrounding the coolant tube carries the higher pressure air flow, and mixes with the coolant at the nozzle, carrying droplets of coolant to the tool.

The coolant tank is made from 2 stainless steel camping tumblers, each about 500ml, tig welded together at their rims. At each end ( tumbler bottoms)  a threaded insert was silver soldered in to fit a pipe fitting at the bottom for coolant outlet, and at the top for air pressure inlet, and tank refilling.

A 12volt solenoid valve feeds both air regulators, cutting air to the the system, stopping the coolant flow

The coolant pressure tank

The solenoid valve and pressure regulators

The complete assembly – 100% proof obtainable from this when fired up!

It works very well, but is a real pain to stabilize. the coolant 'flow' rate is very sensitive to pressure, and the regulators do OK, but are just not good enough.  it is just simply to finicky – does not maintain the same coolant feed rate over 1 or 2 hours .

 

Edited By Joseph Noci 1 on 07/06/2017 22:28:58

[Joseph Noci 1Participant @josephnoci1]

Part 3 – Peristaltic Pumps and others…

Since the problem lies with the control of coolant liquid feed, I felt a Peristaltic pump might work better.

An HDPE housing with 4 wide ball bearings attached to a crude holder were made.

This worked very well, but the hose is a problem. The one used is far to stiff and thick walled – silicon, readily available to me in various sizes, does not work with cutting oils, tygon is only available to me in thicker tubing, unsuitable for this application. It is difficult to get Butyl, Viton, etc tubing in Namibia, and purchasing abroad is equally difficult as I would like a half dozen varieties to try…Not sure what to do here…

Then I lost my head and decide to try other pumps – a petrol fuel injection pump, for example.

I was able to get very low flow rates at about 2 volts, and a 10cm long stream at 5 volts. At low flow rates, since this was not a positive displacement pump, the flow would still vary hugely with pump or nozzle orientation.

Another useless exercise…

I tried a windscreen washer pump as well – I did expect this to be the worst since is has very poor flow performance, etc. A further issue is that the fuel flows back so freely through it that it syphons back to the holding tank, and the when you turn on coolant again, it takes forever for the coolant to exit the nozzle at the low flow rates

I even made up a non-return valve to prevent the syphon process – worked fine, but still has all the same problems of flow rate changing with nozzle height, ie, pump head affects the flow impossibly.

I still think the peristaltic pump is the answer, but the tubing…

And NO, I do not want to use a coolant sodden paint-brush, or a squirty bottle, or whatever..and I hate flooding-

I like the idea of a sort of no-fog lubricator, with variable compressed air to blow chips out of the way, but all the commercial systems I have played with have warts – flow rate is not deterministic at all, and I do not want to lose concentration of a fine machining job, trying to adjust the coolant flow because suddenly it is obscuring the job, or there is no coolant at a critical instant, etc.

Joe

PS – No idea where to post these sorts of long and verbose posts of mine – none of the forum headings seem to leave room to manoeuvre for this? If this is to long or not suitable – let me know! Just thought some folk might be prevented from >>

 

Edited By Joseph Noci 1 on 07/06/2017 22:44:25

[HasBeanParticipant @hasbean]

I use a facet fuel pump with a non-return valve. Just a fast drip feed, mostly for parting, but with neat cutting fluid. Works for me!

Paul

[MuzzerParticipant @muzzer]

Joe – have you looked in detail at how the various "fog buster" type products work and are constructed? I expect you may be able to find an exploded parts drawing / user manual etc. Somebody else has already done the sweat and tears bit.

Mercans seem to really like these systems. They aren't coolant systems so much as oil lubrication systems, so in my eyes they are only really appropriate for aluminium machining where the material can become stuck to the cutter.

That's not so much of a problem with steel – when using HSS with steel, these systems aren't going to cool the tool yet carbide doesn't benefit from (flood) coolant at the cutting rates we see in our workshops. For that application, I don't see fog busters offering any obvious benefit.

One possibility, if you have already got a fuel injection pump, would be to add a fuel injector and fuel pressure regulator (they generally operate at constant flow and need something to regulate pressure by bleeding off excess flow). You would then be operating in constant pressure mode as they are intended and could control the flow rate using a 12V PWM signal. Injectors these days have multiple holes (as many as 12 or more), giving a very fine spray. You could probably use an Arduino or similar microcontroller eval board to provide the drive signal – or a 555 if you are retro old school.

Be interesting to hear if and when you crack it!

Murray

[Mark P.Participant @markp]

Look at this on eBay http://www.ebay.co.uk/itm/121773671723
I’ve been using one of these for quite a few years with no problems.
Mark P.

[Neil WyattModerator @neilwyatt]

Peristaltic pump is what I was going to suggest for simplicity and reliability. Use silicone tubing like they do in hospitals. use bearings to squeeze the tube and avoid excessive pressure.

[SillyOldDufferModerator @sillyoldduffer]

A peristaltic pump should do very well except perhaps that swarf might shred the silicone tubing. Not much of an objection because the tube is easily replaced.

Ideal accessory for a steam-punk workshop – you could drive your peristaltic coolant pump with a Stuart 10!

Dave

[JonParticipant @jon]

Give me flood any day and can use certain through coolant tooling to great effect.

The other week I did break out the Noga mister bought 6 years ago, 3rd time its been used in that time. Originally bought to get me out of the proverbial when the lathe pumps kept going down it didn't happen, useless on boring jobs and does a bad job of cooling the product and tools.
Since then relegated to one of the mills.

Just uploaded a select few piccies of the Noga, medium, heavy and deep cuts it will not handle the heat build up of not only the tool but the cutter and machine castings and vice. Cleanup cuts on mill theyre ok, this is set to near on the max before going airborne.
The roundover part think was 8mm R really struggled and picked up a few times.
Likewise thinning the one side down 1/4" x 1" deep the mister couldn't handle the sheer heat build up, had a right fog with the steam no joke and didn't want to spend all day at it. Flood would have cooled the job and cutter and the heat wouldn't have transferred to vice and head.
However the mister just about done its job cutting the T slot mainly due to the coolant not being able to channel the coolant away enclosed.

After wasting my dosh on the mister and three English 3ph coolant pumps going down, best thing ever did was buy one of those small Vertex coolant kits in 1ph, used daily on the lathe semi pro.

[mark costello 1Participant @markcostello1]

I used a mist system to blow dark Sulfur oil on a slitting saw .053 wide x 3/8" deep into 303 SS. Oil for lube and air to clear chips.

[HopperParticipant @hopper]

I've seen several 12 volt bilge pumps from small boats used as lathe and mill coolant pumps with great success. They are designed to pump a cruddy mix of oil and water and air under harsh conditions so seem well suited to the job. They sell for as little as $25 new.

[John Stevenson 1Participant @johnstevenson1]

Jon, good post.
Coolant flow has been the bane of my working life. Set it up at the start of a job, full tank and clean lathe but by the time you are well into a job with chips in the tray acting as a dam and half the Coolant out of the tank making its way back you have lost the head in the tank with the result that the flow is now down to a dribble.
Open the flow up and eventually when the excess makes it back to the tank, you get drowned.

[HOWARDTParticipant @howardt]

I don't use coolant beyond the odd bit of oil or spray from a can. If you are drip feeding from a reservoir why not over feed the reservoir from another tank and weir back into it, thus retaining a level to drip feed from. Just a thought.

[Author]

Posts