Climb Milling

[CotswoldsPhilParticipant @cotswoldsphil]

Either the spindle is running in reverse or the photo has been flipped horizontally. You can best see the orientation of the teeth at top left.

Phil

[Neil WyattModerator @neilwyatt]

The photo hasn't been flipped, in the original you can see the headstock of the lathe in the background (it's on an old lathe with a milling attachment).

The end will always have approaching 'zero' thickness if climb milling (not sure where the arrow heads went – clockwise cutter moving right to left) …

Neil

[Anonymous]

Can't say I've ever heard the reason for climb milling to be at the chip exit, but I'm not going to gainsay the Sandvik paper. Another reason given for not using conventional milling is that at the start of the cut, where the depth of cut is very thin, the cutter doesn't actually cut. What happens is that the cutter deflects slightly and rubs the cutting edge on the work until the forces build up and the cutting edge eventually penetrates the work. The rubbing does the cutting edge no good at all, and for those materials that work harden creates a very hard thin layer, which makes it more difficult on the next pass and so on. That information also came from a Sandvik paper.

On the CNC mill I normally used mixed milling for roughing and make the final profile pass climb milling. On the vertical manual mill I use both methods interchangeably. As Murray says once your over about 60% radial cut depth it makes no difference. For small radial depths of cut I normally use a bit less depth for climb milling. Here's an example:

The cutter is 10mm three flute carbide cutting full depth (20mm) per pass. Conventional milling was with a 1mm doc and climb milling was with a 0.5mm doc. No way I'm winding that lot back just to avoid climb milling. The Bridgeport has oodles of backlash, about 15 thou, so if needed I do add a little drag with the table lock. However, I have also found when the cutter does grab It has never broken, at least not so far. Which just goes to show that we're probably pussyfooting around with small feedrates.

I do not climb mill on the horizontal mill, despite the fact that it has a proper backlash adjuster, although it doesn't seem to work. Given that the table probably weighs 400+lbs I'm disinclined to take it off to see why.

One material where climb milling most definitely doesn't result in a better finish is plastic.

I always use keys when horizontal milling, irrespective of the type of cutter. I cut under power feed so in the event of a jam I'm not going to able to knock the power off quickly anyway.

Nobody seems to have mentioned a rather more fundamental problem with set up shown. The slitting saw is fine tooth, but is cutting a thick piece of material. That is just asking for the gullets to fill, jam the cutter and bang, a shattered cutter. Much better to use a coarse tooth slitting saw that is designed to take deep depths of cut.

For the record I am not offering to write an article; much too controversial.

Andrew

[AjohnwParticipant @ajohnw51620]

The rational I read advocating it was largely down to efficiency – less HP for bigger cuts / more rapid metal removal. I may be able to find it. Not sure, some books had to go some years ago.

One interesting aspect that I have mentioned before is that it's possible to make a lathe tool self feed. Normally it would be ground to stop short of that for obvious reasons.

I've had loads of papers thrown at me so always start of on the basis that they may be misleading and may contain no useful information at all. They often are like that – both aspects.

John

–

[Neil LickfoldParticipant @neillickfold44316]

With the new rodless pneumatic cylinders that are available today,they should be good enough to have on the axis to eliminate backlash , by preloading an axis with X amount of constant force. It is something that I intend to add to a machine when I get myself motivated to look into it further. Then there will not be any backlash apart from forces above the preloaded pressure. A friend has already done this to the X axis on his lathe. Works real well.

Neil

[KWILParticipant @kwil]

Andrew,

Not a short piece of Al bar either, I suppose you could always open the garage door!

0.015" backlash, you must have very knackered leadscrew nuts, with all the adjustment taken out!!!!!!

[martin perman 1Participant @martinperman1]

Gentlemen,

I've enjoyed reading this thread as it contradicts what I was taught as an apprentice, we were taught that we should "never" climb mill as it was considered an unsafe way to mill and we could be in trouble if we were caught doing it, so forty years on I dont do it.

Martin P

[Involute CurveParticipant @involutecurve]

I'm not sure if Neil can put the Sandvik document up for people to download, however if you PM me Ill email it to you, like I said to Neil its really aimed at CNC users but is very relevant to manual mills, once you have read it you will understand much better how a cutter is influenced by step over etc, its sometimes counter-intuitive, although retired I occasionally make prototype injection mould tools at home, I also do some machining on site for a large plastics company, this entails machining a lot of P20, this article was given to me by a tool rep and it helped no end at increasing tool life, and by a surprising amount, the main point of tool damage is as the cutter exits the job, with conventional milling as the tool exits the job the chip is at its thickest and is unsupported, this allows the chip to bend causing a compressive load and so breaks the edge, with climb milling the opposite happens.

Shaun  

Edited By Involute Curve on 08/10/2015 10:56:49

[Anonymous]

Posted by KWIL on 08/10/2015 09:41:44:

Not a short piece of Al bar either, I suppose you could always open the garage door!

0.015" backlash, you must have very knackered leadscrew nuts, with all the adjustment taken out!!!!!!

Actually 4 stacked lengths of nominally 3/16" thick hot rolled steel. Careful examination will show a 'step' near the lefthand table clamp, as the table travel isn't sufficient to machine the total length in one pass. The carbide cutter was retired after use, but speeds and feeds were such that it was running dull red.

Having just been using the Bridgeport to make some copper busbars I 'measured' the backlash, rather than try and remember. Backlash was estimated by turning the handle one way until there was movement, zero the DRO and then move the handle the other way until the DRO changed. Results were X axis 0.7mm and Y axis 0.9mm. They'd probably come down a bit with some TLC, but basically the mill is pretty worn. Can't say it worries me, and I'm happy with the work that it produces.

Andrew

Edited By Andrew Johnston on 08/10/2015 11:19:14

[MuzzerParticipant @muzzer]

Posted by John W1 on 07/10/2015 23:08:46:

One interesting aspect that I have mentioned before is that it's possible to make a lathe tool self feed. Normally it would be ground to stop short of that for obvious reasons.

Yes – and drill bits too. If you try machining brass with a tool / drill that is ground for aluminium or steel (lots of top rake when ideally it would have none) you need to don your rubber pants. Once it grabs hold, it pulls itself in and you have one of those moments we all know and love. It's even more exciting if you have lots of backlash so it's almost uncontrolled.

[Roderick JenkinsParticipant @roderickjenkins93242]

I use my slitting saws in an arbour with a half inch parallel shank that I mount in an ER collet on the mill. Consequently I tend to put the saw in the arbour first, usually holding it up side down in the vice. It's pot luck whether I get the saw in the correct orientation but since I put a VFD drive on the mill I can easily reverse the direction

By chance, I think I got it right in this photo. As an aside, it clearly demonstrates the result of releasing the locked in forces in BMS.

Cheers,

Rod

[AjohnwParticipant @ajohnw51620]

I was taught to avoid rather than never. Never makes some sense really especially on horizontal millers that can remove considerable amounts of metal in one go. Try climb milling with a setting like that. The miller should cope but the finish on the work may suffer and if not clamped down firmly things will move. Similar when vertical cutters are taking their max cut and feed rate. The need usually only crops up with end mill type cutters are used anyway.

John

–

[colin hawesParticipant @colinhawes85982]

There can be advantages using climb milling on a conventional horizontal mill . One is to obtain a higher finish on a light finishing cut and another is to saw or side face thin sheet material, laid flat on the table, that otherwise would "bounce" with the cutting action unless extensively clamped down. The table lock does need to be applied enough to prevent "grab". Colin

[Nigel McBurney 1Participant @nigelmcburney1]

long ago I was taught not climb mill unless the machine was designed to do it,no cnc in those days,and lightly applying the table lock was frowned by management, The books on milling by Cincinnati and Brown & sharp barely mentioned climb milling, the pages of pictures of horizontal millers with gangs of big cutters are using conventional milling. My Adcock and Shipley has a lever mounted just below the table,to engage climb milling but It does not work the levers loose,somethings broken before I had it and at 74 I don't feel like taking the table off,I don't need it but it would be interesting as I have never tried climb milling on a horizontal machine. The change over to higher speed cnc machines which tend to go very fast with smaller cutters and driven by ball screws to eliminate backlash.I do not know how long a ball screw would last if it was fitted to a Big horizontal with a gang of side and face cutters plus a roller mill,. It is surprising how a relatively small end mill say 3/4 dia can grab a 50 inch table ,so I avoid climb milling on either of my machines, though the finish on aluminium can be improved by climb milling,using the side of an end mill and taking off a few thou. I was also taught to mount thin slitting saws without a key,management considered it better to loose grip rather than shatter a saw particularly if it was a large 6 in dia one, the victoria horizontal mill at our works had a 1/16 th slot across the long horizontal arbour keyway where some one forgot to remove the key, the saw jammed and cut the neat slot,it was a good reminder to remove the key.

[Mark CParticipant @markc]

Nigel,

"the victoria horizontal mill at our works had a 1/16 th slot across the long horizontal arbour keyway"

That was the mode of failure I would have anticipated (as per my comments earlier). The relative strength of the two materials and the lever arms involved would tend towards this mode of failure rather than a "shattered" cutter which would require a buckling failure of the cutter which might happen on a very thin blade but is less likely on anything else (I suppose if the blade had a keyway cut in it there may be a stress point at the key corners that could promote cracking but you would need to be unlucky and need a cutter with a sharp cornered slot – in itself bad practice).

Mark

[MuzzerParticipant @muzzer]

The B&S Treatise on Milling says

"Direction to Move Work Under Cutter. Whenever possible, it is advantageous to feed the work in the opposite direction from that in which the cutter runs. Then the cutter cannot draw the work in as it is liable to do when the table moves in the direction indicated at B. Moreover, when the piece moves as shown at A, the cutter teeth are first brought into contact with the softer metal, and as the scale on the surface is reached, it is pried or broken off.

On the other hand, in milling deep slots, or in cutting off stock with a thin cutter, or saw, it is sometimes better to move the work with the cutter, as the cutter is then less likely to crowd sidewise and make a crooked slot.

When the work is moving with the cutter, the table gib screws must be set up rather hard, for the teeth of the cutter tend to draw the work in, and if there is any lost motion in the table, the teeth may catch and injure the cutter or work. A counter- weight to hold back the table is excellent in such milling."

You need to be able to use both approaches, understand the benefits of each and know when they are appropriate. Many of the posts here start with the words "I was taught that…", without showing any obvious understanding of why. As an engineer that doesn't work for me.

Much of my milling has been done in climb mode, even on my old manual (vertical) Bridgeport clone. It's fine if you make sure the slides are just nipped up enough (you soon get the knack) and if you get the depth of radial cut about right, there's very little force required to feed the work into the table. Most modern machinery seems to default to climb milling – there must be compelling reasons!

[Tony Pratt 1Participant @tonypratt1]

Posted by Muzzer on 09/10/2015 18:22:26:

Most modern machinery seems to default to climb milling – there must be compelling reasons!

Not really sure what you mean by the above, do you mean CNC machines with ball screws which have little or no backlash?

Tony

[KWILParticipant @kwil]

I found the following on one CNC help site, looks rubbish to me, just tighten the gibs, that just makes it stiff to move.

"The best way to correct (at least with box way construction) is to adjust the machine's gibs to eliminate the backlash all together."

All ball screws have some backlash, double ones can be adjusted,  double lead screw nuts reduce it, wear must ultimately happen with both systems.

Edited By KWIL on 09/10/2015 19:17:21

[Neil LickfoldParticipant @neillickfold44316]

The backlash in modern machine tools comes from the compression and tension of the materials it is made from. The higher the speed, the greater the forces and so effectively more backlash occurs. Modern ballscrew nuts are preloaded together, there is no longitudinal movement in the nut assembly. The only way it can get any backlash is for the screw to stretch apart and effectively change pitch. That is why the good gear has some form of dimensional feed back, either off glass scales or laser positioning. The latest in climb milling for cnc uses only about 10% of the cutters diameter as a cut per side, so a lot less forces compared to a heavier cut, but they do it at a very high feed rate and a very high cutter surface speed. Really neat to see.

On manual mills, a finish climb cut should not be that bad on most machines, the only real issue with backlash is if you are climb milling in a pocket, and then that could be a problem. Having one of those air mist blowing over the cutter will greatly improve cutter life and surface finish, either manual or cnc.

Neil

Edited By Neil Lickfold on 09/10/2015 19:49:46

[Neil WyattModerator @neilwyatt]

I've created a short article with the Sandvik Coromant article 'Why Climb Mill' here:

http://www.model-engineer.co.uk/news/article/why-climb-mill/22411

Thanks to Shaun for drawing my attention to this and Sandvik for permission to use it.

Neil

[duncan webster 1Participant @duncanwebster1]

Neil,

any more info on air mist system? If it extends cutter life it would be of great interest to me at least

[Neil LickfoldParticipant @neillickfold44316]

People out here are using an air mist , either through the spindle of the machine,or add on the side misters.

They sell special oils, but rice bran oil or sunflower oils works very well indeed. It is quite thin, alot thinner than canola oil.The amount of oil used is very small, basically you hold a piece of paper in the air stream, and after about 1 min there should be a small circular pattern showing that oil is coming out. If it is wet with oil then that is usually too much. Seems too good to be true but works very well. It is best to have 2 or 3 nozzels blowing down onto the cutter to cover for when it goes around a corner or is in a cavity. The air blast wants to be strong enough to blow the chips clear out of a cavity etc. I think the oil consumption is about 1 liter per month maybe less on the machine center.

Neil

[David Clark 13Participant @davidclark13]

Many years ago a company ( I think) did an electrostatic coolant system that put a tiny bit of oil on a cutter. I think the original was for a lathe tool but it should work on a mill. I think it may have been the company that bought out a round bed lathe in the 1970's or 80's. It was described in Model Engineer.

[EmgeeParticipant @emgee]

Duncan, check in my photo album for the dimensioned drawing and other shots of the misting unit I use on a milling machine.

Emgee

[Douglas JohnstonParticipant @douglasjohnston98463]

In a small poorly ventilated workshop there might be a health issue with mist coolant.

Doug

[Author]

Posts