Radial force on lathe spindle

[mackeym]

Does anyone know typical and approximate radial forces involved in lathe turning. More specifically, turning things on a hobby lathe with normal to conservative cutting rates on common materials (aluminum, 1018 steel, and maybe medium carbon steel). Example: what typical force is required do a 5 mil cut into a piece of 1018 steel with a feed rate of ~6IPM.

[Geof]

Your question is not absolutely clear: Do you mean the force directed outward along a radial direction when the tool is travelling along turning the length of a shaft with a 0.005" depth of cut? Or do you mean the force needed to advance the tool along the work?

In either case with a correctly sharpened tool the forces would be a few pounds; my guess thinking about machines I have used is that it would be less than 10lbs.

Actually on some material if the tool has too much top rake it can tend to get pulled into the workpiece.

Incidentally for turning it is more conventional to express feed in distance per rev. Inches per minute is not very helpful without an rpm figure.

[mackeym]

I was referring to the first of your suggestions: "The force directed outward along a radial direction when the tool is travelling along turning the length of a shaft with a 0.005" depth of cut?" I am, however, interested in both (radial and axial) so the answer to both is even better.

For a spindle speed; 500 RPM. and a work piece that's 1" in diameter.

I'm asked the question b/c i'm designing a lathe spindle and want to make sure the bearings i have have enough capacity. Yes, i know that this is not an easy project, and i've read up on spindle design and bearing fits.

Work pieces that I will machine will be relatively light (<50lbs). In addition to the work piece weight, I want to consider the "cutting forces" (radial, and axial) that will also be on the spindle bearings.

[Geof]

Originally Posted by mackeym ....Work pieces that I will machine will be relatively light (<50lbs). In addition to the work piece weight, I want to consider the "cutting forces" (radial, and axial) that will also be on the spindle bearings.

I think if you did the calculation for that 'imaginary' force...the centrifugal force you will find this is what gives you the largest radial forces when you have a slightly out of balance piece of material spinning at several hundred or a thousand or two rpm.

EDIT: I realised there is one operation that could give very significant radial forces and that is knurling. You could be into the many tens of pounds and possibly into the hundreds.

[mackeym]

Thanks. I wonder why tapered roller bearings are often used in lathes then. I guess the work pieces can get very heavy. There is also the occasionaly mistake that might damage an angular contact bearing but not a tapered roller. and knurling that you mentioned.
Can i ask how you know these typical cutting forces are ~10lbs (or less).

The bearings i have are Angular contact: 7019 phenolic resin cage, duplex universal matching, 15degree, light preload (120lbs), P4Y accuracy (P4 with much higher ID and OD tolerances) (ABEC7).

With such low cutting forces, these bearings should be fine. I have 2 sets; one for the chuck side and one for the pulley side. I know i could also use a cylindrical roller instead of an A/C set for the chuck side. I already have 2 sets of the A/Cs though. I have found a suitable cylindrical roller bearing (C1 clearance, P4 tolerance) which has the same cost as an A/C pair.

Considering work pieces <50 lbs and low cutting forces, two sets of A/C's should work great and have plenty of capacity for my purposes (from the load-life calcs i've done on these bearings).

[mackeym]

Thanks. I wonder why tapered roller bearings are often used in lathes then. I guess the work pieces can get very heavy. There is also the occasionaly mistake that might damage an angular contact bearing but not a tapered roller. and knurling that you mentioned.
Can i ask how you know these typical cutting forces are ~10lbs (or less).

The bearings i have are Angular contact: 7019 phenolic resin cage, duplex universal matching, 15degree, light preload (120lbs), P4Y accuracy (P4 with much higher ID and OD tolerances) (ABEC7).

With such low cutting forces, these bearings should be fine. I have 2 sets; one for the chuck side and one for the pulley side. I know i could also use a cylindrical roller instead of an A/C set for the chuck side. I already have 2 sets of the A/Cs though. I have found a suitable cylindrical roller bearing (C1 clearance, P4 tolerance) which has the same cost as an A/C pair.

Considering work pieces <50 lbs and low cutting forces, two sets of A/C's should work great and have plenty of capacity for my purposes (from the load-life calcs i've done on these bearings).

[Geof]

Originally Posted by mackeym .....Can i ask how you know these typical cutting forces are ~10lbs (or less).....

One way is a guesstimate from the torque that you need to apply to the cross slide handwheel when facing and the carriage handwheel when turning. Another is from a small turret lathe that used to be in my shop. This had a lever to feed the cross slide with about a 5 to 1 mechanical advantage and it was possible to feed the tool when facing or cutting grooves just using light finger pressure.

[mackeym]

ok thanks. I also found a paper that has some force measurements of deeper cuts if anyone is interested. They attached strain sensors to the bit to measure the force in the vertical direction and horizontal (cross-slide travel direction).

I. Conditions: Carbon steel (maybe 1018?)
Feed Rate: 0.21ft/min
Workpiece surface speed: 700 ft/min
Depth of cut: 0.026"

They measured 50lbs in the vertical direction and 20lbs in the horizontal.

II. Conditions: Carbon steel (maybe 1018?)
Feed Rate: 0.34ft/min
Workpiece surface speed: 314 ft/min
Depth of cut: 0.040"

They measured 52lbs in the vertical direction and 30lbs in the horizontal.


III. Conditions: Carbon steel (maybe 1018?)
Feed Rate: 0.68ft/min
Workpiece surface speed: 314 ft/min
Depth of cut: 0.030"

They measured 76lbs in the vertical direction and 40lbs in the horizontal.

[Geof]

Well my guesstimates were in the correct region given that your conditions were smaller depth of cut and slightly slower feed.

But I just realised there is another factor that could have a big influence; positive or negative rake tooling. Positive rake is when the top surface of the tool slopes down away from the tip; you can almost imagine the tool is peeling the chip off. Negative rake is when the top slopes up and the chip is forced through an abrupt direction change and a lot of distortion.

[Pres]

Originally Posted by mackeym ok thanks. I also found a paper that has some force measurements of deeper cuts if anyone is interested. They attached strain sensors to the bit to measure the force in the vertical direction and horizontal (cross-slide travel direction). I. Conditions: Carbon steel (maybe 1018?) Feed Rate: 0.21ft/min Workpiece surface speed: 700 ft/min Depth of cut: 0.026" They measured 50lbs in the vertical direction and 20lbs in the horizontal. II. Conditions: Carbon steel (maybe 1018?) Feed Rate: 0.34ft/min Workpiece surface speed: 314 ft/min Depth of cut: 0.040" They measured 52lbs in the vertical direction and 30lbs in the horizontal. III. Conditions: Carbon steel (maybe 1018?) Feed Rate: 0.68ft/min Workpiece surface speed: 314 ft/min Depth of cut: 0.030" They measured 76lbs in the vertical direction and 40lbs in the horizontal.

Wow! How big of a lathe was that test done on?
What was the size of the tool & cutting surface dimensions?
Any info on rake, clearance angles?

The last time I remember using anything close to that much pressure was a cutoff tool in a 36" LeBlond engine lathe hacking away stainless.

Just wondered.
Thanks,
Pres

[Frank GP]

Why not just look at the bearings that lathe manufacturers use on their machines and use the same ones? Go for a good name, why re-invent the wheel?

[mackeym]

The measurements were claimed to have been done on a standard lathe. I think i remember seeing a bit size of 5/8" but i'm not sure. I tried uploading a zipped file of the paper (it's more of a technical report..95pgs), but for some reason it wouldn't upload. I can email it to anyone that wants it though.