Is it possible to make this feature on a Haas Y-axis lathe?

robert123 · Feb 12, 2018

This part has a peg with 45 degree angles sticking out of the side of a radius. I can't quite visualize this to tell if I can do this with a radial live tool on a Haas Y-axis lathe. The possible issue is the 45's on the peg. The Haas Y-axis lathes do not allow for rotation of the live tool. They will rotate the spindle and move linearly in Z, X, and Y.

Matt@RFR · Feb 12, 2018

Not a lathe guy, but seems to me if you can make the square then you can make the square with chamfered edges as shown.

The real problem is you will have to 3D machine the OD where you couldn't get to with turning tools, and with no apparent radius allowed at the bottom of your square peg, that's not going to work very well.

robert123 · Feb 12, 2018

They do allow a .010" max radius at the base of the peg.

Bobw · Feb 12, 2018

robert123 said:
They do allow a .010" max radius at the base of the peg.

Oh Goody... Regardless of what machine you have, that's going to be a bitch...

You *should* be able to do it.. I don't think its going to be easy, I don't think its going
to look all that great..

What kind of tolerances are we looking at here? The boss diameter mainly.. If I was you, I
would dig into any specs that are referenced on the drawing and read up on allowable mismatch..
It is not going to look as pretty as that solid model..

This is one of those parts that makes me think that engineers shouldn't be driving the
train, they should be thrown under it.

robert123 · Feb 12, 2018

Bobw said:
Oh Goody... Regardless of what machine you have, that's going to be a bitch...

You *should* be able to do it.. I don't think its going to be easy, I don't think its going
to look all that great..

What kind of tolerances are we looking at here? The boss diameter mainly.. If I was you, I
would dig into any specs that are referenced on the drawing and read up on allowable mismatch..
It is not going to look as pretty as that solid model..

This is one of those parts that makes me think that engineers shouldn't be driving the
train, they should be thrown under it.

The boss the peg is sticking out of is +/- .005", so I'm not too worried about mismatch between the turned and end milled parts of the boss. We can always leave stock and finish the whole boss with the end mill if needed.

Ox · Feb 12, 2018

robert123 said:
This part has a peg with 45 degree angles sticking out of the side of a radius. I can't quite visualize this to tell if I can do this with a radial live tool on a Haas Y-axis lathe. The possible issue is the 45's on the peg. The Haas Y-axis lathes do not allow for rotation of the live tool. They will rotate the spindle and move linearly in Z, X, and Y.

View attachment 220415

I would certainly think that it could be made on your machine, but it won't be as pretty as it shows on CAD.

For best results, you would certainly want to turn everything away that you could (10:30 and 4:30 O'Clocks)

Then you could doo a fairly nice job @ C0 and milling 1:30 and 7:30.

For the chamfers tho, for best results - clock your C up and down to get the middle of the chamfer on the apex of the rad. This way you only leave a slight flat near the part. edit - on second thought - that will produce the best floor curf, but will produce a tapered chamfer. I guess someone would hafta decide if they want a nasty floor surface, or a tapered chamfer....

So at this point you should have a decent looking nip like shown, but now you need to try to git rid of all the rest of that material the other 340* around the part and make it look good. I would recommend that you run it around in C with your Y above or below C/L 1/2 the tool diameter - at least for roughing it down. For best results you may want a "true flat bottom" mill to finish the 340* and finish at Y0.

--------------------

Think Snow Eh!
Ox

Tonytn36 · Feb 12, 2018

I would add in time to the quote for hand filing that chamfer from the apex of the boss radius to the intersection of where the peg corners meet the boss. Tedious, but probably quicker and less time than trying to find a machining solution for that small area (x4).

litlerob1 · Feb 12, 2018

robert123 said:
This part has a peg with 45 degree angles sticking out of the side of a radius. I can't quite visualize this to tell if I can do this with a radial live tool on a Haas Y-axis lathe. The possible issue is the 45's on the peg. The Haas Y-axis lathes do not allow for rotation of the live tool. They will rotate the spindle and move linearly in Z, X, and Y.

View attachment 220415

Does the top of the Boss need a Radius? I'm guessing "yes", otherwise it would be fairly simple to Chamfer, the blend into the Turned diameter is going to look ass-y.

Different question; so the Haas lathe with a Y axis doesn't have C axis? Or am I misunderstanding?

SeymourDumore · Feb 12, 2018

Bobw said:
This is one of those parts that makes me think that engineers shouldn't be driving the
train, they should be thrown under it.

That ^^^ Right there!

Or, to put it in another way: The fact it CAN be designed, does not necessarily mean that it SHOULD!

The Diameter to Square ratio on that f@cking thing looks like it can accommodate a nice round recess at the base below without weakening the pin too much
so one could mill all around it and then still blend the dia. axially.

Else: 2 piece construction.

implmex · Feb 12, 2018

Hi Robert:
No you cannot make that geometry on a lathe or on a milling machine for that matter.
You can only make an approximation of it and the problem area is the junction of the octagonal boss with the cylinder.

The only way to make this geometrically correct is to sinker EDM it with an electrode that has the proper shape.
The obtuse angles between the boss and the cylinder could be faked using a Mill-Turn machine like an Integrex, but not for cheap.
The cheapest way to fake it is to accept a semicircular ditch at the foot of the boss and walk around it with a ball cutter.
Any decent Y axis lathe should be able to do that.
The depth of the ditch below the cylinder surface would obviously depend on the diameter of the cutter that was used and how much material can be tolerated at the root of the boss.
To those who commented that it was a dumb design...yeah I agree.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

litlerob1 · Feb 13, 2018

implmex said:
Hi Robert:
No you cannot make that geometry on a lathe or on a milling machine for that matter.
You can only make an approximation of it and the problem area is the junction of the octagonal boss with the cylinder.

The only way to make this geometrically correct is to sinker EDM it with an electrode that has the proper shape.
The obtuse angles between the boss and the cylinder could be faked using a Mill-Turn machine like an Integrex, but not for cheap.
The cheapest way to fake it is to accept a semicircular ditch at the foot of the boss and walk around it with a ball cutter.
Any decent Y axis lathe should be able to do that.
The depth of the ditch below the cylinder surface would obviously depend on the diameter of the cutter that was used and how much material can be tolerated at the root of the boss.
To those who commented that it was a dumb design...yeah I agree.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

I will always defer to you Marcus. I was not referring to the 45º angle around the Post's corners as the "Chamfer", I misunderstood because previous posts were. I thought we were talking about an edge break on the top, sorry.

BUT I don't understand why you couldn't accomplish the Cylindrical and linear (Y and Z) interpolation simultaneously, with a small Ballmill to minimize the gouging into the Cylinder, and several passes, it would not be pretty as a CAD file obviously. That is why I asked the question about Y with no C???

R

always appreciate a correction from you.

gregormarwick · Feb 13, 2018

Definately possible, but would require some awkward toolpaths. X+Y+Z+C simultaneous would allow you to cut a pencil path around the base of the stem to generate a clean floor and rad, the parallel sides of the stem are simple linear or rotary paths. The chamfers would then need to be surfaced with the corner of whatever tool you cut around the base with, using X+Y+Z using C to rotate the stem off-axis to the tool.

Need to make sure your machine can do 4 axis simultaneous.

thesidetalker · Feb 13, 2018

implmex said:
Hi Robert:
No you cannot make that geometry on a lathe or on a milling machine for that matter.
You can only make an approximation of it and the problem area is the junction of the octagonal boss with the cylinder.

I suppose this really depends on what your definition of approximation is...

On a Y-axis lathe, I agree.
On a 5-axis mill, you can tilt away from the walls and reach down to the intersection with a square end mill. Some surfacing should get you close enough, but that depends on the customer, tolerances, etc. Should be good enough for government work, or the 'engineer' who designed this.

gregormarwick said:
Need to make sure your machine can do 4 axis simultaneous.

Yup, a Haas can do 4.

Ox · Feb 13, 2018

Yeah - I didn't understand what he meant in the OP about not having a rotate feature either.
Shirley any Y axis machine has C, so I was wondering if he meant it as a grid shift feature as on a mill, but I don't know how that would apply here.

Sure, if you use a ball mill, and a long plotted CAM generated code, it could be surfaced - if that lathe could handle that amount of code? Most lathes would choke on that much code. ???

A 5X (B axis) lathe, or mill for that matter, should be able to produce as shown with square corners with std mill.

I am one for wiring it out through, and pressing a finished boss down in there.

--------------------------

Think Snow Eh!
Ox

gregormarwick · Feb 13, 2018

Ox said:
Yeah - I didn't understand what he meant in the OP about not having a rotate feature either.
Shirley any Y axis machine has C, so I was wondering if he meant it as a grid shift feature as on a mill, but I don't know how that would apply here.

Sure, if you use a ball mill, and a long plotted CAM generated code, it could be surfaced - if that lathe could handle that amount of code? Most lathes would choke on that much code. ???

A 5X (B axis) lathe, or mill for that matter, should be able to produce as shown with square corners with std mill.

I am one for wiring it out through, and pressing a finished boss down in there.

--------------------------

Think Snow Eh!
Ox

I think he was talking about a B axis when he said he couldn't rotate the cutter.

Should have no problem running long code as a sub off a card/usb, or even drip feeding.

litlerob1 · Feb 13, 2018

gregormarwick said:
I think he was talking about a B axis when he said he couldn't rotate the cutter.

Should have no problem running long code as a sub off a card/usb, or even drip feeding.

This makes more sense. I'm sitting there trying to figure out why the hell someone would even do Y without C, but hey I've seen some funny stuff. I'm still not sure Haas can do 4 true simultaneously.

Really though, Marcus is the guy who does the shit that everyone else says "you can't do that". But he very clearly said "You can't do that", so I am curious to see what he has to say.

R

implmex · Feb 13, 2018

Hi All:
OK here are some pictures.
I think we can all agree on the first 4 steps; easy peasy, all is do-able even though it's a fair number of toolpaths:
This is where we get stuck; nibbling out the little corners marked in red.

implmex · Feb 13, 2018

more pictures

We can't get those corners completely no matter what orientation of the cutter we choose; there will always be a geometry defect somewhere in the corner.
You may leave stock somewhere or you may gouge the part.
If you decide to try to surface the parts you can't reach with simple contouring you can get much closer, but it'll take a lot of time, a lot of code and tiny nibbly cutters.
Here are 4 possible cutter orientations so you can see what I mean:

implmex · Feb 13, 2018

a common way to fake this

Here's the way I often recommend to designers when they send me stuff like this

Ox · Feb 13, 2018

Even that last one will require a boat load of code to produce the compound angle/rad on the chamfers.

---------------

Think Snow Eh!
Ox

Is it possible to make this feature on a Haas Y-axis lathe?

Stainless

Titanium

Stainless

Diamond

Stainless

Diamond

Diamond

Diamond

Diamond

Diamond

Diamond

Diamond

Stainless

Diamond

Diamond

Diamond

Diamond

Attachments

Diamond

Attachments

Diamond

Attachments

Diamond

Similar threads