What's new
What's new

Manually programming TNRC

ab1sket

Plastic
Joined
Sep 23, 2022
I've got a question for the lathe programmers out there.
I am used to machines that are programmed with the tool set tangent with the work. I am now discovering that some programming is done by setting the tool according to it's center line. meaning you would add the tnr for Z and double it for X.

So my question is do you still have to subtract or add the tool nose radius to the arc to get the correct arc geometry since you're still not using tool nose radius compensation? or does programming to center line save you from having to do that? And if you do still have to subtract (or add) the radius, can you do the same thing when using I & K instead of R?

I can't find any good info concerning this specific type of question. Thanks in advance for the help!
 
Why should we think so much about something which we do not need to!
We just need to store the tool radius and tip number in the geom/wear offset tables, and invoke G41/G42 as needed. No need to re-define the profile.
 
Why should we think so much about something which we do not need to!
We just need to store the tool radius and tip number in the geom/wear offset tables, and invoke G41/G42 as needed. No need to re-define the profile.
Exactly!
I use G41/G42 all the time. On both lathe and mill. On lathe it's just too easy, enter the tip radius, correct tip direction and call up when needed.
I get superior results on radii, chamfers, angles, tapers, etc.
 
Because I do have to. Aerospace industry does not rely on TNRC. The engineers program to center-line. Not my choice. So anyone have an answer to my question or just peanut gallery stuff?
 
I made up a drawing that might answer your questions. It shows the centerline of a 1/32 radius tool. note the X dimensions (or Y according the axis in the drawing) are radius measures not diameter.

1663946251146.png
 
Yeah a pretty good idea. If it deflects the question it's peanuts. How do you think I've gotten this deep into this profession without the answers to these questions? Noone will answer them.
 
Thank you Kenton. That actually helps alot! Can I ask what platform/software you drew that on?
 
Autocad. Most any cad software should work, or even a drafting table if such things still exist. The picture is a screen shot
 
I've used Solidworks but no longer had access to it when I ended school. This really helps me get a better idea of what's going on. I appreciate it.
 
FYI the first drawing assumed you would make a .0313 radius move around the 45deg corner.
This is for linear moves around the corner:

1663947652079.png
 
So I'm noticing the difference between the radial move and the linear moves is .0221. So this means you must add .0221 to the arc or the R code?
 
I take that back. I was looking at the wrong coordinate. This seems to show that there isn't a difference?
 
first drawing approaching the outside corner would be
G01 Z-3.9979 X4.0442 (2.0221*2)
G03 Z-4. X4.0626 R.0313 (2.0313*2)
G01 Z-......

Second would be
G01 Z-3.9871 X4.0626 (2.0313*2)
G01 Z-......

If that helps, and I can do my math right.
 
This definitely helps. But how would it compare to programming tangentially? That's what my main confusion is. Because I know that you have to compensate for the radius when programming that way without tool nose radius comp.
 
I am thinking that what you're showing is that to center-line you wouldn't need to compensate in any direction because everything is already set to center-line. It feels like I'm over complicating it and it's probably pretty simple.
 








 
Back
Top