[westerfieldcct]

I have always been a center person thould I would get other peoples input.

[o1dakota440]

i'm just getting into the cnc hobby and am new to all of it, but so far i use the center of the tool.

[307startup]

I have always been a center person thould I would get other peoples input.

What the hell are you talking about?

Cutter compensation by definition is an offset either to the LEFT or the RIGHT of the center line, to compensate for the difference between the measured OD of a cutting tool and the dimensions of a cut surface, whether it's due to incorrect coding of a tool path, mismeasured tool, excessive runout, or a cutting edge wearing out or being replaced.

How exactly would you compensate ON CENTER?

A tool isn't going to get any more centered by adjusting cutter compensation , nor is a line on a tool path going to become more centered than it was programmed...if your tool is off-center, you have a bad tool-holder and it should be replaced, if a cut on your workpiece isn't where it should be in relation to a dimensioned drawing, your program should be checked for errors.

[Torsten]

Been a bit confused by the terminology.
1) Do you mean to the Center of the Tool by using the Toolradius as compensation Value?
2) And to the Edge of the Tool by compensating for the difference of the actual Toolradius from nominal Toolradius only?

Tough choice
For the short Programs I manually enter into the controller I often use the 1st option it is far easier to drive the drawing dimension directly without having to do the math and i can decide on the tool size later.
It also helps with macro cycles where the actual size of the tool is known to the program.

Cam generated Programs are a different story to avoid problems with verification and the different ways compensation is invoked by varying Controllers I usually use the 2nd methode.

[307startup]

i'm just getting into the cnc hobby and am new to all of it, but so far i use the center of the tool.

I will excuse you for being a newbie, but nobody cuts with the center of the tool, everybody cuts with the periphery of the tool.

A tool rotates about it's axis...at the center, there is no rotation. Even with a center-cutting endmill, this is a misnomer, as the center isn't cutting anything, there is a gap between the webs of the tool, like on a split-point drillbit.

[glovebox20]

When I use the "Dumb Thumb" programing method, I program to the edge of the tool. That way you can program to the print demotions and be done with it.

But when using a Cam system I use the "Center" like Torsten dose to avoid cutter comp issues.

I only program mills, so lathes maybe another story.

[westerfieldcct]

What the hell are you talking about?

Cutter compensation by definition is an offset either to the LEFT or the RIGHT of the center line, to compensate for the difference between the measured OD of a cutting tool and the dimensions of a cut surface, whether it's due to incorrect coding of a tool path, mismeasured tool, excessive runout, or a cutting edge wearing out or being replaced.

How exactly would you compensate ON CENTER?

A tool isn't going to get any more centered by adjusting cutter compensation , nor is a line on a tool path going to become more centered than it was programmed...if your tool is off-center, you have a bad tool-holder and it should be replaced, if a cut on your workpiece isn't where it should be in relation to a dimensioned drawing, your program should be checked for errors.

All you need to do is ask for more information. I am amazed that others can read and understand but you don't. The questions could be stated by saying do you program to the print without using the radius of the tool, or do you program computing in the radius of the tool you wish to use. Others have answered with out issue, why can't you?

[307startup]

All you need to do is ask for more information. I am amazed that others can read and understand but you don't. The questions could be stated by saying do you program to the print without using the radius of the tool, or do you program computing in the radius of the tool you wish to use. Others have answered with out issue, why can't you?

Because this: I have always been a center person thould I would get other peoples input. doesn't mean anything in English or any other language.

If someone else wants to communicate a point, it is their responsibility to write intelligibly and effectively, it isn't my responsibility to guess at what they are trying to say and interpret gibberish. Especially when they are conducting a poll. Otherwise what bearing would the results have on anything?

That's the problem with the world today, no one has or maintains any standards, it's just a free-for-all, do-whatever-I-wanna-do kinda world, where criticism is bad and justifications & excuses are ok.

Being a "center person" doesn't have jack-squat to do with actually USING cutter compensation, now does it? Your program is either correct, your tool was indicated properly and your workpiece hit it's target dimensions, or it didn't. And that is why cutter compensation would be used...to alleviate one of these conditions.

Again...you can't use cutter compensation ON center, because it's an offset FROM center. That whole communication thing, where words have definitive meanings and thus a common understanding can be achieved...

[Geof]

........That's the problem with the world today, no one has or maintains any standards, it's just a free-for-all, do-whatever-I-wanna-do kinda world, where criticism is bad and justifications & excuses are ok..........

Down boy down. You are encroaching on the territory of crotchety old farts like me. But I guess it is good you encroach because this old fart gave up trying to get young twits to use correct spelling and grammar years back.

[fizzissist]

Down boy down. You are encroaching on the territory of crotchety old farts like me. But I guess it is good you encroach because this old fart gave up trying to get young twits to use correct spelling and grammar years back.

....and old farts like me as well :cheers:

I was taken aback by the tone of the response to what was a simple and innocent question....and an answer that to me at least was inaccurate.

Cutter comp on a lathe for a tool with a radius tip takes into account the theoretical tip location. That takes into account the X AND the Z compensation needed to give accurate contours, correct?

Least that's the way I have to program my cutter comp on my lathe.....

....but wait, we're talking about milling comp, where the tool rotates?? A rather different nature to the compensation?? Ok, like the ability to compensate to the edge OR center of the tool for wear?

My Fadal Format 1 allows not just zero compensation, but negative compensation so you compensate for a tool that is undersize, right on the mean, or oversize. Regardless, you program the toolpath according to the part's dimensions directly, and invoke either left or right cutter comp with G41 or G42 so the control has to suffer the calculations of the actual path.

I always (with RARE exception) program the toolpath using dimensions right off the print, input the tool's diameter into the control, and use the appropriate comp. It makes reading the program feasible for the inevitable editing .....

[holbieone]

307startup , WHAT ?

[fizzissist]

Maybe to simplify.....

You really need to program to the part, NOT the tool....and use the part dimensions as your toolpath (all this presumes you have the ability on your control to use cutter comp??), so you're free to change the tool to a different diameter if need be.

For example.....a simple part that is 3"tall in Y by 5"long in X....with origin at lower left corner 0,0 and you want to go clockwise from the origin (we'll ignore approach here for simplicity)

You've called G41, left offset, in the approach move, then you have; (we'll program in absolute)

G1 X0 Y0 (tool starts at origin
G1 X0 Y3 (tool goes to upper left corner
G1 X5 Y3 (tool moves right to upper right corner
G1 X5 Y0 (tool moves down to lower right corner
G1 X0 Y0 (tool moves left back to origin

Now, if you don't use cutter comp, you'll have to change the values of the X and Y coordinates in each and every move to allow for the radius of the tool, and you'll have to change the entire program if you break your precious and only 1/2" dia endmill, the customer is screaming for his part, and the only tool you've got is a 7/16".....

To people who've been doing this for awhile, the poll could seem silly. But if you're like me and at one time had absolutely NO clue (arguably still the case), there's no stupid questions.

When I teach Aikido, I always remember the frustration of wondering where the hell are my feet supposed to go when I first started. The beginner before me is going through that very same experience, and I am bound to get him past that as quickly as I can. I don't need to beat him up, he'll do enough of that on his own.

Same thing with this skill. We don't need to be discouraging newbies while a skilled labor force is so crucial to our future.

[Geof]

........ a simple and innocent question.........

But the simple and innocent question is a bit nonsensical. Someone with programming experience can conclude, and I think this in supported in subsequent posts, that the question should really have asked something like does your CAM output program to the toolpath or the cutpath.

All programs define the path of the tool centerline, it cannot be by other way. When the program is written to the part coordinates this path does not create the desired shape and it is necessary to invoke tool compensation so the controller compensates for the radius of the cutter by internal calculations on-the-fly and creates the correct cutpath from the programmed path. This is probably always the case when hand programming, the tool size must be entered into the controller and the appropriate left (G41) or right (G42) compensation invoked and cancelled through the program. In addition, during running of the program, if tool wear occurs this can be accomodated in the compensation calculation by using a (tool) wear value entered into the controller; the wear value is added to the tool value to get the value used in the calculation.

Alternatively the CAM output can incorporate the radius compensation so the code in this case is the actual path needed to create the shape and the controller does not have to do any compensation. It is possible to run the program without invoking either G41 or G42 and create the desired shape (approximately). However, if the actual tool radius is a little different to the radius used in the CAM output the part will not have the correct dimensions. Because tool compensation (and wear) is not invoked the only way to adjust for a discrepancy in the actual tool size is to create another CAM output using the actual tool size.

Therefore, in one sense, it is probably best to produce a CAM output which uses the nominal tool size but also includes the G41 and G42 commands for tool compensation in the controller. This means that the tool size entered into the controller is zero, because the CAM path incorporates it into the code, but any values needed to correct for small differences between the nominal size and the actual size of the tool can be entered into the wear value in the controller.

[fizzissist]

But the simple and innocent question is a bit nonsensical. Someone with programming experience can conclude, and I think this in supported in subsequent posts, that the question should really have asked something like does your CAM output program to the toolpath or the cutpath.

All programs define the path of the tool centerline.......

Ok, I have to concede the point of nonsensical...I was trying to be nice...(WHAT!?! You say??)...It struck me that way at first too.

But I think I understand what he means....and he really did need to be more clear.

When (and if) I program in Notepad manually, I always use the method I described above, where the code describes the line of the part's nominal dimensions. ... I have a library of 'canned cycles' that are easy to cut-and-paste for simple programs, and sometimes I'll put them into a CAM-generated program because I like my own toolpath better.

I use Mastercam, and it gives you the option of cutter comp by Control, Computer, or None.

Posting with the Control option results in G-code that has the G41/G42 embedded and the code reads as though it describes the outline of the part...."to the centerline", if you will. Same result with None, but no G41/G42 output, which is useful for engraving.

On the other hand, posting using the Computer option gives code without the G41/G42 and the X,Y addresses are the actual coordinates the comp would yield. Next to impossible and tedious at best to edit manually... You have to be absolutely confident that your tool dia and cutting conditions will keep you in tolerance.

So, maybe that's what he meant, do you let the computer calculate the path and post the actual path like you'd get using the Computer option, or use the Control option that inserts G41/G42 into the code and the X,Y are "centerline".

No matter, I think we're probably all done here.

[slucas_clmach]

I generally program most of my cuts to the edge of the tool. But on occasion a center line program will be needed. The problem is usually one of clearance between part features or workholding when I use center line programming. The other problem is remembering when I am having this problem that the center line option is available.

[DSIF]

There are advantages to both methods. I prefer part-edge for ease of reading the code. For tight ID rads.,centerline is preferred to avoid comp. errors. It depends on the situation.

[dertsap]

I use Mastercam, and it gives you the option of cutter comp by Control, Computer, or None.


On the other hand, posting using the Computer option gives code without the G41/G42 .

you forgot wear

my preference is to use a wear compensation while using cam or not , this tends to leave less room for error between jobs . lets say a guy uses d comp from center line and uses a value for a 1/2" end mill and the next job he's using a 3/4" end mill , if the guy forgets to change that d comp value then there are going to be problems . if a wear value is used then usually the most that will be forgotten in the control will be a few thou .
there are remedies for it like using g10 , small program to clear everything in between jobs or not being forgetful , but stuff happens .
The company where I learned used cam output code based from center line , eventually after a number of destroyed parts, fixtures, tools etc (not mine) , they decided to post from the tool edge and if needed the cutter comp was changed within a few thou .
I'll use center line when I'm at the control and I've got some funky moves to make based on my drawings , this saves me loads of time and calculations . So in doing the type of work that I do , my programming can be all over the place in order to get that one off .

[christinandavid]

I put all cutter rads in table whether program uses them or not. Often find that I need to rewrite bits of offline programs and use cycles instead, for which the machine requires the size of the cutter.

I even go to the length of putting in the radius of a drill/tap into the tool table to make it easier to search through...

Now then - what is the preferred method for offsetting a chamfer tool? Everyone at our place has a different approach - comp/no comp, minor radius (my preference - again due to the mc cycles), major radius, shank radius plus-a-bit - even random figures in some cases....this has been a major issue recently.

DP

[dannystooblue]

All you need to do is ask for more information. I am amazed that others can read and understand but you don't. The questions could be stated by saying do you program to the print without using the radius of the tool, or do you program computing in the radius of the tool you wish to use. Others have answered with out issue, why can't you?

When I answered the poll I was in error as to what was meant, different geographic areas raise different schools of though regarding terminology. The question where I work would be asked " do you program online and comp over or do you program actual cutter path?"


Dan

[pimpbike]

If someone else wants to communicate a point, it is their responsibility to write intelligibly and effectively, it isn't my responsibility to guess at what they are trying to say and interpret gibberish. Especially when they are conducting a poll. Otherwise what bearing would the results have on anything?

That's the problem with the world today, no one has or maintains any standards, it's just a free-for-all, do-whatever-I-wanna-do kinda world, where criticism is bad and justifications & excuses are ok.

You're right, it isn't your responsibility. Do what I do most of the time...ignore the crap and move on, or if you really do care, ask politely what they meant and answer proactively.

Off topic, what you said did strike a cord for me though. My friend became a teacher this year here in Ontario Canada. He told me that you're not allowed to fail students anymore. What genius thought of that? Apparently we don't need our kids to try anymore. Sad direction that Canada is going. Apparently you can still fail in the US. I thought we had enough dough head Politicians. It's going to be a sad place to live when the kids of today are making the decisions for us when I'm trying to retire.....