CMM vs on machine probing

Hi,

Admittedly new to machining and metrology so apologies if I've missed something obvious.

With the availability of Renishaw probes on CNC machines, what benefits does a separate CMM offer? Or to put it the other way, what useful metrology features are missing from CNC based probing?

Are there particular tolerance ranges or lack of GD&T that might make CNC probing an affordable/cheaper alternative to a separate CMM? At what level of inspection or accuracy should a CNC shop add a CMM machine?

I realize that probing on the same machine and setup that produced the part would be subject to issues with the machine itself. I'm wondering when putting a CMM in the shop "makes sense".

Expecting a broad range of responses to an overly broad question of course.

Thanks in advance.

The machine can't make you money if it's stuck running probing cycles on a part it just made.

Also unless things have changed things like flatness, parallelism, perpendicularity and profile aren't usually included in the probing cycles on the machine. Macros can be written and verified but sometimes it's just easier to put it on a dedicated machine.

What sort of features are you trying to check?

As Hazzert notes, the CNC machine can't make parts and do inspection at the same time. Any built-in part probing capability is typically very basic, and if you have inspection of any sort of form profiles or relationships between features, it seems to me that a CNC milling machine is not any reasonable substitute for a CMM or other off-machine inspection. And as you note yourself, the CNC machine will bring its own errors in terms of orthogonality, leadscrew pitch, travel straightness, etc., into the results.

In broad terms, CNC part probing is meant to give you an indication of part size variation, and part presence, rather than as an inspection for specification compliance on a finished part. Putting a CMM in your shop "makes sense" when you can afford it, or when you can't afford NOT to have it. A CMM will tell you many things about parts and features (and your machines) that you can't find out easily in other ways; in some cases, things you may not want to know.

Hazzert said:

The machine can't make you money if it's stuck running probing cycles on a part it just made.

Click to expand...

Ah. So I think you're saying that the probing cycles are typically much longer in duration than the actual machining?

Hazzert said:

Also unless things have changed things like flatness, parallelism, perpendicularity and profile aren't usually included in the probing cycles on the machine. Macros can be written and verified but sometimes it's just easier to put it on a dedicated machine. What sort of features are you trying to check?

Click to expand...

I don't have any specific features in mind. I've read thru the "shop owner" forums on here quite a bit, and didn't see many using CMM's.

Outside of the lost spindle time doing measuring CMMs and machine tools are designed and built differently.
Even if they had a spindle a CMM would make a terrible metal cutting machine, metal cutting machines make terrible measuring machines.
The real use of a probe in a machine tool is to control a size not measure it.
Bob

specfab said:

As Hazzert notes, the CNC machine can't make parts and do inspection at the same time. Any built-in part probing capability is typically very basic, and if you have inspection of any sort of form profiles or relationships between features, it seems to me that a CNC milling machine is not any reasonable substitute for a CMM or other off-machine inspection. And as you note yourself, the CNC machine will bring its own errors in terms of orthogonality, leadscrew pitch, travel straightness, etc., into the results.

In broad terms, CNC part probing is meant to give you an indication of part size variation, and part presence, rather than as an inspection for specification compliance on a finished part. Putting a CMM in your shop "makes sense" when you can afford it, or when you can't afford NOT to have it. A CMM will tell you many things about parts and features (and your machines) that you can't find out easily in other ways; in some cases, things you may not want to know.

Click to expand...

You mentioned "when you can afford it, or when you can't afford NOT to have it".

Do you have some examples or situations of when probing on a CNC would be good enough?

Making motorcycle parts +/-.10"? +/-.001"?

I see many shop threads where there is a large surface plate, height gauges, pin gauge sets, etc, etc.

At what point does setting all that up by hand and checking justify using a CMM instead?

Sorry for the bit of an unusual question, I've thought of starting a small CNC shop and wonder how the metrology end of things should work.

Some of the shop owner threads also mention large rejections of parts which makes me wonder what could/should have been done different from a QC perspective.

Often the shop owner threads discuss fixturing, tool paths, etc etc. and how those might calculate into profitability/margin of a job.

But I haven't seen any discussions regarding actual QC time for a job.

I understand using a CMM in tight tolerances in Aerospace is a must. But there have to be some varying degrees of inspection (another example, a ball bearing pocket for a spinner toy)...and at some point between spinner toy and Aerospace a CMM makes sense.

Thanks in advance.

CarbideBob said:

Outside of the lost spindle time doing measuring CMMs and machine tools are designed and built differently.
Even if they had a spindle a CMM would make a terrible metal cutting machine, metal cutting machines make terrible measuring machines.
The real use of a probe in a machine tool is to control a size not measure it.
Bob

Click to expand...

Can you give an example of what you mean by "The real use of a probe in a machine tool is to control a size not measure it.".

Thanks for your patience regarding my 'new guy' questions.