Renishaw HAAS TM-3P ASSISTANCE

[not a machinist]

My HAAS TM-3P is in Germantown, MD. I am looking for an expert machinist to verify the diameter probing problem that I just can not figure out. The folks from HAAS/Phillips can not figure it out either. The problem is that the probing diameter routine is generating a tool diameter that is too high on all tools probed. The Renishaw Wireless OTS Probe has been set, reset and calibrated many times both by me and the HASS/Phillips personnel. Some of the values for the diameter are off by .0030" and all are on the oversize side. I am willing to offer up to $100.00 per hour for the right person.

 

[thesidetalker]

If instead you tried posting some of the details about what you're doing, you might receive a helpful response and possibly learn a thing or two. You wouldn't have to waste money for someone to show you how things deflect under cutting forces and how to use wear comps nor bother with your "class action suit" you talked about previously.

How can you blame the tool when you don't know how to use it?

 

[not a machinist]

Thank you for the reply. You have made a number of inaccurate assumptions:
1. I have learned a great deal from the helpful posts that I did receive.
2. I am aware of "deflection". My assumption was that, given the diameter variation of .003", using 6061-6 4"x6" solid block squared and mic'd to within .0003" on all sides, a 2" diameter cutter with new rectangular inserts from Maritool and taking a .002" cut on each of 4 sides; that the deflection would be in the tenths, if that.
3. Wear comps are not an issue. Using them would be a work around. The purpose of the probe is to reduce set up time and provide MORE accurate length and diameter measurements. Wear comps are not for taking up the error in the probed diameter. What use would the probe be if you had to enter offsetting values in the "wear" column every time you loaded a new tool. Further, the technicians at Renishaw agree with me that something is wrong and the service folks at HAAS/Phillips have duplicated the problem and do not know what is wrong.
4. You are correct that I am not educated regarding all of the features of the machine. However, I have decades of experience working with expert CNC operators that worked for me. We built racing motors among other automotive work. I have worked with various manual mills and lathes for decades. Also, the CNC operations required to do the testing required to determine that the probed diameter is off are easily done and require little expertise.
5. While true that I am not an expert machinist (nor will I ever be), I do know what I don't know and must get unbiased assistance in those areas. This problem is one of them.

Thank you again for your post.

 

[mhajicek]

Have you checked the cutters for runout? If you have a tool that mics nominal across the flats, but it has runout, the spinning diameter probe routine will return a larger value; it measures the most protruding flute. I can only pontificate where the error might be, since you haven't told us anything about how you're getting your numbers to compare to.

 

[not a machinist]

Have you checked the cutters for runout? If you have a tool that mics nominal across the flats, but it has runout, the spinning diameter probe routine will return a larger value; it measures the most protruding flute. I can only pontificate where the error might be, since you haven't told us anything about how you're getting your numbers to compare to.

Yes, I have checked the cutters for runout and they have been .0002" - .0005" for the tested cutters. My thinking was that if the cutter did have runout , it would reflect in the probing result but, would also reflect in the actual cut. Therefore, I thought there would be no impact on the actual cut?? My comparisons are between the probed diameter and the actual cut.

This problem has been present since the machine was new, almost 2 years. HAAS/Phillips has been out numerous times and duplicated the problem, but were not able to determine the cause. ALL probed tools return an oversized reading and it is not consistent.

 

[mhajicek]

Yes, I have checked the cutters for runout and they have been .0002" - .0005" for the tested cutters. My thinking was that if the cutter did have runout , it would reflect in the probing result but, would also reflect in the actual cut. Therefore, I thought there would be no impact on the actual cut?? My comparisons are between the probed diameter and the actual cut.

This problem has been present since the machine was new, almost 2 years. HAAS/Phillips has been out numerous times and duplicated the problem, but were not able to determine the cause. ALL probed tools return an oversized reading and it is not consistent.

Ok, when you say the results are oversize, do you mean it leaves the part oversized, or it measures the cutter oversized and leaves the part undersized?

 

[AJ H]

I have never had one of these be particularly accurate on setting tool diameter, so I don't use it to do so. It's still worth every last cent of the $6000 price tag for the rest of the things it can do. Very often in this industry things are advertised with lofty capabilities.

 

[PROBE]

In general, I wouldn't relay on tool diameter measured on tool setter. Proper way is to take a cut with the end mill and to measure the width of the slot with spindle probe.
Nevertheless, the diameter measurement should give some measure of continuity.
In order to revue the correct operation of the system I suggest following procedure:
1. Prepare for yourself the calibration tool, loading to tool holder either pin or upside down loaded end mill.
2. Measure the tool diameter using the micrometer.
3. Determine the calibration tool length using this procedure:
- using the HANDLE place the indicator under the spindle face and lower the
Z to read 0.
- reset Z RELATIVE to 0.
- using the handle raise the spindle.
- load the calibration tool the spindle.
- using the HANDLE place the calibration tool tip over the indicator and lower
to read 0 on the dial.
The length of the calibration tool is now displayed in Z RELATVE readout.
Write id down.
4. Perform complete tool setter calibration procedure using just measured values of calibration tool length and diameter.
5. Once calibration process is finished, perform complete MEASUREMENT of the calibration tool. Of course, the values loaded to length and diameter/radius offset registers should correlate with the values given during the calibration process.
6. If the values differ - something is wrong with macro statements.

Stefan
Cogito Ergo Sum

 

[not a machinist]

Ok, when you say the results are oversize, do you mean it leaves the part oversized, or it measures the cutter oversized and leaves the part undersized?

sorry for the poor wording. The part is oversized.

 

[GiroDyno]

At least its oversized, easier to cut extra material off than cut it on!

I've never counted on one of those plunger tool setters to be 100% accurate before, I bet every tool offset on each of the 12 machines here has some compensation value entered.
If your probe is properly calibrated it should get you close enough to start cutting, but 'dialing things in' is a fundamental part of machining.

It's pretty easy to cut a feature, inspect with a probe/mic, and adjust tool offsets accordingly. Lots of examples of macros out there that you can modify to suit your needs, a 3 second Google search pulled up this one. Fusion360 (and I'm sure most other CAM packages) even have "Probe Geometry" functions that can generate these adjustments for you. What a time to be alive!

 

[mhajicek]

sorry for the poor wording. The part is oversized.

Ok, that does sound a lot like deflection. Keep in mind that it's not just the cutter that can deflect, and a TM-3P is not the heaviest and most rigid machine. A 2" indexable is probably pushing the limits, especially if you're running it hard and not skimming. Maybe try an experiment: use a new, sharp, 1/4" or 3/8" solid carbide endmill (designed for aluminum, and not overly long), probe the diameter, then use it to finish something to size with two skim cuts and a conservative feed rate. Then see how close the result is to what it should be. If it comes out really close to nominal, then it is deflection that's causing your problem. If it still comes out .003" oversized, then you have another issue that will be harder to diagnose.

I run a CM-1, which is also not the most rigid machine. A single pass with a 3/8" endmill can be .002" oversized, but I skim it a couple times and it's dead nuts.

 

[thesjg]

Ok, that does sound a lot like deflection. Keep in mind that it's not just the cutter that can deflect, and a TM-3P is not the heaviest and most rigid machine. A 2" indexable is probably pushing the limits, especially if you're running it hard and not skimming.

I would second this, I have a lot of hours on a TM-2P and a 2" indexable is actually a pretty large tool for the machine.

To add to this though, one wouldn't think a 0.002" bite would cause that much deflection... but... this makes sense to me, based on my experience with a similar machine. I would think you would do better taking a larger bite on the final/spring pass with that type of tool. Too small a step and those inserts have trouble taking a bite (especially if they're coated) and you deflect instead. My TM-2P did the very best work in 6061 at the bleeding edge of or just beyond the tool makers recommendations. ~1350 SFM for most of the polished Mari aluminum tools, and taking a big bite, running it hard tended to reduce deflection and chatter, and taking it easy was often unsatisfactory/just "ok". Once you step into the mass of a Haas VF or similar it seems like you get a wider sweet spot.

If it behaves similarly (leaves parts oversize) with different sizes/styles of end mills and different configurations of spring pass obviously this may not apply.

 

[tteitgen]

I am guessing that you have ran an indictor across the pre-setter foot. After you calibrate the pre-setter, have you checked the position of the pre-setter with a .500 dowel pin. What version controller is this in. Do you use the built in VPS to run the probe.

 

[EMTech]

Put a gage pin in a collet and check the diameter of the pin.

Does It measure correctly?

If not correct, then it is a probe issue. Possibly calibration.

If it measure correctly, then the issue is machine related. As others have said, it could be deflection.

 

[not a machinist]

Ok, that does sound a lot like deflection. Keep in mind that it's not just the cutter that can deflect, and a TM-3P is not the heaviest and most rigid machine. A 2" indexable is probably pushing the limits, especially if you're running it hard and not skimming. Maybe try an experiment: use a new, sharp, 1/4" or 3/8" solid carbide endmill (designed for aluminum, and not overly long), probe the diameter, then use it to finish something to size with two skim cuts and a conservative feed rate. Then see how close the result is to what it should be. If it comes out really close to nominal, then it is deflection that's causing your problem. If it still comes out .003" oversized, then you have another issue that will be harder to diagnose.

I run a CM-1, which is also not the most rigid machine. A single pass with a 3/8" endmill can be .002" oversized, but I skim it a couple times and it's dead nuts.

All cuts were in 6061-6 aluminum block 4"long x 2"wide and x 3"thick. I have run the majority of cuts at .0010" - .0025" per side (4 sides) and 12.0 ipm. a couple at .0100" per side at 12.0 ipm. Same results. Thus, I have ruled out deflection. I guess I have another issue that has not been determined, as yet. Of note; when the tech from HAAS/Phillips completed his testing, duplicated the problem and made changes to the macro program, he came out with a variation of only a few tenths using the 2" shell mill. A few days later I got the results listed here.

 

[not a machinist]

I am guessing that you have ran an indictor across the pre-setter foot. After you calibrate the pre-setter, have you checked the position of the pre-setter with a .500 dowel pin. What version controller is this in. Do you use the built in VPS to run the probe.

yes, I have indicated the foot to .0001" or less. this is a "new generation" controller. Machine is 2 years old. Problems began shortly after installation. I do use the VPS probing and cutting routines.

 

[EMTech]

Sounds like this may be a machine issue then.

You should run an outside contour and an inside contour with the same tool. The results will tell you if it is a cutter comp issue or a machine scaling issue.

 

[not a machinist]

Put a gage pin in a collet and check the diameter of the pin.

Does It measure correctly?

If not correct, then it is a probe issue. Possibly calibration.

If it measure correctly, then the issue is machine related. As others have said, it could be deflection.

I put a Maritool Master Calibration tool (.3000 diameter with .0003"TIR) and the diameter measured .3000".

 

[not a machinist]

Sounds like this may be a machine issue then.

You should run an outside contour and an inside contour with the same tool. The results will tell you if it is a cutter comp issue or a machine scaling issue.

I ran a .25" end mill to cut a .3750" hole and it was undersized by about .0025" using Meyer Z- gage pins.

 

[EMTech]

Has this machine been ball bar tested or laser mapped?

From the information you have given so far, I would suspect the machine.

1: A table probe that measures correctly.
2: tool leaves material on inside and outside contours.
3: light cuts taken to eliminate deflection.

My only other thought is the machine calibration itself.

After that I am out of ideas...