X axis offset after Probing

[christopera]

Hi Dave,

I have an issue that I thought was cam related but am realizing it's related to something else. After I run a probe routine to find a hole center and then mount a new tool (M6t16 for example where as probe is M6T101) and then enter X0Y0 and then lower the tool to the hole the tool is obviously offset X- visually. It's hard to measure the exact offset as the hole that I caught the issue on is a .332" diameter hole and I am probing with the Tormach probe (diameter .2"). After mounting the probe and entering XOY0 and lowering the probe to the hole it is dead center visually. This is true even after Mach 3 closures that include fixture saves, the probe will go visually to a true X0Y0 with no visible runout in the center of the hole. When I rotate the probe 360 degrees it shows no visible runout, on the test indicator it shows less than 2 ten thousandths runout. I am very perplexed. The only thing I can think of is that I have an odd tool offset parameter that simply I am not aware of.

My mode is listed as: G15 G0 G17 G40 G20 G90 G94 G54 G49 G99 G64 G96

I've run into this issue with multiple probes now, multiple programs, no program at all, put simply when I try to find center on my 0 coordinate as established by a probe I end up with this slight X- offset. I am missing something I'd say, but after more than a few days of searching for answers I am coming up blank, hopefully you can provide insight!

Best Regards,
Chris

[christopera]

So I was digging around in the forum and I see that in May of last year another posted a similar issue, though he identified his issue to be a controller or plugin of some variety. I'm using ESS so I don't think my problem is there. That said, I know my tip wear compensation for my probe is around .02", which if I had to guess is around about the amount that I am seeing my tool being offset in X-. This wear compensation was calculated by means of the calibration function using a known diameter hole. I think this evening I am going to remove that offset and see if it doesn't correct the issue. It was nice that the probe was reading consistently and accurately the size of holes within +/- .0015 or so. It's important to note that I get very consistent results no matter the rotation position of the probe, initially I was thinking that there must be a probe issue where in the X axis movements I was getting bad results. So I tested by turning the probe in 90 degree increments and re-probing. This made effectively no difference.

If changing offset doesn't work I think I may attempt to reinstall the software. I need to make sure to back up my setting first though, which worries me, as this could be a setting issue somewhere along the line.

[christopera]

Well after further playing I'm at a loss. The prob tip has .0001" runout, the probe body has less than .0005" runout. The repeatable holder and spindle are both under .0001" runout. I probe a hole to find center, the op runs. I manually run the spindle up, change the tool, no X or Y movements made, run the spindle back down and into the hole and the tool offset described before is visibly evident. I can turn the probe any number of ways in the 360 degree revolution, the offset is always visible in the X- direction, and I think it's probably there in Y+ as well, it's just hard to see behind the tool. How the probe appears perfectly center not just by eye but by measurement as well but I end up with an obvious tool offset I do not know. It seems impossible. I think I'm going crazy.

[DaveCVI]

Hi,
Alas, I don't have a ready explanation for what you are seeing. The circle probe is a complex operation which is made up of smaller probe operations. For example the X/Y center find is done by invoking X and Y single axis probes. So, If it were me, I think I'd try to manually position the probe in the hole and do some simple X- probes to the edge of the hole to see if the offset shows up as part of the simpler X probe operation. This may help narrow down what op is associated with the offset.
Dave

[christopera]

I figured this one out. After running the scenario through my head many many times the only conclusion I could come to is that it was a geometry issue, specifically that of the milling machine itself. Though I had recently trammed the mill and squared the head I knew something had to of moved. So yesterday I had a complex part to make and was resolved to solving the issue. Sure enough, the head had rotated out of square. I couldn't recall crashing any tools so I kept investigating. I use a gas strut to aid in lifting the head, the upper mount connects to head not the saddle and the head rotates on the saddle, apparently the spindle head lowered low enough that the strut bottomed and resulted in rotating the spindle head slightly and bending the upper strut mount in the process. Yikes! I can only guess that it did this during a TLO measurement of a very short tool that I have no recollection of. I'm going to remove the strut and monitor to see if I even need it. I'm glad to say after squaring the spindle head to the table I was able turn the part successfully. It required 5 setups, each of those I found the WC using the provided probing operations. It feels good!

Being a self taught home machinist is always an adventure...

[DaveCVI]

Hi,
It's good to hear that you found the problem.
If you end up needing to use the strut, you might raising the touch plate with a block (and thus reduce the amount of Z travel required for short tools).
Dave

[christopera]

Well, running sans gas spring was a failure. I couldn't audibly detect step loss but after a tool crash and a some analysis it's obvious that it did occur. I guess two years ago when I installed it because I believed the machine would benefit I was on the right path. Reinstalling it and trying again confirmed it, everything was nice and consistent. Nevertheless, probing has been superb. Your recommendation on the Tormach unit has resulted in a massive improvement in my machine operation and the products I produce. Thanks again!