XXT probes dropping from rack

[[Do...]]

Hi everyone,
Hoping someone can point me in the right direction on this one. All of our RDS CMMs are set up in the same fashion with the rack in the far right corner and facing towards -X and all have roughly the same approach parameters (some tweaks to approach direction before/after based on stylus length). I have one machine that now has started to have an issue when racking styli. As it disengages the adapter the plate, the dust cover on the rack catches onto it and slides it forward as the machine moves back into position in front of the rack. This results in either the stylus being shot out onto the granite or being slammed into by the XXT head.
The documentation within Calypso was not entirely clear to me so before I destroy the machine I wanted to see if anyone else has had this issue and how they resolved it. My current train of thinking is that the XXT needs to retract further up in +Z before moving away from the rack. Would that be accomplished by editing the value for "Top" in the "During Change" row or do I have that incorrect? Any help would be appreciated!

[[Je...]]

Check to make sure the tab on the plate is seating into the pocket for its location. Often if the tab on the XXT plate gets bent, it wont seat in the pocket properly.

[[Do...]]

Hi Jeff,
I took a closer look at the tabs on the adapter plates. Some had a minor bend to them, however, I am having the dropping issue on both bent and unbent plates. I even tried straightening the bent tabs to see what would happen, but it did not improve the issue. What should my next steps be at this point?

[[Da...]]

The probe rack may have been bumped and moved. Realign each bay as you would if you added a new bay. If it still continues, I would contact Zeiss.

[[Je...]]

If you are having an issue with the plates being pushed out by the door then the underlying cause is that the plates are not being seated in the rack properly at drop off.
Typically the causes for this are (on XXT rack specifically):
- Broken/damaged XXT rack (typically causing flex in the top cover so the position of the pocket offsets are invalid/inaccurate)
- Improperly aligned rack using the MasterProbe
- MasterProbe length being set incorrectly
- XXT rack has been moved since locating or is skewed/crooked
- MSR rack is crooked. Should have no more than 0.2mm XY skew and XZ tilt over its entire length.
- Bent stylus plates

If you are seeing the plates being pushed out, the above conditions should be triple checked. If no other option exists then you may add an offset to the "Offset" column. Each field should be tested EXTENSIVELY and documented for each individual pocket. An offset should only be applied if all other checks pass as these should almost never be necessary in a properly configured setup.

Additionally, for an XXT approach parameter, I advise a Z clearance BEFORE and AFTER change to be set to 9999 so the machine navigates to Z soft limit before performing an RDS indexing move. A "DURING" change safety distance is not usually necessary. Approach direction (yours are set to 70/100/100) should be set to about 10mm + length of stylus protruding in the same direction as the approach. IE If you have a star-probe setup with 1x30 styli, you should have an approach of 40mm. A single vertical Z stylus (like MasterProbe) should have an approach of ~10mm. These are not "firm rules" but general guidelines to get your navigation assigned for a more efficient use of volume as well as reduce risk of machine collisions. If the probe rack setup is overly complex (like XXT set up on a 2 or 3 tier rack plus MT pockets) then more complex assignments must be created on a per pocket basis

[[Jo...]]

Please sign in to view this quote.

I've had this exact issue going on with my Contura for years now. I've been down every road and avenue; re-squaring the rack and relocating the sockets, shimming the rack to get even more square, been on the phone with multiple Zeiss reps and gotten nowhere. I've swapped out sockets to other machines to see if the problem repeats elsewhere and it doesn't. It's been super frustrating, time consuming and expensive.

A coworker of mine, Odin bless him, MacGyver'ed the problem while he was filling in for me one week by using Scotch tape to keep the dust covers permanently open (keeping them from pushing the probe back out) and I haven't dropped a probe since. It looks super gross and unprofessional, but I haven't had to spend $3,000 on a new rack and the machine runs great. I just use a lint-free cloth and some solvent to clean the contacts off on the adapter plates in lieu of the cheap plastic cover. There are some parameters you can mess with under the Master account (if you have that level of access) that can alter the behavior during probe changes (such as the retract in Z before it moves out of the socket) but I haven't had the time to mess with that.

[[Ja...]]

I have had this problem before. First thing I would do is recalibrate the master probe and relocate the rack with the master probe if you haven't already. I have found that the XXT rack with 3 sockets is not perfect. The bottom of the center socket looks parallel but the bottom of the two outside sockets are not. I have an 8x150 mm. long probe that wobbles more that the others because it is so long. I have trouble with that one once in a while. I found that it was worse in one of my two racks so I labeled my racks left and right and always put that probe in the rack on the right.

[[Mi...]]

Edit: I should add a potential cause nobody else has listed. Clean the BOTTOM of the XXT rack especially the little nub that interfaces with the Adapter Plate, you'll see what I'm talking about if you flip it over. Any buildup on that will make the Stylus System take longer to settle and that could cause it to get kicked out. It's a common occurrence for us with our shop floor machines, less often in the labs.

This forum wont like me saying this but, Zeiss needs to do better with these XXT Racks, they are sub par compared to practically everything else branded Zeiss. Yes there are many things we can do to make them suck less, most are listed in the thread, but the fact remains they should be better considering their job and cost.

QMark makes a knockoff holder that is stiffer, and the adapter plates sit a bit deeper and more securely, it's also a good chunk cheaper and due to the much stronger/thicker plastic it lasts much longer. The only downfall is the top plastic tends to crack around the dimple for locating. Switching to that holder saved us a lot of headache, and saved cost from the holder itself and probes being spit out.

Caveat, don't use QMark for anything else. We tried, went back to Zeiss and ITP for Probes and Zeiss only for adapter plates. It's fun having brand new probes fall apart trying to calibrate...

[[Do...]]

Thanks for all of the replies. I've performed each solution listed below, but haven't had any luck. Any further suggestions are welcome.
- Verified and straightened any bent tabs on adapter plates
- Verified that the holder top cover was not warped or bowed
- Re-aligned the MSR/holders to be within .2mm in XY and XZ
- Shimmed under the MSR to check for any change in behavior
- Set the stylus length (we use a copy of the masterprobe rotated 180º)
- Set stylus length of masterprobe as a test
- Qualified the masterprobe and the copy used for aligning the holders
- Re-defined holder locations at each step along the way
- Replaced holder with a "working" one from another machine
- Checked for cleanliness or damage to the holders
- Fiddled with a multitude of values for the approach parameter and offsets
- Taped a piece of shimstock underneath the holder so the adapter plate would stay low enough to not get caught by the cover (almost got that to work, may revisit later)
- Taped a piece of shimstock to the dust cover face hoping the extra clearance would keep the cover from catching the plate (note: it did not)

I did notice that if I turned the speed way down (maybe 5% or so) the probes would intermittently drop off successfully. I am at my wit's end trying to troubleshoot this issue and may wait until after the holiday to take another crack at it.

[[No...]]

Please sign in to view this quote.

How far up does it move currently? I watched a stylus change at our O-Inspect today and there the head moves as far up as it can afford without the dust cover snapping shut beneath. It stops maybe 1 or 2 mm below the cover's top edge before moving out.

Can you possibly post a cloesup video of a failing stylus change for comparison?

[[Ni...]]

Sounds like the machine is dropping off the Stylus into the rack but its not staying there? The spring cover is pushing it back out which makes it fall or makes the stylus hit the head sensor?
IF possible, could you post a video?
And you're using the RDS which are the smaller M3 stylus correct?
Because the racks for the M5 stylus have a little tab you need to rotate so the stylus sits in there properly.

You did qualify the master probe with the reference sphere and THEN qualify the copy of the master probe to the reference sphere right?
Then you did the "Define Holder Location" right?

because that should be able to fix it.

[[Ca...]]

I'm not the originator of this post, but we have the same problem. Here's a video of a partial probe ejection. There's a satisfying "crunch" at the end. This is the machine Jon Drown referenced on 11/22.

Again, after months of troubleshooting, the problem was remedied with tape.

Probe Crash.zip

[[No...]]

So the actual problem is that the probe plate doesn't settle back down into the bay after being seperated from the magnet. It stays in the upper position and is then pushed out by the snapping lid. Hmmmmm.....

[[Ca...]]

Please sign in to view this quote.

That's "our" problem here. The dust cover pushes the plate back off the rack. Holding the dust cover back has negated the problem here.

[[Do...]]

The video Casey posted is exactly what we're seeing on our end, with the occasional probe launch onto the table when the head doesn't catch it in time. I'd hate to tape the covers back, but that may be the best option to avoid machine downtime for now. Hope to hear if anyone else finds a solution.

Nixon, I qualified the MasterProbe then qualified the copied Master at 180º. Defined length of copied Master in the auto change window and then used it to define holder locations. I believe that's the correct order, but please let me know if I've got that mixed up.

Thanks all

[[Ni...]]

The issue I see is that the Stylus is not dropping into place after the sensor head dislocates from it.

Few things come to mind,
Using alcohol or some liquid that's not going to destroy the plastic, clean out the inside of the rack. Clean around the adapter plate and under.
Clean where the stem of the stylus connects to the adapter plate.
Feels like there might be gunk that's holding up the adapter plate and not letting it fall into place.

Another thing to try would be to get a new adapter plate with a random stylus attached to it and have that load into that same rack.
If the new adapter plate falls freely into place, then could be that the adapter plate might be a little bent so it is not freely falling into place.

Another thing would be to unscrew and screw the rack back. Perhaps a lot of tension was being applied when screwing it in next to the other rack. It might have caused a small bent on the rack itself, so the adapter plate won't freely fall into place.
You'll need to define the holder location again if you do this though.

That's all I can think of. Maybe contacting Zeiss support to have them take a look at it might give better results.

[[Mi...]]

Another potential cause is if the internals of the XXT head have been rotated, they won't drop the probes off correctly. If you are like many people relying on mostly down probes this won't be very obvious, but stick a probe system with a +/- X facing or +/- Y facing probe in and either visually line it up to something already in line with the machine axis (Front or side of the XXT Rack) or calibrate and check the coordinate values after.

This is very easy to see if you're using an integrated cube as they should be "Perfect" for all the side probes.

A normal amount of error we find with 60mm long probes is about 0.3mm.
Eg Front probe has a 0.3xxxx Y value
Back probe has a -0.3xxx Y value

Longer or shorter probes will change what values you should expect.

If that is the only thing wrong with the head, you can likely loosen that collar and straighten it out then tighten it back up if you must, but replacing the head is the better course of action.

Here is a very extreme example of what I'm talking about, that is an integrated cube so all those probes should be in line with the machine axis.. I was originally called for probe change issues...*it just happened* 🙄

[[Ni...]]

Another thing you could do that i don't think you've done is probe the ends of your rack to make sure it is straight.
They should be very close on the Y axis or X axis. Depending on where it is sitting.
If it is not, you'll need to unscrew the rack from the table and push a bit and screw it back on the table.
But only if there is a big difference.
After doing that you'll need to define the holder location since you did move the rack.

I used the image from Michael and modified to show how the whole rack would look and where to probe. Should be straight.

rack probe location.png

[[Jo...]]

Please sign in to view this quote.

Can you explain where to find that setting? I know how to set the sensor to retract all the way to the top before and after a probe change, but cannot figure out how to change how far it retracts during the actual probe change routine, immediately after it puts the probe in the rack. I have master account access and can get to some of the stuff that is greyed out on the "Approach Parameters" screen, but I don't know which field to edit.

[[Jo...]]

Please sign in to view this quote.

They will just say everything that has been said on this thread already, then suggest purchasing a new MSR once all else has failed. Been there, done that.

[[No...]]

Please sign in to view this quote.

I wasn't trying to say that I had changed any settings. I just described what I saw the O-Inspect do (maybe 1-2 mm was a bit optimistic, but seeing the video I guess this isn't your real problem anyway). By the way, I think these "inside the bay" movements are controlled by a macro stored in the controller.

Have you checked for differences in the relevant dimensions between a working and a failing rack? Plastic is susceptible for a number of influences (tension being one, humidity another) that may change the geometry of the rack considerably.

[[Ni...]]

Please sign in to view this quote.

I don't know. This seems like a failure that happened on its own. This should be something Zeiss should take responsibility for. Wouldn't hurt to reach out to them and see.
If they truly try to upsell them, then just end the conversation there. I'm also invested in this and would like to find a solution haha.

[[Do...]]

Please sign in to view this quote.

I believe you are correct, those movements are controlled elsewhere and are likely locked down by Zeiss (unless someone knows how to access that). In the approach parameters I tried messing with the "Top" value "During Change", but that just raises or lowers the head in Z as it moves from one pocket to another, honestly not sure what the benefit of that is.

In terms of squaring the rack to the machine, I used the same method Nixon described by probing off the ends of the near and far holders and got it to ~.006" in yz axis over the four holders.

I had placed a functioning holder from another machine onto this rack, but continued to have the same issue. Visually the holders look identical up close, but I'll see if I can find dimensional differences between the two.

[[Do...]]

Progress.
Went through the whole process again, starting from the beginning and noticed that the head was not entering the pocket on centerline, but rather it was shifted about .030". We found this quite strange, considering we had just defined the holder location along with everything else.
To remedy this though, we began entering offsets in the "side" approach parameter until we got it to drop off and pick up the Masterprobe without error. Unfortunately, the offset did not seem to translate well for other pockets within the same holder, again strange.
Next, we tried swapping the XXT from another machine. After running through the standard checks and balances we immediately saw an improvement. The head was now entering the pockets on center without the need for additional offsets, but we hit yet another snag. Now when a stylus was being dropped off, the drives would shut off.
We had seen this issue in the past, and the only thing that seemed to help was wiping down the Z shaft well with IPA and a lint free cloth. Gave it a shot and the next change was successful, but the next one failed. Wiped it down again (avoiding the scale of course) and it's now cleared several changes without error.
We are running our calibration routine and are hoping for the best. I'll continue to monitor the situation and follow up with more information about our progress.

Appreciate all of the feedback and replies so far.

[[No...]]

Please sign in to view this quote.

They are not really locked down, its more like "security through obscurity". You can easily FTP to the controller IP, log in anonymously, find the macro directory and even look at the macros (simple ASCII files). In the past I even edited one of them a tiny little bit when we had remotely similar issues with our ProMax rack after the machine had been moved.
But without fair knowledge about the macro language and the commands of the CNC controller I wouldn't recommend to follow my lead. .

Please sign in to view this quote.

What I had in mind are the dimensions of the parts inside each bay where the probe plate touches the rack.