hexagon logo
Register Sign in
  • Replies 40 replies
  • Subscribers 221 subscribers
  • Views 377 views
  • Users 0 members are here
Options
Related

Hexagon Wrist Failure After Less Than Four Years - Is This Normal?

mbatten
A bit over three and a half years ago we bought our first continuous-scanning-capable CMM, a Global S Blue 9.12.8 with a HP-S-X1 scanning head (I believe that's just Hexagon's rebranding of the Leitz LSPX line) and Hexagon's 7.5 degree indexable auto wrist, the HH-AS8-T7.5. What the machine has allowed us to do has been amazing, and on the whole we're very happy with the system.

We've never crashed it badly (as in, nothing big enough to make any visible dents on either the scanning probe or the wrist), though I did put too much torque on the scanning head trying to set up a star probe at one point and had to replace that.

Over the course of the last week or so we've had repeated instances of the wrist not seating - sometimes it will try multiple times and then succeed, other times it will not succeed and then just give up. This morning I had to power cycle the entire system to get it to work after it gave up. From what I've seen this behavior points to a failure of the wrist that will require a replacement.

We use the machine frequently over two shifts, five days a week, but beyond the regular use I don't think we do anything that should be particularly hard on it. I see pictures in Hexagon's marketing literature of these heads waving around huge laser scanners that must weigh a good bit more than the scanning head we have on it. Is this seemingly-short lifespan normal? Did we just get a bad egg? Is there something in our use or programming that could be causing more than typical wear? My boss is not very happy at the thought of having to replace these so frequently. Any ideas would be appreciated.
  • in reply to KatmandudaMeow
    Some variant of "Wrist not seated", usually.
  • in reply to KatmandudaMeow
    typically there will be an error number associated with the problem. there may be a timing issue with the controller and cmm. double check your input air and the regulator is set to ≈70 psi. Just spitballing here because ive seen this happen to ours also in the past. may need to call in a service technician to check it out
  • "We've never crashed it badly (as in, nothing big enough to make any visible dents on either the scanning probe or the wrist), though I did put too much torque on the scanning head trying to set up a star probe at one point and had to replace that."

    visible damage isn't the ONLY indicator of problems. repeated forcible unseating of the pivot does internal damage. I've not seen the inside of any head newer than a PH9, but the 'lock' of the PH9 is a flat plate that one of the servo pulls to 'force' the locking indents into position to secure it into that rotate. Now, on the PH9, if you repeatedly forced it to unseat, especially at a specific angle, then the plate would get warped at that indent inside the head, and you would start getting seating errors. On the PH9, there is a set screw that you can tighten in order to get more 'power' on the locking plate and that could get you along for quite some time before you had to RBE the head. That set screw is long gone since it allowed people to get more life out of the head. I have no idea if that plate locking mechanism is still in play in the newer heads.

    Also, there is a life span of rotations. If your programs are constantly changing angles more than they should (as in, using the same angle in multiple places in the program instead of all at once) you are wearing the head out faster than you should be.
  • I have had a PH9 fail after like 17 years, and I had a 7.5 degree hexagon head fail after 10 years - it WAS hit.
    I have a functioning 18 year old PH10, a five year old 7.5 degree hexagon head and a 11 year old Revo.

    I would say four years is abnormal.

    With what Matthew said above me, if you are running the thing 14 hours a day and it is rotating after every feature, maybe four years isn't abnormal, but I am more inclined to think someone might have crashed while no one was looking (we always blame the night shift lol), or changing probes manually and applying force that is causing it to unlock (again as stated) which probably led to the shorter than normal life span.

    I will say, I do my absolute best to program for efficient run cycles. I load a probe and do EVERYTHING I can with that probe before I change it. No probe checks the part with any rack time in between touches.
    I rotate an angle and do EVERYTHING I can with that angle.
    I don't look for an optimal angle, I look for an angle that will do as much as possible so I don't have to rotate the head.
    I don't want to move the machine clear to rotate the head, and I don't like sitting while it changes off the rack.
    This will also extend head life (at least on the heads that lock)
  • another thing is using TOO much length, that will send a head to an early grave.
  • KatmandudaMeow: It did its thing a few minutes ago and I was able to capture the full error through the web portal: 

    
Error code:0X1500601A Module:AI - Axis Interface (layer 5)
Label:PHM datum error (not seated)
File:../../src/aiCAxPhm.cpp - Function:axInterpPhmData - Line:1084
Error type: Strong2 - Motors immediately turned off
Answer from Probe Head Module = HDA0.0B0.0
Indexable motorized probe head controller - datum error/wrist not seated
Current command: NOINPUTCOMMAND
Position: 528.909 1099.638 -394.754 0.000 0.000 0.010 0.000 0.000 0.000
Velocity: 0.000 0.000 0.000 0.000 0.000 0.000
Auto mode
CNC mode
Probe seated
Safety barrier closed or not present
Occured:03/20/2023 13-46-57.810
Reset: 03/20/2023 13-46-57.814​


    The air pressure currently reads about 60psi - there's no air going to the wrist, though, is there?

    : There has been occasional forced unseats, but until the various messes associated with this issue they were very rare. Today it bombed out with a seating error in the middle of a probe change, which then confused it about which probe it had, causing a tool changer crash that resulted in a forced unseat. That sort of event has not at all been the norm until this mess. I've always been conscientious about not changing angles more than I need to, mostly for the sake of cycle time.

    The annoying thing about this is we have some quite old Brown and Sharpe CMMs running Renishaw PH10T's (I think) with TP20s on them that have been chugging away for years and years despite gouges in the side (one or two of which might have been my fault Blush) and various such abuse. By comparison this Hexagon head has been living a pretty cushy life, and here it konks out already?!
  • in reply to mbatten
    shouldn't be any air going to the head.

    yeah, if the strength of the magnet holding the probe in the rack is enough to unseat the head, I would say the head is toast.
  • in reply to mbatten
    no but it does check its position and times out if it hasnt been met.
    ya you may need a new head it appears.
  • Other notes:

    We had a Hex-a-tech in here at one point that told us it would improve our accuracy to have the head unlock and re-lock immediately after it picks a probe out of the changer. He changed some Settings Editor thing to enable this. It's not changing angles, it just unlocks the A0B0 position and then locks it again. Could this cause excessive wear?

    Also, , we did for a couple of years have an absurdly long star tip build for getting into some holes on one particular part for which we didn't have an index angle close enough. We never did anything with it that wasn't at A0B0, though. Would just carrying such a tip around cause a problem for the head, or just the excessive torque that would be required to swing it to, say, an A90 position?
© 2024 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement