How can I remove the x axis scale on a PM25MV?

[WobblyHand]

My x-scale is missing counts. It loses over 8mm on 168mm of travel. How can I remove the scale from the mill? I didn't install it and there's no documentation on the scale installation. I removed the aluminum L cover and the M5 screws at the ends of the scale and it doesn't budge. I realize that the read head is attached still but it seems like there's something else holding it on the mill, like glue or other fasteners.

Have any suggestions? Had to scrap some parts because of this problem. Once I get the scale off, maybe I can clean it and test it. The scale seems to not count in a certain region. In any case, it's not possible to return to a datum, which means that the axis is NFG.

 

[Jim F]

Which is the reason to know how to use dials...............

 

[7milesup]

Which is the reason to know how to use dials...............

I'm guessing he does, and that doesn't answer his question at all.

 

[markba633csi]

It could be glued or if it's on a painted surface maybe the paint is acting like an adhesive. Some careful prying would be ok I think

 

[WobblyHand]

I will have to follow up with a call, but I do have to mention this here. Sent an email to PM this evening around 5:30pm. Got a real technical support email back in less than an hour later. I'm impressed. This was not a perfunctory response, but a real technical one. Didn't quite understand something in the email, but I'll clear that up tomorrow. I will keep you all posted.

 

[WobblyHand]

PM was quite gracious, stating they would make good for me. They made some suggestions for me to try. Unfortunately I got delayed in working on this. I was busy on my ELS. Today I was going to install a piece of 80/20 on my lathe. I had printed some toolholder holders. It seems that I needed to slightly open up a hole so I could thread it to 3/8-16. This would be nice to do on my mill. But it has be down due to the DRO. So the round tuit had come due...

So I swapped the inputs to the display to confirm that the problem followed the read head, not the display. Check.

I had removed the screws that attached the DRO to the mill table. Over time, the DRO slid downwards. It would seem it was glued to the table by cosmoline! Removed the DRO housing and the read head from the mill. Had to remove the way cover to get access to the screws. There were some chips up near the read head. Cleaned that all off gently. Remove chips, check.

Removed one of the ends and slid the read head out. I cleaned stuff with isopropyl alcohol and put things back together again. Took me a couple of tries before I got things in the right order. Cleaned the read head, check.

Oh yeah, lost one of the really tiny screws for a while. Looked everywhere for it. 15 minutes worth. Nothing. Swept a magnet in the slots looking for it, still nothing. Eventually found it inside a tee nut. It's still there - I need to find my tweezers to get it out.

The DRO has a dead zone where it is not counting. The region is about 2" long and it is near the table origin. Outside of the dead zone, the DRO seems to be fine.

Is there anything else I can try? I was able to remove one of the shields partially and look inside. The other shield seems to be stuck, although I suspect it is due to cosmoline. Using a light shining on the glass I didn't see any particulate matter on the surface. Can you clean the glass surface? Or will that damage it?

 

[B2]

@WobblyHand

I have no idea how they stick it to you machine other than with screws, but bending it is a no-no! However, I doubt that they intentionally glued it to the machine as one usually wants it to be free standing so that it determines its own straightness. PM should know. So chemicals may have melted the paint or other chemicals may have bonded it. Cutting the paint around and a little behind it may be the trick.

For proper operation, both the scale and the transducer must be clean. Otherwise, counts will be lost and the reading will be in error. Likewise, they must be properly aligned and uniform spacing is important. If enought paint, oil, or some other chemical has gotten on the housing in sufficient quantity to glue the scale housing to the mill then who is to say that it is not also on the optical scale inside the housing... or on the rails the transducer rides on... to cause the errors? The transducer on most, if not all, DRO's merely sends pulses to the electronics box to be counted. Missing pulses means missing steps of distance. So if there is noise that gets on this signal line or the connections are poor then the counts will be off (high or low). So you may want to check out the cable.

PS. We buy these things (DROs), but most of us have no way to check out their accuracy other than by comparing their reading to the mechanical motion of the dials etc. I have often wondered if when my cross slide on my lathe says it has traveled even a short distance, like 1 inch, how accurate it is? Its resolution is suppose to be 1 micron, but accuracy is another issue..... I put a electronic digital dial indicator on it and compared the two and they did not agree exactly. At difference distances one would read higher than the other and then at other distances this would reverse.... But which one, or both, is wrong? TBD.... I intend to further investigate this.

Dave L.

 

[RJSakowski]

Grizzly furnishes calibration graphs for all their scales. DRO's need to be calibrated for accuracy. It is possible to accurately calibrate your scales against gage block or a micrometer. Set up a guide rail parallel to the axis being calibrated . Sweep it with a test indicator to verify its alignment. At one end of the rail, fix a block and zero a test indicator and the DRO against that surface. Now move the table and place a block of known length against the fixed block and move the table to zero the test indicator. Your DRO should indicate the block length. If not, the DRO shoulfd have a built in calibration procedure.

 

[WobblyHand]

@WobblyHand

I have no idea how they stick it to you machine other than with screws, but bending it is a no-no! However, I doubt that they intentionally glued it to the machine as one usually wants it to be free standing so that it determines its own straightness. PM should know. So chemicals may have melted the paint or other chemicals may have bonded it. Cutting the paint around and a little behind it may be the trick.

For proper operation, both the scale and the transducer must be clean. Otherwise, counts will be lost and the reading will be in error. Likewise, they must be properly aligned and uniform spacing is important. If enought paint, oil, or some other chemical has gotten on the housing in sufficient quantity to glue the scale housing to the mill then who is to say that it is not also on the optical scale inside the housing... or on the rails the transducer rides on... to cause the errors? The transducer on most, if not all, DRO's merely sends pulses to the electronics box to be counted. Missing pulses means missing steps of distance. So if there is noise that gets on this signal line or the connections are poor then the counts will be off (high or low). So you may want to check out the cable.

PS. We buy these things (DROs), but most of us have no way to check out their accuracy other than by comparing their reading to the mechanical motion of the dials etc. I have often wondered if when my cross slide on my lathe says it has traveled even a short distance, like 1 inch, how accurate it is? Its resolution is suppose to be 1 micron, but accuracy is another issue..... I put a electronic digital dial indicator on it and compared the two and they did not agree exactly. At difference distances one would read higher than the other and then at other distances this would reverse.... But which one, or both, is wrong? TBD.... I intend to further investigate this.

Dave L.

There was no bending. I had removed the screws at the end and could not remove the unit. It turns out there was cosmoline in between the scale and the table edge acting as glue. I put a few drops of WD40 on the junction and over time it softened and the scale dropped due to gravity. After that the scale came right off. There was a bit of that cosmoline on the ends where it was screwed in. I removed all of it with WD40.

The only place where there seems to be a problem is in the center of the DRO. This is of course, the highest wear or use region. Its also where the concentration of chips would be highest. From shining light on the edge of the glass, I didn't see any chips or things on the surface. The cable was laced (tie wrapped) rather tightly, but the cable and the head don't move when the table moves.

All I know is I have return to zero problems that are randomly between 0.020 - 0.5". Most of the time it is skipping counts when the table goes left.
Can one clean the optical scale? If so how? What about the reader?

 

[WobblyHand]

Grizzly furnishes calibration graphs for all their scales. DRO's need to be calibrated for accuracy. It is possible to accurately calibrate your scales against gage block or a micrometer. Set up a guide rail parallel to the axis being calibrated . Sweep it with a test indicator to verify its alignment. At one end of the rail, fix a block and zero a test indicator and the DRO against that surface. Now move the table and place a block of known length against the fixed block and move the table to zero the test indicator. Your DRO should indicate the block length. If not, the DRO shoulfd have a built in calibration procedure.

The issue is not calibration. It is skipping counts and not returning to zero correctly in a 4" span. The return to zero error is random and about 0.020-0.5". Other regions seem reasonable, although I have not checked them with a 123 block.