Update from older software version

[LeChuck]

Yeah, seems like it. I'm not sure about the parameters in DMMDRV and because my specific servo is not listed in that program, I can't run an autotune (I have the parameters from the online doc but that application is not accepting a 0.xxx value so I can't enter them). I don't know if this is causing an issue. I'll contact them about it.

I spent a good amount of time earlier with a dial indicator. Set it up on Z and zeroed it, then traveled left for exactly 4 revolutions, which is 0.4 inches, or 10.16mm, on this indicator. Went back and forth many times and tweaked the mm/rev until I got exactly 10.16 on the DRO for 4 turns. End result is 2.495, which is very close to what I had already been using(I also tried the theoretical 2.54 but no matter what, the zero still creeps). I'm not getting a dramatic change on the back/forth that I've been doing, but it still creeps to the left every time by a visible amount. I haven't tested to see if the creeping happens on the G30 or the G28 so I'm not sure which direction it happens in. The thing that ticks me off is that each time I home after that, it goes to the same spot that keeps creeping forward, and DMMDRV always shows the same absolute position. Meanwhile, the X, with a similar value and same motor and settings, is going back to the same spot always, and always homes the same.

I'll probably just pull the motor tomorrow if I get a chance and see if I can spot anything weird. I can't imagine that this problem has existed before. I have cut a lot of parts while tweaking my X position and thread fitting, and profiling, and this was the first time this happened. When it started happening, I noticed it right away when it started cutting by doing rapids into the piece instead of 1mm away like I had set it. Then did a rapid 2mm in and just broke it.

[LeChuck]

Here's the latest in this saga:

I pulled the Z motor this morning and didn't find anything obvious with the mechanics. Everything seems tight and smooth, although I ended up slightly tightening the axis pulley, and making sure there was no backlash in the movement (I didn't check that before I started on the nut so I'm not sure if there was any to start with). In the end, it turns with a bit more resistance than before and more like the X axis. I also swapped the motors (and belts), so now the old Z motor is on X.

Some things have changed:
- now X is homing in a different spot than before. Looks like it's just finding the pulse in a different place 2 or 3 mm lower. I'm not sure if it's something I can change (by turning the axis without the belt on I guess, but I don't want to remove it again, or would that work if I just unplug the drive encoder then just adjust manually?) or if I just need to start my homing a little after the mark now, but it's not a big deal, I don't need that extra bit of travel on X. I need to redo my tool #1 position I guess, hoping this will stay consistent.
- I did a bunch of G30/G28 back and forth movements like yesterday, and it looks like it's now returning to the same spot on my tape markers, at least no visible change.
- So far I'm no longer getting any Z scraping or X knocking during those movements.

I'm not sure that I'm ready to trust it but I'm going to continue playing with this later today. Going to have to redo the Z positioning in a few of my programs and watch that this doesn't change again.

[tblough]

How are the pulleys held on to the ballscrews? Just set screws? Keyways? Keyless bushings?

[LeChuck]

I didn't completely remove the pulley so I'm not sure how it locates itself on there, but there is a locking nut at the end. If you tighten that too much, it will lock the whole ballscrew. Loosen it too much and the screw still turns fine but then there's backlash, so I tried to do it just right to where the screw turns smoothly but there isn't any backlash when quickly jiggling the pulley back and forth.

[tblough]

Something is terribly wrong with that design. You should be able to securely tighten and mounting screws/nuts without binding the bearings. My guess is there is nothing securing the pulley to the shaft other than this one nut and therein lies your problem. Your pulley is ultimately slipping on the shaft.

You're going to need to disassemble the shaft end and see how the bearing are mounted; how the pulley is secured to the shaft, and how these components all come together. Post some photos of the disassembly and we can make some suggestions on how to fix the problem.

[LeChuck]

You're the glass half empty type, aren't you

There is something on this page about the ballscrew and pulley and how the nut holding the pulley also sets the preload on the thrust bearing. We'll have to see how it holds up. Maybe I over greased it and some made its way there.

[tblough]

Nope. Just a realist. Hope is not a reliable strategy. You'll be chasing this for a few more months if you don't fix the underlying cause.

[LeChuck]

I get that but there isn't much I can do about how the pulley mounts on the ballscrew.

[LeChuck]

It took me a while to reset my positions, but I decided to trust it with doing profiles on some parts I cut yesterday on the manual lathe.

It worked perfectly. I eyeballed the position (and left the tape on) each time it returned to home and it's not losing any distance so far, so I'm going to continue and keep an eye on that. If I see it drift again, I'll go straight to have a look at the pulley nut without touching anything else. I had been having trouble with getting exact results on X but spent more time with the dial indicator today and these profiles came out to exactly 15.50mm wide as programmed.

I appreciate the help, will keep an eye on that pulley as I haven't given up on this yet and it seems like it's looking better now

I really like the outside profile operations because they need NO COOLANT!

If anyone has any recommendation on good coolant for plastics, please feel free to mention it. I use TRIM 9106 at 5%-7% dilution and it smells like the antifreeze in my car and leaves sticky residue on everything, parts, collet etc. I'd just like something that's closer to water but with rust inhibitors.

[martyscncgarage]

Here's the latest in this saga:

I pulled the Z motor this morning and didn't find anything obvious with the mechanics. Everything seems tight and smooth, although I ended up slightly tightening the axis pulley, and making sure there was no backlash in the movement (I didn't check that before I started on the nut so I'm not sure if there was any to start with). In the end, it turns with a bit more resistance than before and more like the X axis. I also swapped the motors (and belts), so now the old Z motor is on X.

Some things have changed:
- now X is homing in a different spot than before. Looks like it's just finding the pulse in a different place 2 or 3 mm lower. I'm not sure if it's something I can change (by turning the axis without the belt on I guess, but I don't want to remove it again, or would that work if I just unplug the drive encoder then just adjust manually?) or if I just need to start my homing a little after the mark now, but it's not a big deal, I don't need that extra bit of travel on X. I need to redo my tool #1 position I guess, hoping this will stay consistent.
- I did a bunch of G30/G28 back and forth movements like yesterday, and it looks like it's now returning to the same spot on my tape markers, at least no visible change.
- So far I'm no longer getting any Z scraping or X knocking during those movements.

I'm not sure that I'm ready to trust it but I'm going to continue playing with this later today. Going to have to redo the Z positioning in a few of my programs and watch that this doesn't change again.

Just FYI. If you are using the index pulse then yes, its home position will change based on the new index pulse point.