[wtopace]

Hey Guys, just wanted to report back on a successful gib adjustment. I was having lots of chatter doing circular interpolation at 12 o'clock and 6 o'clock positions. Everything else seemed perfectly fine. Even at 50% of my commanded feed rates (which weren't anything too crazy, 42 IPM on 12L14 I believe), I was still getting chatter rounding the corner. Through some searching online, it seemed like it could be my y-axis gib.

I adjusted the gibs on all three axes and the problem completely went away. In fact, I cranked the DOC up and the feed rate and the entire cut still sounds great! Surface finish seems to be a little bit better as well, although that could just be my eyes playing tricks on me.

If you decide to adjust your gibs, I would definitely recommend using a small ratchet with a screwdriver bit for the back side of the y-axis and a very long flat blade screwdriver for the front. Once I started using the correct tools, it went very quickly.

The process I used:

1. Attach an indicator to the table on a mag base and zero on the spindle nose (I used a 0.0005" DTI I use for gun work, seemed to work well for me here)
2. Zero the DRO for the axis you are working on
3. Jog the table IN THE OTHER DIRECTION in small increments until your DTI reads 0.001" of movement (I used 0.0001 steps in PathPilot)
4. Verify the amount of movement on the DRO (this will be different from the actual movement, the difference is your backlash)
5. IMPORTANT: Create a written table with the following columns (at least this worked well for me): | DTI Reading | DRO Reading | Number of Gib Turns |
6. Loosen the gib quite a bit (Tormach recommends 8 turns). For the y-axis, this means loosen the front screw and tighten the back screw. IMPORTANT: Do not move the table while doing this, you must loosen one side and tighten the other with everything stationary. The right tools for the job REALLY help here (small ratchet / short screwdriver / long screwdriver / etc...)
7. Jog to zero the indicator, zero the DRO, jog THE OTHER DIRECTION until the indicator reads 0.001", record the indicator reading, DRO reading and -8 turns on your chart
8. Tighten the gib 1 full turn
9. Jog to zero the indicator, zero the DRO, jog THE OTHER DIRECTION until the indicator reads 0.001", record the indicator reading, DRO reading and 1 turn on your chart
10. Tighten the gib 1 full turn
11. Jog to zero the indicator, zero the DRO, jog THE OTHER DIRECTION until the indicator reads 0.001", record the indicator reading, DRO reading and 1 turn on your chart
12. .... repeat until you notice your DRO reading increasing for three consecutive turns on the chart
13. Back the gib off until you get back to the starting number (or slightly above it)
14. Jog to zero the indicator, zero the DRO, jog THE OTHER DIRECTION until the indicator reads 0.001", record the indicator reading, DRO reading and 1 turn on your chart
15. If you're happy with the number being "back to the start" - then you're good to go!

Notes:

• I was very frustrated until I slowed down and charted everything to visually see the changes, rather than just sweating in a Texas garage in the summer spinning screws in and out! Once you get to see the backlash on a chart, it helps a lot.
• The whole "other direction" thing is important for measuring backlash. If you have done manual machine work, this will come naturally to you as you will feel "wrong" about jogging both ways during a zeroing operation =)

Hopefully this is helpful to someone else, I wish I had the above list when I started. Also, feel free to correct me anywhere I went wrong, but my machine is running much better now than before and the dreaded y-axis directional change chatter is gone!

[keen]

Hi - Yes that's similar to the method I use to set the gibs too. Too tight and slid friction gives too much movement 'delay' or 'grip delay'. Too loose and the slides slop around.

I don't actually chart it, but fit dial indicator, change axis +/_ directions, zero DRO, note indicator reading, change +/- directions and make small step movements until the indicator responds, note difference in the readings between the indicator and DR0, tighten gib - if the difference increases, loosen gib - until the difference in the readings is at the minimum. On my first machine the X and Y is nom 0.0008" and on my new machine nom 0.0005" from memory.

Was the minimum backlash you could get for the X and Y 0.001"? Thats about normal I gather.

Did you check the Ballscrew end thrust bearing preloads also? If they are not at optimal you may be able to reduce the backlash more.

Keen

[mountaindew]

Hey Wes, Thanks for posting.
I copied this to my notebook and filed it under machine adjustment procedures to compliment the procedure Tormach PP manual details.

Recently I performed gib adjustment in my shop with the temperature around 80+ degrees. Anyone have experience with or how this could effect machine performance in the coming months when my shop cools down and starts hitting low temperatures of 50 deg?

[wtopace]

Hi - Yes that's similar to the method I use to set the gibs too. Too tight and slid friction gives too much movement 'delay' or 'grip delay'. Too loose and the slides slop around.

I don't actually chart it, but fit dial indicator, change axis +/_ directions, zero DRO, note indicator reading, change +/- directions and make small step movements until the indicator responds, note difference in the readings between the indicator and DR0, tighten gib - if the difference increases, loosen gib - until the difference in the readings is at the minimum. On my first machine the X and Y is nom 0.0008" and on my new machine nom 0.0005" from memory.

Was the minimum backlash you could get for the X and Y 0.001"? Thats about normal I gather.

Did you check the Ballscrew end thrust bearing preloads also? If they are not at optimal you may be able to reduce the backlash more.

Keen

Hi Keen, sounds like our methods were pretty similar! From looking at my notes, it looks like my backlash (after adjustment, I attempted to add no more than 0.0001" for wear-in) is as follows:

y-axis: 0.0013"
x-axis: 0.0006"
z-axis: 0.0002"

I'm pretty happy with the x and z axes, but the y-axis could use a little TLC. I would work on the bearing preloads, but I don't have the Tormach spanner wrenches.

I think CNCDude lives near the DFW area, maybe I could trade him some beer to borrow his =)

- - - Updated - - -

Hey Wes, Thanks for posting.
I copied this to my notebook and filed it under machine adjustment procedures to compliment the procedure Tormach PP manual details.

Recently I performed gib adjustment in my shop with the temperature around 80+ degrees. Anyone have experience with or how this could effect machine performance in the coming months when my shop cools down and starts hitting low temperatures of 50 deg?

For all of the Sprutcam help you have given me, I have an awful lot of notes to share with you to repay the favors!

[CadRhino]

Thanks so much for posting this detailed write up. Gib adjustment is something I want to know how to do but there isn't much info out there on.
I would love to see tormach do a video on this sometime.
Thanks again
Nathan

Hey Guys, just wanted to report back on a successful gib adjustment. I was having lots of chatter doing circular interpolation at 12 o'clock and 6 o'clock positions. Everything else seemed perfectly fine. Even at 50% of my commanded feed rates (which weren't anything too crazy, 42 IPM on 12L14 I believe), I was still getting chatter rounding the corner. Through some searching online, it seemed like it could be my y-axis gib.

I adjusted the gibs on all three axes and the problem completely went away. In fact, I cranked the DOC up and the feed rate and the entire cut still sounds great! Surface finish seems to be a little bit better as well, although that could just be my eyes playing tricks on me.

If you decide to adjust your gibs, I would definitely recommend using a small ratchet with a screwdriver bit for the back side of the y-axis and a very long flat blade screwdriver for the front. Once I started using the correct tools, it went very quickly.

The process I used:

1. Attach an indicator to the table on a mag base and zero on the spindle nose (I used a 0.0005" DTI I use for gun work, seemed to work well for me here)
2. Zero the DRO for the axis you are working on
3. Jog the table IN THE OTHER DIRECTION in small increments until your DTI reads 0.001" of movement (I used 0.0001 steps in PathPilot)
4. Verify the amount of movement on the DRO (this will be different from the actual movement, the difference is your backlash)
5. IMPORTANT: Create a written table with the following columns (at least this worked well for me): | DTI Reading | DRO Reading | Number of Gib Turns |
6. Loosen the gib quite a bit (Tormach recommends 8 turns). For the y-axis, this means loosen the front screw and tighten the back screw. IMPORTANT: Do not move the table while doing this, you must loosen one side and tighten the other with everything stationary. The right tools for the job REALLY help here (small ratchet / short screwdriver / long screwdriver / etc...)
7. Jog to zero the indicator, zero the DRO, jog THE OTHER DIRECTION until the indicator reads 0.001", record the indicator reading, DRO reading and -8 turns on your chart
8. Tighten the gib 1 full turn
9. Jog to zero the indicator, zero the DRO, jog THE OTHER DIRECTION until the indicator reads 0.001", record the indicator reading, DRO reading and 1 turn on your chart
10. Tighten the gib 1 full turn
11. Jog to zero the indicator, zero the DRO, jog THE OTHER DIRECTION until the indicator reads 0.001", record the indicator reading, DRO reading and 1 turn on your chart
12. .... repeat until you notice your DRO reading increasing for three consecutive turns on the chart
13. Back the gib off until you get back to the starting number (or slightly above it)
14. Jog to zero the indicator, zero the DRO, jog THE OTHER DIRECTION until the indicator reads 0.001", record the indicator reading, DRO reading and 1 turn on your chart
15. If you're happy with the number being "back to the start" - then you're good to go!

Notes:

• I was very frustrated until I slowed down and charted everything to visually see the changes, rather than just sweating in a Texas garage in the summer spinning screws in and out! Once you get to see the backlash on a chart, it helps a lot.
• The whole "other direction" thing is important for measuring backlash. If you have done manual machine work, this will come naturally to you as you will feel "wrong" about jogging both ways during a zeroing operation =)

Hopefully this is helpful to someone else, I wish I had the above list when I started. Also, feel free to correct me anywhere I went wrong, but my machine is running much better now than before and the dreaded y-axis directional change chatter is gone!

[jcposada]

Thanks for posting this!

[widget_maker]

Yes, thanks. I have boat loads of slop and one day soon will give bearing preload and gib adjustment a go.

[MichaelHenry]

Maybe John at NYCNC can address gib adjustment in one of his videos. He is supposed to be doing a maintenance video sometime in the near future.

[wtopace]

He has one out already, but he doesn't show the actual adjustment process itself.

If you guys would like a demo video, I could probably put something rough together when I get some time, for some reason a newborn seems to consume all of it right now =)

https://www.youtube.com/watch?v=paCaT-M07qg

[CadRhino]

That would be awesome. Nothing like seeing it first hand.
Congrats on the new addition.

He has one out already, but he doesn't show the actual adjustment process itself.

If you guys would like a demo video, I could probably put something rough together when I get some time, for some reason a newborn seems to consume all of it right now =)

https://www.youtube.com/watch?v=paCaT-M07qg

[wwendorf]

This whole process is frustrating. I can't believe that Y can only be adjusted to .0013", but that seems to be the best I can get.

I'm fine with my X at .0006", but there HAS to be something to do to get better on Y.

I've been going rounds on this for weeks.

I'm nearly at the point of having a machine rebuilder tear it apart and get it totally tweaked in to the tolerance of the ball screws which are .0003".

Wade

[keen]

Hi - Did I mention on this thread excess backlash in the Y can also be due to a twisted Y slideway due to uneven base stresses - I found this once - it was causing the Y to 'yaw' around and give some really strange errors. Have you blocked the base front up off the stand in the middle to remove any base bias and measure the corner gaps with the bolts loose - if uneven then the base is twisted on clamp down, and the Y Gib cannot be adjusted to fix the twisted fit gaps.

Keen

[wwendorf]

Yes. I shimmed one of the four mounting points to eliminate twist.

Wade

Hi - Did I mention on this thread excess backlash in the Y can also be due to a twisted Y slideway due to uneven base stresses - I found this once - it was causing the Y to 'yaw' around and give some really strange errors. Have you blocked the base front up off the stand in the middle to remove any base bias and measure the corner gaps with the bolts loose - if uneven then the base is twisted on clamp down, and the Y Gib cannot be adjusted to fix the twisted fit gaps.

Keen

[keen]

OK - Another possibility is Y slideway racking. One quick test is do the Y backlash test say a foot apart on the table, using different indicator mounting or indicating positions. If you get different backlash readings it is probably a racking issue and may be due to a taper error eg with the Gib (chips, paint, burr behind it, bowed or incorrect tapered gib etc )

Keen

[wwendorf]

Can you clarifiy what you are saying a bit? Are you asking me to setup 2 indicators and do a backlash test on both of them at the same time?

Or, are you saying to jog the table on X to different positions and do the backlash test?

I get your concept, but am not sure how to accomplish it.

Wade

[keen]

Can you clarifiy what you are saying a bit? Are you asking me to setup 2 indicators and do a backlash test on both of them at the same time?

Or, are you saying to jog the table on X to different positions and do the backlash test?

I get your concept, but am not sure how to accomplish it.

Wade

OK, so you have a Y backlash issue that shows when moving the Y travel in a direction change. Now to see if the Y slideway is racking.... first mount the indicator on the spindle and contact eg central with the Y slide ie in the middle on the table and do the test.

Now move and extend the indicator say a foot to the left and after that the right that will allow tests off centre on the table - if the Y slideway is racking, you will get different results. (you could also mount the indicator on the base at each side).

Keen

Keen

[wwendorf]

Ok. I think I did what you told me to do.

I put the table in the middle of X, then mounted the indicator on the left side of the table. I measured .0015" backlash.

Then I put the indicator on the right hand side of the table, and I got .0020" backlash.

Now what in the heck do I do???

Wade

[keen]

Hi Wade.

I composed a full reply then the website error 522'd on me I lost it ! Damn.

OK so at least part of your problem is Y slideway racking, as you have adjusted the gib to the point where friction increases the backlash, so it is not just a poorly adjusted gib.

Before I rewrite it, just thinking about your test - can you test it also in the middle, and let us know how far offcentre the left and right tests were.

Keen

[wwendorf]

Ok, just tested again:

Right side: .0016"
Middle: .0014"
Left: .0012"

I can't seem to get consistent values.

In addition, this is what I all did.
- Tightened the blocks that the ball nuts into (didn't really tighten them, they were already tight) - This was per Tormach's request
- Adjusted the castle nuts for ballscrew preload.
- creeped up on the gib adjustments per the instructions in this thread.

One word to note, I get absolutely zero wiggle on X. It's nice and tight, and i'm getting next to no backlash whatsoever on that axis right now.

However, if I adjust the gibbs on the Y so I have zero wiggle, I get backlash on Y of .002 to .0029. When I get the backlash down to about .0013" (the best I have been able to achieve) I have about .0005" wiggle on the Y axis.

Wade

[keen]

Hi Wade - OK so the X seems fine.

OK before you pull out the Y gib to check for burrs, paint or metal flecks, and gib errors etc.

One more check will be worthwhile . Put the indicator at least a foot off centre and grab the end of the table and heave ho in both directions of push pull.

You will notice two types of displacement...slop...and flex. Flex will return after letting go, slop will stay.

Let us know what you get.

Keen