[jake hoback]

So I called Tormach on Friday and did a bit of trouble shooting with John in Tech, but I had to check the ball screw BL with a dial test indicator and when i called back I missed them by 2 mins at 5pm So Ill bring it hear for some help and Ill call them Monday as well.

Here is the deal. Went to adjust the gibbs on my older machine (1.5 years) series 3 though. Noticed I was getting excessive backlash from my Y axis. Could not get the gibbs adjusted to a point where I could get any better than .003 I then noticed if I pull or push on the table with about 30lb pressure y+ or Y- I can get the table to move .007-.008 and stay there. (machine is on and stepper powered up so I'm not back feeding it)

I pulled the Y axis gib and still get the same readings.

1. Tried adjusting the Gibbs again, no luck.
2. Pulled the way covers and checked the ball screw nut bolts all where tight
3. Pulled Y axis stepper and coupler (all screws where tight)
4. verified all ways and screws were getting oil

I think the ball screw nut is bad? I can turn the screw by hand and there is slack before you can feel the nut engage the screw. About 1/16th turn before the table will move in either direction.

I'm sad about this, as I have only been running it for 1.5 years now (not long) largest cutter is 1/4". and I'm really nice to the machine, lots of oil and love. Machine has been run on average 3-4 days a week for 4-6hrs a day. I have a 1983 bridgeport anilam CNC retro fit with original ball screws on it that has less than .0004 backlash!!! 30 years of production use at a cereal factory!!!


Anybody have any thoughts on this? ill be intermittent on here tomorrow to check on this thread.
God bless!

[SCzEngrgGroup]

Most likely, the bolts holding the ballnut to the saddle have come loose. Or the bearing retainer or bearing block on the Y ballscrew have come loose. I can assure you the problem is almost certainly NOT a failed ballscrew or nut. You need to remove the bellows and figure out what has loosened up.

Regards,
Ray L.

[philbur]

Have you checked the adjustment/preload of the ball screw mount bearings, inside the motor coupling housing.

It seems that lots of people think that you adjust the backlash by adjusting the gib. Checking backlash when adjusting the gib is a means of setting the correct frictional resistance between the table an the saddle. What it does is reduced lost motion due to excessive frictional drag. It will not get rid of backlash caused by other issues.The primary reason for backlash, especially when it appears suddenly is slack between the back screw mounting/thrust bearings. It could also be a loose ball screw nut mount but I don't think this happens very often. A worn ball screw will develop backlash slowly and will not be the same at both ends of the travel.

Phil

[rowbare]

I am with Phil, my guess is the bearing preload is off. You have measured the play relative to the table so if you measure it relative to the screw, you can narrow things down. If you have a DTI, put the base on the table and indicate on one of the screw threads. The screw is fixed to the table by the ball nut so it should move with the table i.e. you are looking for near zero movement. If that is the case, when you push or pull the table, the bearings are loose.

If you then indicate the motor coupler, you should see the 7 or 8 though movement you measured earlier.

bob

[jake hoback]

Thanks for the help guys! I appreciate it.

I measured the end of the screw with the bellows removed. I got .001 play pulling front to back with the DTI on the main casting. Doesnt seem that the screw is moving much, seems as though its the table moving without the screw.

I checked the ball screw nut retaining bolts and the bolts holding the bracket to the underside of the table, all seemed tight but Ill check them again to verify for sure.

Any one know the exact sizes of the allen keys for the bracket that holds the ball screw nut onto the bottom of the saddle? I cant see them since I haven't pulled the saddle yet. I just went through all my allen keys (half million it seems, lol) just want to make sure I got the right size when I checked the tightness.

[adamvs]

There is one more thing to check: if your X-axis gibs are too loose ( this would be *way* too loose for that kind of backlash ), you would show min backlash on X, but it would create backlash on Y as the table wouldn't move until the gib slop was taken up by a reverse in motion.
If you check the preload on the Y axis bearings, be careful about the washer in between the two jam nuts, you will need to re-use it. Tho Tormach has spares if it gets damaged.

Allen keys are metric. Most likely 4mm, 5mm, or 6mm.
Good luck, would be very interested in what you find.

[jake hoback]

Very good idea man! thanks for that. Ill give that a try tomorrow am.

Was waiting for Tormach to call back all day but no dice, guess Ill have to play phone tag with them tomorrow.

There is one more thing to check: if your X-axis gibs are too loose ( this would be *way* too loose for that kind of backlash ), you would show min backlash on X, but it would create backlash on Y as the table wouldn't move until the gib slop was taken up by a reverse in motion.
If you check the preload on the Y axis bearings, be careful about the washer in between the two jam nuts, you will need to re-use it. Tho Tormach has spares if it gets damaged.

Allen keys are metric. Most likely 4mm, 5mm, or 6mm.
Good luck, would be very interested in what you find.

[mountaindew]

Thanks for the help guys! I appreciate it.

I measured the end of the screw with the bellows removed. I got .001 play pulling front to back with the DTI on the main casting. Doesnt seem that the screw is moving much, seems as though its the table moving without the screw.

I checked the ball screw nut retaining bolts and the bolts holding the bracket to the underside of the table, all seemed tight but Ill check them again to verify for sure.

Any one know the exact sizes of the allen keys for the bracket that holds the ball screw nut onto the bottom of the saddle? I cant see them since I haven't pulled the saddle yet. I just went through all my allen keys (half million it seems, lol) just want to make sure I got the right size when I checked the tightness.

Tormach manual exploded assembly drawing shows this as a 6x26 screw.

Hope this helps
md

[adamvs]

Tormach manual exploded assembly drawing shows this as a 6x26 screw.

So if it's a 6mm cap screw it takes a 5mm allen wrench.

Here's a thought also: REPLACE the chinese black oxide cap screws with real cap screws. The chinese black oxide cap screws *should* be grade 10.9 ( our grade 8 ) or better yet grade 12.9, but most likely they are the equivelent of our grade 5 or worse.

[mountaindew]

So if it's a 6mm cap screw it takes a 5mm allen wrench.

My chart shows it to be a 4 mm hex key

Wish I could help more
Mostly following along trying to understand your problem and where it is.
md

[jake hoback]

OK so I went through yesterday and adjusted the rest of the gibs. Got my backlash down to .0034 on the Y after I adjusted the x axis gibs! that's a bit better now thanks AdamVS!!! Although not even close to what its supposed to be its better than .007!

I got ahold of Tormach and spoke with another guy (forgot his name) after I told him what was going on he recommended I adjust the bearing preload on the screws. I ordered the spanner wrenches for it and Im just waiting on that to see if it will take up the remainder of the slack.

ill keep ya'll updated when i get the wrenches in and get the bearings adjusted. Hoping to get this thing down to at least .0015. I can machine more accurately than the backlash if I'm careful with how I setup parts and approach them, but it makes it much more difficult!

[hayes]

Be sure to check your bearing blocks are properly torqued to the base as well.

[mountaindew]

OK so I went through yesterday and adjusted the rest of the gibs. Got my backlash down to .0034 on the Y after I adjusted the x axis gibs! that's a bit better now thanks AdamVS!!! Although not even close to what its supposed to be its better than .007!

I got ahold of Tormach and spoke with another guy (forgot his name) after I told him what was going on he recommended I adjust the bearing preload on the screws. I ordered the spanner wrenches for it and Im just waiting on that to see if it will take up the remainder of the slack.

ill keep ya'll updated when i get the wrenches in and get the bearings adjusted. Hoping to get this thing down to at least .0015. I can machine more accurately than the backlash if I'm careful with how I setup parts and approach them, but it makes it much more difficult!

Any update on bearing preload adjustment?
md

[jake hoback]

Any update on bearing preload adjustment?
md

Just got the spanners in tonight, hope to get it adjusted on Sunday, when I head back into the shop.

On a side note, these 1100s must need to be broken in for a few hrs then everything adjusted because I went through and adjusted the gibs on my brand-new (2 months 2nd machine) 1100 and I couldn't get any better than .0023 on my Y and .0022 on my X thats using a DTI and adjusting until i have lost moment. Gonna have to adjust the bearings on that one as well kind of disappointed!

[mountaindew]

Just got the spanners in tonight,!

Small spanners wrenches, I found I could not tighten or adjust much using them !
In short I couldn't adjust the bearing preload and change much of anything on backlash anyway and only increased drag in bearings. I spent days messing with this and made no improvement. Might even say I made things worse.
I even tightened to the point I could hear a difference in table motion "not good" stepper lock up was next and still no change in backlash.

Reason for asking above question, wanted to know if you had better methods, skill and or luck.

Also when I purchased my machine I didn't have good enough tools to check it against the inspection sheet that come with it. "Wish I did" What does your newer machine inspection sheet show compared to what your seeing? Might be a conversation topic with tormach.
md

[adamvs]

The ones Tormach supplies, do the job, but they are really hard to work with. I was thinking of cutting some out of 1/4 7075 that fit the diameter of the nuts. It would help. What worked for me was to:
1) take the bearings out and clean them really well with Brakleen, compressed air, then oil them with a very light oil, like gun oil or sewing machine oil.
2) get the preload 'feel' just so on the first nut, and mark the position with a white felt tip pen.
3) take out the bearing, grease them and re-install them. Put the first jam nut back to the mark, then carefully tighten the second against it.

The tabbed washer is really important. I ended buying about 10 of them from tormach for future use. The washer tends to keep the forward nut from moving around when you tighten up against it.
It's a painstaking process any way you look at it, and you have to be in a very patient frame of mind.
Hope that helps.

All my tinkering didn't help my backlash either. I'm beginning to suspect the gib, once loosened doesn't quit seat right when you go to tighten it up. It would be nice if the screws that tighten against the ends of the gib also drew them to the fixed side of the dovetail. I'm toying with the idea of making some that took a hex key, not the annoying slot, and had a small ball bearing that would push the gib against the fixed side. Just some thoughts.

[SCzEngrgGroup]

It would be nice if the screws that tighten against the ends of the gib also drew them to the fixed side of the dovetail. I'm toying with the idea of making some that took a hex key, not the annoying slot, and had a small ball bearing that would push the gib against the fixed side. Just some thoughts.

If your gibs are so loose that you feel they need to be pushed to either side by the adjusting screw, then that is your whole problem. There should be very close to zero clearance on EITHER side of the gib. That's the whole point of the adjustment is to virtually eliminate clearance, leaving only enough (a few ten thousandths) clearance so the axis still moves very freely. If you have measurable clearance, they it's not adjusted anywhere near right.

This is: Center the table n X, then tighten the X gib, so there is NO clearance, and NO movement between the saddle and table. Now grab one end of the table, and push/pull it to the backfront of the machine, and measure how much movement you have. It should be no more than about 0.001". I'm guessing you'll find it's quite a bit more.

One method of adjusting gibs used on commercial machines is to adjust for a specific amount of motion under the above conditions.

One thing to keep in mind: a static backlash measurement is of limited value. The "dynamic" backlash, when the machine is actually running, will typically be less, as the vibration will help the axes move close to the target position. I typically see between 0.0005"-0.001" of static backlash on my machine, but I don't see it in the parts I make. Also, your part accuracy has as just as much to do with how the toolpath is programmed, as with the basic machine accuracy. A bad toolpath will create cr@p parts on a very accurate machine, and vice-versa. CNC doesn't mean the operator doesn't still need to know what he's doing. Things like the number and depth of finish passes, "spring" passes, climb vs conventional, etc. milling ALL make a difference. No machine in this price class is going to turn out perfectly round pockets every time. There are just to many variables at play. Even getting close most of the time requires a very skilled operator, and a certain degree of luck. If you want perfectly round, use a boring head or reamer.

Regards,
Ray L.

[mountaindew]

The ones Tormach supplies, do the job, but they are really hard to work with. I was thinking of cutting some out of 1/4 7075 that fit the diameter of the nuts. It would help. What worked for me was to:
1) take the bearings out and clean them really well with Brakleen, compressed air, then oil them with a very light oil, like gun oil or sewing machine oil.
2) get the preload 'feel' just so on the first nut, and mark the position with a white felt tip pen.
3) take out the bearing, grease them and re-install them. Put the first jam nut back to the mark, then carefully tighten the second against it.

The tabbed washer is really important. I ended buying about 10 of them from tormach for future use. The washer tends to keep the forward nut from moving around when you tighten up against it.
It's a painstaking process any way you look at it, and you have to be in a very patient frame of mind.
Hope that helps.

All my tinkering didn't help my backlash either. I'm beginning to suspect the gib, once loosened doesn't quit seat right when you go to tighten it up. It would be nice if the screws that tighten against the ends of the gib also drew them to the fixed side of the dovetail. I'm toying with the idea of making some that took a hex key, not the annoying slot, and had a small ball bearing that would push the gib against the fixed side. Just some thoughts.

Thanks for the detailed post!
I can still address this "the machine is in parts all over the garage". I think I will try what you mention, if for no other reason to get a better feel for how much preload they have.
I did all my adjustments from the access cover and it lacks much access or a way to judge what your doing because its kind of tight working in there. I didn't damage the tab lock washer much, but could use a couple extras and might order a few. I did ding up the lock nut spanner slots a bit from wrench slippage. Might order a extra one of them also.

I try to address all this with a patient frame of mind. I have been spending time at the drawing board and the machine is not getting as much use. I figured it was great time to do some work on it and not be pressed to get it working at end of day. I can Tinker as you said then research and ask questions then go back and tinker with it some more.

Again thanks for posting what you have done and results. helps a great deal to know what others are doing!
md


Edit: looked at drawings and you have to do all adjustments from the access cover opening anyway duh , no idea what I was thinking above.

[TurboStep]

This is all sounding very familiar. The lost motion on my Series II was out of specification on all 3 axes but particularly high on Y. Adjusting the gibs according to the method described in the manual I was able to reduce the lost motion, but not bring it within specification. Adjusting the bearing pre-load didn't bring any significant improvement either. The side effect of this work was that the Lateral Displacement test (Certificate of inspection, test G12) was now well out of spec due to the fact that the gibs were now much too loose. Tightening the gibs improved the lateral displacement but increased the lost motion! Tormach just maintained that my gibs were too loose but we couldn't find a solution.
I suspected debris between the ways, so when the guarantee ran out I stripped all 3 axes down, flushed out all lubrication channels, cleaned all running surfaces and rebuilt the axes. Surprise, surprise, I was able to bring both the lost motion and lateral displacement into specification! Both X and Y axes have since been at about 0.018mm (0.0007") lost motion, which I'm quite happy with.
While I found significant debris in places that would have no affect on lost motion I didn't find anything obvious on the running surfaces. However, I now believe that it just takes very little, fine debris to cause problems.
Looking back, other than the lateral displacement problem, there were two other symptoms that I believe indicated the presence of debris:
1) The lost motion wasn't constant. Successive measurements gave different results. Then after believing that I had optimized the settings on one day, the results were unacceptable again the day after.
2) Although we should not expect all stepper motor microsteps to be the exactly same size I was seeing 2 or 3 very small step increments (measured statically using a dial indicator) followed by a MUCH larger step increment. My interpretation of this effect is that the ways were "sticking" for 2 or 3 microsteps until the forces were high enough to force a "jump". After rebuilding the axes the steps are not exactly equal, but much more even.

I put off stripping everything down until I was confident that I could rebuild and readjust everything correctly. As it turned out the only hard part was removing the Y axis ballscrew from underneath the carriage without removing the mill from the stand. It can be done, but it was a real pain. If I were ever to buy another mill this is the first thing I would do after removing it from the crate.
Step

[mountaindew]

Turbostep!
Your posts contain excellent information! Most of my concerns and focus has been on the Y axis at this point and what you just described is the exact symptoms I experience. Get everything set and then 2 days later I was getting different results. Making me question my indicators and methods. At this point the way covers are off and I can see excellent lubrication on both sides. After cycling the y axis in its range of motion a number of times checking and verify bearing preload I can also see a discoloration in the lube on the Y gib side of ways indicating to me a very fine debris much as you describe above. Now I need to think about what to do next. Maybe look over drawings and machine to come up with a new plan.

Thanks again for insight!
md