Adjusting Gibs and Angular Contact Bearing on a Tormach! WW189

Tormach Maintenance: How to adjust your x-axis gib & x-axis angular contact bearing scientifically!
Tormach PCNC Maintenance | NYCCNC: bit.ly/2Eky8Av
Checking and setting your gibs periodically is key to getting the most out of your machine, surface finishes, and tool life. We'll introduce a torque tool that can help us get consistent, reliable results!
NYC CNC Publishes weekly content on CNC, CAD, CAM & Manufacturing including the Wednesday Widget CNC project and Fusion Friday Fusion 360 tutorial! 5 Reasons to Use a Fixture Plate on Your CNC Machine: bit.ly/3sNA4uH

[@illustratedlefty213]

I know I’m 6 years late to the party but I just got a hand me down 770 and the x gib is a bit loose. This video is a life saver.

[@matthewlennox2488]

We do this very differently where I work assembling mills. For gibs we indicate the table in whichever direction we are adjusting and with a prybar move the table back and forth and tighten the gibs until we see .0002" movement, basically just tight enough that there is room for a little bit of oil between it and the bearing surface. As for the contact bearing with an indicator on the very end of the ballscrew we tighten the nut until we see about one tenth in endplay when jogging back and forth.

[@rmkscrambler]

Matthew I'd agree. Some servo machines even have a nice feature where you can generate a drag graph after your done setting the Gibbs to show tight spots in travel. The other thing I noticed, which is a no no on a servo machine but probably ok with stepper's is using jog to check movement. Most older servo machines don't really move the set increments they just generalize however if you punch in an MDI move they are dead on.

[@tacitus101010]

That's my weekend plans sorted. Thanks for this John, can't wait to give it a go.

[@adithmart]

I had to do this when I was first programming a lathe. I read that it is as much as a science as it is an art. I decided to read the amps of the motor to make the adjustments. I never thought about sticking. I just learned something, and thank you John!

[@mastert217]

Thanks for the vid John, I really, really can't wait till you get to the Z axis.

[@prodesign8189]

You make things very understandable. Thanks for another great informative vid. Btw, thanks to you for the superglue tape trick and I have to say the work holding class was great. I have since bought two thread mills from Tormach and I am looking to use one this weekend in steel to cut an M12 thread in steel. Nervous, but this is how we learn. Thanks again for the spreadsheet.

[@tedfry236]

That was awesome! way to geek out over getting the table adjusted!

[@beachboardfan9544]

YES! love these machine maintenance vids

[@EZ_shop]

Fantastic video John.

[@DavidCook42]

I learned a TON from this video. Thank you John!

[@ele4853]

Great Job! thanks for sharing!

[@deemstyle]

This was super interesting!

[@drienkm]

Great video! I periodically clean ways & screws and then adjust gibs on my Grizzly, logging the lost motion; usually about 0.002" on all 3. Not sure why I never thought to take the gib out of play first to check the ballnut and bearing preload. I'll try that. I really do need more accuracy on some parts. Thanks!

[@bcbloc02]

I like these hands on style videos. :-)

[@seansmith4886]

Another option to the pulley is a torque tester. We use them all the time to test the pre-load of bearings for aircraft parts. Same Idea but may be more accurate with static holding in a pulley system.

[@johngover1455]

Some machines have a "Drag" measurement feature. This is a test operation that runs the table back and forth while measuring drive current (I believe it is the current, or some other means of measuring load). he results are plotted and saved. Neat way to monitor condition by comparing results over time.

[@SmallShopConcepts]

John, Love the vids! However on this one I think you overlooked backlash as part of the problem in your end readings. To get the lost motion from stiction due to drag you would measure it in one direction with all backlash taken up and preloaded to move in the test direction. This would be the best method to determine actual lost motion from stiction and use that then to determine your actual backlash without stiction being involved. For instance, you had 0.001" movement on that first test move with 0.0011-0.0012" commanded and stated it was preloaded. This would be the more accurate lost motion from stiction reading. When reversing direction, you have both the lost motion of stiction plus lost motion from mechanical backlash in the ball screw. This is a very good method for accurately obtaining a backlash compensation setting because in most machining circumstances, the machine movements will be fast enough to overcome lost motion stiction on a properly adjusted machine which would leave you with just the need for that accurate backlash comp amount to have accurate positioning. Just a thought! Might be a good follow up video comparing the two and seeing what impact a properly adjusted gib has on actual backlash. I have seen too often abnormally high backlash numbers from poorly adjusted gibs. This would help many to solve that issue! Keep up the great vids! Chris

[@NateUE]

Been waiting for this. :D

[@mc-eschx8972]

Maybe I'm an army of one, but whenever I hear, the specific stinger at ~3:50, I think of Sinefeld haha. Thanks for the vids!

[@spkrman15]

Awesome video. To me this had the old Wednesday Widget feel to it. Liking them all though. Keep them coming 👍 Basically the fish scale is the same as a torque wrench on the shaft. The advantage is you can pull a long time and see where there is binding by the fish scales value. Nicely done. Btw how is that machine so clean? 😉👍

[@BenjaminGShaw]

Awesome video. Also, How do you get the dark interface for Fusion 360 as shown in the intro clip?

[@nickdhrones6425]

If the controller is micro stepping, then each ten-thou step does not have the same amount of torque. This could explain what appears to be striction being overcome after a number of steps.

[@OriginalJetForMe]

Is it not possible to take the housing around the coupling off to make those adjustments?

[@ltborg]

Great video John! What made you decide on the torque value you used for the angular bearings? Was that info from Tormach or a bearing manufacturer?

[@seanwolfe9321]

There is very little info on these adjustments for the 440. Being the bearings adjustment is much different, how do these steps apply?

[@mobo9391]

What is the name of the KZbin link or site to find the info to make this pulley widget?

[@eng.zaherelshaar5868]

Is there is any gudie file can we get it..

[@snowgoer540]

Christmas!!!!!!

[@lavachemist]

What about using a dial style torque wrench? I could be wrong, but it seems like it would be even more accurate. I’m thinking that the elasticity of the string and the diameter of the pulley will both affect your measurement.

[@MojoMfg]

Even though I just did this to my 770 I'm about to do it again for "fun" after seeing this.

[@deeremeyer1749]

Nothing better to do with your tools and time?

[@MojoMfg]

I do have better things to do but I would rather make quality parts with my tools and time.

[@deeremeyer1749]

Then buy quality tools you don't have to "adjust" repeatedly "for fun" and can put to work and keep there and trust to do the job you purchased them to do until scheduled, regular maintenance/adjustment is required. Endlessly jacking with machines "for fun" is a really good way to do create problems if they don't already exist. Machines aren't made to be endlessly "fiddled with" and "adjusted" if they require tools and complex procedures to do so. When they are, they can be adjusted by hand. And they're typically even cheaper junk than machines that need "fiddled with" and "adjusted" when "brand new". Nothing made to be "accurate" and "precise" and made to MAKE accurate and precise "parts" should need "adjustment" before its even "broken in" much less used enough to have any "wear". And I'm aware this is NOT "normal routine adjustment" for "new" machines and the "manufacturer" emails a link to THIS "how-to video" to its "new customers" who are sharp enough to double-check the machine before putting it to work and who do "test work" on the machine ASAP after delivery. Meaning it must not be in the operators/owners manual and that manufacturer must have pretty piss-poor "customer service" and "product support" if this "machinist" is their "trainer" on how to "adjust" machines that are supposed to be "right" right out of the box. What does Tormach pay YOU to act like this is all just "normal" and "fun" and how did YOU manage to do it without THIS HOW-TO VIDEO?

[@artmckay6704]

Another approach - other than the string and fish scale - would be to monitor the stepper motor current. For this, since the motor is driven by pulses, you'd need an oscilloscope to monitor the current pulses. Too tight a gib and you'd have excessive current pulses. This would be far more accurate and more meaningful since you know the manufacturers limits for motor current. In fact, I would strongly suggest you do get an oscilloscope. They are very cheap these days and it would show you if you're overloading your motor.

[@jeromevuarand3768]

Isn't the computer taking into account that lost motion and adjusting motor commands accordingly?

[@williamsquires3070]

Hi, John. Quick question: don’t machines like this have a door interlock that prevents the machine from operating while the door is open? Also, I think one of your machines there in the background wants your attention; a white flashy light-on-a-post was blinking, saying “John, feed me!” 🤗

[@sp1nrx]

It's probably the Haas wanting its gibs checked too! 😉 Machines also can get jealous! 😉😉

[@tacitus101010]

The enclosure is an optional extra on Tormach mills. They don't have door sensors, but if they did it would be easy to disable them.

[@Lorgoth117]

This isn't a technique for setting preload that I've ever seen. Most machine manufactures use a stretch dimension, you then use two indicators on each end of your screw to measure that dimension.

[@belair_boy6035]

John, I am having problems with the sound on this video. The intro is fine but your voice sounds like the microphone is in the other room.

[@belair_boy6035]

Not sure if you did anything but the sound is good now.

[@beachboardfan9544]

So why do machines run dovetails rather then profiled linear rails? Just a cost thing or is there advantages to the dovetails that rails dont have?

[@Lorgoth117]

Ways are much much more rigid than linear guides. They are generally not as fast as guides though.

[@beachboardfan9544]

Seems like most modern machines use rails, whats up with that?

[@wknoxfarms]

Its a cost thing from what i understand linear rails are superior. Much like running steppers vs servos. Not knocking the machine they are great just not a full blown vmc, but they dont cost what a vmc does to repair!

[@multiHappyHacker]

Box ways are more rigid, linear are faster. Most CNCs are plenty rigid for most of the work they do with them (though, most mfgrs sell box way machines too), so the linear rails are just fine. They also have like 50 HP motors and can easily weigh 15,000 lbs.

[@urgamecshk]

Price for sure. Being adjustable is an asset though

[@johnpilagonia5238]

The readings you got from the pulley set up are not correct. You must take in to account the leverage advantage you get from the size of the pulley. The reading should be taken with a dial torque wrench directly attached to the shaft being turned for the readings to be accurate for the bearing pre-load. Also new bearings and used bearings will require different torque specs.

[@outkast187]

It doesnt matter as much as you think, but i agree. Plus flex in string. But...the only thing that matters is if its repeatable.

[@gregfeneis609]

He keeps calling the X motor a stepper motor. Is there only a stepper motor and no encoder?

[@robertschmitt9160]

Yes the Tormach is all steppers.

[@jodyolivent8481]

Greg Feneis correct. Tormach uses open loop controls.

[@gregfeneis609]

Then if he eliminates backlash and jams the gibs up and commands 4 thou in steps and only gets 1 or none, that could be step loss, not the lead screw (ball screw?) winding up. Not really consequential to the content, I guess

[@outkast187]

@@gregfeneis609 you hear all the know it alls speaking of step loss, but after 1200hrs on mine, never have i run into it. Possible? Sure... I just machined steel at .4", and was less than .001" off. Ive never adjusted anything. I just put new spindle bearings in (greasing old ones would have been enough, easily). Its funny how many people say oh no...dont buy that, buy a haas. And then they do nothing. My tormach makes $1000 a day pretty easily. I dont use it daily, just when i need to.

[@joashparker8271]

Am I the only one struggling with audio lag?

[@jlippencott1]

Lost motion and backlash are two names for the same thing. Lost motion is the motion lost when changing directions. Backlash is the motion lost when going either way, which is the same amount.

[@Freetheworldnow]

Not sure about your pulley and rope test. How do you know about the torque value versus your pulley OD? This should be explained. The bigger its D. the easyer to pull, therefore offsetting your torque values. Making the test non acceptable.

[@ww55d]

notice how he did everything relative to the same setup. with the same diameter, he measured before and after adjustments, and was able to see the torque change.

[@drienkm]

Pierre Flamand Perfectly acceptable, just scaled😊. To avoid some simple math, he just made the pulley grove about 2" OD. You can use any pully and just multiply the scale values by (R_actual / Preferred_torque_length_unit) to get true torque values. Example: 3.6" dia pully reading 21oz. So R_actual 3.6" / 2 = 1.8", and 1.8" / 1" = 1.8" ... actual torque would be 1.8" x 21oz = 37.8 in-oz of torque.