Adjusting Lathe Y-Axis "Home" Position

I'm working on an OKUMA LB3000EX-MY. We had a tech come up a couple months ago because our turret had sustained enough small impacts over the last 5 years that the y-axis casting actually moved out of position, and we couldn't get the inclination back into spec. Anyway, he adjusted everything back to where it should be, but never spent the time to reset any of the y-axis turning position or y-axis "home" position information. I have the turn position fixed, but we are finding that we always need to have a -0.015 y axis offset in milling or manually adjust the y values in the program by -0.015 for things like drilling. It is a real pain in the rear. Does anyone know how to adjust the y-axis zero position when milling is enabled? I have scanned the parameter pages but I don't speak Janglish, none of the descriptions appear to relate to the y axis zero position.

Thanks!

i will share the method for the osp300; i am busy now; pls wait :)

loading ....

please, will you share your alignment method ? i have used a centro watch

if i know how you align, i can give more specific advices, and not only the parameters that need to be edited ( sys.para.2 \ base zero offset )



there are several ways to alling a non-Y radial turret :
... in respect to ID holders : thus for turning ( more precise : aligning the ID holder bore coaxial with the spindle )
... in respect to live holders : thus for milling coaxial with the spindle ( more precise : aligning the taper of the holder coaxial with the spindle )
... without using holders, but directly on the turret ( it is kind of a compromise )

there are some things about Y lathes :
... messing with the Y origin will also alter the X origin, so is good to edit those parameters by having a centro ( or something ) clamped in the cnc, during an alignment procedure
... if you wish to align while Y is on, pls be aware that there is a cca0.03 axis slide between G138 & G136, because the Y axis break engages, etc



so ... do you have a centro ? if yes, i may guide you step-by-step :) kindly :)

---

one more thing : the radial play + holders tolerances + sleeves + tools, etc, generally is much more than 0.015; i don't know the conditions over your place, but in a "normal" shop 0.015 is ok :) 0.015 < axis slide caused by the Y axis break engage/disengage

if you put on each turret post a live holder ( align it on Z ), and the centro shows less than 0.015 deviations inside all of them, than you are a lucky guy :) i suggest you do that before continuing, so to know the numbers; if you have only 4 holders, put them at 1 4 7 10

I typically use a coaxial indicator to quickly align tool holders in the lathe. If I need to be more precise I'll use a .00005" indicator on a swivel shank. I like to indicate the tool holders, since it is more accurate to ACTUAL machining conditions vs using the turret itself. In this case, it is a BMT turret, not VDI, so I can't use the turret bores themselves.

What you're saying is that I can't just assign a "y-axis milling position" like I would a "y-axis turning position", to move ALL of the milling tools -0.015 in Y when G138 is enabled?

hi :) don't use swivel shanks for rotating the interapids around a horizontal axis, because you may experience errors, lack of rigidity, etc

the best way to use the interapid horizontaly is by clamping it rigid ( pls check image 1 )

only if you rotate arround a vertical axis, you may use articulations, because their position is affected by gravity in a constant manner ( when you rotate it arround a horizontal axis, the articulations are affected different by gravity, depending on rotation angle )

---

What you're saying is that I can't just assign a "y-axis milling position" like I would a "y-axis turning position", to move ALL of the milling tools -0.015 in Y when G138 is enabled? : i don't know how you assign the Y turning ( from code, from offset, etc ); i just installed the Y_turning_function, and so far i have not played with it; so please, how do you adjsut the Y turning ? ( if you simply edit the VSZOY, this will work for both turning & milling; again, i don't know how you adjust Y for turning, but i guess there is a chance to make it work also for milling : pls share more details )


if it will work, you may leave the encoders origins as they are; well, is good to know how to adjust them, but, until then, pls explain how you adjust Y for turning :)

---

about alignments ( encoders origin ) : if you have time, put more live holders into the VDI turret, equally spaced among the turret, than indicate only one of them ; after that :
... move the turret right ( only on Z; don't move it on X;Y should not be active )
... index the turret to the next holder
... go Z left
... check the values that the centro shows; don't move the turret on X, don't try to reduce the indicator value ( don't setup the centro, but only read what it displays )
... for each post you should have the indicator value for X+ Y+ X- Y- positions ( you may use the C axis to position at 0 90* 180* 270* )
... repeat this with all live holders ( to make things easier, thus to indicate on a cilinder instead of a cone, clamp inside each holder a collet + bar_shank_dia_12mm for example )

* if you wish to indicate on a cilinder, be sure that you have in all live holders same collet + same shank ( don't move the collet&shank between posts, because you may end up by a slight dissanlignment of the holder ); for perfect conditions prepare all live holders before, and after that only indicate them ( if tir on the tool_shank clamped inside the collet is big, than forget about this method, because you simply sum up errors from the taper+collet+shank; stick only to the taper of the live holders )

...... the data deviation will show how "accurate" you are :) there is no point to wish for an alignment value that is lower than the data deviation :) is it ok ? does this make sense ? did i explained it well ?

I completely understand what you're saying, and I have used this technique to align turret angles in VDI machines (to some extent). When I set the y-axis turning position, I first ensure that the turret angle and inclination are properly set, then face off a piece of stock. I measure how far above or below centerline the turning tool holder is, then adjust the y-axis turning parameter as necessary. This has always worked well for me, but I can't seem to find the parameter to accomplish the same thing for milling. Basically, if I were to try to mill a hex on the end of a piece of bar in the spindle, it will always be ~.015 off in Y (negative direction), regardless of which tool pot I put the holder in. It is the same for hole patterns, milled pockets, etc. Does this mean that my ENTIRE Y-Axis is off -0.015 and needs to be adjusted, THEN my y-axis turning position needs to be re-set to center?

Also, THIS is what I use to hold indicators when I tram tool pockets...
Attachment 392834

When I set the y-axis turning position, I first ensure that the turret angle and inclination are properly set : i never setup a VDI turret, but i guess that turret angle & inclination are steady, unless a crash ocurrs; however, only for turning, turret may be a bit missalingned; for drilling, long reaming than a better alignment is required

I measure how far above or below centerline the turning tool holder is : please, how do you measure this ? eye ball ? specific instruments ?

then adjust the y-axis turning parameter as necessary : please, where is this parameter ? i really don't know what parameter you are reffering to ...

on osp300 you may modify the origin for the Y axis ( is near the classical program_z_zero ); i will share a photo tomorrow : you can edit this value from code, so to keep it 0 for turning/G136, and -0.015 for G138

if I were to try to mill a hex on the end of a piece of bar in the spindle, it will always be ~.015 off in Y (negative direction), regardless of which tool pot I put the holder in. It is the same for hole patterns, milled pockets, etc. Does this mean that my ENTIRE Y-Axis is off -0.015 and needs to be adjusted :so, if you mill a circle dia 10, you will get a circle with dia 10, but excentric 0.015, thus tir 0.03 ? whatever the turret post ? if yes, than Y is missaligned :) however, at only 0.015mm that value is small : a normal tool offset / wear corection may be 0.2 ... i don't know, but i feel like you are looking for too much; however, if this behaviour is all arround the turret, than Y needs to be adjusted. At such a small value, it is needed to check the entire turret before, otherwise, you may end up having 0.005excentricity between posts 1-5 and 0.025 excentricity between posts 7-10

there is a thing : is easy to reduce a missalignment from 0.3 to 0.05; going from 0.05 to 0.02 and then to 0.005 may take much more time; again, 0.015 is ok :) and how do you know that the value is 0.015 ? to be sure, you need an instrument that has it's precision pretty high, for example are you sure that the dimension is 0.015±0.003 ?

THEN my y-axis turning position needs to be re-set to center? : if i was in your situation, i would simply leave Y origin as it is for G136, and change it's value from code when G138 is there :) pls, where is this Y parameter that you keep talking about ?

---

look, i have also done alignments below 0.01mm, targeting 3um, etc : such a precision is not for a lathe ... lathes go hard, roughing, milling, shiet happens, and sometimes only high specs may lead to turret missalignment

also materials work, for example of you allign something at 0.01, tomorow it may be 0.005 - 0.03 depending on the force used when aligning, and on the residual force that remains after the alignment; this residual force may lead to soft slides during the night


ps : that thing from the image : use it to indicate a big diameter, like 50-80mm; after that recheck with a centro watch : if same value, than perfect; check with an instrument and recheck with another, especially for important things :)

Also, THIS is what I use to hold indicators when I tram tool pockets... : is not ok to tram tool pockets : in that pocket you insert a reduction + a tool ?

so there is the "play pocket-reduction" + "play reduction-tool", and this value is <>0

ideally is to align the tool :)



where is north ? that way :) something like that ....

ok, I might be confusing myself a bit. It has been awhile since I worked on this machine. I think I need a better description of the y-axis zero offsets. There are two columns (YI , YS) and two rows (ZERO OFFSET, ZERO SHIFT). There is also a parameter for the y-axis turning position which I mentioned before. So zero offset is the y-axis "home" position, right? The zero shift is a temporary alteration to that home position based on individual requirements, right? What are YI and YS?

yup, pretty close : 'zero offset' is the 'program zero' :
... X : the calibration holder is coaxial with the spindle ( it may be an id holder, or, on a vdi turret, i guess it is the turret bore )
... Y : same
... Z : ' normal ' zero, you know ...
... C : i set my cnc so if this origin is 0, than C0 will put one jaw paralel to X axis

you may call it ' home position ', but it may sound a bit confusing : for example :
... on a lathe, a ' home position ' may be considered the ' turret safe position where it indexes ' ( upper right )
... on a mill with atc, it may be the location where the spindle changes tools ( upper left )
... on any cnc, you may declare a position where the tool will retract, so to allow the operator to do something ( this is a ' home position ' )
...... thus, a ' home position ' is somekind of a ' safe, colision proof position ' : it may the index position, or somewhere away from the part; if you wish, you can consider the ' home position ' right where ' program zero ' is, but this is less common; normally 'program zero' <> ' home position '; is just a matter of how you call them :)

The zero shift is a temporary alteration to that home position based on individual requirements, right? : yup : is a temp shift of the ' program zero '; if you reset the cnc, the ' program zero ' = ' zero offset ' will remain, but the ' temp shift ' = ' zero shift ' should vanish

What are YI and YS? : on osp300, at the ' zero offset ' is only one Y, but inside the parameters are both ... i knew something, but i forget it ... however, if i remind it right, those values are linked, so if you change one of them, the 2nd will update automatically



one more thing : that 0.015 excentricity is it aligned with the Y axis ? well, realY is not perpendicular on realX, so that 0.015 exc is it perpendicular on realX ? if it is perpendicular, than you may reduce it by messing with the Y_encoder_origin; if it is not perpendicular, than what you see is a combined effect from both X&Y axis

clamp a centro in the chuck, and indicate into a live_holder or id_holder or turret_bore ( i guess they are all coaxial ): move X&Y until it shows 0 : after that, you need to edit the encoder_origins, so the machine to display X0Y0 ( x0y0 G138 <> x0y0 G136 because the Y axis break slides a bit the axis : when you active Y, G138, the servo starts to hold the axis and the break is disengaged; when you disable Y, G136, break is engaged; break clamp/unclamp may lead to a tiny shift of only few hundreds )

editing the encoders origins may shift your y_turning_parameter with 0.015-0.030, but i guess is ok :) i still have no clue about this y_turn_para


if you turn something, and input into the finish_tool a corection of 0.015, is it too big ? if not, than why would it matter for milling ? kindly :)

on osp300 is not possible to edit Y encoder origin while Y is active; this means that if the centro shows 0 with G138 Y<>0, it is required to go G136 Y0 : this move will input an error into the centro

however, now you should edit the Y encoder in such a way, that when you activate G138Y0, the centro to show 0 : this is tricky :)

each time you go G138-G136-G138 : the break will clamp / unclamp, and this will input more deviation into the centro

---

about that Y turning parameter of yours : are you reffering to the Y offset ? thus that parameter is specific to each tool ? so there are many more such parameters, that are not unique, like the ' zero offset ' ? if so, than this may work also for the miliing operation :)

i thought that there is a parameter labeled ' Y turning origin or something ', but if you don't see a label that specifies ' turning ', you should not assume that the parameter is for turning only; just saying, i don't know what you have there / kindly :)