[OkumaWiz]

But if you set the new Z origin, it is some difference from the previous #. This amount should be added to your W for the tailstock axis as well and should follow along with Z origin.

Should be accomplished easily by shifting Zero Set numbers automatically. Even a kitten can do it... :-)

Best regards,

[deadlykitten]

yup, as you said, normal steps are :
...contact & detect material
...calculate some differece & move W origin
...call tailstock

i wanted only to " call tailstock " ... it would be nice

[deadlykitten]

hello

... 1) chuck is open
... 2) i put the bar inside the lathe cilinder
... 3) turret comes in front
... 4) i pull the bar, by hand, out of the cilinder, until it reaches the turret
... 5) i close the chuck
... 6) turret leaves

like this, bar is always out at same distance from the chuck

before step 4, Z effort is 1...2% why after closing the chuck, Z effort raises ? ( 24% in my case )

i don't understand how the effort is monitored when turret is not moving ...

1. NC torque limiter - always ON monitoring for crashes.
2. Load monitor - for detecting worn or broken tools.
3. Torque limiter - used to limit load to a maximum value such as during a part transfer.
4. Torque skip - used to feed to a certain load, and then stop such as when feeding to a part or using a bump stop.

i guess that "NC torque limiter" has senzors not only for motors power consumption ...

ps : if someone is curios why i post this in this thread, is because real steps are :

... 6a) turret not leaving, but "sensless gauging" occurs ( z0.6 ), as described in this thread thx again mr Wizard
... 6b) cut frontal / face at z0.3
... 6c) reset counter from previous bar
... 7) if frontal is not clean, than repeat step 6
... 8) preset number of loops / parts, relative to bar length
... 9) main program start

[OkumaWiz]

before step 4, Z effort is 1...2% why after closing the chuck, Z effort raises ? ( 24% in my case )

i don't understand how the effort is monitored when turret is not moving ...

Your chuck has clearance in it in order to operate. As it clamps, the tolerances are taken up and the part moves slightly in Z. This shows up as motor load as the axis sees the motion. This occurs even on a dead length collet chuck. That is why after clamping, you may want to back off slightly with a W axis to avoid pinching the cuttoff tool when the load is released. Use the load meter to determine how far to back off. Move in W+ until the load decreases to zero and then starts to climb again to about 20%. Then record the amount W moved. Put a G91G94 W.00xx F50 in your program to accomplish this after clamping. It will make your groove tool last MUCH longer.

Best regards,

PS> of course not needed if no sub spindle...

[deadlykitten]

Hello and good day to you mr Wizard ! and everybody else

i did not fully understood your answer, so i will reformulate the question, to eliminate doubts

... image 1 ) chuck is open, bar is inside ( i can slide / move the bar by hand )
... image 2 ) turret comes in front the chuck, and i move the bar to "bump" on the turret

at this moment Z_effort is 0 - 2 %

after i close the chuck, material will be pressed on the turret; thus there is a tension Z_effort will be ~25%

how is that tension sensed by the machine ? especially when turret is static ?

far as i know, load monitor monitors "electrical current" before the motor, thus monitoring will work only if motor is active

this behaviour that i don't understand is based on monitoring after the motor; how ?

how this load monitor works ?

if monitoring works when motor is active, as well as when motor is passive, than monitoring should be after the motor ? kindly !

[deadlykitten]

Your chuck has clearance in it in order to operate. As it clamps, the tolerances are taken up and the part moves slightly in Z. This shows up as motor load as the axis sees the motion

how does the "axis sees the motion" ? where are the eyes ?

That is why after clamping, you may want to back off slightly with a W axis to avoid pinching the cuttoff tool when the load is released. Use the load meter to determine how far to back off. Move in W+ until the load decreases to zero and then starts to climb again to about 20%. Then record the amount W moved.

as you also said, this is about part transfer between spindles ? well, i don't have 2nd spindle, but is good to know

Put a G91G94 W.00xx F50 in your program to accomplish this after clamping

what means W.00xx ? i guess is inches ? let's see : 0.12mm / 25.4inch-mm = 0.0047 got it

Best regards

you too

[deadlykitten]

That is why after clamping, you may want to back off slightly with a W axis to avoid pinching the cuttoff tool when the load is released. Use the load meter to determine how far to back off. Move in W+ until the load decreases to zero and then starts to climb again to about 20%. Then record the amount W moved. Put a G91G94 W.00xx F50 in your program to accomplish this after clamping. It will make your groove tool last MUCH longer.

i think i get it what is this about :
... turn something in main spindle
... 2nd spindle comes and grabs the part
... bought spindles are rotating syncro
... cutting tool comes in and starts cutting
... when cutting is almost finished, cutting insert may get damaged because of axial tension between spindles

and you, mr Wizard, suggested how to avoid / minimize this, so to protect the tool

i understood this 5 minutes ago hope i got it right

from the 1st moment i read your answer and until now i was wondering why are you talking about W, than cutting tool, than W again, than cutting tool : it was a mess in my head

[deadlykitten]

and you, mr Wizard, suggested how to avoid / minimize this, so to protect the tool

what about cutting, but not all ... only a big groove, so part will still hang in there

... and after, you just break the syncro rotation between spindles ... so you twist the material cutting tool will work as nothing happens

so you may search for the W.00xx, or control the groove depth, so only a small/thin layer remains ...

sounds ok question is if syncro can be breaked when bought spindles are clamping same stuff kindly !

[deadlykitten]

stuff from last thread may work if groove depth is not so big; now i will start writing variants :

v1
... 2nd spindle comes + clamps
... tool cuts a groove, leaving a thin layer at bottom
... tool goes out
... syncro is broken
... 2nd spindle leaves

in this variant all is ok if tool can cut / retreat without being affected by axial tension between spindles

a good sugestion for protecting the tool is "feeding it out" instead of "rapid out" : out_feed = cut_feed x something

v2
... 2nd spindle comes + clamps
... tool cuts a groove, leaving a thin layer at bottom
... tool goes out
... 2nd spindle moves a bit alternative left - right
... syncro is broken
... 2nd spindle leaves

this variant delivers lower material resistance, thus less torque needed when "syncro is broken" ; if coded right, than much required torque from v1 is drastically reduced

v3
... 2nd spindle comes + clamps + moves a bit to right ( as W.00xx from mr Wizard )
... tool cuts a groove, leaving a thin layer at bottom
... tool goes out
... 2nd spindle moves a bit alternative left - right
... syncro is broken
... 2nd spindle leaves

this variant keeps advantages from v2 + protects the tool

v4
... 2nd spindle comes + clamps + moves a bit to right ( as W.00xx from mr Wizard )
... tool cuts
... tool goes out
... 2nd spindle leaves

this is what mr Wizard said

variants v1 .. v4 balance the energy required to cut the part between tool and spindle ( between rotary and linear axis )

if i would have a 2nd spindle, i would try v3

[deadlykitten]

please, i need help finding an answer for this questions :

1) if chuck is open and bar is pulled on the turret ( z effort : 0 % ), than barr will push the turret when chuck is clamped ( z effort : 24 %)

... what mechanism makes the effort to increase ? how does the turret "feel" that the bar is pushing her ?

2) can i continue using the lathe like that ? will the turret get a bit rotated if i repeat clamping the material like that ? kindly !

[deadlykitten]

what mechanism makes the effort to increase ? how does the turret "feel" that the bar is pushing her ?

i think i understood this : turret tends to move right, rotating the screw ball; encoder records motion, and requests more tension to keep the turret where it is

actually, i never understood how okuma's "absolute position encoders" work ...

however, hope i got it right / i am not an electronist, so this is enough for me

ps : and the "following error" as well, i guess, is controlled by this encoders ...

[OkumaWiz]

You are correct in that the encoder is always checking the position.

The control is always trying to maintain 0 diff between commanded and actual position. (RDIFF & RAPA)

The motor is always on and holding position unless e-stop or control off. So as force is applied to an axis,the DIFF becomes greater and the greater the distance, the more the motor will try to reach 0 DIFF. This shows up as motor load or current applied. This can increase until max current is reached at which point the DIFF will continue to grow until about 1mm is reached. That is when the DIFF over alarm will occur.

Make sense now?

[deadlykitten]

You are ... This shows up as motor load or current applied

thank you mr Wizard it is more clear now

each mechanical chain has some mechanisms ( more or less ); here are 2 mechanisms types:
... "control mechanism" : checks movement
... "comand mechanism" : creates movement

after this thread i started understanding "control mechanism" : absolute encoders

if the turret is pushed, than the screw ball tends to rotate; thus the screw ball is rotated :
... normally by the motor
... accidentaly by the turret
i guess that the screw ball must be designed to be rotated by the turret, so to avoid bad things

a few time ago i was curios about "comand mechanism", because increased accuracy will delay the motors, making the cnc respond slower; for example a linear axis is moving smoother among 10mm than among 2 segments of 0.001 length each i was targeting "optimal accuracy" : http://www.cnczone.com/forums/okuma/...precision.html

This can increase until max current is reached at which point the DIFF will continue to grow until about 1mm is reached. That is when the DIFF over alarm will occur

i encountered a "diff over" alarm when feed was to high

also, when a crash occurs, thus "NC torque limiter" drops in , there is :
... 1181 - diff over
... 1098 - motors off

if i would see this error during machining, i guess it would mean "get ready to pay"

Make sense now?

yup, thx again

[deadlykitten]

if chuck is open and bar is pulled on the turret ( z effort : 0 % ), than barr will push the turret when chuck is clamped ( z effort : 24 %)

can i continue using the lathe like that ? will the turret get a bit rotated if i repeat clamping the material like that ? kindly !

i think there is a god in everything :

i needed OD holders so to start a setup; Okuma dealer did not replied, and i am not allowed to buy from the producer

there are some games arround here about Okuma dealer as unique suplier, etc prety messy

so i find alternatives at Kintek and i had them sent by plane prety fast : these ones solved the raising effort

as you can see in attached images, newer holders clamp the tool from side, while older clamp the tool from underneath

on new holders, when the material is pulled by hand, it touches those screws that are used to clamp the tool inside the holder thus contact area is no more intersecting the spindle axis, but is excentric : in this way, when the chuck is closing, the material may rotate a bit ( i think ) and slide ...

z effort after clamping is 0 % if i pull harder, i reach 17% ; on older holders, 24% was reached pretty easy, and i could feel how the tension goes into the turret ... now all is softer, lighter

ps : and i was thinking to craft sleeves with springs, and other nasty solutions pfff

[deadlykitten]

these new holders have increased coolant holes : o8.5

and price is almost the same as the ones that i got from my dealer

in attached image can be seen, with blue, the contact area on older holders ( left ), respectively on newer ones ( right )

i shared also photos with them and the Kintek holders for Okuma

[deadlykitten]

hello all it is almost a year since i started this thread, and with the help from mr Wizard i have written a code that detects part/material frontal, when a barr-puller is used, and a lot of slippery occurs, thus the material is not always pulled the same, even if puller travel is constant; code minimizes errors and checks the goldilocks zone, so to avoid a crash : post #17

meanwhile, i have been writing code that cuts the bar frontal, so to begin clean, when machining many parts from a single bar, and a bar-puller is used

i will share soon a procedure that is meant to do this start-up task simple, safe and fast

---

i have trials over trials, and there is a lot of mess before code gets simpler sometimes is hard for me to focus, because guys arround here are welding, hammers, etc ... but these are the conditions there are moments when i open a code that i wrote a while ago and it takes me too much time to reenter that state of mind that i had when i initialy wrote it ... only after that i can modify it

lately i stoped sharing fresh code ... i consider a code good to be shared only after it was running on the cnc for a while

so i try to avoid sharing 2nd quality but only the best that i can deliver kindly !

[deadlykitten]

... chuck has clearance in it in order to operate. As it clamps, the tolerances are taken up ...

hy mr Wizard if i may, how do you verify this clearance ?

i do something like this :
... unclamp the chuck
... with the inside key, i rotate the nut, so to put the jaws in the upper limit
... i put a comparator on the jaw, paralel with jaw_radial_movement ( comparator_value=A )
... i turn the key in the opposite direction, so to create a bit of play ( comparator_value=A-0.1 )
... clamp & unclamp the chuck
... comparator value will be smaller, because of backlash ( comparator_value < A-0.1 )
... now i raise the jaw by hand, and see if i can reach the A-0.1 value
... i adjust the nut inside, until each jaw has a minimal comfortable/admisible play; 0.1 for example, or sliding the jaw by hand feels ok

- at this point, the chuck is free of tension when it is opened
- also there is enough jaw travel left, so the chuck will behave normal
- thus i reduce a bit the jaw travel, so to gain clearance
- if jaw travel is maximal, than there is no clearance and internal tension appears when chuck is unclamped


this aplies only to normal ( OD ) clamping

if there is no play when chuck is open, than it means that the chuck is in tension, also when lathe is powered off this behaviour is similar to when using the chuck for ID clamping

when a chuck is used for :
... OD clamping, it is in stress only at clamp position, and at unclamp position may be free
... ID clamping, it is in stress at bought clamp & unclamp position

achieving no tension at unclamp position when ID clamping is tricky

i recomend always leaving the chuck unclamped, before power off, at the position where jaws are up, and be sure that the chuck is in a free position, thus only the hydraulic cilinder is in tension

kindly !

ps : here is a contraexample for the behaviour that i try to suggest : imagine a classic lathe, which has clamped normal a part ; when you unclamp it, consider that you dont leave the jaws free, but clamp a pipe, thus the chuck remains in tension, until next part is clamped

[deadlykitten]

hello i would like to share the procedure that i use for "senseless gauging" technique, thus using torque skip to detect Z0, when material feed is done by using a bar-puller

such a procedure is subject to the following :
... fast editable for different materials
... fast editable for frontal machining, depending on material frontal condition after cutting ( tilted face after saw / circular / etc )
... safety bounds ( goldilocks zone )
... log
... no extra movements
... existing movements are as short as possible
... "senseless gauging" occurs on knife shank ( specified by 2nd offset ), and cutting ocurs normal, on insert ( specified by 1st offset )

i tried to minimize the number of input variables without compromising functionality

these are the input parameters :

    V1 = 0.3    ( Z position for last frontal cut )
    V2 = 0.4    ( frontal cut width / how thick is the salami slice )
    V3 = 3      ( how many slices is desired to be cut >= 1 )
    V4 = 2      ( number of tries >= 1 ) (*1)
    V5 = 11     ( T - tool )
    V6 = -123   ( X contact )

    V21 = 1900     ( n )
    V22 = 0.15     ( f )
    V23 = 35       ( X start )
    V24 = -2*0.8   ( X  end  ) ( *r )

for the above example, machine will behave like this :
1) safe position
2) will position, using 2nd offset of tool V5, at X=V6 and a specific Z
3) M0 will occur, for the operator to pull the bar on the turret and close the chuck without the need to keep the bar pressed on the turret, thus no (extra) force is needed
4) torque skip will occur
5) if detected Z is ok, than machine will change to 1st offset of tool V5 and cut V3 slices, each slice with width=V2; last slice will be cut at Z=V1
... rpm, feed, X+ and X- for slicing ( frontal cut ) are specified by V21 .. V24
6) safe position
... if after inspection frontal is not clean, than operator may rerun the program without the need to pull the bar on the turret at M0
... in this way, program will perform faster
... program will stop if number of repetitions exceeds V4

---

ok, now lets go deeper into this

program cuts a specified number of salamy slices (V3), with a specified width (V2)

last slice is not at Z0, but at Z=V1, thus an editable value :
... if this program is not used with a bar puller, thus it is used for a single part, than V1 may be 0, and there may be no frontal cut inside the program
... if this program is used with a bar puller, than V1 must retain same amount of material, like the bar puller had been used
...... in this way, bar puller will pull the part at same Z as the final cut of this program it sounds normal

about the repetability :
... this program cuts 3 times, each time a slice width=0.4; thus a total of 3*0.4=1.2
... this amount should be specified in such a manner that will deliver clean frontal bars, acording to their state ( how rough and tilted are before machining )
... same thing occurs on 2*0.6 or 1*1.2; however, is good to use more tiny slices instead of a big one, so to protect tool nose
... in this case, program uses a repetability input of 2 times ( V4 ); thus second time when program will be runned, it will cut same amount : 3*0.4=1.2; if operator runs the program a 3rd time, no cutting will ocurr, because program will detect that actual Z is too low
... this behaviour is because this program acts like a presseter : thus it runs only once, before the main program that loops to cut the parts; this program runs on an absolute zero, and it checks the detect frontal to be within some limits; if limits are ok, than slices will be cut, and Z will be updated, delivering a ready state for the main program

thus this presseter detects part frontal, checks limits and logs data, just in case

---

safe position
come with tool shank in front of chuck
orient chuck & change offset
feed to position
M0 ( operator should pull the material on the turret )
torque skip
log data
check limits and continue only if all ok
change offset & cut slices
safe position

---

so far so good ? lets develop

safe position without broken vector
activate main offset
update X in 2nd offset, so to eliminate the need to manually update X in 2nd offset each time a X offset is changed ( thus a corection occurs )
come with tool shank in front of chuck at Zmaxim+2.5
orient chuck & change offset
feed to position Z=Zmaxim
M0 ( operator should pull the material on the turret )
move to right
torque skip
move to right, so to disengage contact
log data : zero before, z contact, computed zero
check limits and continue only if all ok
change offset & cut slices
safe position without broken vector

[deadlykitten]

lets continue

safe position without broken vector
activate main offset
update X in 2nd offset, so to eliminate the need to manually update X in 2nd offset each time a X offset is changed ( thus a corection occurs )
come with tool shank in front of chuck at Zmaxim+2.5
... Zmaxim=V1 + V2 * V3 * V4
... i dont go directly at Zmaxim, but at +2.5 in front, so to avoid contact when i rerun the program, and during previous session no cut occured
orient chuck & change offset
feed to position Z=Zmaxim
... feed, not rapid, so to achieve a gentle touch in case that material is allready there
M0 ( operator should pull the material on the turret )
move to right
... moving at Zmaxim+5
... +5 so monitoring_to_be_effective; 5 is a minimal recomended value
...... more than 5 : useless motion
...... less than 5 : monitoring may be ineffective
torque skip
move to right, so to disengage contact
... move right with 2.5
log data : zero before, z contact, computed zero
check limits and continue only if all ok
... Zminim<detected_z<Zmaxim
... Zminim=V1 + V2 * V3
change offset & cut slices
... using not a cycle, but optimized G code
safe position without broken vector

[deadlykitten]

the thing :

OS000 ( pulling bars on the turret )

    V1 = 0.3    ( Z position for last frontal cut )
    V2 = 0.4    ( frontal cut width / how thick is the salami slice )
    V3 = 3      ( how many slices is desired to be cut >= 1 )
    V4 = 2      ( number of tries >= 1 ) (*1)
    V5 = 11     ( T - tool )
    V6 = -123   ( X contact )

  ( * ) ( LV01 , LV02 , LV03 : used lower )

  ( IF [ VRSTT NE 0 ] NEND ) ( this remains in paranthesis; this code will never be restarted )

  ( M867 )                   ( no shortening restart sequence )
    G00 X+LVXP-VETFX Z150-VETFZ
    T+V5*101 M66
    VTOFX [ VETLN+20 ] = VTOFX [ VETLN ]              ( 2nd offset )
    V4 = V1 + V2 * V3 * V4                            ( Z contact maxim )
    V7 = V4 + 5                                       ( Z start G22 = Z contact maxim + 5 ) (*2)
    V8 = V1 + V2 * V3                                 ( Z contact minim )
    V9 = 2.5 + VTOFZ [ VETLN ] - VTOFZ [ VETLN + 20 ] ( retreat after torque skip ) (*3)

    CALL OSG
    CALL OZCHK
    CALL OQ000

  ( NEND )

RTS

 ( . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . )

OSG ( senseless gauging )

    G00 X+V6 Z+V4+2.5    ( Z contact maxim + 2.5 )
    M19 T+VETLN+20
    G01 X+VSIOX Z+V4    F2000   G94
    M0
              ( Z+V7+5 ) (*4)
                Z+V7
    G29 PZ=60
    G22 PZ=25   Z-V7    F+100*5 G94 D+V7*2 ( L0 still going ) (*5)
    G28

RTS

 ( . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . )

OZCHK ( origin Z axis check )

    G00 Z=VSIOZ+V9 M18
    CALL OSGL
    IF [ [ [ VSIOZ-V9 GE V8 ] AND [ VSIOZ-V9 LE V4 ] ] EQ 1 ] NJUMP
        NRPT M0 ( out of bounds )
             GOTO NRPT
        NJUMP VSZOZ=VSZOZ+VSIOZ-V9-V8
              T = VETLN ( Z = Z contact minim + 2.5 )
    NEND

RTS

 ( . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . )

OSGL ( senseless gauging log )

    FWRITC contact-1.txt;A

    LV01 = VSIOZ-V9
    LV02 = VSZOZ+LV01-V8

    PUT '    '
    PUT VSZOZ       ( origin before contact )
    PUT '    '
    PUT LV01        ( contact )
    PUT '    '
    PUT LV02        ( origin after contact )

    WRITE C
    CLOSE C

RTS

 ( . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . )

OQ000 ( straight face )

    V21 = 1900     ( n )
    V22 = 0.15     ( f )
    V23 = 35       ( X start )
    V24 = -2*0.8   ( X  end  ) ( *r )

   ( * )

    G97 S=V21 M42 M03 M08 G00 X+V23 Z=VSIOZ M63

    NAGAIN V3=V3-1
               Z+V1+V2*V3
           G01 X+V24 F+V22 G95 
               Z+VSIOZ+0.3
           IF [ V3 EQ 0 ] NBREAK
           G00 X+V23
    GOTO NAGAIN

    NBREAK G00 X+LVXP-VETFX Z150-VETFZ M05 M63 M09

RTS

 ( . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . )

(*1)
( Z contact minim = V1+V2 )
( Z contact maxim - Z contact minim : multiple of salami slice )
(*2)
( 5 is minimal recomended distance, so  torque skip will work )
( Monitoring is invalid   if : [APPROACH] <= [WORK POSITION]+[+OK]+2mm               )
( Monitoring is effective if : [APPROACH] >= [WORK POSITION]+[+OK]+[CENTER HOLE]+5mm )
(*3)
( while 2nd offset is active, turret is moving at a position specific to 1st offset )
(*4)
( eliminates backlash effort )
( to be used if this effort is big and may lead to wrong results )

	   105.555	     2.512	   106.967
	   105.555	     1.752	   106.207
	   105.555	     1.353	   105.808
	   105.555	      1.26	   105.715
	   105.555	     2.502	   106.957
	   105.555	     1.953	   106.408
	   105.555	     1.779	   106.234
	   105.555	     2.259	   106.714
	   105.555	     2.378	   106.833
	   105.555	     2.592	   107.047
	   105.555	     2.514	   106.969
	   105.555	     2.673	   107.128
	   105.555	     2.206	   106.661
	   105.555	       1.9	   106.355
	   105.555	     2.247	   106.702
	   105.555	     2.593	   107.048
	   105.555	     2.687	   107.142
	   105.555	     2.633	   107.088
	   105.555	     2.259	   106.714
	   105.555	     2.273	   106.728
	   105.555	     2.044	   106.499
	   105.555	     1.873	   106.328
	   105.555	     2.647	   107.102
	   105.555	     2.246	   106.701
	   105.555	     2.153	   106.608
	   105.555	     1.833	   106.288
	   105.555	     2.366	   106.821
	   105.555	      2.14	   106.595
	   105.555	     2.206	   106.661
	   105.555	     2.406	   106.861
	   105.555	       2.3	   106.755
	   105.555	     2.273	   106.728
	   105.555	     2.593	   107.048
	   105.555	     2.139	   106.594
	   105.555	     2.046	   106.501
	   105.555	     1.766	   106.221
	   105.555	     2.006	   106.461
	   105.555	     2.687	   107.142
	   105.555	     2.419	   106.874
	   105.555	     2.472	   106.927
	   105.555	     2.526	   106.981
	   105.555	     2.206	   106.661
	   105.555	     1.633	   106.088
	   105.555	     2.166	   106.621