[gunda]

Hi all
Does anyone has any suggestion on how to turn a cam or ellipse (viewed from the XY plane) controlling the X value in accordance to the position of the C axis on an Okuma LB4000/2000EX with P200 control?
Thanks

[budgieW]

Something like this

[ame=http://www.youtube.com/watch?v=Enze_8LlCgc]LB4000EX MY x750 - YouTube[/ame]

[gunda]

Exactly!
I've only seen it on videos but I can't put it to work. How is this programmed?

[Tancuda]

I'm pretty sure you need a software for it, but only my opinion !!

I would love to try this on my LB4000ex too.

[underthetire]

I saw it done on a standard Cadet once. Our old app engineer wrote a macro for it years ago. Needed user task II i believe. Way beyond my programming ability.

[OkumaWiz]

Cams can also be cut using the old G88 threading function. I typically use this to do oil ring grooves, but with the right calculations and RPM it can cut cams too since it syncs axis drives with spindle rotation.

X and Z coordinates can be used on each line and large feedrates to allow for cam diameter variations per revolution. RPM is limited since axis feed per rev is pretty intense, but as long as you stay below the rapid rate, you can do some amazing things.

Best regards,

[fordav11]

I believe Okuma has 'cam turning' as an option?

[OkumaWiz]

Yes, they make cam lathes that have been used to make automotive pistons (they are not round) Those machines are a specialized animal with special slides and turrets and special software that can learn and compensate for following error. VERY high tech and $$$! They made TM-APT cam software that went with it. RPM's were in the thousands and accuracy was in the microns. You could produce a piston skirt with a changing lobe amount and taper at the same time.

But I don 't think this post was about that. Your standard lathe has some "camming" ability using G88.

Best regards,

[gunrights]

Yes, they make cam lathes that have been used to make automotive pistons (they are not round) Those machines are a specialized animal with special slides and turrets and special software that can learn and compensate for following error. VERY high tech and $$$! They made TM-APT cam software that went with it. RPM's were in the thousands and accuracy was in the microns. You could produce a piston skirt with a changing lobe amount and taper at the same time.

But I don 't think this post was about that. Your standard lathe has some "camming" ability using G88.

Best regards,

You are correct about automotive piston. They are not only oval but they are also barrel shaped. The tolerances we generally hold at the highest point of barrel diameter are +- 7 microns. We use specially designed machines because even the specially designed Okuma cannot get us the 23 sec. cycle time that we need.

[gunda]

How exactly is the G88 used to define the exact geometry of the turned part?
I think the LB4000 as shown on the video is using some control of the C, X and Z axis but I didn't find any function to do the job in the manual of P200L. How could it be achieved with a macro?

[budgieW]

Yes XZC is the answer

[MCR-B II]

For cams LU 15 V1 CAM or V8 CAM lathes are suitable.

[gunda]

Yeah, the LU might do the work but I'm trying to achieve the same result as on the video below as I have the LB4000. I'm still trying to figure out how make it

[budgieW]

Try this
DRAW
VZSHZ=-155
CALL OCALL
VZSHZ=0
VZSHZ=-185
VZSHC=180
CALL OCALL
M02

OCALL

T020202
M110
M808
NLP1
ZZ1=0
PTC=0.1
G94 M146
G00 X400 Z0
G138
G0 X400
FF1=95000
XX1=111.5/2
G17
G0 Y0 C0
X=XX1
NLOOP
ZZ1=ZZ1-PTC
M15
G01 C180 X=XX1 Z=ZZ1 F=FF1
ZZ1=ZZ1-PTC
G01 C0 X=XX1-6 Z=ZZ1 F=FF1
IF [ZZ1 GT -13] NLOOP
G00 X400
G0 Y0 C0
G136
G00 X400
M109
RTS

[gunda]

Many thanks for the example budgieW! I just tried it with some minor changes as my machine doesn't support all the features from your example program but it does work perfectly. Thank you again for sharing this!
Here is what I edited as my machine does not have Y axis.

N100 T010101
N104 M808
NLAP1
M110
N108 ZZ1=0
N110 PTC=0.1
N112 G94 M146
N114 G00 X400 Z0
N118 G0 X400
N120 FF1=95000
N122 XX1=111.5/2
N126 G0 C0
N128 X=XX1
N132 ZZ1=ZZ1-PTC
N134 M15
N136 G01 C180 X=XX1 Z=ZZ1 F=FF1
N138 ZZ1=ZZ1-PTC
N140 G01 C0 X=XX1-6 Z=ZZ1 F=FF1
N142 IF [ZZ1 GT -13] GOTO N132
N144 G00 X400
N146 G0 C0
N150 G00 X400
N152 M109
N154 M02

[budgieW]

You may need to adjust the feedrate to get a round part depending on what options your machine has

[gunda]

I'm only concerned that the rotation using C axis control is limited to 200rpm and if I change the parameter to a larger number it may affect accuracy.
I think a very good variety of convex shapes could be turned this way, you probably know better

[budgieW]

I would not change that parameter. I doubt it will actually let you anyway.

I would try to cut it around 80-100 RPM

[MicMeadow]

Try this
DRAW
VZSHZ=-155
CALL OCALL
VZSHZ=0
VZSHZ=-185
VZSHC=180
CALL OCALL
M02

OCALL

T020202
M110
M808
NLP1
ZZ1=0
PTC=0.1
G94 M146
G00 X400 Z0
G138
G0 X400
FF1=95000
XX1=111.5/2
G17
G0 Y0 C0
X=XX1
NLOOP
ZZ1=ZZ1-PTC
M15
G01 C180 X=XX1 Z=ZZ1 F=FF1
ZZ1=ZZ1-PTC
G01 C0 X=XX1-6 Z=ZZ1 F=FF1
IF [ZZ1 GT -13] NLOOP
G00 X400
G0 Y0 C0
G136
G00 X400
M109
RTS

This is a great example! Wouldn't be possible to make this moving Y axis as well? I guess the G1 should be swapped to G3 right? Also, what if I'd have to make eccentric conical turning?

If you could give an example it would be absolutely helpful because I'm facing this problems at the moment.

Thank you very very much for the codes you are sharing!
M.

[deadlykitten]

Cams can also be cut using the old G88 threading function

please mr wizard, will you develop ?

---

hy mic promising as that codes looks, is not ok, because motion speed is close to math speed, exe vs buffer

it will input discontinuities into an allready unsynced motion, as there is needed a custom thing to syncronize C with linear axis, as allready stated

this does not mean it can not work, only that there is room for better : for example, for dense toolpaths, i simply generate linear code variants with different resolution, then test to find the "one"; linear code goes smoother, as there is no costtime associated with the loop

moving Y axis as well

may not be a must, as is possible to turn with the tool a little offcenter, and also such motions are discontinued, and generally inserts may break, so surface is not always nice

also, you can use a turning tool with a round insert, and compute the code accodringly, so to avoid y

thus is possible to put the y, also without it ... issues are not related to y, but others

I'd have to turn some diagonal (cone) still moving eccentrically

just share what you need, sketches, if you wish leave an email / kindly