OKUMA FL OSP-E100L

[tntokuma]

we are having problems machining wheels,the fine finish on face of wheel we seem to be getting transition lines in the same position rewrote program many times. We replaced the ballscrews on the x and z axis and all bearing on the ballscrews. Adjusted the backlash but it seems to make no difference. Our machine has the upgraded linear slides designed for HQ finishes. Can we adjust anything on the z axis servo to help this. And how to do this.
Thanks for all and any replys.

[Algirdas]

if You would search a little information about surface finishing, You will find out, that only way to avoid 'transition marks' is to make the surface without any stop by the same tool.
There are other surface finishing techniques, which, probably, are suitable for wheel face.

[tntokuma]

hi thanks for the reply, we are currently running at 2000rpm with a feed rate of 0.1 mm, I feel it is a servo or gibbs problem due to the fact it happens on the transition while cutting a concave arc it seems to happen when the machine changes direction.

[Algirdas]

so, You don't learn. There is no hope to cover high-finish issues here. It's entire theory. Okuma provides Hi-NURBS, Hi-cut and more other solutions. Cutting feed acceleration/ deceleration rate to cutting feed is crucial here.
Regular surface polish may help at Your case, hope.

[tntokuma]

I am trying to get rid of the major steps as we will be sanding and buffing the face afterwards with robots but atm we are getting a three inch box on the face , I have tried to adjust the backlash and found leaving it at x & z zero that we get the best finish . and yes I have read the surface finish forums but they seem to be about mills (in theory I guess they are kind of the same).

[Algirdas]

we will be sanding and buffing the face afterwards
what's the problem then? Do You have more, than 5µm difference at junction?

[OkumaWiz]

I did see on more than one occasion that the spindle encoder belt had the teeth gone due to bad coolant and slow decay. This caused the spindle rpm to go up and down by belt slippage and since they were feeding in IPR, the feedrate was varying and producing bad surface finishes on wheels. You can eliminate the issue by programming in IPM to test this theory. Most newer machines have gone away from that design though...

If the finish problem occurs only when the axis reverses, backlash is most likely the culprit. They are set at at about .0004" from the factory, but it needs to be adjusted slightly as your ballscrew nut wears. Since you replaced them, did whoever replaced them adjust the backlash properly? Did they test using laser or by what method? Another thing that can affect the finish is if the ballscrews were installed crooked. This can cause a waveform with each 1 revolution of the ballscrew. Typically this happens after a crash hard enough to bend the screw rather than poor installation.

The other thing to try in your programming is to try to do your cutting without the axis reversal...such as always X- and Z- for cuts with no X+ Z+ moves. Then blend them at the axis reversal point.

The biggest thing I've seen help with surface finish is to use natural diamond tools with large nose radii. I have gotten mirror finishes that would be damaged by wiping them off using a shop rag! Unfortunately they are expensive.

Hope these ideas help. If you have any more specifics, let us know- I'm unclear what you mean by " a three inch box on the face".

Best regards,

[chimbote]

Please could someone tell me where I can have the problem on my lathe Okuma Captain L-370 OSP-E100L, I get the following message: 811 ALARM-P FCP I / O LINK COMMUNICATION ERROR 11020000.

thanks

[old timer okuma]

This is fiber optic link alarm. This machine has orange cables from the FCP board to the FUB DR board and the lower machine panel. If you can get a long double connector cable you can connect directly from the FUB to the FCP and also from the FCP to the lower machine panel. the machine should boot up past the alarm, it won't run but will boot further if the shortened loop is good. then you will be able to see if the issue is cable or board.

[chimbote]

thank you very much for answering Old Timer Okuma:
Please could you send me the manual Alarms & Errors List para el OSP-E100L

thanks.

Chimbote

[old timer okuma]

you will need to contact your local Okuma distributor for the alarm book.