Mesa 5i25/7i77 - Lathe spindle – Servo positioning

More
10 Apr 2012 15:44 #19102 by ROG
Hi all

I’m currently evaluating LinuxCNC for a retrofit on a Hardinge HXL CNC lathe I’m in the process of rebuilding.
The GE control, DC motors, DC spindle motor, resolvers / encoders all stripped.

The general nuts and bolts of the control / hardware are at present:

A PC running LinuxCNC ver 2.5
1 X Mesa 5i25
1 X (probably 2X) Mesa 7i77
2 X Omron 400W R88M-U40030VA-S1 servos for X and Z axis
2 X Omron 400W R88D-UA12V drives (+/-analogue input)
2 X 2048 count quadrature encoders incorporated into servos / drives
2 X Heidenhain LS403 linear encoders (11 microamp sinusoidal) with 10 micron grating for Z and X
1 X Heidenhain Rod 1080 3600 PPR 1VPP encoder, possibly for spindle indexing positioning
2 X Heidenhain Rod 426 500 PPR TTL encoders for MPG
1 X 5 channel interpolation unit based on ICHaus iC-NQC interpolator ICs (homebrew job)
(www.ichaus.de/NQC)

I’m completely green when it comes to Linux& LinuxCNC.
Presently, I have an old PC, Ver 2.5 installed and Axis setup in lathe mode.
The Omron drives / servos are connected and “working”, but feedback is coming from the built in encoders not the linear encoders, just for the purposes of testing. Homing is being done from the index pulse that the servo encoders put out.

As far as axis is concerned, I can power up, home the axis, jog and turn a chess pawn with the servos spinning away happily.
The big conundrum is the spindle drive.

The 5KW+ DC motor that drives the spindle will be replaced with an AC motor / servo / drive.
Whatever drives the spindle will controlled from the +/- 10v from the 7i77 with encoder feedback.
One way or another, I have to have this lathe doing C axis stuff with live tooling in the turret. I’ve read a few threads on this subject but found nothing conclusive or specifically servo related.
The 3600 line encoder can be interpolated by any factor 8X/16X/32X ... 8000X+ to get the required resolution (input frequency limitations) from my DIY interpolation unit.

If someone could direct me in the general direction, has any thoughts or comments, any input would be very much appreciated.

Roger

Please Log in or Create an account to join the conversation.

More
13 Apr 2013 23:18 #32685 by spangledboy
Correct me if I'm wrong, but isn't lathe C axis machining with live tooling basically A axis machining on a vertical mill - at least on the fundamental level?

If so, then I'd assume that it would be relatively easy to do with LinuxCNC as the mill setup certainly supports the A axis.

Unfortunately I'm pretty new to Linux CNC myself and I'm no expert in lathe CNC, but perhaps that's the direction you need to start looking in....

Very nice project BTW - you're clearly a perfectionist!

Ben

Please Log in or Create an account to join the conversation.

More
15 Apr 2013 22:33 #32752 by ROG
Hi

Sorry for the delayed reply.

The lathe C axis operation has been addressed to some degree here: www.linuxcnc.org/index.php/english/forum...stion?start=12#30946
Although I haven’t had a chance to try it yet.

There are quite a few “issues” to address when trying to get this working.

Yes, fundamentally it would be exactly the same as an A axis on a vertical mill.
So you would end up with a 2 linear and one rotary axis scenario, versus 3 linear and one rotary on a 4 axis mill.

The difference is that you don’t generally ask a rotary table to rotate with a high degree of accuracy for one operation and then ask it to spin upwards of 3000 RPM while sending tens of thousands of encoder pulses every second to the control. It will lose count in spindle mode and, in any case, you will get a positioning error as the lathe spindle will be rotating the encoder for the C axis, which as far as the control is concerned, shouldn’t be going anywhere as it’s an axis, and it hasn’t told it to move.

Thanks for the compliment - The project is taking forever. I’m partially to blame for being too picky.
That and the fact it’s been too cold to work on the lathe for the last few months.
Should be in a position to take some cuts all being well in the next 6 weeks or so.

Please Log in or Create an account to join the conversation.

More
16 Apr 2013 05:49 #32773 by andypugh

The difference is that you don’t generally ask a rotary table to rotate with a high degree of accuracy for one operation and then ask it to spin upwards of 3000 RPM while sending tens of thousands of encoder pulses every second to the control. It will lose count in spindle mode and, in any case, you will get a positioning error as the lathe spindle will be rotating the encoder for the C axis, which as far as the control is concerned, shouldn’t be going anywhere as it’s an axis, and it hasn’t told it to move.


The HAL config that you linked attempts to keep track of this, though a follow-error limit of at least half a turn is probably needed.

Please Log in or Create an account to join the conversation.

Time to create page: 0.096 seconds
Powered by Kunena Forum