Synchronised Hobbing

I'd like to setup lathe spindle and rotary table sync for hobbing on the lathe as per AndyPugh's instructions at wiki.linuxcnc.org/cgi-bin/wiki.pl?Hobbing .

I've got a few questions:

1. My hobbing gear ratio is fixed, so I'm happy to hardwire it into the .hal file to avoid having to configure the Pyvcp panel mentioned in Andy's example. What modifications to the Andy's .hal file do I need to make to hardwire it?

2. My current ini file has X Y Z A for my lathe setup. The Pico-Systems Universal Stepper Controller .hal files use A for the spindle encoder, and I'm using X & Z for the two lathe axes. If I specify LATHE = 1 in the .ini file to give me a more helpful display in AXIS, I can see the X, Y and A axes in the AXIS display panel, but not Y. Because of this, I'm thinking of using Y for the rotary table/worm wheel blank axis - I don't need to see anything in AXIS, just have it synced to the spindle behind the scenes. So I guess I need to replace AXIS_3 references in Andy's .hal example with AXIS_1 (my Y axis)?

3. Pico-Systems Universal Stepper Controller has several .hal files - where should I put the .hal stuff from Andy's example .hal file?

4. Does any of the other stuff in Andy's example .hal file conflict with existing entries in the various Pico-Systems .hal files (for thread sync)?

Many thanks,
Richard

llrjt100 wrote: 1. My hobbing gear ratio is fixed, so I'm happy to hardwire it into the .hal file to avoid having to configure the Pyvcp panel mentioned in Andy's example. What modifications to the Andy's .hal file do I need to make to hardwire it?

adding for a 15-tooth gear ought to work.

I'm thinking of using Y for the rotary table/worm wheel blank axis - I don't need to see anything in AXIS, just have it synced to the spindle behind the scenes. So I guess I need to replace AXIS_3 references in Andy's .hal example with AXIS_1 (my Y axis)?

It might be more normal to map it to B or C, but in many ways the gear blank axis isn't an "axis" at all, if you don't want to be able to jog it then it can be a perfect slave, invisible to G-code.

4. Does any of the other stuff in Andy's example .hal file conflict with existing entries in the various Pico-Systems .hal files (for thread sync)?

it's hard to say. I am afraid you will have to read enough about HAL to understand exactly what is going on.

It does help to have a basic understanding of how HAL works when you start to modify configurations.

linuxcnc.org/docs/html/hal/basic_hal.html

As for conflicts between various HAL files I would assume yes.

HAL files are loaded in the order you list them in the ini file.

linuxcnc.org/docs/html/config/ini_config...id_sub_hal_section_a

JT

Thanks guys - after a fair bit of hacking I've got it working - I'm getting a joint following error though for the rotary table if the spindle rpm is around 1000 rpm with 10,000 ppr spindle encoder (USC max theoretical input freq of 250 kHz occurs at 1500 rpm) - I've tried various FERROR and MIN_FERROR values and the joint following errors only really get subdued when the FERROR values are around 1000!

even though this is an old thread, i found the solution for the following error. Maybe someone find it usefull sometimes.
You have to cut the Connection
net zpos-fb stepgen.1.position-fb =>axis.2.motor-pos-fb

Regards Michael

Do you really mean Z?

And if you break any feedback loop, you have no idea if the table is lagging behind the hob. If it is then the gear will not come out right.