What's new
What's new

Fanuc G31 Torque Skip, and displaying additional load meters

npolanosky

Cast Iron
Joined
Apr 9, 2016
Location
USA, FL
Hey all,

I have a 2003 KIA SKT21LMS with a Fanuc 0i-TB, and I'd like to start using the subspindle. I have two use-cases in mind. One of those being a conventional slug transfer application (I *think* I understand how to do this), and the other is as a support for long parts.

I don't know whether I'm best off getting a spring loaded center or just leaving some extra stock and chucking on that for the turning of somewhat longer parts, but either way I would like to be able to monitor the servo load so I don't burn up a servo/drive or ruin parts.

This brings me to my problem though- on the servo monitor screen, I can only see X, Z, and Spindle 1. I have a B, Spindle 2, and milling spindle that I would love to be able to keep an eye on. Is there a parameter to enable display of the load on these additional axes?

If I can't see the axis on this screen, does that mean I can't use G31 torque skip with it? I'm tempted to program a slow feed move with a low load set to test this, but I figured I'd be best off asking here first before breaking something.

I'm also not sure how to program the subspindle efficiently; Should I use a macro variable to control the position, assign it a WCS, use absolute coordinates, or something else? I don't have (as far as I know) an M-code for simple subspindle turning like I know some HAAS and Okuma machines have, and the manual doesn't even begin to explain how it's expected to be used. I'd be interested to hear everyone's favorite subspindle workflow.


In any case, I get the feeling that I'll want to write some macros or subprograms to simplify operation.
I'm programming with fusion 360 CAM, but I am comfortable editing my post to make it cooperate with whatever I settle on for workflow. I can also just do a Manual NC passthrough to "hand code" from within CAM, so anything should be fine.
 
I recall that I noticed a little while back that you liked a post I wrote a decent time ago about doing torque skip on a lathe that didn't have built in MTB functionality for it.

Does your machine have this feature built in, or do you intend to do it my way? I ask because while my way works great for me and I've been using it frequently since posting that, I'd strongly recommend you get the green light on that from someone like Bill (angelw) or Vancbiker who know Fanuc inside out and would be better aware of any potential pitfalls.

Regarding your actual question, on the 18i there are parameters to select which axis are displayed on the monitor screen, I assume that the Oi has them too but I wouldn't like to assume that they're the same. Do you have a parameter manual for your control?

The axis are assigned numerically internally (X=1, Z=2 etc.) but the actual order beyond that looks like it is probably arbitrary and defined by the MTB, so I wouldn't like to assume your B is the same axis number as my B.

You should be able to check which axis number is B by doing a simple #001=#500x in mdi, where x is 1 to however many axis' you have, and checking #001 every time to see if it matches the B position.

My B axis is number 6, which corresponds to P6 in the following:

%
O9014 (SUB PUSH ON)


(R B-AXIS RAPID PLANE)
(W B-AXIS PUSH PLANE)
(H PUSH TORQUE - 800)
(F PUSH FEEDRATE MMPM)


G0 B#18 (RAPID TO R PLANE)


G10 L50
N2060 P6 R#11
G11
G98 G31 P99 B#23 F#9 (PUSH ONTO PART)
G10 L50
N2060 P6 R8010
G11


M99
%


Also, FWIW I can confirm that on my lathe this doesn't care if the monitor screen displays the B axis or not.
 
A spring loaded live center is much more trustworthy on the LMS. I dicked around with the torque skip on mine and realized I don't trust anything Kia pushes. I tried that on some 17-4 PH and found the spring loaded live center was the way to go.

I chucked the part then ran the B axis until it alarmed out then jotted down the numbers, backed it off .050 and set that as position in the program. There is something about Kia that just doesn't give me a warm fuzzy feeling when it comes to what that wonderfully translated manual has to say.

no reason to leave extra stock.....

3047.jpg
 
G-Coder- Duly noted with thanks! Which live center did you get? Are you running it in a 5C collet, a 3 jaw chuck, or something else?

My KIA didn't come with a 3-jaw for the subspindle, but I got a Royal 5C collet chuck with it instead. If you can, let me know which 3-jaw you have for it, I'd like to pick one up.

I need to turn the adapter for the 5C chuck in place or something, I get about .007" of runout at the collet face, but the 110mm spindle face itself runs true, so it's in the 110mm to A2-5 adapter. I figure carefully skimming that can't hurt, and I'll just mark orientation so it goes back on the same each time.
 
I somehow glossed over the part where you wanted to use torque skip to use a non-spring loaded centre.

I do that occasionally, but only on larger parts where I don't want to abuse my spring loaded centre, which I consider to be a light duty item for small, precision parts. Also, when I run a dead centre with torque skip, I do it with the spindles synchronised to prevent the centre spinning in the part. It works fine tbh, but once set, the sub is just stationary. It doesn't compensate for expansion of the part the way a spring loaded centre does.

The centre I have is a Royal, p/n 10527, and it is very nice, but it really is a light duty part.

The royal has a scribe line on the spindle to show you where the limit of compression is, so no need to test and back off as g-coder describes.
 
I recall that I noticed a little while back that you liked a post I wrote a decent time ago about doing torque skip on a lathe that didn't have built in MTB functionality for it.

Does your machine have this feature built in, or do you intend to do it my way?
...
Also, FWIW I can confirm that on my lathe this doesn't care if the monitor screen displays the B axis or not.

As far as I know mine doesn't have it built in, so if I go that route I will try it your way (with caution, of course).
Good to know about the monitor screen. I'd still love to have a live readout, but I can go without.

And yeah, I've been thinking about this for some time and doing research on it. PracticalMachinist has really good SEO and a wealth of info, it's easy to find years-old threads with golden nuggets of info (This is also why in a lot of cases I never get when people complain about dredging up an old thread. I'd rather have one good thread with a 10yr lifetime that shows up #1 on the results, than 5 different ones, of which 3 get lost)


I'll do some testing soon with the chuck pressure set low so the stock can slip if anything's not right.
 
G-Coder- Duly noted with thanks! Which live center did you get? Are you running it in a 5C collet, a 3 jaw chuck, or something else?

My KIA didn't come with a 3-jaw for the subspindle, but I got a Royal 5C collet chuck with it instead. If you can, let me know which 3-jaw you have for it, I'd like to pick one up.

I need to turn the adapter for the 5C chuck in place or something, I get about .007" of runout at the collet face, but the 110mm spindle face itself runs true, so it's in the 110mm to A2-5 adapter. I figure carefully skimming that can't hurt, and I'll just mark orientation so it goes back on the same each time.

I use a Dorian long nose heavy duty straight shank in the 5C. I prefer the 5C since I can put a back stop to eliminate it being pushed out of the center.

You say you need to turn the adapter, Did yours not come with a "Set True" 5C chuck?
 
I didn't see a currently made Dorian spring loaded center in a shank size that would fit a 5C, and the Royal one is the same situation- I did some digging and found one from Riten (#17303) that should do the trick though. It's a 1" shank and supposedly good for a 400lb workpiece. I think that will do just fine, and I'll make a hard stop for it in the 5C collet.
For less than $400 I think it'll do just fine.

I also got around to playing with the sub a bit today. I think I've got it more or less figured out, and even though there isn't much clearance between the turret's sheet metal and the sub, there does seem to be a small range of tool lengths that will simultaneously cut to center on the sub without colliding with the subspindle or the rear way covers. I cut down a couple of tool shanks and it looks like we'll be in business shortly.

Next steps are getting a working postprocessor for Fusion that knows what to do with a subspindle. I have half a working post adapted from a HAAS ST20, and half a working post from a Doosan Lynx. The Lynx post handles drilling cycles and live tooling/C Axis work beautifully, in proper fanuc-speak. The HAAS post knows what the heck to do with a subspindle and handles a lot of the basics better (doosan has some weird m-code sequencing going on) but doesn't speak fanuc well when it comes to C axis work.

I'll figure out how to smash em together to get a working one. Supposedly they both know how to use a subspindle, but there will certainly be some differences I'll have to re-engineer or work around. Til then, adding a Manual NC Passthrough function should get the right code where it needs to be.

Still better than FeatureCAM, whom I paid to provide me a working KIA post and sim model (they said they had one) and it turns out it was only half finished and had so many bugs I went back to fusion (Which isn't exactly bug-free itself!)

Oh well, if you want it done right, do it yourself :)
 
If you need a post for Featurecam send me a PM. Got ya covered on the SKT-LMS. Kia's code for the sub sucks so I had to build the post from basically scratch. Its a bit quirky but some quick hand code will weed out the bugs.
 
I use Concentric brand spring loaded centers.
Someone bought them out a few years back. Possibly Royal?
They still use the Concentric name tho.


------------------

Think Snow Eh!
Ox
 
I use Concentric brand spring loaded centers.
Someone bought them out a few years back. Possibly Royal?
They still use the Concentric name tho.


------------------

Think Snow Eh!
Ox
I asked my distributor for a Concentric center and they came back with a Riten brand center of the same part number, so I assume they're the ones who bought em.
 
Getting back to this today, had a fixturing failure last week. I don't think I applied enough pressure with the center or I didn't have it seated deep enough. On top of that, the first threading pass was too deep, I had a chuck in a chuck on the spindle end....it all came out so I changed plans. Proper 5c chuck instead of russian nesting doll-ing it, larger center hole, etc. And first op is done with a 4" stickout (3/4" bar) before pulling the bar out to 10" for the left hand acme thread on this part.

I'll position B by hand this time around, but I think the torque skip is going to work for the future. I just don't have time to write and test it all right now.

I was looking at the sample program posted above and I like the idea of using a macro variable to control B position. It seems obvious, but I just now put two and two together. If I do that I can use standardized code in CAM and then just tweak the position at the machine, instead of hard coding it into the program. I guess I could maybe get Fusion to post out the correct B position and assign it to that variable by pulling the parameter for the model stickout from the chuck or something.

Lots of possibilities. Thanks for the help so far!
 








 
Back
Top