PLC Update - New Features

[glbreil]

Hello, I have a novice question about updating/merging PLCs to get some new features. I have a custom PLC that is very specific to my machine it does not have any fluff in it to allow it to function for multiple machines. It all works very well, and I when I upgraded to version 5.30 I was content to stay with my PLC because it does all I really need, but once the update was done I started looking through the Hickory Lathe PLC that was in version 5.30 and thought for more of an education than anything else, I might like to try merging my PLC into the new one to get a few of the new things, but what I noticed is that the 5.30 included PLC is much, much more complex than mine, not so much from a function stand point, but because it includes code for everything out there that is imaginable including using a plasma torch.

After studying it for a while it looks to me like it might be easier to pull out a few of the things that I have an interest in like SSV and FRV and add them to my PLC rather than trying to merge my PLC into the 5.30 included PLC.

I went as far as searching out what I think is all of the code for both of those functions and inserting it into a copy of my .src file and once finished it complies ok, although I haven't tried it on the machine.

I guess what I am asking is if I just want a couple features does the method I am describing makes sense, or is there some reason that I would want to definitely add my PLC into the newer one? I may go ahead and try it on the machine and see if it works, just wanted to hear thoughts from someone else before I go any farther.

After separating those two items I decided to look at diagonal jogging, and what I found was code to work on about every format of machine out there, like inverted axis, swapped axis, swapped and inverted axis, and so on.

The question here is to ask if I just have a normal machine, then am I good to just insert the code for the normal machine and leave out all of the other formats? I think all of the extra is just to make it work for any machine, but again just wanted to make sure I understand.

Thanks Gary

[cncsnw]

I am obviously not an unbiased source, since I provided your original PLC program.

I maintain my own library of "stock" PLC programs (mill/lathe, oak/allin1dc/hickory/legacyadd, atc/non-atc, etc.) which I keep as internally consistent as possible. I gave up trying to make them similar to the factory programs some years ago.

Unfortunately, I have not yet had time to incorporate some of the more useful of the newer features -- such as SSV and diagonal jogging -- into my programs yet. Once I do, that would provide an easier template for the task you have been working through.

Because I like shorter, simpler programs, I would definitely recommend adding the desired new features to your existing program, rather than trying to add your Hardinge-specific features to the new Wizard-oriented factory program.

One tool you can use to isolate and identify the changes in a new factory program, is to compare it with an older-version factory program (such as the January 2024 "Centroid_Hickory_Lathe_Standard-r2.src" that was on your control when you first set it up). Then your comparison doesn't get bogged down with quite so many formatting/spelling/style differences.

For what it's worth, jogging on your machine is not "normal", but instead is with the 2nd axis (X) inverted. "Normal" on a lathe is with X+ pointing up, toward the back of the machine. That is typically used on slant-bed turret lathes. Toolroom lathes with a manual toolpost in front, and Hardinge lathes with the turret in front, require that the X axis jog buttons be reversed so that the downward-pointing jog button jogs X+.

[glbreil]

Thanks Marc, I was definitely thinking the same way. With all that extra stuff in the factory PLC it is difficult to even compare them on notepad++, I believe I was able to isolate and move everything for SSV & FRV, so I will probably go ahead and try it on the machine and see if it works.

For the diagonal jogging, on my machine with no more travel than it has it probably isn’t even worth the effort, but the part I was trying to get to, is if it is acceptable or even preferred to move only the code for the orientation that fits the machine rather than all 5 or 6 options in the factory PLC?

To me once you get to that point it looks like completely separate code sections that are not needed if your machine doesn’t fit that orientation?

Another thing I notice is everything now has a section related to external usb, which I think is USB-Bob. Seems like that isn’t necessary either if you have no intention of adding that kind of hardware? I am learning a little, but the extra stuff just adds to the complexity and makes it harder to understand.

Thanks Gary

[cncsnw]

To me once you get to that point it looks like completely separate code sections that are not needed if your machine doesn’t fit that orientation?

For a PLC program that only needs to run on one control and one machine, yes: there is no need to include the code for features you will not be using.

[glbreil]

Well I am happy to say that tested the function of both SSV and FRV after moving it over from the factory 5.30 PLC into mine. and they both worked like they were supposed to. I haven't tackled the Diagonal Jogging yet because it is a little more complicated and and it probably isn't worth it for no more travel than my lathe has, but I who knows I may give it a go after I have some more time to study it.

I do have a few questions about FRV. It is changing from 0 to 100 percent every half second the way I have the timing set and that seems a little extreme. The amount it varies is hard coded in the PLC, and at 0% the chip isn't really breaking it just stopping the cut. I but I am wondering if anyone has tried just going to say 50% so it doesn't vary so much and if that would be sufficient to allow the chip to actually break? I guess the other alternative would be to slow it down by increasing the time? I saw on another post that 500ms was a good starting point, is that pretty much what everyone is using?

[suntravel]

Well I am happy to say that tested the function of both SSV and FRV after moving it over from the factory 5.30 PLC into mine. and they both worked like they were supposed to. I haven't tackled the Diagonal Jogging yet because it is a little more complicated and and it probably isn't worth it for no more travel than my lathe has, but I who knows I may give it a go after I have some more time to study it.

I do have a few questions about FRV. It is changing from 0 to 100 percent every half second the way I have the timing set and that seems a little extreme. The amount it varies is hard coded in the PLC, and at 0% the chip isn't really breaking it just stopping the cut. I but I am wondering if anyone has tried just going to say 50% so it doesn't vary so much and if that would be sufficient to allow the chip to actually break? I guess the other alternative would be to slow it down by increasing the time? I saw on another post that 500ms was a good starting point, is that pretty much what everyone is using?

I am using 150 - 300 ms in most case. The best setting depends on the spindle rpm, higher spindle rpm -> shorter FRV time.

You can change the time with G10 in the g-code for different tools.

Chip breaking is ok with most metals, for Polyamide pecking is better.

Uwe

[lavrgs]

What is the difference between FRV and Pecking?

[suntravel]

What is the difference between FRV and Pecking?

Centroid FRV varies the feedrate from 0-100% in a given time in ms in a sine curve.

Pecking is same feedrate forward and rapid backward, forward to the same position minus clearance, than feedrate again to the defined peck depth.

100% chip breaking with soft material.

Centroid FRV will work on any machine, but is not 100% perfect for chip breaking, to be perfect in this regard it needs to be a function of the servo drive, considering rpm and feed per rev with a more complex calculation to make sure there is negative chipload once per rev.

This is a xxxx$ option on fanuc servo systems, but we are enjoying 80% performance for free on all Centroid CNCs

Uwe

[glbreil]

Thanks I understand I probably just need some to do some testing, but at 500ms it seems pretty jerky and you are actually going 0 - 100 even faster at 150ms and 300ms, does that actually have the effect of slowing down or smoothing it some since it’s doubtful it and accelerate and decelerate that quickly?