Author Topic: G83Extended High Speed Asymmetrical Parametric File Converter  (Read 3806 times)

Offline dave benson

  • CNC Jedi
  • *****
  • Posts: 1511
    • View Profile
G83Extended High Speed Asymmetrical Parametric File Converter
« on: December 09, 2019, 06:49:24 am »
I watched this video the other day  (watched it a few times since).
It's very good despite being a little corny.
Hass video about peck drilling

https://www.youtube.com/watch?v=AM6nVgKjBQo

I recommend you watch it ,as it has all the information on how to set the settings.

I'm sure your settings will be different be conservative with the I value at the start.
They Recommend  3 to 5 times the Diameter of the tool for the first peck.

What they say is a good starting point  for the settings, although I've set mine
down a bit.

For example the I value I use in practice is 1.5 not three as my Mills Spindle speed
is set by pulley change  and is either a bit fast or a bit slow for a 3mm Drill.

Remember Just because you can do state of the art pecking doesn't mean you should
use it everywhere as a default, then Don't it's a folly, as the wrong settings
can pummel the Drill Bits to  an early death Use only when necessary.

In Light of the Video I've implemented all of the features mentioned plus a
couple of extras.
Like a settable “Anti Chip Packing Depth” and “Anti Chip Packing feed Rate”
thus clearing the chips better and slowing the feed the feed at the bottom of the hole at breakthrough if required.

It's runs nice, but has no error checking for the textbox's and no translation so
when I get a bit of time, I'll Finish those things and post it.

I've just left a G83 Extended Test file for you to simulate. If you get the angle
of view just right in Camotics you can see the pecking tool paths.
It's very small file, so you can manually go through the Gcode to see it works.
If you have any more suggestions for any extra features then let me know.

The settings are Fully Parametric Values in Drill Diameters or Ratio's of Drill Diameters.

I    =  Depth of first peck.
J   =  Amount to reduce peck depth each pass.
K  =  Minimum depth of peck.
Q  = Peck Depth.
R =  Position above the R plane (position above the part)
Z =  Hole Depth.
E =  End Feed rate.
D =  Depth at which to apply end Feed rate.
(this saves the tool as it exits the bottom of the hole and keeps the heat down.)
B =  Back off amount at each peck.   


Dave

Offline Garyhlucas

  • CNC Jedi
  • *****
  • Posts: 1422
    • View Profile
Re: G83Extended High Speed Asymmetrical Parametric File Converter
« Reply #1 on: December 09, 2019, 17:56:53 pm »
Pecking is a chip control function. So it is 100% dependent on how the specific drill and the specific material form a chip. So the goal is to peck as little as possible while avoiding a chip wrapping around the drill and staying there to catch others. Pecking is rough on the drill, breaking the chip by pulling it up while cutting, then sliding on the material to get a new chip started. So for me the first peck depth is controlled entirely by how that first chip forms and breaks.

Another reason for just a chip breaking peck instead of a full withdrawal is having a chip fall back into the hole which you crash into on the next peck.
Gary H. Lucas

Have you read my blog?
 http://a-little-business.blogspot.com/

Offline dave benson

  • CNC Jedi
  • *****
  • Posts: 1511
    • View Profile
Re: G83Extended High Speed Asymmetrical Parametric File Converter
« Reply #2 on: December 10, 2019, 11:55:42 am »
Garry I set out to do some practical tests and drilled plenty of holes in wood 40 x 40 mm pine last night,but it was a bit of a failure as I could not get the chips to pack up enough to see if the clearance move to the retract plane worked well of not.

I've made it so that you can select the retract plane value (from the mop)
or retract to a value inside of the hole, so the drill does not clear the stock surface of the
work piece. In practice I liked this a lot.

I also added Rapid to stock Height from the retract plane. To save a bit of time.

I've found some small pieces of 25 x 25 mm Al and mild steel bar and will use this file posted to
do some tests.

I produced these holes at 10 times Deep, the only thing is this is Gummy Al, I think I could have used more feed rate to make the stringers that come off heavier so they break easier rather than get caught up around the Bottom of the collet.

The second file is of the optimized settings after drilling quite a few holes and adjusting them to suit.

Basically upped the feed rate until I could hear the spindle start to slow and backed off a bit
200 % overfeetrate that I started with 25 mm/plunge feed rate to 50 mm/min then back down to 40 mm/min  at the end of the run.
I also backed off the depth of the first peck, and pecked more often.

All in all it worked pretty well in Al, as the holes came out straight (no drill wandering) and the finish is quite clean.
I did some tests with Bright Mild Steel bar  and same Gcode with normal feed rate and that went ok too.

I've done a fair few tests of the Gcode engine and think that it is pretty good now, I did run into sort of an anomaly where if you don't set the Retract Height to 0 in the mop, the default PP will out put
G83 codes instead of G81 even though the Target Depth of the mop is say 0.5mm for a spotting operation meaning the pecking cannot and does not take place,  which making it seem like you want to do a pecking cycle (by looking at the Gcode) when in fact you were wanting to do a normal drilling cycle.
This tripped up the Gcode engine for a while as it wanted to modify these spot drilling operations
even though there was no need to.
In the next couple of day's I'll try it with some of my more complex (many parts,even more mops) files to make sure it plays well with other mops.

Dave

Offline dave benson

  • CNC Jedi
  • *****
  • Posts: 1511
    • View Profile
Re: G83Extended High Speed Asymmetrical Parametric File Converter
« Reply #3 on: December 10, 2019, 11:57:29 am »
Ran out of room for the files
Dave

Offline dave benson

  • CNC Jedi
  • *****
  • Posts: 1511
    • View Profile
Re: G83Extended High Speed Asymmetrical Parametric File Converter
« Reply #4 on: December 11, 2019, 06:10:40 am »
I've completed some more testing,using the same file as previously and added some more Parts and Mops.

The first Part and Dill Mop (Spot Drill) won't be touched as it is a G81.

The second Part and Drill Mop will Be converted into a G83 extended ToolPath and the nice thing is, that it shows up
in camotics. my version of camotics can not do this natively.

The Third Part:  Profiles Mops.
Mop 1 is an Normal profile mop won't be Feed Rate Adjusted.
Mop2 is an Troc mop and will be Feed Rate Adjusted.

Part Four: Pocketing Mops.
Mop1 is a TrocPocket mop and will be Feed Rate Adjusted.   
Mop2 is a Normal Pocket mop  and will be Feed Rate Adjusted also.   

Part Five: Engraving Mop
Mop 1 is an Engraving mop and won't Be Feed Rate Adjusted.

Part Six:  Engraving Mop
Mop 1 is an Engraving mop and mop Will Be Feed Rate Adjusted.

It works fine, want I wanted to know was, if the linking tool paths were correct, and they are.

The .nc file can be ran in Camotics (although not from within CB though) and with the show stock turned off
you can very clearly see what's going to happen and go and regenerate the CB file and run the converter on it again adjusting your parameters to suit. In practice this is a lot quicker to do then say  and a bit of fun as well as being able to see the outcome of your setting changes helps a lot when you're starting out.

The plugin generates a separate file with the extension to the file name like this "Your File Name"+"G83Extended.nc"
so the original is not touched.
It will refuse to convert an already converted file.



Dave

Offline dave benson

  • CNC Jedi
  • *****
  • Posts: 1511
    • View Profile
Re: G83Extended High Speed Asymmetrical Parametric File Converter
« Reply #5 on: December 16, 2019, 21:47:51 pm »
I got around to some more testing today after adding a few more features, there were some issues to work through with multiple mop files, right now you can use global default for use with files that have only one G83 mop, mostly the files I work with have at lest a couple of G83 mops so there is an option to  use individual settings per mop. The only caveat is that you have to set the settings of all of the G83 mops in the file.

Too make it a little easier, I've included a calculator to help work out the settings for each G83 mop.
After you have done a few, the process becomes quite  intuitive  and I have put a few error traps for some of the variables,but not all, as a setting suitable for one mop say (Large diameter drill, shallow Hole) will be quite different from  a (small diameter drill, very deep hole).
The plugin will start up with a generic 3 mm x 30 deep hole programmed into the settings, which serves as an good example because it has been checked on the machine and works well.
Just hit the calc button and using your drawing units, the lengths of your pecks in (mm/imp) will be displayed --> rinse and repeat until the pecking cycle looks good. Then hit convert and a file with the CB.filename + G83extended added to it will be produced  for use on the machine.
 

I've also added a special pecking routine for when the Drill flutes are not long enough to drill a particular hole perhaps a 30 mm deep hole using a drill with  flutes 25 mm long. And am going to try this out later Today
with a new sharp drill  to see what happens.

Dave

Offline dave benson

  • CNC Jedi
  • *****
  • Posts: 1511
    • View Profile
Re: G83Extended High Speed Asymmetrical Parametric File Converter
« Reply #6 on: December 17, 2019, 06:16:11 am »
Well, I tried out the “Drilling holes deeper than the flute length thing”, and at first it didn't look
all that promising see pic 1 .What happened was the Depth of each peck setting was too high
at 1 Drill diameter 4 mm and this filled up the flutes causing binding in a big way.

The Second pic is Showing the Drill and how much stick out there is, it has 20 mm of flutes
and roughly 20 mm of shank projecting from the collet.

The third picture is of almost the bottom of the hole with the flutes well under the deck.

I did modify the Gcode Engine to give it more granularity at the bottom of holes and Halved the pecking step  from 4 to 2 mm, and set the retract to the clearance plane function on, as this helped with the air blast in cleaning the bit, then generated the Gcode and drilled the hole, it went well so I'm going to leave that function in. It's really only an emergency last resort type of thing though
as it takes more time to do as compared to the other pecking cycles.


Dave

Offline dave benson

  • CNC Jedi
  • *****
  • Posts: 1511
    • View Profile
Re: G83Extended High Speed Asymmetrical Parametric File Converter
« Reply #7 on: December 20, 2019, 23:33:59 pm »
As it was (44 Deg C) here yesterday, I got a couple of hours to do some more work tidying up things with the plugin.
I reformatted the output so it's easier to read and added the ability to embed comments in the Gcode
The G83 is fully programmable and all of the settings used to make it are output in the Gcode.
I tried most of the PP's for compatibility Mach3,Linuxcnc, and the Default.

The only one that the Parser choked on was Mach3+cutviewer (which I've never tried before)
and had to add another rule for it to work. I've vague feeling that the Troc mops feed rate adjuster
 might not work with it either and will have to check when I get time.
I also re-arranged the UI and added some more checks.
 
I think I'm pretty happy now, I've made two test\tutorial CB files, one of which must be run
the first time you run the plugin and shows how to load and work with the .nc file CB produces
and generate your first G83Extended canned cycle.

I just have to write a few more notes and take a few more snapshots and I'll post it in the plugins.

 
Dave

Offline dh42

  • Administrator
  • CNC Jedi
  • *****
  • Posts: 6604
    • View Profile
    • Cambam V1.0 French Doc
Re: G83Extended High Speed Asymmetrical Parametric File Converter
« Reply #8 on: December 21, 2019, 07:36:51 am »
Hello

Quote
As it was (44 Deg C) here yesterday, I got a couple of hours to do some more work tidying up things with the plugin.

 :o I'm starting to overheat to half that temperature !  ;D ;D

++
David

Offline dwc

  • CNC Jedi
  • *****
  • Posts: 545
    • View Profile
    • corson.ch
Re: G83Extended High Speed Asymmetrical Parametric File Converter
« Reply #9 on: December 21, 2019, 10:14:01 am »
As it was (44 Deg C) here yesterday
OMG, you will have to start compensating the expansion of your machine!
I hope you are at least far from the fires.
Don

Offline dave benson

  • CNC Jedi
  • *****
  • Posts: 1511
    • View Profile
Re: G83Extended High Speed Asymmetrical Parametric File Converter
« Reply #10 on: December 21, 2019, 23:35:15 pm »
David I spent my early life up, to about 10 years old in a hot dry dessert, my farther was part of the construction of a National Highway
linking the coast to a vast underground mine  in  Mount Isa QLD,this meant that we stayed at the small towns in between the coast and Mt Isa in the outback of Queensland, everything out there including the weather wants to kill you.
 
I now live down south in a coastal 'Farming and Dairy Farming' town 2600 KM away and have not the slightest urge to go back even for a Holiday.

Don I was determined, too at least get some work done for the day and went out to the shop to start, but when I opened the door
I was greeted with a fiery blast of even hotter air from inside the shop and so beat a hasty retreat back to the house.

My spindle preload is settable from the outside and is easy to get at, and I do adjust it on average two or three times a year usually a few weeks into summer or winter proper, and occasionally sometimes during the late summer after a hot day where the temperature can precipitously drop 10 to 20 deg C  by the early evening because our city is on the ocean.

Dave

Offline dave benson

  • CNC Jedi
  • *****
  • Posts: 1511
    • View Profile
Re: G83Extended High Speed Asymmetrical Parametric File Converter
« Reply #11 on: October 02, 2021, 10:09:09 am »
A little while ago someone inquired about a post processor for a woodworking machining center, so I had a quick look at the
machine and manual, it's has a proprietary controller that uses it's own macro language and so would need a PP and a PPT.

While looking through the manual I discovered that the machine doesn't have asymmetrical drilling, surprising considering that
this is the very machine that would be capable of and benefit from them. Maybe it's available as an upgrade over the base price
for your other arm and leg.

Anyway I looked at the drilling plugin again added some  text which indicates when your adjustments to the
variables turn on  Symmetrical\Asymmetrical tool paths.

The other thing was to stop pecking once the slow feed rate at end of hole is active.
I found that in practice that when the slow feed rate at end of hole option was in effect, that there was no benefit in pecking,
while in this part of the cycle.

To test the plugin out  I made a randomised fixture plate. I have some fixture plates that have a rectilinear hole patterns,
but sometimes for odd shape or low profile castings, where clamps would get in the way, I want to bolt them on
to the plate and sometimes find I'm some fraction of a hole out and have to use a clamp anyway.

I sized a bit of 125 x 160 x 12 mm HRS stock, in reality it was 11.78mm and ended up at 11.04 mm after skinning both top
and bottom faces. I did check it a couple of times with a mic.
The end result was although there will be some error in the short axis I could not measure well enough as the mic only
reads .01 mm div so I'm going to call that ok.
The long axis was out (see pics) 0.01 mm over 160 mm or about 4\tenths. This error was avoidable as its indication error
just my shabby indicating.

I made up the CB file with a randomised hole pattern of 126 holes.
The file consists of a spot drill mop and a deep drill mop and a Break edges mop.
The holes are though holes and the work is set up in a  vice on parallels.
The holes will be rigid tapped with a spiral point machine tap.

I've posted the CB file in a folder along with a copy of the .nc file and a copy of the plugin generated drilling and tapping file.
 Along with folder containing some example settings. Remember the first thing you must do is extract the file folder to your
cad\cam folder then open the file and immediately save it to set the file associations to your folder.

You can then generate a .nc file from CB, and run the plugin and load up the .nc file then press update mops and select the
mop to modify from the drop down box.
Make your adjustments to the variables and press calc (rinse and repeat) until your happy with the resulting cycle.

Then press include this mop in output file button and press create file.
You'll then get a new .nc file with the extension drilling or tapping.
This is the file you use on the machine and load up into Camotics but remember
that you can not run Camotics from with in Cambam you must start a new instance
of Camotics and then add the file and press the regenerate button.
edit to change pics where text was fuzzy:







 
« Last Edit: October 02, 2021, 10:25:09 am by dave benson »

Offline dave benson

  • CNC Jedi
  • *****
  • Posts: 1511
    • View Profile
Re: G83Extended High Speed Asymmetrical Parametric File Converter
« Reply #12 on: October 02, 2021, 10:10:16 am »
dll and examples
edit: added pics
« Last Edit: October 02, 2021, 10:27:55 am by dave benson »

Offline Bubba

  • CNC Jedi
  • *****
  • Posts: 3174
    • View Profile
Re: G83Extended High Speed Asymmetrical Parametric File Converter
« Reply #13 on: October 02, 2021, 23:03:27 pm »
Pretty neat. Dave.
My 2¢

Win10 (64 bit, CB(1.0)rc 1(64 bit) Mach3, ESS, G540, Endurance Laser.

Offline dave benson

  • CNC Jedi
  • *****
  • Posts: 1511
    • View Profile
Re: G83Extended High Speed Asymmetrical Parametric File Converter
« Reply #14 on: October 03, 2021, 01:10:02 am »
Thanks bubba
I can't think of any more features to add, so this is it I think.