Drag-Knife Plugin

[dave benson]

Hi Kelly

I had a look at the file, and the geometry exhibited some of the characteristics of an imported
SVG file ie. Various Z heights. This is more than likely the cause of the error.
I put it through the file cleaner and deleted the extra polylines.
The holes were all over the place so I took the liberty of attaching them to a centre line.
It sims fine. Don't forget to check the hole spacing.
Just as an aside, I would do this on the large format Laser 600cm X 600 cm, I did use a drag Knife
for a couple of years but for ease of use, speed and very good results the laser is a good
addition to have. 

Dave

[Tool-n-Around]

Hi Kelly, I had a look at the file, and the geometry exhibited some of the characteristics of an imported SVG file ie. Various Z heights. This is more than likely the cause of the error. I put it through the file cleaner and deleted the extra polylines.

Thanks Dave,

I did flatten and clean the polylines both before and after applying the DK plug in. Maybe I didn't have File Cleaner settings right? What settings did you use?

To clarify, the error is being thrown when I attempt to execute the drag knife plugin on the mounting hole polylines. Those unaltered mounting hole polylines in my file will execute just fine with a simple engrave MOP before applying the DK plugin, and as I said, the two DK adjusted polylines contained in the "Drag Knife Profiles" layer in my file were generated just fine, and also if I just create a new hole of the same size in a new location and apply the same drag knife parameters, the plugin will generate an adjusted polyline path for that hole just fine. I included the drag knife settings used in image in the previous post, but changing them had no affect....same error.

Have I got this right or did you somehow apply the DK plugin? Looking at the tool paths on the file you posted, they appear to just be the tool paths resulting from applying the engrave MOP to the mounting polylines before they are changed to the drag knife paths by the plug in. If I try to apply the drag knife plugin to the "working layer" in your file I get the same error. 

The holes were all over the place so I took the liberty of attaching them to a centre line. It sims fine. Don't forget to check the hole spacing.

None of the holes are actually centered between the inner and outer profile. In fact, only the three interior holes on the upper and lower runs share the same Y coordinates.....but they are not centered between the profiles. The holes that fall outside the outer profile are in the correct location. I just haven't altered the outer profile to accomodate them because I was first going to cut a piece of paper/cardboard to check fit and alignment of my measurements before I expended the effort. It sure would be nice if the features on these old parts were symmetric, centered, and logically layed out, but they almost never are.

Just as an aside, I would do this on the large format Laser 600cm X 600 cm, I did use a drag Knife for a couple of years but for ease of use, speed and very good results the laser is a good addition to have.

I did consider a laser, in fact it is how many of my present higher volume gaskets are made (not by me), but only mineral paper and a few other materials can be used for those which is a just a so-so material but it burns fairly clean and quickly. My source wont use other/better materials probably for all the same reasons I won't. I could have just hung a laser on my present machine but the problem is they (at least those within the realm of a hobbyist) don't cut many of the better gasket materials very well or at all depending upon material thickness and the reinforcement fiber, and they absolutely stink to high heaven. It's bad enough wearing ear plugs in my shop when the router is running, and health ramifications notwithstanding, working in a shop that perpetually smells like a nasty burn pit is a nonstarter anyway. Ventilating in Iowa Winters is impractcal and marginally effective. Lasers also leave ugly charring and friable debris on the cut path edges which worsens as the material thickness increases and is completely unacceptable as thickness approach ~>.063" and greater.


On the atatched file I just added a randomly placed hole of larger diameter, then converted to polyline, then executed the drag knife plugin.....and all was fine.

I did the same thing with a randomly placed hole of the same .282 diamter thinking maybe there was a conflict with the radius of curvature and the knife offset, but not so, all was fine.

However, if I create that same .282 holewith the center located in the upper left hand corner of the gasket, convert to a polyline, and execute the drag knife plugin, it fails in the same unhandled exception error. ??

Best,
Kelly

[Tool-n-Around]

The Detailed report in the dialogue box for the unhandle exception:

See the end of this message for details on invoking
just-in-time (JIT) debugging instead of this dialog box.

************** Exception Text **************
System.ArithmeticException: Function does not accept floating point Not-a-Number values.
   at System.Math.Sign(Double value)
   at DragKnife.Dragknife.MyMain()
   at DragKnife.CamBamInit.plugin_clicked(Object sender, EventArgs e)
   at System.Windows.Forms.ToolStripItem.RaiseEvent(Object key, EventArgs e)
   at System.Windows.Forms.ToolStripMenuItem.OnClick(EventArgs e)
   at System.Windows.Forms.ToolStripItem.HandleClick(EventArgs e)
   at System.Windows.Forms.ToolStripItem.HandleMouseUp(MouseEventArgs e)
   at System.Windows.Forms.ToolStripItem.FireEventInteractive(EventArgs e, ToolStripItemEventType met)
   at System.Windows.Forms.ToolStripItem.FireEvent(EventArgs e, ToolStripItemEventType met)
   at System.Windows.Forms.ToolStrip.OnMouseUp(MouseEventArgs mea)
   at System.Windows.Forms.ToolStripDropDown.OnMouseUp(MouseEventArgs mea)
   at System.Windows.Forms.Control.WmMouseUp(Message& m, MouseButtons button, Int32 clicks)
   at System.Windows.Forms.Control.WndProc(Message& m)
   at System.Windows.Forms.ScrollableControl.WndProc(Message& m)
   at System.Windows.Forms.ToolStrip.WndProc(Message& m)
   at System.Windows.Forms.ToolStripDropDown.WndProc(Message& m)
   at System.Windows.Forms.Control.ControlNativeWindow.OnMessage(Message& m)
   at System.Windows.Forms.Control.ControlNativeWindow.WndProc(Message& m)
   at System.Windows.Forms.NativeWindow.Callback(IntPtr hWnd, Int32 msg, IntPtr wparam, IntPtr lparam)

Best,
Kelly

[Tool-n-Around]

Here I added a couple randomly placed holes, converted them to polylines and executed the drag knife plugin and all was well. The second circle is the same diamter at .282 so there doesn't seem to be a problem/conflict with the DK plugin as far as the radius of curvature and drag knife offset.

When I created another .282 circle in the upper left hand location of the gasket, converted to polyline, and executed the DK plugin the plugin fails to execute for same unhandled exception. ?

Best,
Kelly

[dave benson]

Kelly

I used the 32 bit version as I was on my Laptop and have issues with the 64 bit version
running too slow.
The file runs well with this version.
I opened the file with my PC running the 64 bit version and it failed too (on the same geometry)
What I did was open the file and entered a value in the 'dent' textbox and the plugin failed even before clicking
OK so this value is not being handled by the plugin properly.


With the Laser I mostly do Brown paper gaskets (I buy it by the roll in various thicknesses)
for industrial gearbox's and for this purpose it great, my laser is open frame rather than the enclosed
type like the K40 and the like style lasers where they have filters, because of this I never cut neoprene, when cutting
materials like fibre re-enforced there's no substitute for power.
my machine has 4 watt optical power and I would not even attempt to cut it, a sure sign of this is the cut edge will be
burned  and a little ragged .

32 Bit plugin creation date = Monday 30 March 2020 11:00:42 AM  works well.

64 Bit version creation date = Tuesday 8 February 2022 6:45:55 PM  <--- doesn't work.
edit to add year for  creation date 64 bit version.

Dave

[EddyCurrent]

Kelly,

Have a go with the attached version 7

[Tool-n-Around]

Kelly, Have a go with the attached version 7

Thanks a bunch Eddy. Dont know what you did but that seems to have fixed it. Now I'm off to make some test cuts. I'll report back!

FWIW, on the previous version, if I created a circle of any size near the origin, it worked. If I moved that circle a diameter or so away from the origin in any direction it would fail in the same mode.

Best,
Kelly

[Tool-n-Around]

Kelly I used the 32 bit version as I was on my Laptop and have issues with the 64 bit version running too slow. The file runs well with this version. I opened the file with my PC running the 64 bit version and it failed too (on the same geometry) What I did was open the file and entered a value in the 'dent' textbox and the plugin failed even before clicking OK so this value is not being handled by the plugin properly.

Thanks Dave, as always much appreciated. Eddys revision seems to have resolved the issue. Happy Day!

Best,
Kelly

[Tool-n-Around]

I’ll post this here in case there are any drag knife users here. I’ve still have some learning and work to do.

I made a vacuum hold down table from MDF and sealed it. I tried using ¼ MDF for a spoil board because all the YouTube builds said it was porous enough to use but mine was not. It had a very weak hold even after I sanded each side and sealed the edges. Maybe I just didn’t get deep enough through the resin surface, but I had forethought to lay out the grid on my vacuum table to accommodate peg board on 1” grid center. I can get 2 spoil boards out of one $8 piece peg. It’s melamine on one side so seals well against the gasket and is fabric like on the other side which seems to distribute vacuum from the through holes very well. This holds sheet stock to be cut very well, but the peg board seems to vary about .010” in thickness. My router table is ~.005” of flat where the vacuum table is mounted.

I cut some paper and cardboard test samples and then moved onto cut a small gaskets from some 1/32” and 1/16” scrap gasket material. The material is a Kevlar/Aramid fiber reinforced Buna blend. In general, as long as the radius of curvature is not too tight I get good results. I need to cut ~.28” diameter holes in most of my gaskets.

The settings I used for the drag knife plugin were:

Knife Tip Offset: .093 (Measured and set by dial caliper on the drag knife)
Cut Depth: .042
Swivel Angle: 5
Dent Length: .01
Retract Enable: On
Retract Angle: 40
Retract Depth: .015
Polyline Direction: CCW

The above was for 1/32”t material and the only thing I changed for the 1/16”t material was the Cut Depth to .093 because and retract depth to .030, plus .010” overcut into the waste board, plus .20 of additional clearance to insure the drag knife holder did not drag on the stock due to varying thickness of the peg board and the router table surface. 

Initially, the holes cut significantly undersize but the other features were accurate. I had to increase the Knife Tip Offset setting to .175” to get the correct hole size. Not sure why? Perhaps it’s error because the actual KTO of .093” is not small compared to the .14” radius of curvature of the hole? Might that be the case? This could be reduced for the 1/32”t stock but not sure it will make a great difference.

Attached are pictures of my set up and the sample cuts. The 1/32”t pieces are ok. I’d like to do better on the swivel marks. The holes on the 1/16”t stock are not acceptable. They were cut in two depth increments. I could always drill/mill them, but it would require a tool change and if I had a lot of parts/holes nested in a sheet, I may lost vacuum and holding force.

Any thoughts on improving the holes or cause of the dimensional inaccuracy?

Best,
Kelly

[dave benson]

Kelly

I battled with those little tear outs especially in thicker materials and smaller holes.
So I set up an experiment watched the cut, and it turned out that the knife was not
orientated normal to the cut, ie. The knife could have any angle and wasn't clocked to the
impending motion around the cut. Industrial cutting machines can clock the knife angle.

What I tried then was a profile mop with the appropriate roughing clearance (to centre the knife)
and used a spiral lead in, this worked well as the knife was gently orientated into the cut
rather than simply plunging to full depth and then trying to orientate the blade possibly tearing out
small pieces of material.

Dave

[Tool-n-Around]

I battled with those little tear outs especially in thicker materials and smaller holes. So I set up an experiment watched the cut, and it turned out that the knife was not orientated normal to the cut, ie. The knife could have any angle and wasn't clocked to the impending motion around the cut. Industrial cutting machines can clock the knife angle.

Were you actually cutting stock or simulating? Either way, thanks for going the extra mile. In my sample cuts I did notice the little swivel move at the end of a cut didn't always get the knife reoriented properly, especially on thicker material. I played around with retract depth but wasn't much help.

What I tried then was a profile mop with the appropriate roughing clearance (to centre the knife) and used a spiral lead in, this worked well as the knife was gently orientated into the cut rather than simply plunging to full depth and then trying to orientate the blade possibly tearing out small pieces of material.

I actually had a fleeting thought about the profile MOP too, specifically for lead in function but didn't try it. I have a couple questions about the method you describe above.

1. I presume the knife still had offset. Is that so? If so, what was it?
2. Must have been a fairly shallow lead in angle. What was it?
3. Appropriate roughing clearance? How is that determined? Are you using the unaltered toolpath for the shape or the one generated by the drag knife plugin?

I also think using a smaller tipped blade possibly along with smaller offset could be helpful, because the various sources suggest drag knifes can cut radii comparable to but no smaller than the knife tip offset, so smaller would seem better for small holes?

Those utility blades oriented at 45 degrees have a lot of (straight edge) blade engagement/width in contact with the stock and I could see it was causing some tearing especially on the smaller holes and thicker materials,

I bought my drag knife because it was inexpensive and I already had to make the vacuum table. The popular hobby knifes are 1:1 with offset and depth of cut, but it doesn't have to be that way, and I'm not opposed to making what I need. It would be very easy to make a simple holder to test with a smaller blade.

The alure of being able to cut any gasket, or number of gaskets on a sheet, complete, without a tool change, is quite appealing to me. Plus, cutting and loading a sheet on the vacuum table is very fast and the cut cycle times are very short!......I'll get there.

Thanks again Dave.

Best,
Kelly

[dave benson]

Hi Kelly

Yes I was cutting gaskets for industrial gearbox's for about 2 years with a home built drag knife and then a much smaller one
bought of ebay to suit a Roland vinyl plotter, early on in my career as an apprentice we did many by hand by laying the gasket
material over the gasket surface and gently tapping around the periphery with the flat face of a ball peen hammer
and using the ball end to do the holes.

Using the the drag knife was a huge plus over this method although it did have issues with too, with
thick or thin materials. Any material as thin as baking paper would often pucker up and tear out, thicker
materials exhibit the same thing you are seeing.

I presume the knife still had offset. Is that so? If so, what was it?
2. Must have been a fairly shallow lead in angle. What was it?
3. Appropriate roughing clearance? How is that determined? Are you using the unaltered toolpath for the shape or the one generated by the drag knife plugin?

With the home built one, I used a repurposed dremel size chuck and 1/8 drill rod ground and sharpened
(I put a fairly shallow angle on the blade 22.5 deg) 
about half the angle of the bought one this helped to stop the material lifting up as the material was being cut.
The tool steel was much more robust than the carbide blades than came with the vinyl plotter
blades which were quite delicate.


I would show it to you but had a shop fire on the 23 dec and the demolition crew are coming around
today to knock the rest of the building down.

The rouging clearance is half the blade thickness. I only used the profile mop on holes and
used the original geometry for this. Incidentally you can also use a drilling mop with spiral milling set for holes.
Making the depth increment smaller lessens the angle of attack.

I found in practice that less offset was better, you only need enough offset to orientate the blade
against the friction of the blade holder.
For example sealed bearings need more offset, shielded bearings less and open bearings even less.

Dave

[Tool-n-Around]

Thanks again Dave.

That's all pretty much as I suspected and had the same experience as you related above.

Since the last post,as a quick and dirty test, I just clamped a small razor knife balde in the slot of an ER16 collet and used the sprindle (router) bearings. It initially cut one great hole, then the fine tip on the blade broke, the blade was too flexible with much stick out, and the spindle bearings had too much drag to keep the knife properly positioned.

I'll make a small holder and grind my own blades......and I have an inexaustable supply of small tool steel and carbide. I'll see if I can cut the entire gasket with it, but if it requires a tool change to a larger knife so be it. I know I can cut those holes very nicely with a small end mill, and probably the hole gasket profle as well for that matter, but they expose the hole and with multiples cause vacuum loss on my vac platten. The knife cut holes seem to (mostly) stay in place.

.......I would show it to you but had a shop fire on the 23 dec and the demolition crew are coming around today to knock the rest of the building down.

That's terrible. So sorry to hear this. I hope you can recover quickly and build back better than before.

Best,
Kelly

[Tool-n-Around]

Hi Everyone,

Me again. I’ve been using the drag knife plugin with good results but still learning and refining. I re-read this entire thread again and have a few questions. I’ll try to reference other posts in this thread that I think are relevant to my questions.

I’m cutting automotive gasket stock with thicknesses that range from .015”-.063”.  I was having trouble cutting smaller holes (~<.25”, but would like to cut as small as .125”), particularly on the thicker materials. Because of the material thickness, I am using the retract feature. My knives are fixed in their holders and not spring loaded.

Part of the problem with the smaller holes was the style and size of knife I was using. Dave Benson suggested I look at the small vinyl cutting knives. I purchased some of the “deep cut” variety and with their small offsets, made knife holders, and was able to cut the smaller holes with the paths generated by the drag knife plug in. I also was successful using the Profile MOP to do same as described by Dave earlier in the thread.

I realize I could mill or drill the small holes, but they collectively make for vacuum leak and loss of holding on the vacuum table, and optimally, I’d like to cut a large number of features and nested gaskets without a tool change. The use of a smaller knife and offset accomplished this for me.

My question is, how do I get the plug in to generate the knife repositioning move on the waste side of the material? On the attached file, they always land on the gasket side of the material not the waste stock.

If I could get the repositioning move to land on the waste side of the stock, I could use greater knife engagement in the stock to get more reliable lead out knife positioning without scarring the gasket.

Per post/reply #28 on page 2 of this thread, I thought the answer was through combination of polyline direction and starting point. But, on the attached file, if I change the polyline direction in the DK plugin dialogue box, it changes the direction of cut, but the lead out move still lands on the gasket not the waste stock side. Changing the direction of the original polyline used by the plugin to generate DK profiles has no affect, I presume because the plugin overrides direction per that selected in the dialogue box as remarked in post/reply #78 on page 6 of this thread.

I don’t know how to change the starting point. I see on the engrave MOP I can enter a new starting point, but hasn’t the DK plugin already pre-determined the required starting/ending point with the placement of the lead-out knife positioning move? Also, because the DK plugin generated tool paths are not closed, the direction is stated as unknown and is not able to be changed.

As aside, on the attached file, if I examine the DK profile polylines, I expected to see a different Z-heights at the lead out moves driven by the retract depth and angles selected, and the engrave MOP would follow accordingly, but all Z coordinates are at the depth of cut. Why?

I am using the latest revision of the plugin per the date of this post.

Here is a YT video I made of my learnings thus far using the drag knife.

https://youtu.be/FxsfDxR8CD4?si=tdivA4LVzXlxKlGZ

Best,
Kelly

[lloydsp]

Kelly,
I cannot respond directly to drag-knife use, because I've never done it.  However, I have made clean, unburred holes in gasket material with my CNC.  You do 'drill' them, at a very low rpm, using a hollow hole punch, not a drill or a mill. 

They come out perfectly smooth and sharp-edged, if you use the sort of punch tapered on the inside, rather than on the outside.  (In other words, a "hole" punch, not a "disc" punch.)

Lloyd