Cutting foam sheets... with a needle!

jhitesma

Some guy in the desert
Mentor
Thank you jhitesma for taking the time to explain that process. Your video is very helpful. I keep the URL in my favorites for quick reference.

Love hearing that it helps people! Going from PDF is a bit more involved, but once you get the hang of it it's not that bad. That new process you sent me a message about sounds interesting as well - I need to give it a try but don't get to my windows computers very often so it will be awhile before I can give it a go. Maybe you can share it here so others can give it a try as well ;)
 

TEAJR66

Flite is good
Mentor
So it comes down to preferences and what works for you. The best process is the one that you can use, right?

I have CutePDF Writer on my computer. If i have a document open, i can select CutePDF Writer as a printer. When i tell it to print, it writes whatever i have open as a PDF file and i get all the SAVE AS selections. So, here is how i use CutePDF Writer to clean up the FT plans.

I open an FT plan and turn off the layers that i don't need to see. CUT, SCORE and sometimes REFERENCE are the important layer to leave on. If you turn off the "Full Size" layer, you lose the scale drawing. After you have done a couple conversions, your settings are pretty much set and the scale drawing becomes less important. I recommend leaving that layer turned on for one or two iterations, till you are comfortable. Turning the "Full Size" layer off gets rid of a lot of stuff that clutters up the drawing and helps the import/export process move faster. My computer can take a couple minutes to import to inkscape if the file is too big.

So, once the layers are what i want, i select print. Choose CutePDF Writer as the printer. Before you print (save to file) go to PROPERTIES. On the PAPER/QUALITIES tab, select ADVANCED. Select the drop down menu for PAPER SIZE and POSTSCRIPT CUSTOM PAGE SIZE. Under PAGE SIZE DIMENSIONS specify width 40 and height 30 (this gives a 5 in border that CutePDF Writer wants to make a 30x20 sheet). Click OK and under GRAPHIC, next to PRINT QUALITY, select 300DPI (this helps reduce the file size), and click OK. Click OK again and you are back to your initial print window. Under PAGES TO PRINT, select PAGES and specify the pages that have the parts you want to cut. Click PRINT and a save dialog box appears. Give your new PDF a working name for the project.

Now I open Inkscape and go to DOCUMENT PROPERTIES. In CUSTOM SIZE, i specify mm and set 762 width, 508 height. import the new PDF. Select the page you want to work with. Then i position the drawing so that the parts are in the work space. I don't worry about the stuff from the "Full Size" layer. Newer FT plans are arranged in 20x30 sheets on the full size sections. I don't bother rearanging them in Inkscape. Even if the "FULL SIZE" layer is turned on, i do not ungroup and delete. Those items will be outside of my work space in Estlcam. So i just save the file once as a DXF and once as an SVG. When saving as a DXF, a dialog box will come up and you can specify mm for the BASE UNITS. Once I changed it, it has remained the default.

Once I open Estlcam, it is set up just like jhitesma has his. I open either the DXF or SVG file, select and move the drawing to verify the scale against the grid. Then i move the drawing to where i want the parts in the workspace. I select engaving, center cut, and I do the cut lines at 5.5mm for cut through, 2.5mm for score, and (i am still playing with depth) ~.5mm for etch and reference. Skewer holes are done with drill and i use the intersection of two lines option.

After i save the Gcode, i open that file in Repetier Host just to make sure it looks as it should. Before i close RH, i save to file again with a different file name. That gives me Gcode both from Estlcam and Repetier Host. Just in case there is a problem with the one saved from Estlcam, i know RH produces MPCNC freindly code. Both of those files go to an SD card and carried over to the machine.
 

ironkane

Member
You may be overthinking this. I am using a MPCNC with Ramps and it's connected to a laptop via USB.
I use Inkscape and Estlcam. Nothing else.
Set Inkscape default document size to 762 x 508mm. Use Import and not Open to bring a PDF page into Inkscape. Right-Click and select "Ungroup". On average around 3 times for FT plans. There will be times where it doesn't look like it did anything. Just repeat. Once things are selectable, delete the graphics, text and non-Cut lines. This is mostly just for cleanup of clutter. Save As .SVG.

Estlcam is all you need for the rest. Estlcam not only generates gcode, but can run it on Ramps too via USB. Or you can put the gcode file on a chip. But with Estlcam, you get more interaction and control. FYI, Estlcam saves gcode, but it wants a .nc (default) file and not a .gcode. But the contents are the same, it's just the file extension and that's easy to change. There's a setting in Estlcam that denotes the file extension of the saved gcode when you select "Save your CNC Program.

Once you've gotten the hang of it, it takes me about 30 minutes to process a page from PDF to ready to run gcode. Prepping the SVG right will save you a ton of time. Beyond removing the clutter, the most important thing is to break unintended intersections like a bevel cut line and the exterior cut line. Doing this will reduce the time you spend playing connect-the-dots in Estlcam.

Depths:
6.5mm for 100% cut. I may tweak it a little deeper, but I don't want parts falling out. Or lose too much vacuum while cutting.
2mm for 50% cuts. More than enough. Still requires a break.
0.5mm for reference marks. Deep enough to see, not deep enough to weaken.
 

jhitesma

Some guy in the desert
Mentor
Depths:
6.5mm for 100% cut. I may tweak it a little deeper, but I don't want parts falling out. Or lose too much vacuum while cutting.
2mm for 50% cuts. More than enough. Still requires a break.
0.5mm for reference marks. Deep enough to see, not deep enough to weaken.
[/quote]

I actually did some testing last weekend now that I have my vac table working to try and determine "optimal" cut depths.

On through cuts I don't want to cut deeper than I need to since it just puts more wear on the waste board, but I want to make sure I get full cuts. 6.5 feels like overkill since the foam is only ~4.5mm thick...but it varies so a bit of leeway is always going to be necessary (I've seen foam sheets over 5mm thick and barely 3mm thick - in the same freshly opened carton!) I've done 5.5 in the past but found sometimes I didn't get a full cut across the full sheet. 6 has been a good compromise but I wonder if I can do better.

Score cuts I usually used 2.5 but lately I've been doing 3. If things are setup well there's little danger of going too deep at 3 and it makes for an easier to break score.

For shallower score cuts (where the foam doesn't have to open but you want to be able to bend it on the line) I use 1mm now.

For marking cuts along bevels (to help guide me when I cut the bevel) I use 0.75mm. Deep enough to make sure it goes through the paper but not so deep it causes a weak spot.

For most marking cuts I'll use .5mm - but I've been experimenting with 0.25 lately which if things are dialed in well is really nice for marking cuts since it will sometimes only dent the paper and not actually cut through it. And if I can get that working consistently it would be ideal for marking cuts since then they'd truly only mark and not risk the integrity of the foam at all.

However....

I'm still trying to determine what it is about my machine that's causing my Z to vary over the surface of the workspace since my outer rails are all level but I still have about a 1mm variation in certain areas. (I suspect one of my gantry rails is warped slightly...but I also noticed one of my rollers may have issues as it doesn't "clamp" with all of the bearings touching the rail, at some points one of them stops touching the rail so something is off there but I'm not sure if it's the cause or a symptom.)

So I made up a test in estlcam, I just enabled a 1/4" grid and then used the engraving tool to draw "freehand" snapping to the grid. I made a series of lines a few inches long: 6, 5.5, 5, 4.5, 4, 3.5, 3, 2.5, 2, 1.5, 1, .75, .5, .25 (IIRC - I don't have the files here at work and shockingly didn't take any photos yet.) I then drew a box at 6.5mm deep going through the middle of the lines and extending 1/4" past them on one side.

Next I selected all of that and grouped it, then stamped it - once in each corner and once in the middle.

Basically I'm cutting 5 combs with each "tooth" of the comb cut to a different depth so I can see how consistent they are.

The results were interesting. Turns out the low points aren't where I expected them to be - I thought the variation was only left to right but it turns out there's a bit of variation front to back as well so it's more complex than I expected.

6mm was always a great cut.
5.5mm went through in 4 out of 5 places (IIRC) and almost went through in the 5th. 5.75 as max depth is probably safe on my setup.

I need to look at it more closely to determine which cut was the deepest that didn't go through in any position - as that should give me a safe max score cut.

The marking cuts - .25 just broke the paper in some spots but in others was barely visible...the Z variation in my machine is too much for it to be reliable. .5 however at least marked everywhere.


I'll post more details and photos later but I've been meaning to talk about this for almost a week and keep forgetting so figured I should get something down now while I'm thinking about it :D

The big thing bugging me is still why I've got that Z variation. The conduit I used for my gantry rails has always bugged me...it was from a different store than my side rails and not quite as smooth and I had a harder time finding straight pieces. So that's where my suspicion is focused. But that one roller acting goofy has me concerned again...it's the only one printed out of that particular brand of PETG so it may be slightly different than the rest.

I really need to stop slacking and make some calls about SS tubing to see if I can afford enough to finish the low-rider AND redo my MPCNC with 525 parts and SS tubing or if I'll just go SS on the low rider and stick with conduit when I print 525 parts.
 

TEAJR66

Flite is good
Mentor
rotate rails

I actually did some testing last weekend now that I have my vac table working to try and determine "optimal" cut depths.

On through cuts I don't want to cut deeper than I need to since it just puts more wear on the waste board, but I want to make sure I get full cuts. 6.5 feels like overkill since the foam is only ~4.5mm thick...but it varies so a bit of leeway is always going to be necessary (I've seen foam sheets over 5mm thick and barely 3mm thick - in the same freshly opened carton!) I've done 5.5 in the past but found sometimes I didn't get a full cut across the full sheet. 6 has been a good compromise but I wonder if I can do better.

Score cuts I usually used 2.5 but lately I've been doing 3. If things are setup well there's little danger of going too deep at 3 and it makes for an easier to break score.

For shallower score cuts (where the foam doesn't have to open but you want to be able to bend it on the line) I use 1mm now.

For marking cuts along bevels (to help guide me when I cut the bevel) I use 0.75mm. Deep enough to make sure it goes through the paper but not so deep it causes a weak spot.

For most marking cuts I'll use .5mm - but I've been experimenting with 0.25 lately which if things are dialed in well is really nice for marking cuts since it will sometimes only dent the paper and not actually cut through it. And if I can get that working consistently it would be ideal for marking cuts since then they'd truly only mark and not risk the integrity of the foam at all.

However....

I'm still trying to determine what it is about my machine that's causing my Z to vary over the surface of the workspace since my outer rails are all level but I still have about a 1mm variation in certain areas. (I suspect one of my gantry rails is warped slightly...but I also noticed one of my rollers may have issues as it doesn't "clamp" with all of the bearings touching the rail, at some points one of them stops touching the rail so something is off there but I'm not sure if it's the cause or a symptom.)

So I made up a test in estlcam, I just enabled a 1/4" grid and then used the engraving tool to draw "freehand" snapping to the grid. I made a series of lines a few inches long: 6, 5.5, 5, 4.5, 4, 3.5, 3, 2.5, 2, 1.5, 1, .75, .5, .25 (IIRC - I don't have the files here at work and shockingly didn't take any photos yet.) I then drew a box at 6.5mm deep going through the middle of the lines and extending 1/4" past them on one side.

Next I selected all of that and grouped it, then stamped it - once in each corner and once in the middle.

Basically I'm cutting 5 combs with each "tooth" of the comb cut to a different depth so I can see how consistent they are.

The results were interesting. Turns out the low points aren't where I expected them to be - I thought the variation was only left to right but it turns out there's a bit of variation front to back as well so it's more complex than I expected.

6mm was always a great cut.
5.5mm went through in 4 out of 5 places (IIRC) and almost went through in the 5th. 5.75 as max depth is probably safe on my setup.

I need to look at it more closely to determine which cut was the deepest that didn't go through in any position - as that should give me a safe max score cut.

The marking cuts - .25 just broke the paper in some spots but in others was barely visible...the Z variation in my machine is too much for it to be reliable. .5 however at least marked everywhere.


I'll post more details and photos later but I've been meaning to talk about this for almost a week and keep forgetting so figured I should get something down now while I'm thinking about it :D

The big thing bugging me is still why I've got that Z variation. The conduit I used for my gantry rails has always bugged me...it was from a different store than my side rails and not quite as smooth and I had a harder time finding straight pieces. So that's where my suspicion is focused. But that one roller acting goofy has me concerned again...it's the only one printed out of that particular brand of PETG so it may be slightly different than the rest.

I really need to stop slacking and make some calls about SS tubing to see if I can afford enough to finish the low-rider AND redo my MPCNC with 525 parts and SS tubing or if I'll just go SS on the low rider and stick with conduit when I print 525 parts.


Jason, Loosen some clamps and rotate suspect rails 90 degrees. Do one at a time and see what changes. Just an idea.
 
Last edited:

TEAJR66

Flite is good
Mentor
Jasons video

You may be overthinking this. I am using a MPCNC with Ramps and it's connected to a laptop via USB.
I use Inkscape and Estlcam. Nothing else.
Set Inkscape default document size to 762 x 508mm. Use Import and not Open to bring a PDF page into Inkscape. Right-Click and select "Ungroup". On average around 3 times for FT plans. There will be times where it doesn't look like it did anything. Just repeat. Once things are selectable, delete the graphics, text and non-Cut lines. This is mostly just for cleanup of clutter. Save As .SVG.

Estlcam is all you need for the rest. Estlcam not only generates gcode, but can run it on Ramps too via USB. Or you can put the gcode file on a chip. But with Estlcam, you get more interaction and control. FYI, Estlcam saves gcode, but it wants a .nc (default) file and not a .gcode. But the contents are the same, it's just the file extension and that's easy to change. There's a setting in Estlcam that denotes the file extension of the saved gcode when you select "Save your CNC Program.

Once you've gotten the hang of it, it takes me about 30 minutes to process a page from PDF to ready to run gcode. Prepping the SVG right will save you a ton of time. Beyond removing the clutter, the most important thing is to break unintended intersections like a bevel cut line and the exterior cut line. Doing this will reduce the time you spend playing connect-the-dots in Estlcam.

Depths:
6.5mm for 100% cut. I may tweak it a little deeper, but I don't want parts falling out. Or lose too much vacuum while cutting.
2mm for 50% cuts. More than enough. Still requires a break.
0.5mm for reference marks. Deep enough to see, not deep enough to weaken.

Some of my point was missed because it was out of context.

This process was in response to Jason's video and my dislike for working in Inkscape. Jason produced a great video tutorial that i keep handy as a constant reference.

CutePDF was a means of cleaning the file before Inkscape so that there was no need to work in that dreadful program. Do not get me wrong, i completely appreciate the FREEness of Inkscape. If i had another (FREE) program that would import PDF and output DXF/SVG files, i would use it.

The hardest part of cleaning before inkscape was keeping the ruler drawing since it is on the layer that contains the most editing. Eliminating that layer all together eliminates any need for editing in inkscape, unless you want to rearrange parts. But the newer FT plans don't need to be rearranged because Sponz has already optimized for 20x30 sheet. Since the ruler, page border, and other little logos are outside the Estlcam workspace, i just leave that layer alone, but that increases the amount of time it takes my computer (I am impatient and those seconds while my computer "thinks" about it, annoy me) to import into inkscape. Either way my goal, and i achieved it, was to not edit in Inkscape.

I would like to know how you set the default page size to 762x508. I dislike having to reset that everytime i open Inkscape. I am using "IMPORT" not "OPEN" when i bring a file into inkscape.

I save both DXF and SVG just so that i have those formats for use in other projects.

I can go straight from Estlcam to the MPCNC, I have the file extensions set and i save the cnc program. But Repetier Host is a proof for me. My machine is in a shop, not co-located with the computer in my room. I like the checks and balances of seeing the file in another program before i trek across camp.

If you are using the needle cutter, the shallower your cut depth the faster you can go on feed rate. If you use a .020" wire for a needle, you will get clean cuts with parts that stay in place without tabs. You will also retain your vacum.

So it comes down to preferences and what works for you. The best process is the one that you can use, right?
 

TEAJR66

Flite is good
Mentor
Who is using a spindle?

Which spindle do you recommend?

I want to cut delrin, kydex and some light plywood.
 

ironkane

Member
I actually did some testing last weekend now that I have my vac table working to try and determine "optimal" cut depths.

On through cuts I don't want to cut deeper than I need to since it just puts more wear on the waste board, but I want to make sure I get full cuts. 6.5 feels like overkill since the foam is only ~4.5mm thick...but it varies so a bit of leeway is always going to be necessary (I've seen foam sheets over 5mm thick and barely 3mm thick - in the same freshly opened carton!) I've done 5.5 in the past but found sometimes I didn't get a full cut across the full sheet. 6 has been a good compromise but I wonder if I can do better.

Score cuts I usually used 2.5 but lately I've been doing 3. If things are setup well there's little danger of going too deep at 3 and it makes for an easier to break score.

For shallower score cuts (where the foam doesn't have to open but you want to be able to bend it on the line) I use 1mm now.

For marking cuts along bevels (to help guide me when I cut the bevel) I use 0.75mm. Deep enough to make sure it goes through the paper but not so deep it causes a weak spot.

For most marking cuts I'll use .5mm - but I've been experimenting with 0.25 lately which if things are dialed in well is really nice for marking cuts since it will sometimes only dent the paper and not actually cut through it. And if I can get that working consistently it would be ideal for marking cuts since then they'd truly only mark and not risk the integrity of the foam at all.

I'll tell you what I think is causing the Z variation and I don't know if it can be addressed...random variations in the amount that the needle bows as it makes it's circuit around the flywheel. When it's not under load, the needle is going to be pretty consistent in it's flex. But when cutting you're not only getting vertical resistance from puncturing the paper/foam sandwich, you're also getting lateral forces adding drag to that vertical movement.

It's really strange when you run the same code through two pieces of foam and get slightly different results of cut. It's usually more pronounced in my 50% cuts than my through cuts. I'm curious as to what will happen when I get off my lazy ass and swap out my HicWic for my new quick change mount and consequently my new slower, higher torque motor. Will having more "oomph" behind the needle make any difference? I'm sure that a real engineer could provide some insight.

I only recently had increased my cut depth from 6.2 to 6.5mm. For me, that's starting to hit the sweet spot. I'm perforating the bottom, but not so much that all the parts just fall out. It looks like stitch marks on the bottom while the top is clean. Are you getting some slight melting in your cuts? I get enough that it's best to run a razor through to release the parts. It feels kind of like it's running through gravel. I actually like this as I don't need to add any time consuming holding tabs. And my vacuum table likes it too. I've had some cuts that was letting too much air flow and had to slap some scrap on top to keep a good vacuum.

I love my MPCNC, but I'm planning on building a new frame using OpenBuilds V-Slot rails. This will really eliminate any questions about trueness. It's a lot easier to see if something square is flat than something that's round is sitting flat. I figure the frame is going to run me around $200. I'm hoping to eliminate the Z screw drive in favor of a servo controlled Z. Also want to add a plotter pen mount and a laser for cutting thin ply. I still think the needle is better than a laser for cutting foam.
 
Last edited:

jhitesma

Some guy in the desert
Mentor
I'll tell you what I think is causing the Z variation and I don't know if it can be addressed...random variations in the amount that the needle bows as it makes it's circuit around the flywheel. When it's not under load, the needle is going to be pretty consistent in it's flex. But when cutting you're not only getting vertical resistance from puncturing the paper/foam sandwich, you're also getting lateral forces adding drag to that vertical movement.

I get the same thing with my router, so it's not flex in the needle. There is definitely something about my machine that's off. I've also mounted a dial indicator to my tool mount and confirmed with that that it's the machine.

I've been meaning to try rotating the gantry rods like TEAJR66 suggested trying for some time - but wanted to do this test to determine if it was only in the X plane as I suspected. The test however has shown that there's variation in both X and Y.

It's really strange when you run the same code through two pieces of foam and get slightly different results of cut. It's usually more pronounced in my 50% cuts than my through cuts. I'm curious as to what will happen when I get off my lazy ass and swap out my HicWic for my new quick change mount and consequently my new slower, higher torque motor. Will having more "oomph" behind the needle make any difference? I'm sure that a real engineer could provide some insight.

I don't really have that problem. My problem is inconsistent depth on the same piece of foam - but it happens consistently from piece of foam to piece of foam.

The only times I've had inconsistent cuts between sheets are when either:

1) The foam thickness varies - this has been happening a LOT with the latest changes by adams to their redi-board and is now something I check for carefully.

2) I goof up and mess up my initial Z - either because I forget to check it or because I set it wrong.

I've learned I get best results setting my z height near the front middle of my work since that's the average height due to the Z issues inherent to my particular machine. I set Z by manually rotating my flywheel to get the needle as low as it will go - then slowly lower z manually while pushing the gantry forward and back (or side to side) until the needle just starts to touch. Then I back off the z about 1/8 of a turn so the needle doesn't quite touch.

Set like that at the home position my needle is about .25-.5 into the paper at Z of 0 and at the far right side it's about .25-.5 above the paper. Which is usable - but I know I can do better if I work out this mechanical issue with my machine.

I only recently had increased my cut depth from 6.2 to 6.5mm. For me, that's starting to hit the sweet spot. I'm perforating the bottom, but not so much that all the parts just fall out. It looks like stitch marks on the bottom while the top is clean. Are you getting some slight melting in your cuts? I get enough that it's best to run a razor through to release the parts. It feels kind of like it's running through gravel. I actually like this as I don't need to add any time consuming holding tabs. And my vacuum table likes it too. I've had some cuts that was letting too much air flow and had to slap some scrap on top to keep a good vacuum.

Sounds like you may be setting your z higher than I do, and like you may have more friction in your needle guide. I was getting some melting when I bent my needle a few weeks ago - but once I straightened it back out that's gone and my parts all but fall out now. I don't need any holding tabs as the pieces stay in place while on the table (partly due to the vac table but mostly just because of the super fine kerf of the needle) and if I rotate the sheet of foam before lifting it off the vac table (after turning off the vac of course) the pieces stay in place but can be knocked out with a very gentle tap of the finger. If I just lift the sheet from an edge without rotating it first then sometimes a few pieces will stay on the table.

I love my MPCNC, but I'm planning on building a new frame using OpenBuilds V-Slot rails. This will really eliminate any questions about trueness. It's a lot easier to see if something square is flat than something that's round is sitting flat. I figure the frame is going to run me around $200. I'm hoping to eliminate the Z screw drive in favor of a servo controlled Z. Also want to add a plotter pen mount and a laser for cutting thin ply. I still think the needle is better than a laser for cutting foam.

Yes, V-slot is great - but considerably more expensive :) For my budget I'm willing to put up with an ~1mm variation in Z across 3.5' of distance. And I'm fairly sure that eventually I can get that cleared up as well since I have a few suspicions about exactly where it's coming from. It's just not been a high priority for me to track down since the variation is slight enough I can live with it for the work I'm doing currently.

I'm really not a fan of laser for foam. I've built some of FT's laser cut kits and I really don't like the undercutting the laser causes. And for the home builder a needle cutter is considerably safer and more affordable than a laser capable of cutting foam. I do love my laser for cutting poster board though!
 

moebeast

Member
I'm still trying to determine what it is about my machine that's causing my Z to vary over the surface of the workspace since my outer rails are all level but I still have about a 1mm variation in certain areas. (I suspect one of my gantry rails is warped slightly...but I also noticed one of my rollers may have issues as it doesn't "clamp" with all of the bearings touching the rail, at some points one of them stops touching the rail so something is off there but I'm not sure if it's the cause or a symptom.)

Jason,
I wonder if your variations in the z axis are actually in your bed and not your rails. The DTFB vacuum pads we made surely have some variation in thickness. A good straight edge and feeler gauge will confirm.
On my MPCNC, I milled the entire cutting area flat with a 3/4" router bit. That was on the blue foam waste board. You could remove the paper and mill the top of you vacuum pad. Then add a fourth sacrificial layer if you want.
Mark
 

jhitesma

Some guy in the desert
Mentor
Jason,
I wonder if your variations in the z axis are actually in your bed and not your rails. The DTFB vacuum pads we made surely have some variation in thickness. A good straight edge and feeler gauge will confirm.
On my MPCNC, I milled the entire cutting area flat with a 3/4" router bit. That was on the blue foam waste board. You could remove the paper and mill the top of you vacuum pad. Then add a fourth sacrificial layer if you want.
Mark

I actually haven't measured to the vac pad...though I agree there is a strong possibility of variation there.

When I've tested with the dial indicator it's been to the baseboard of my machine - which is pretty darn flat. I added more support than probably necessary in the base because I didn't trust the 3/4" particle board to not droop if it wasn't supported. A straightedge on the table seems to indicate it's nice and flat all across. But when we're talking about a 1mm variation...it is possible there could be that much variation just in the thickness of the particle board - though I'd expect a straight edge to show it. (I have a long 4' aluminum straight edge I used to check it so I was able to get edge to edge on the 4'x4' table.)

I've considered just milling the surface...but I often fixture things in a way that they extend past the area that would be milled. If I was to mill it I'd put down another board first and mill that (like the waste board I use with my router.) I actually did mill a sheet of pink foam like that with a 1' square pocket which I set a mirror tile into for use with the vinyl cutting drag knife since trying to cut vinyl but not the backer paper did require extremely consistent Z ;)

And remember...I'm on the old pre-525 parts so my Z isn't as stiff as it could be. And these old parts were harder to print reliably - I had some issues with supports failing on the big snake like Z pieces so I wouldn't be surprised if there's some issues there. And I have that one roller that acts a bit goofy. I really need to just bite the bullet and do the 525 parts....but need to decide if I also want to upgrade to SS.
 

jhitesma

Some guy in the desert
Mentor
But the newer FT plans don't need to be rearranged because Sponz has already optimized for 20x30 sheet.

Well, they've been laid out for 20x30 sheets...but I don't know that I'd agree they've been optimized. Which is by no means a swipe at Sponz, his work is terrific and to be commended. But he's not designing the plans for automated cutting as much as he is for most users who will be hand cutting.

I didn't do it in most of my videos because I was trying to keep them short...but I regularly re-arrange the parts and manage to eliminate a sheet or more of foam.

I can pack pieces very closely together - closer than I'd want them if I was cutting by hand. I also eliminate pieces I don't need like motor pods (I either make them in batches or 3D print them now,) the poster board pieces, the 90 degree angles (since I have a nice plastic one) and on some designs optional parts or variations. The FT-29 I was able to eliminate a sheet but still keep all of the parts for example.

I've also noticed things like on the Simple Scout - in the build video I noticed that the laser kit has the pushrod guide holders inside one of the waste pieces in a fuselage part. In the plans they're separate. This isn't a great example because the SS can't be done in fewer than 3 sheets because each wing half NEEDS it's own sheet as does the main fuselage - and there aren't enough parts to fill all the rest of the space on 3 sheets so there's going to be wasted space and nesting those little bits doesn't really help on this particular plane. (It probably helps FT due to how they package for shipment - even if it means wasting more foam that likely saves more money over having to ship larger packages.)

So while Sponz does an outstanding job on the plans I would hesitate to call them optimized as I'm pretty sure that's not his intent. If they were optimized for our use they wouldn't be as friendly for people manually cutting and those are the primary target of the plans so those are who they should be optimized for. So - to be fully fair I could stay they're optimized...just not optimized for mechanized cutting ;)

Of course they can be cut as-is just fine. And at $1 a sheet it feels silly to spend an extra 20 minutes to save one sheet of foam...but I enjoy the process and detest waste (which is why I have a big pile of foam scraps that are still big enough I could cut "something" out of them - I use them to cut repair parts out when needed.) So I take that little bit of time because I enjoy it - but I fully accept that it's not for everyone!
 

ironkane

Member
You set your Z while not running? I get mine running at speed and jog it to the center and lower until it just starts to touch. Then jog it in the X or Y to see how much it's touching by looking for the trail and adjusting. Now I'm curious to see if there is a discrepancy between the two methods. Since Estlcam remembers Home, I don't set it often.

V-Slot doesn't have to be all that expensive for this application. I think my pre-tax & shipping price for my frame build is $185. I spent $165 for the printed parts, $285 for the hardware bundle and then the conduit and assorted other details. I'm shooting for sub-$500 with V-slot since the typical Z tower is a couple hundred and I'm wanting to eliminate that with the servo Z. MPCNC is still awesome. Great platform to learn on.

But check out the new OpenBuilds Acro
OpenBuilds_ACRO_laser_Build_Pic_transparent_blue__85150.1505940342.500.659.png


If it was secured to the table instead of free standing, it'd be almost and instant foam cutter for $445.99 (30" x 30" work area), just add needle cutter instead of the laser mount. My build is going to use a 20x60 rail instead of the 20x20 so I can secure it to the work bench. So I'm not buying the bundle. and it'll be a 20" x 30" work area.
 

TEAJR66

Flite is good
Mentor
Meh. "Optimized", poor choice of words. More like, "fit". Yeah, thats better.

On the upside, i did a test today. I imported the PDF to Inkscape and set nothing. Just saved as DXF and ensured the base unit was mm. It was already set, so just a matter of clicking Ok. That DXF opened in Estlcam to scale with no problems. Only the parts were in the workspace. Since Estlcam ignores all the lines that you do not define as cuts, i just ignore the outer border stuff.

In the end, i'm not tiling plans, tracing, and cutting by hand. So i am pretty happy. The post office can't get foam to Afghanistan fast enough.
 

TEAJR66

Flite is good
Mentor
You set your Z while not running? I get mine running at speed and jog it to the center and lower until it just starts to touch. Then jog it in the X or Y to see how much it's touching by looking for the trail and adjusting. Now I'm curious to see if there is a discrepancy between the two methods. Since Estlcam remembers Home, I don't set it often.

V-Slot doesn't have to be all that expensive for this application. I think my pre-tax & shipping price for my frame build is $185. I spent $165 for the printed parts, $285 for the hardware bundle and then the conduit and assorted other details. I'm shooting for sub-$500 with V-slot since the typical Z tower is a couple hundred and I'm wanting to eliminate that with the servo Z. MPCNC is still awesome. Great platform to learn on.

But check out the new OpenBuilds Acro
OpenBuilds_ACRO_laser_Build_Pic_transparent_blue__85150.1505940342.500.659.png


If it was secured to the table instead of free standing, it'd be almost and instant foam cutter for $445.99 (30" x 30" work area), just add needle cutter instead of the laser mount. My build is going to use a 20x60 rail instead of the 20x20 so I can secure it to the work bench. So I'm not buying the bundle. and it'll be a 20" x 30" work area.


Where is the drool emoticon?!
 

rockyboy

Skill Collector
Mentor
Of course they can be cut as-is just fine. And at $1 a sheet it feels silly to spend an extra 20 minutes to save one sheet of foam...but I enjoy the process and detest waste (which is why I have a big pile of foam scraps that are still big enough I could cut "something" out of them - I use them to cut repair parts out when needed.) So I take that little bit of time because I enjoy it - but I fully accept that it's not for everyone!

Hey - I resemble that remark too! :black_eyed:

Happy to report my first three 'full size' plane parts cut on my Lowrider this week. The test power pods don't really count :)
Still using Easel for gcode generation, but will be switching to Estlcam for my next plane.

I did discover if I go into the little control pad on the Marlin software and kick up the speed there to 150% I start missing a lot of steps on my Y axis. So that's not the right place to tweak speed on my machine - need that tool and travel speed control from Estlcam :)
 

jhitesma

Some guy in the desert
Mentor
But check out the new OpenBuilds Acro
OpenBuilds_ACRO_laser_Build_Pic_transparent_blue__85150.1505940342.500.659.png


If it was secured to the table instead of free standing, it'd be almost and instant foam cutter for $445.99 (30" x 30" work area), just add needle cutter instead of the laser mount. My build is going to use a 20x60 rail instead of the 20x20 so I can secure it to the work bench. So I'm not buying the bundle. and it'll be a 20" x 30" work area.

Saw their announcement about that - nice stuff for sure. I also saw their announcement about the minimill giveaway and entered :) Would be fun to have a small super stuff machine for making little aluminum bits after all! Would also love something portable so I could share it with the maker club at my daughter's school since she just joined it this quarter.

But...that's still pricey to me.

I didn't really keep an accounting as I built my MPCNC...but I think the single biggest expenditure I had was my steppers which was $58 (and pricing steppers for my low-rider build I see the price has actually come down even further, same set from same seller is now only $51.80!)

A big part of why I was able to build my MPCNC was I was able to buy parts $20-$40 at a time over a period of a few months. (attempts to save up funds for hobby purchases always seem to result in some kind of household emergency coming up just as I'm about to reach my savings goal and wiping it out :( ) A number of the parts I had on hand already like a spare RAMPS (though I later bought a new one because I thought I fried mine before realizing I just had the serial port speed wrong.) The hardware I got locally at a wholesaler for about $25. The belts and pulleys I got cheap off ebay and same with the bearings. Only the stepper purchase was over $50 which is kind of the limit of what I can get away spending without having to seek spousal approval ;)

Someday I'd love to do a rail based machine...but not until my daughter is out of college and my wife had had another trip to Europe ;)

I did put a call in today to a friend who has a driveline shop and connections with metal suppliers to find out what kind of deal he can get me on SS tubing though....
 

jhitesma

Some guy in the desert
Mentor
In the end, i'm not tiling plans, tracing, and cutting by hand. So i am pretty happy. The post office can't get foam to Afghanistan fast enough.

Yep! And it's great that there are a number of toolpaths available so each person can do what they feel most comfortable with. I really don't care for sketchup so I fully understand someone not liking inkscape and wanting to avoid it in their tool chain.

Thankfully we live in a time when there are tons of both free and paid solutions to work with so everyone can find something that makes them comfortable!
 

wild.bill

New member
ACRO

I have not posted here for quite a while but I bought an ARCO system. I got the one that is 1000mm wide and 500mm deep. All I needed was to buy the kit because I already had motors, controller and laser sitting around. I did add a Z axis to it for a low cost. BangGood has a light A axis that was only $35 plus shipping. It's not heavy enough for a spindle but more than enough for a laser. Some pictures of it along with my old image that I like for testing.
IMG_0473-resizedM.jpg
IMG_0508-resizedM.jpg
IMG_0500-resizedM.jpg
 

ironkane

Member
Thanks for posting. Could you clear up something for me about the design?
It looks like it's not secured to the table from what I can see. It just sits on acrylic legs. Great for lasers, not so much for pushing a needle around. My thought was to replace the 2020 rails with 2060 rails to fill the gap that the legs provide. Will that be enough to provide clearance for the plates from the deck? My plan is to use the plate kit and build the frame from parts instead of the complete kit. Then I can bolt it down and get the 20 x 30 (working) size I want.

I have not posted here for quite a while but I bought an ARCO system. I got the one that is 1000mm wide and 500mm deep. All I needed was to buy the kit because I already had motors, controller and laser sitting around. I did add a Z axis to it for a low cost. BangGood has a light A axis that was only $35 plus shipping. It's not heavy enough for a spindle but more than enough for a laser. Some pictures of it along with my old image that I like for testing.
View attachment 97484
View attachment 97485
View attachment 97486