jaredstrees
Well-known member
I'm sure whatever you end up with someone around here can help!
I'm tired of cutting Foam by hand
- Thread starter foamtest
- Start date
Hi all, so brand new to this hobby , still building my first scratch Explorer FPV setup, , , so yes cutting foam while mostly relaxing is tedius and a littil back breaking, I have not bought this yet, but thinking about to pull the trigger on this Logan Foam werks system, Good luck,
https://www.amazon.com/dp/B00UUEA85S/?tag=lstir-20
Logan Graphics Foamwerks Deluxe Cutting Kit for Foam Board ,
Just a line of thought:
For instance, I'm building another cartesian gantry based cnc hybrid machine, that basically works a bit like the MPCNC or maybe more like SPONZ's Rascal CNC, and is made from steel only. The MPCNC and the Rascal are probably a bit easier to assemble though. The advantage of mine is the possibility to fairly easily add an extra x and z axis at one end for hotwiring. Reason for choosing this type of system is that I really do want a hybrid system, that I can use for many things, like plate foam cutting with needle or laser, 2D/3D routing, hotwire foam cutting, coarse 3d printing, maybe plasma cutting and whatever else I might think of along the way. This is however a somewhat complicated solution, that takes up a lot of space, both in use and storage. Doesn't make my solution or MPCNC or the Rascal bad in any way, they are just like that to be able to do more different things.
BUT! If what you want to do really only is to cut flat foam sheets, something along the lines of a Phlatprinter might be a better option? It's a transportable box, about a yard by 1 ft by 1½ ft in size, that works a bit like a large sheet plotter, with the foam moving back and forth through a pair of rollers, and the router or needle head cutting from the underside moving from side to side in the box. It would take up far less space when not in use, and possibly also be a bit cheaper to build.
You can find the opensource instructions on openbuilds.com; Main structure is a box made from a set of plates cut from MDF, a number of rollers and a cutting head inside the box moving along a rail of sorts with the cutting end pertruding through a slot in the surface the foam slides forth and back on.
Given the limited need for movement in the Z direction, that could probably be handled by a simple servo moving the cutter maybe up to half an inch up or down. Then it just needs 2-3 stepper motors for the x and Y axis. If you build it smaller to just take DTF, 2 might do.
an example of one in use. I'd use Linuxcnc or machinekit on an old pc in stead of Mach3, as they are free/opensource and just as good, or alternatively run it from an arduino board with a GRBL variant or similar, but that is generally more about personal preference.
Another space saving option could be an adaptation of a maslow 2D or 2.5D CNC router system (https://www.maslowcnc.com/). Ditch the heavy router for a much smaller one or a needle cutter, place the 2 servo motors at each end of a 5-6ft 2by4, hang it firmly and accurately(possibly with finely adjustable height at one end?) from the ceiling, and have a removable ~ 2 by 3 ft work surface centered at or below table height a bit in front of the 2by4 and tilted back to line up with the pull lines. Oh, and replace the heavy chains with GT2 toothed belts, they are both lighter and cheaper. After end of use remove or fold away the cnc work surface, and your workbench, etc. is cleared.
For instance, I'm building another cartesian gantry based cnc hybrid machine, that basically works a bit like the MPCNC or maybe more like SPONZ's Rascal CNC, and is made from steel only. The MPCNC and the Rascal are probably a bit easier to assemble though. The advantage of mine is the possibility to fairly easily add an extra x and z axis at one end for hotwiring. Reason for choosing this type of system is that I really do want a hybrid system, that I can use for many things, like plate foam cutting with needle or laser, 2D/3D routing, hotwire foam cutting, coarse 3d printing, maybe plasma cutting and whatever else I might think of along the way. This is however a somewhat complicated solution, that takes up a lot of space, both in use and storage. Doesn't make my solution or MPCNC or the Rascal bad in any way, they are just like that to be able to do more different things.
BUT! If what you want to do really only is to cut flat foam sheets, something along the lines of a Phlatprinter might be a better option? It's a transportable box, about a yard by 1 ft by 1½ ft in size, that works a bit like a large sheet plotter, with the foam moving back and forth through a pair of rollers, and the router or needle head cutting from the underside moving from side to side in the box. It would take up far less space when not in use, and possibly also be a bit cheaper to build.
You can find the opensource instructions on openbuilds.com; Main structure is a box made from a set of plates cut from MDF, a number of rollers and a cutting head inside the box moving along a rail of sorts with the cutting end pertruding through a slot in the surface the foam slides forth and back on.
Given the limited need for movement in the Z direction, that could probably be handled by a simple servo moving the cutter maybe up to half an inch up or down. Then it just needs 2-3 stepper motors for the x and Y axis. If you build it smaller to just take DTF, 2 might do.
Another space saving option could be an adaptation of a maslow 2D or 2.5D CNC router system (https://www.maslowcnc.com/). Ditch the heavy router for a much smaller one or a needle cutter, place the 2 servo motors at each end of a 5-6ft 2by4, hang it firmly and accurately(possibly with finely adjustable height at one end?) from the ceiling, and have a removable ~ 2 by 3 ft work surface centered at or below table height a bit in front of the 2by4 and tilted back to line up with the pull lines. Oh, and replace the heavy chains with GT2 toothed belts, they are both lighter and cheaper. After end of use remove or fold away the cnc work surface, and your workbench, etc. is cleared.
Last edited:
Hello all!
Ive also been looking at the similar options today, specifically retrofitting a drawbot with a needle cutter.
There are a couple of design available on thingiverse, this looks like one of the more popular ones:
https://www.thingiverse.com/thing:798076
If the gondola could be modified to hold a needle cutter instead of a pen and a way of precisely controlling the cutting dept it could become the go to for easily accessible automated foam plane cutting.
There have been a couple of comments from people having issues with the ball chains slipping so maybe going for belts, like those used in 3d printers, with idlers to hold the belts in an S shape across the drive pulley and counterweights instead of the chains would be useful for accuracy.
From what i understand of the drawbot design, it uses a servo to lift itself off the surface to break the lines its drawing, does anyone know or are able to figure out if the amount of "lift" can be configured for through cuts and score cuts?
Im very much a n00b when it comes to arduino, gcode reading machinery, programming and all sort of machine logic.
Ive made stuff that only requires hooking wires up and installing preconfigured packages like an adalight setup, a 2 axis HID joystick and putting together a prusa i3 clone kit.
Another option might be a variant of something like this:
https://www.thingiverse.com/thing:2881396
Perhaps it would need a little bit of beefing up but it could also be a great option of its accuracy is good enough, doesnt look like it from the vid though.
Ive also been looking at the similar options today, specifically retrofitting a drawbot with a needle cutter.
There are a couple of design available on thingiverse, this looks like one of the more popular ones:
https://www.thingiverse.com/thing:798076
If the gondola could be modified to hold a needle cutter instead of a pen and a way of precisely controlling the cutting dept it could become the go to for easily accessible automated foam plane cutting.
There have been a couple of comments from people having issues with the ball chains slipping so maybe going for belts, like those used in 3d printers, with idlers to hold the belts in an S shape across the drive pulley and counterweights instead of the chains would be useful for accuracy.
From what i understand of the drawbot design, it uses a servo to lift itself off the surface to break the lines its drawing, does anyone know or are able to figure out if the amount of "lift" can be configured for through cuts and score cuts?
Im very much a n00b when it comes to arduino, gcode reading machinery, programming and all sort of machine logic.
Ive made stuff that only requires hooking wires up and installing preconfigured packages like an adalight setup, a 2 axis HID joystick and putting together a prusa i3 clone kit.
Another option might be a variant of something like this:
https://www.thingiverse.com/thing:2881396
Perhaps it would need a little bit of beefing up but it could also be a great option of its accuracy is good enough, doesnt look like it from the vid though.
Last edited:
One thing you will have to consider is the fact that unlike drawing, you are creating cuts in the material. You need to be able to slide over the trenches without the contact points getting caught. You also need to be able to cut fairly perpendicularly into the foam at any set height.
If you put a large servo("standard" size or bigger) at each of the 3 "corner" positions, maybe with some sort of fairly large plastic slide foot with nicely rounded up edges on each and with a fairly long servo arm on it, i don't see why this shouldn't work. You should of course wire the 3 servos to the same control channel, so that they move in sync.
Another option could be to make a single large slide plate that stays on the surface and have the cutter part move up and down, moved by a single large servo in stead. You only really need 3 or 4 different height settings (above - halfway - thru or above - 1/3 - 2/3 - thru), so the Z values could be handled as a set of experimental arbitrary values for each of the different foam thicknesses you will be working on, which could simplify software setup a bit (Servo rotation will not directly translate into linear height adjustment. This can be handled by a bit of trig math added to the software. But it means you will have to do some programming)
Another thing to consider is that you, unlike for a simple pen, will need a certain amount of force to hold the cutter onto the surface to be able to pierce the paper, or the needle will simply bounce the cutter off the surface instead of cutting. Leaning the work surface further back, like on the Maslowcnc, would be a simple approach. Adding weight to the "stationary" part of the cutter unit for increased inertia probably won't hurt either.
The idea of using an S shaped idler construction, I assume a bit like on a cam belt tensioner on a car, and weights is spot on. Wish I had thought of that. The belts generally slip a lot easier than the chains would, but that should certainly help keep them in place. You can likely also make this idler system so that it will twist the belt a bit to still have it fully supported where it leaves the idler, hanging straight down to the weights without kinking at the top (I hope you understand?)
For weights I'd suggest water bottles, easy to adjust for weight balance, so the forces at the belt drive wheel can be minimised (The needle cutter doesn't generate much in the ways of sideways forces, unlike a router, so you're mainly dealing with friction btw. foam and slide surface and the inertia of moving the cutter unit). And if you put them inside a vertical tube of sorts, they won't flail around and get cought on stuff.
If you put a large servo("standard" size or bigger) at each of the 3 "corner" positions, maybe with some sort of fairly large plastic slide foot with nicely rounded up edges on each and with a fairly long servo arm on it, i don't see why this shouldn't work. You should of course wire the 3 servos to the same control channel, so that they move in sync.
Another option could be to make a single large slide plate that stays on the surface and have the cutter part move up and down, moved by a single large servo in stead. You only really need 3 or 4 different height settings (above - halfway - thru or above - 1/3 - 2/3 - thru), so the Z values could be handled as a set of experimental arbitrary values for each of the different foam thicknesses you will be working on, which could simplify software setup a bit (Servo rotation will not directly translate into linear height adjustment. This can be handled by a bit of trig math added to the software. But it means you will have to do some programming)
Another thing to consider is that you, unlike for a simple pen, will need a certain amount of force to hold the cutter onto the surface to be able to pierce the paper, or the needle will simply bounce the cutter off the surface instead of cutting. Leaning the work surface further back, like on the Maslowcnc, would be a simple approach. Adding weight to the "stationary" part of the cutter unit for increased inertia probably won't hurt either.
The idea of using an S shaped idler construction, I assume a bit like on a cam belt tensioner on a car, and weights is spot on. Wish I had thought of that. The belts generally slip a lot easier than the chains would, but that should certainly help keep them in place. You can likely also make this idler system so that it will twist the belt a bit to still have it fully supported where it leaves the idler, hanging straight down to the weights without kinking at the top (I hope you understand?)
For weights I'd suggest water bottles, easy to adjust for weight balance, so the forces at the belt drive wheel can be minimised (The needle cutter doesn't generate much in the ways of sideways forces, unlike a router, so you're mainly dealing with friction btw. foam and slide surface and the inertia of moving the cutter unit). And if you put them inside a vertical tube of sorts, they won't flail around and get cought on stuff.
Last edited:
As long as the tension is alright the teethed belts are quite grippy.
I havnt tried a needle cutter but i would think the force of the "punch" of the needle is so quick that it slices through the foam without much counter force, if that isnt the case the surface where the foam rests could be tilted out further to make more of the gondola weight rest on it.
For the gondola i though it might rest on ball casters as are common as the third wheel on robots, omnidirectional and low friction.
Lifting the cutter could be done with some sorta lever activated lifty bit, like a scissor lift perhaps, that the servo moves.
I havnt tried a needle cutter but i would think the force of the "punch" of the needle is so quick that it slices through the foam without much counter force, if that isnt the case the surface where the foam rests could be tilted out further to make more of the gondola weight rest on it.
For the gondola i though it might rest on ball casters as are common as the third wheel on robots, omnidirectional and low friction.
Lifting the cutter could be done with some sorta lever activated lifty bit, like a scissor lift perhaps, that the servo moves.
Having built a 3dprinter, cnc machine, eggbot, and polar drawbot.....I would recommend going with a polar bot.
The slipping is an issue, but there's just so many other aspects of it that can go wrong as well. The mechanics look simple - but there are a good number of things that make them less simple than it appears. And if you want things to be accurate you need a MUCH larger machine than the area you want to cut because they loose accuracy big time near the edges of their cutting areas.
But the bigger issue I see - I don't think a need cutter head is heavy enough to work well like that. Even if you make it heavier it doesn't press it into the surface a lot more because you need it mostly vertical. So you'd have to make it quite a bit heavier - that's going to slow it down and make slipping a bigger issue.
And the software for working with polar bots isn't nearly as nice. I haven't looked in a year or so but last time I messed with mine I wasn't able to find anything that would accept normal gcode. It would either have to be processed into some other construct or you'd have to use a dedicated host program to send it commands.
You might be able to get away with something like an axibot https://www.instructables.com/id/AXIBOT/ But making it big enough to do full sheets of foam I don't think it would scale very well.
The approach a few others have taken of starting with a low cost diode laser setup that's either large enough or expanding it with new rails seems like a good low cost approach - as does the MPCNC or the Rascal or a number of other existing CNC designs. It's just a matter of what you can get make or buy parts for at the lowest cost given your abilities and time.
The slipping is an issue, but there's just so many other aspects of it that can go wrong as well. The mechanics look simple - but there are a good number of things that make them less simple than it appears. And if you want things to be accurate you need a MUCH larger machine than the area you want to cut because they loose accuracy big time near the edges of their cutting areas.
But the bigger issue I see - I don't think a need cutter head is heavy enough to work well like that. Even if you make it heavier it doesn't press it into the surface a lot more because you need it mostly vertical. So you'd have to make it quite a bit heavier - that's going to slow it down and make slipping a bigger issue.
And the software for working with polar bots isn't nearly as nice. I haven't looked in a year or so but last time I messed with mine I wasn't able to find anything that would accept normal gcode. It would either have to be processed into some other construct or you'd have to use a dedicated host program to send it commands.
You might be able to get away with something like an axibot https://www.instructables.com/id/AXIBOT/ But making it big enough to do full sheets of foam I don't think it would scale very well.
The approach a few others have taken of starting with a low cost diode laser setup that's either large enough or expanding it with new rails seems like a good low cost approach - as does the MPCNC or the Rascal or a number of other existing CNC designs. It's just a matter of what you can get make or buy parts for at the lowest cost given your abilities and time.
That's why I'm pointing at modifying the Maslowcnc, which is a ready to use solution for a quite heavy duty fully scalable polar cnc system, where the majority of these potential problems have already been adressed and solved. PC Software(Called "Ground Control") and arduino firmware as well as electronics are ready to use, you pretty much just need to
1) load a ready made sketch into the arduino (no programming needed!)
2) install the dedicated PC software
3) mount the gear motors to a stable structure that keeps the distance btw them constant - mentioned 2by4 - and attach the powersupply and cables from the arduino driver shield to the motors. Done with the original maslow parts. The rest is :
4) placement of the motor-2by4 and the work surface
5) possibly drilling the hole out in the belt drive wheels and/or attaching them to the motors, and figuring out the idler stuff
6) making your cutter/sled assembly, likely mainly based on the existing ideas of the maslow, just a bit lighter and with a needle cutter. Do take a thorough look at the triangulation link kit they have thought up to have the chains/belts always line up with the tool center. This simple solution removes a lot of the most complicated calculations and improves accuracy quite a bit. Their kit is likely well made, but can easily be implemented by other means, like a bit of meccano or something.
7) putting on a long belt piece and going through the full calibration sequence, which also takes care of setting up distance between all parts and size of the work surface in the software - automatically fully scalable. Vid example here:
8) put on both belts and attach them to the sled, ask maslow to center on the work surface......and you should be ready to go.
You don't need to buy the full Maslow package to do this. What is needed from Maslowcnc.com is Arduino and power shield, cables, powersupply, 2 of their gear motors and if you like possibly motors mounts. I beleive this amounts to just over 200$ or so. The rest can be sourced elsewhere as you prefer. An angle could of course be to buy a full kit, giving you the option to change it into a full size one later.
Only tricky bit should theoretically be to either find tooth belt drive wheels with the exact same diameter at chain/belt center......or figure out where in the arduino sketch to adjust this. Latter seems the more sensible solution, and I'm very sure a ton of help for this can be found by asking on the maslowcnc forums on www.maslowcnc.com. It's surprisingly active.
Last edited:
Foam cutting can be carried out by multiple machines, take a look. Obviously this is an industrial scale. Cheapest would be needle or hot wire. 3D printer wise, I've heard good things about the Creality Ender-3. It's apparently a very good entry model.
Similar threads
