Thanks for this thread - it’s about half an hour in front of where I am, and all this talk of svg and paths is a bit frightening! 
In a week or three we will see how much I think I’ve learned from it!
I am wondering if I actually need Inkscape? I am reasonably proficient in OnShape, and if I use Kiri-moto perhaps I can just get them to talk to one another? That will be an issue if I want to use graphics or lettering I suppose, as OnShape doesn’t really do that simply.
Thanks for the kind words.
Tbh I’ve worked few times with CAD in the past, but for some reason I find those programs counterintuitive. I rarely find the function I need, and the damn thing never seems to want to do what I want it to do 
even when I am watching tutorials:D
It’s probably me, more used to photo/designer/publisher software, and most of all: my limited patience:) But… I installed Fusion 360(?) yesterday. Let’s see what happens next
Thanks for the suggestion Darwin, tbh I wasn’t planning on using Inkscape, but in an effort to find the reason why my dxf wasn’t properly scaled, I figured to give it a try to convert my files.
Must say that Inkscape improved in the 5 years since I’ve last used it.
So I am going to follow your advice and play with it.
Don’t worry Peter, I am sure you’ll manage it!
No not at all, from what I can read Onshape can export flat files to dxf.
I am taking the long path here because my CAD skills are lacking, and I wanted to start using the machine so it no longer collects dust from standing still
Pretty sure that you don’t need Inkscape.
It is one tool, but only one, and if you’re familiar with OnShape, enough to do what you want for drawing in it, then you probably don’t need a different tool.
I’ve used InkScape to convert PNG/JPG files to SVG for stuff, but most of the time, I get lost enough in Inkscape to just close it without actually getting anything done. (I might have to go through some sort of primer to use it at some point.) The only thing that I’ve managed is to convert some files, but not creating my own.
FreeCAD has been pretty good for 2D, and not bad for 3D, so I use that, and it works well enough for everything that I’ve wanted to make
haha, user preferences 
I got triggered by your post to boot up FreeCAD once more, yesterday evening, got lost in it. Didn´t manage to do anything so I decided to stop trying
So, I know I had it coming, but this happened while trying to mill a larger oval shape of approx. 1000mm;
My machine seems not to be square. I got away with it milling some text, but for the job I intend to use it, it´s not okay.
I experience 2 issues. Most likely related;
My machine seems to travel sideways. At first I thought it came because the wheels don´t have enough traction on the board. But I think my 3D prints aren´t square;
I´ve tried to change pressure on the tubes on the XY part, but I can´t get it square. If I loosen the screws to much, the tubes just drop to the ground. If I tighten I hear some cracking noise in the parts. So that´s no option either.
I´ve read someone in the forum used tape on the tubes to get it squared. But I can not figure out where to place them to solve my issue. Does anyone have an suggestions on this?
Another thing I notice, is that I have no idea how far the wheels should be from the edge. I read the FAQ multiple times now, but I seem to miss an indication for this distance.
Basically, the track is that the belt is as straight as possible going past the tubes to the idler pulleys. At least that’s what I used as a guideline. The belts are connected to the holders, and go from the holder to the idler, then over tot he motor pulley, then the other idler, and the far holder. I lined stuff up and tightened my X tubes when I could sight down the belt from one holder to the opposite side, and the belt was as straight as possible by sight. This leaves a very small gap between the table edge and the Z tubes.
I put a washer between the printed parts and the wheels, which increases the space a little, but as long as the wheel spins freely, probably isn’t important.
This may not be possible with your setup but a lot of people end up setting up some type of track for the wheels to run in. It helps stabilize the x drift when the machine is moving longer stretches in Y. My version is in the link below, along with my solution to the squaring.
Well, I thought it was fine too based on my observations, but reading your input on the small gap, I might try to bring them closer to the table - so close they just don´t touch. It might improve my issue of going sideways, but probably not enough to run square.
I am sure it is possible to do, however, to find a piece of wood in my area that´s straight - that´s going to be a challenge.
I was already looking to add some linair rail as a solution, but I suspect that even then I would need to get my tubes straigth. No?
My tubes had about an eighth inch of space to the table.
Far more important that the belts are straight. The 2 belt segments from holder to idler pulley should form a straight line. If there is any difference in angle of these 2 segments, it will cause linearity issues that are impossible to address in firmware or gcode
I’d just have it spray the entire insides of the enclosure with flame retardant expanding foam.
I think you’re right. I found that not having a parking spot that was square compounded the problem. So linear rails and a squared stop block on the “bottom” end of the table that I could manually reset against helped a lot.
Ok, I’ve checked it and the seem to be 5 mm out of line.
So will adapt this whilst I am going to try and straighten my tubes once more. But I think it will be a bit waste of time, as I tried it few times now and always seem to end in exactly the same situation as I started:)
Yes, I figured, I always start by bumping my parking spot, it helps to align, but somewhere during travel it seems to move.
Going to think about how I can solve this. Was looking at the alu XY main too, as this would always be square, but the quotes I received were pretty expensive 
I hate to say this, but I don’t believe you can straighten tubes once cooled. You’d have to somehow heat them up, straighten them, and then keep them straight while they cooled again.
tbh, I don´t know if it will be the tubes or the 3d printed part.
I noticed that some bearings don´t touch the tubes, and I can see some warping on the 3d printed parts. That makes me believe it isn´t the tube but the printed parts.
Small update while I am looking on how to get my machine squared.
My wife thinks I am getting nuts by ordering 15 meters of vacuum hose 
The delivery guy handed her a large box containing this beauty;
It´s a perfect fit with the original LR2 part; so if anyone needs a hose, the Nedis vacuum hose seems like a good deal for a very cheap price!
Now I am waiting for my cyclone filter to arrive this week so I can give it a test
Yesterday the delivery guy handed me an early Chrismas present
My dust bin and cyclone filter arrived. Yeeey!
Some pictures, for those that like to see them:
As you can see, I opted for a metal 60 liter bin, after reading @BingBadaBoom 's feedback on the PVC version. I wasn´t keen on securing the bin from the inside
I will have to say though, that after testing the metal bin, I nearly got a heart attack, as the lid of the bin actually did got sucked in because of the vacuum. It made a hell of a noise
Good thing that I can open the valve on my vacuum cleaner to avoid further issues.
As you can see on the pictures, I added adapter mounts to perfectly fit my 15 meter hose, and the hose of my vac cleaner. What I don´t like is that the mount of the vac cleaner isn´t that solid squeezed in, so I´ll have to keep an eye on that so it doesn´t disconnect. Time will tell.
During testing I did not notice any decrease in performance, so that´s a good thing
Now I´ll need to start thinking on how to mount it to the ceiling, so the tube isn´t in the way of the moving LR2. If anyone has great idea´s on that part - shoot!
What happened there? You know that you could add a spring valve in the lid, so it opens instead of imploding:)
Yes I can
I was actually watching your video this afternoon! It’s great inspiration for me.
There’s enough room at my ceiling, that’s why I actually needed a long vac hose. However I am not so keen on adding a swingarm (would need to drill into the wall, the bricks are really rock solid and takes a lot of effort to align 2 hinges)
So I’ve seen your solution of using a zero gravity tool holder. How did this work out, and by looking back to it, do you have a preference between the two?
It´s time to move on in my journey and have a look on how I am going to try and square my L2. I suspect that part of my experiences are caused to bad printed LR parts, but also due the fact that my wheels are slipping on the slippery surface I´ve got.
I´ve seen that some users add a beam to keep the wheels aligned, some make grooves for the wheels to run on. I was considering linear rails as a solution.
I was never planning on building a full sheet CNC machine. It just happened because I can do it And that´s now causing an issue to find a linair rails long enough to fit.
So, I found this earlier on Thingiverse:
It is basically something bolted on the stock LR parts, and using the same principle for the Z/X axis. For the rails to mount they use this CNC alternative:
Now, I am considering to tackle this by a different approach.
I was hoping to mount M8 bolts /threaded rods to the tube, and mount those to the table. That way I can create some overhang to reach the outer sides of the full sheet without making my table larger.
I would add nuts between the rail/table to stabilize the whole on my required height.
A sketch to make it more clear:
Like you can see, I would overhang the “rail” ± 20 cm on each side, that should be enough for the bit to go from 0 to 244cm (full plate lenght).
For mounting M8 bolts I contacted my local metal shop, they did an experiment for me, to check if they could mount blind rivet nuts to a 25mm round tube
This is their result:
So I believe my idea of mounting this using bolts is feasible. What do you guys think of it?
I am a bit worried on the distance I would choose between each of the holders (39cm) but since the X axis is longer and uses a similar principle, I believe that I might get away with it.
any suggestions?
