Yes it’s an ITEM Profile 8 120x40 currently the light version. Not yet decided if I may change it to the heavy version to have more reinforcement in the axis. If the heavy version fits in CAD I probably use it. Did not see the heavy version could be ordered online easy as well… If your not a commercial buyer it’s pretty hard to order at ITEM directly but there are enough online vendors selling ITEM compatible profiles. But not any version of them.
You should check out MISUMI. They have a heavyweight option for the 40x120.
Thanks Andre,
already got some EU vendors over here selling the ITEM compatible parts with easy shipping to Germany… At the beginning of the design I wanted to buy the original ITEM parts because of accuracy and the available CAD files but looks like the sell only to commercial buyers directly… so I will give it a try with the compatible ones…
1 Like
My German is very poor but;
Widerstandsmoment W x: 20,26 cm³
Widerstandsmoment W y: 54,47 cm³
Seems to be saying that the moment of intertia of that beam is 20cm3 about the narrow axis and 54cm3 about the wide axis.
It would be worth comparing that with the carbide3d extrusions before designing anything on this extrusion IMHO.
EDIT - Sorry, I think I should have chosen by units not google translate;
Trägheitsmoment I x: 40 cm4
Trägheitsmoment I y: 326 cm4
But still, I think these are small numbers compared to the C3D extrusions
It’s a good point I did not compare those numbers. I was looking for profiles used by others for comparable CNC builds. Do you know the numbers of the r carbide extrusion? Compared to the structure of the heavy version it looks like the section modulus should not be a big issue. But that’s maybe just the opinion of the noobie building he’s first CNC axis… that’s why I did this post in hope of getting some things I missed or good tips.
Do you thing it will be a problem?
I’m not a mech engineer, so the calcs below are subject to correction by people with the right qualifications…
If I have the calculations correct (and a correctly assumed Aluminium alloy) for the C3D extrusions then the X beam would be approximately;
58cm(4) along Y axis
106cm(4) along Z axis
Given that this alternate extrusion is 150% the height and 70% the width with a similar wall thickness (the stuff in the middle doesn’t do that much) that all seems reasonable.
This extrusion would therefore be 300% of the stiffness in up / down flex but only 66% of the stiffness in front / back flex under load.
If you were looking to upgrade the performance of the machine, a larger cross section in both directions is likely to pay dividends. I would not replace the C3D extrusion with this one.
The 120x80 profile they list on the other hand;
Trägheitsmoment I x: 277 cm4
Trägheitsmoment I y: 580 cm4
would make for a pleasingly solid frame and is sufficiently stronger than the C3D extrusion to be worth the cost and effort of replacement.
There are various online calculators for determining deflection under load of a beam once you know the moment of inertia and the material which this vendor helpfully provides.
It might be worth having a read of the discussions on DIY builds over here;
http://www.mycncuk.com/forums/282-Gantry-Router-Machines-Building
and looking at some of the tidier designs such as this;
http://www.mycncuk.com/threads/10658-El-Beast-Initial-design-phase-comments-and-critique-welcomed!
I have found it to be a great source of ideas and examples of what has worked for people and what hasn’t
Are you planning to replace the steppers and control electronics on your machine to drive the additional rotating and moving masses?
I would also suggest looking closely at the Shapeoko Pro before doing any upgrading, decisions the Carbide team took such as running a longer X extrusion that spans across the Y rails to provide extra running room and space for protection bellows over the X rails all make a lot of sense. If you’re buying new extrusion you really shouldn’t be restricted by the constraints of the old parts.
HTH
2 Likes
Yeah me neither so i might have been a little to easy in some of my design decisions. I guess that was also somewhere in the back of my mind as I post my design first before i started to build the mockups 
My first idea was to use a 120x80 but i wanted to gain more space on the Y Axis… but after soem research and your calculation i think i try to fit the 120x80 heavy profile in my design. This also will be more easy to mount with 6 screws to the plates… I am a little concerned regarding the mounting of the carriage plate the new profile may block off some mounting screws for the carriage plates.
Item has a defelction calculator on their website will give it a try but dont really know operating force will be on the profile… maybe i need some more research in that…
I will also look on the websites you mentioned looks like a good source for ideas…
Yes i started planning to replace the GRBL board even before the axis idea… Currently I have to options in mind.
Option 1 Estlcam Based Controller:
Great controller solution runs in a lot of hobby machines…
Option 2 EdingCNC Based Controller:
It a real advanced controller with great software but not very cheap 
For the motion itself I also have two options in mind:
4 of these Sorotec Online-Shop - Stepping Motor 1.8 Nm 4.2 A with matching digital stepper drivers
or
4 x of these closed loop systems https://www.sorotec.de/shop/2Nm-Closed-loop-Integrated-Stepper-Motor-IHSS57-36-20-.html
The goal is that the new overall controller solution may also fit in a future machine without too much modifications. At the moment I tend to use the Eding solution with the Closed Loop steppers. Will stretch my budget a lot but also the experience i get out of building it.
I will also give the Pro a closer look… The announcement started that hole thing here again for me
very appreciate all the input here so far… i really like the shapeoko community
I have upgraded my XXL to a Gecko G540, UC-300 eth and UCCNC and it is a great combo definitely worth considering. I’m not sure what the cost is compared to your other options.
Bit of a write up here
Here’s a thread where Mike summarised the controller and stepper parts he used;
I decided some time ago though that I would not bother with open loop steppers again, the incremental cost to closed loop steppers from people like stepperonline or to DC servo motors is small enough compared to the time and other costs involved that I don’t think it’s worth the trouble of open loop systems any more.
The Y plates can really be any shape you like, if you look at the X Carve Pro you’ll see their Y plates lean back quite a long way to keep the Z axis above the Y rail blocks. This increases the torque on the X beam but on the 80x120 deflection is likely small. There’s no need for the Y plates to stay within the bounds of the machine footprint, the constraints are your chosen X beam position and maintaining a reasonable distance between the Y rail blocks to minimise the leverage on them. What limits Y travel is the total length of Y block - gap - Y block, just as on the X rail.
Do you plan to completely replace the new Y plates when you go over to Y axis ballscrews? Based on the rendering, you could add bracing between the horizontal and vertical to stiffen up the side to side and rotational flexing there.
As you’re using an XL I suspect the carbide extrusions are more than sufficient for your Y axes.
Do you plan to deal with the lack of rigidity in the machine base?
Regarding your design, how will you attach the rails to the extrusion? Just nuts? Do you have any plans for keeping the rails precisely flat?
I’m thinking about doing similar things to my Nomad and I was thinking of extrusion but ruled it out for lack of precision. My thought was to buy a pre-milled 15mm or 20mm EN AW-7021 plate, bolt the linear rails to it and stick it where the Nomad’s X-axis rods are with a bracket or by putting some beefy threads into the plate.
In case it helps, I’m using the CNC720 on my Nomad and I love it 
You might find my thread helpful, I compare controllers and drivers.
Since you’re cost-constrained, it might also make sense to keep the steppers that came with the Shapeoko. If you’re in Germany I could probably send you one of the adapter PCBs I made for the Nomad, though I’d have to think some more about how useful it is to you. My first board might work out of the box, depending on which stepper drivers you use. The later board is nicer and more compact but also very focused on the ESP32 and TMC5160.
LinuxCNC was also on my list of controller solutions a while ago but dropped it in favor of the Eding CNC. The Controller and Simple BOB Breakout board look like the better way to go al least for me. But I guess LinuxCNC is also a really cool solution. How happy are you with your solution?
I am not really cost constrained for the mod… Cost will just dictate the pace if the mod I will definitely replace the steppers. The controller and steppers could be used in a future machine as well. But thank you for the offering.
The only way I decided not to go was buying a complete new machine for now. Since I love my Shapeoko and the modding of it.
I will use my shapeoko to cut the plates and also the surface the material for the plates. if that does not work as intended I will look for a way to get it surfaces somewhere. I have more concerns how flat the cut of the extrusion will be. I dont have a belt sander or something similar if the cuts are not flat. Bujt the first comments of others ordered at the webshop said they are good cuts so we will see…
I mean the linear rails. They need to be as straight as possible to keep your machine accurate. Since they’ll be attached to the extrusions, if the extrusions aren’t perfectly straight, the rails won’t be either.
I was wondering whether you had any ideas on how to mitigate that. Like will you face the extrusions or use shims or something to get the rails perfectly flat?
I already have a aluminum t-slot bed as replacement for the mdf board. This gives the base a pretty good rigidity for now. No plans (so far) to do more here
Hopefully that should be covered with tightening of the screws… if you look on other machines using extrusions like this your can do a lot of alignment by adjusting the screw tightening.
This works well for the positioning of the rails to each other. Tighten on and the just hand tighten. Then you the mounted plate to positioned the other rails correctly with its blocks.
If the extrusion if really uneven this might be a problem. But I’ve seen many builds with extrusions like this and it does not seem to be a problem.
With UCCNC that I’m currently running… literally could not be any better. I’m learning to make my own screenset for it which is made really easy in UCCNC
I was running LinuxCNC for a few years, and it was good, but really hard to set up, it took me a good week to get a shapeoko 3 on a Gecko g540 working… It has a huge amount of functionality and customisation, more than any other software I’ve seen, but it all has to be done in code and that started to drive me nuts. I wanted to set probing cycles a certain way, add buttons to the screen to do things like raise the Z and go to back right corner while loading stock… very possible with LinuxCNC but I couldn’t get it figured out.
UCCNC has been a dream since then… <10mins to set up, you want a button, go to settings and click ‘Add button’… you can move buttons around, create your own macros linked to buttons… It’s all Very easy
A lot of people say good things about MASSO controllers too
I finally was able to rework the first design with the 120x80 Profile… Also changed the side of the ball screw motor to have the maximum drive way of the Z plate to the right where the zero point should remain.
Hope I can start the mockup in December!
3 Likes
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.