[Admiral]

My last post was a shambles and I apologize, I wanted to get some feedback but did not have the time to do it properly.

This should be a more comprehensible post.

Lets begin, I started working operating cnc machines in 2007 and was bitten hard as soon as I found out I could actually build one for myself I started planning.

My first design was pretty basic a desktop design but I quickly realized that I wanted something larger, This was back in 2008 as I got my hands on a probotix 3 axis kit.



My next design was severely over complex and would have been difficult to access, I wont bather to much whit this design but I worked around whit it for close to 2 years because I was hung up at a above below support system for the X-axis.



Now back before the summer I was finally able to get hold of cheap material provider threw my workplace and started from scratch.

I wanted to build for stability and rigidity, Extruded aluminum profiles was the choice to go in my mind, easy to work whit and strong, my third design was a fairly straight forward gantry model, biggest difference from the majority would be using 4 guide rods instead of the traditional 2 on the X-axis.





As I continued I realized that 20mm guide rods would have a significant deflection and switched to fully supported rods and I made one major change and departure from the norm, on the Z-axis I made the guide rods be the moving part to increase rigidity when moving at full extension.



I also added an extra slide to each, mostly because I had extra from previousa projects and after tests there was a noticeable gain in stability.



Next up is going to be actual build pictures.

[Admiral]

The beginning of the build was both fun and frustrating as noting usual works out the way you plan, but at this point there was not really any major problems.

[Admiral]

Now i ran in to a bit of a problem the bars I made connecting the rods to the profiles worked great however the pillows did not work they would not properly seat in the and whenever I tried to tighten them down they would twist in the socket, I decided to rebuild them more sturdy.

[Admiral]

This worked allot better, I simply treaded the pillows on the rod and then screwed them on to the bars and perfection this is also where I changed from 2 to 3 pillows.

[louieatienza]

Pretty cool design. I'm a firm believer in using moving rails for the Z though it does require more planning. What fo you plan to machine primarily?

[Admiral]

[Admiral]

Alu, plastic and wood planed but it also depend on how it behaves.

[xaruum]

It certainly appears that you have put a lot of thought into the building of your new machine. It looks very clean, and you have put in a lot of good work.

Judging by the over and under supports for your gantry, it is obvious that you are conscious of issues with rigidity with a gantry that tall. However, I believe your gantry is so thin vertically that all benefit is lost from the over/under rails.

If I understand things correctly ( I am still reading through "Machine Tool Design and Numeric Control ), given that the force from the leadscrew is applied all the way at the bottom and the resulting force from the endmill pushing through material is being applied about ~30 inches above that, it will cause the gantry to act like a ~30" long lever with the side rail supports being the fulcrum. In other words, both the stepper(servo?) and the cutting force are going to cause the gantry to twist against the support rails.

You may want to consider making the extrusions that bolt to the pillow blocks longer and adding 45degree supports to reinforce the uprights. Especially if you are expecting to make deep cuts in hardwood or thick aluminum.

Other than that it looks like great start. Keep up the good work.

[dbsharp]

you'll be making chips and dust in no time!

[Admiral]

xaruum your absolutely correct, it's a problem that I struggled whit designing it because i wanted at least 200 mm vertical movement in my design sense I will be (possibly) operating on articles that have some height.

The 4 vertical 'pillars' holding the gantry are 90*45 mm extruded aluminum bought for one of the leading manufacturers i talked to them about my problem and this is what he recommended, try look weak but are amazingly strong I don't think that general flexing from material weakness is going to be a problem.

But your are absolutory correct about the lead screw, it was a problem, I got a hold of that long lead screw a while back so cheap that it had to be it, now I wish I would have taken the time to design the idea around a twin 25mm lead screw design, but this is a first machine and it's been expensive enough.

There is going to be strain on the lead screw and it will suffer from both flexing and the small jog you will inevitably get whit that big a difference between your point of contact and drive

But like you point out structurally it could be a problem but there pictures are a little deceiving in that point, the 90 angle 's are not the only thing holding them together, in each of the T slots there are locking bolts holding the profiles together, so the bottom of the gantry is held together at 8 points on each side any twisting will be from metal flexing structurally the profiles have to literally fail. (not impossible but looking at the numbers it's unlikely.)

I'm more consigned over vibrations and over time fatigue.

I did consider a 45 support at the bottom and have more material if i need to do that but for right now the feel of it is that it might not be necessary, the top horizontal bar are 90*90 mm pure over till those two will each hold more than a ton before failing but they make for a very strong top of the gantry.



Thanks for the input most people here know more about building machine then me so please comment I want to had both praise and criticism, this post made me think and have to reconsider structural issues and I now think I need to do a test and see how the gantry behaves under pressure how much will it flex at the bottom when I put pressure at the point of contact.

[louieatienza]

I don't think you should have too much problems with the single leadscrew moving the gantry , provided that the linear bearings keep their relation when a force is applied obliquely to the gantry. Namely any force that causes one side of the gantry to move forward or back in relation to the other side. It looks the gantry assembly is pretty strong however. You could probably test this easily by disconnecting the nut carrier, blocking off one side of the gantry, and pushing the other to see how much force it takes to move the free leg in relation to the fixed one. Possibly a plate below the lower stretcher of the gantry would prevent any tendancy for that section to "paralellogram" ...

[Admiral]

Just thought i would post one or two pictures on the progress mostly drawing cables at the moment easy but time consuming.

Also did some simpel run test, worked proboerly so far some tunning issues and need to calibrate the exact distance travel.

[steve123]

louieatienza:

Possibly a plate below the lower stretcher of the gantry would prevent any tendancy for that section to "paralellogram"

I agree with louieatienza. I think that adding skins is the cheapest, easiest, lightest way of stiffening the structure. I would also add (thin) aluminum plates to the outside of the gantry verticals, and to the back of the gantry. Use lots of screws. Fortunately you have t-slots there!

One other thing I notice is that it looks like the bar on top of the top gantry linear bearing is held on only by angle brackets. If that is the case, it is going to want to twist. You could also put screws through the center of each end into the t-slots of the verticals.

You have a nice clean, strong build. Good work!

[Admiral]

Thanks steve123, I think there is a tendency to underestimate the strength of these materials but adding plates to stiffen the "undercarriage" up is one idea that I have thought of.


Right now however less is more, there is no reason to stiffen it up more then you need to alu is light but it gains weight quickly so before the machine is actually going to be doing work there is no need to start adding unnecessary structural material, I'm going to buy a whole load of more angels to add on and those things are crazy strong believe me they won't flex but they might all out break if the strength is there.

Jonas

[louieatienza]

Thanks steve123, I think there is a tendency to underestimate the strength of these materials but adding plates to stiffen the "undercarriage" up is one idea that I have thought of.


Right now however less is more, there is no reason to stiffen it up more then you need to alu is light but it gains weight quickly so before the machine is actually going to be doing work there is no need to start adding unnecessary structural material, I'm going to buy a whole load of more angels to add on and those things are crazy strong believe me they won't flex but they might all out break if the strength is there.

Jonas

I used similar if not the same extrusion for my larger router, 4545 and 4590 from 8020, in conjunction with ROLLON TLV43 rails and blocks. Even though the rails are steel, and bolted to the extrusion, I still get some flex though very little. You do have it configured a bit stronger than mine.

As to the skins, you don't really need a lot of thickness, and it probably would end up being cheaper, lighter, and stronger than going crazy adding gussets or angles all over. Even a composite material like phenolic would work while adding some vobration damping as well.