[peteeng]

Hi Redding - I don't think the lifting gantry design changes the footprint over a moving gantry, moving Z design. It may make it narrower due to the removal of a saddle.

Nic - I'm just throwing up ideas at the moment since the design is still on the drawing board. Redding has to make his own decisions. He won't be a test subject, they have already been built by DIYers and they work. In your prior entries you do not mention a lifting Z, you do mention moving gantries. Nic its the nature of this sort of communication to upset people occasionally due to the poor "communication" it actually offers. Its not my intent to upset anyone just want to help people out. The lifting gantry has the same amount of moving parts as a "normal" machine just one extra drive screw. I'll leave it at that. Peter

[NIC 77]

Yep, I probably over reacted a bit there.

It is what it is, I can't seem to edit it. Looks like it's gonna be stuck there. Probably why I don't use facebook.

A couple points I want to mention.....

Any machine with a dual column rising gantry will need to be homed in the Z axis to prevent the gantry from being skewed in the Z. If it's dual driven in the X as well, then that should be homed also to prevent skew in the X. Some boards can't handle this.

From my perspective, a good DIY machine cutting aluminum needs a stiff, damp gantry, and a decent spindle, and that means heavy, making it a good idea to counterbalance the system if the entire gantry moves in the Z.

Both add complexity.

I don't have anything else to add to this.

Redding,

Don't mind us, this is your thread, please carry on and continue. .

[cnc99-de]

learn it!

[redding]

That's all good NIC. The design is pretty much the same as my last post. I still need to add some guard rails etc, but I think I'll just get the right size unequal angle and get it in the right sections. There are still some items obviously not finished, but so far good enough to order linear rails, ballscrews etc. I know that's not the perfect way to do it, but just keen to get hands on!

For those interested, All the parts I'm ordering total $1140 USD. Only things not included so far is most of the heavy plates. Check out my links below in PDF (you may need VPN - location HK or China, to access Taobao). I think you can't beat this pricing, but you need a smart way to get product to you.

[redding]

Minor Update - So all parts are being ordered now then may be a 2-3 month wait for a pallet to ship from China. May be quiet until then.

After that I'll measure up all of the parts, more CAD modeling then being assembly and buying some laser cut stuff. Talk soon!

[redding]

While waiting for parts to arrive - table design!

So I'll need to make a table specifically for this machine. An obvious choice is to weld something together, but I don't have a welder! What do people think is best for a machine like this (assume will weigh approx 100-150kg).

What is the best option here:
1. Just design an make a steel table. Buy cheap welder or pay someone to weld.
2. Design a table from wood. Perhaps add steel supports and a thick steel plate as the base.

Is a wood table acceptable? It would be a solid and very thick hardwood table.

[peteeng]

Hi Redding - If you make a steel table bolt it together. You then can adjust it. If your not an experienced welder, a welded table will distort badly on you and will be very hard to correct. If you make a timber table make it very heavy and stiff and glue it together so it does not move. Peter

[cyclestart]

steel, bolts, braces, gusset plates
I went nuts and built a wood table from 5 torsion boxes, strong as all get-out and but a really dumb idea in hindsight. Oh well at least it doesn't shift around, ( even if you want to move it )

[redding]

Minor Update - Detailed design almost finished. All Off-shelf parts ordered and looking to get ~10 parts CNC overseas. The table also built, ended up going the wood road, it's super solid so happy with it.

[peteeng]

Hi Redding - You have rails & motor on the tool plate- a design after my own heart. Should be more of them. How are you going to stop the Z from falling when the powers down? I use braked motors or gas struts. Peter

[redding]

Good point peteeng. After some research, it seems it is a common issue that I didn't account for. I think until I get setup, I'll put a simple endbit plastic protecting cover (magnets in the top) and just put it on when finished, but I will do the gas strut option considering I have already purchased the motor.

Added another picture also!

[peteeng]

Hi Redding - if your motor is double ended shaft type you can buy a brake to fit. Easier and same cost as a strut. You loose travel with struts and springs. A brake is easy to do, clean and solves some problems. Worth investigating. One day the power will blip (blow controller fuse for instance) and your Z will crash down gouge your job, break your bit, break your heart. Peter

[awerby]

I think your original design, with a fixed bridge and moving table, would be more rigid. I'd support the table from underneath, though, and not hang it from the sides of the bridge. That way the support points would be spread wider, and the whole thing would be more stable. Moving that high structure with all that stuff attached to it is asking for trouble, especially if you want to cut metal. Here's a nicely-documented build that shows what I'm talking about: https://www.wadeodesign.com/design-details.html

[peteeng]

Hi Redding - You have started ordering things a little early, but that's common as we do get excited about these things Its time you reviewed what you are trying to do before you go too much further. Your original intent was machining aluminium (aluminum) so you need a very stiff machine. Most twin gantry machines end up adding a web between the two beams, so you need to think about having provision for that. Your transverse extrusions, how are you going to cut them to length accurately? This impacts how parallel the rails will be. Have you left enough room to assemble the saddle? Have you thought through the assembly logic of the Z axis, saddle and gantry? Your design traps the Z axis assembly so it cant be assembled on the bench checked for square them slide onto the gantry. The saddle cars have to be on the gantry prior to all of that, so access to bolt heads and even the length of the bolts have to be checked. On my first machine I had to pull it down and add holes to plates so I could get access to these things. They do come loose and you don't want to have to pull the whole thing down to get to a carriage bolt just to tighten it! Keep at it. Peter

[Pipik]

Fully agree with peteeng. Also adding that you will have problem with establishing Xaxis ballscrew bearings with a ballscrew nut and Xaxis rails, because you place each other in a different plane.
And for god sake - why you have a hole between X-rails?? How do you want to establish parallel between them? Do you realise that it need to be at very good precission (catalogue 0,02mm i guess)?? If your gantry is one solid plate, then its a piece of cake to do that, but two independent pieces...?
Also - I dont think that screwing a Y-axis ballscrew nuts like that is a good solution. You must consider that there will be a lot of vibrations and force, so I dont think that four screws holding that sidepiece with nuthousing is a solid and firm solution.
What is the advantage of placing spindleholder in a distance of lower edge of the Zaxis plate? Otherwise than that you want to make some obstruct with that Zaxis plate? And it will obstruct, you can bet ;-)

You ordered parts very early, peeteng was right.

[Pipik]

Hi Redding - if your motor is double ended shaft type you can buy a brake to fit. Easier and same cost as a strut. You loose travel with struts and springs. A brake is easy to do, clean and solves some problems. Worth investigating. One day the power will blip (blow controller fuse for instance) and your Z will crash down gouge your job, break your bit, break your heart. Peter

I think that until this day happen, there will be a huge amount of broken bits anyway, so one more... :-D :-D :-D

[redding]

Most twin gantry machines end up adding a web between the two beams, so you need to think about having provision for that.

Yes this is good point. I had been planning to use the 8080 profile slots and adding some simple, thick Alumun L-Extrusion bolted between the two as a intrum solution. Once the CNC is up and running, hopefully I can can slowly cut something to add between these also after that. Is there any common joining method used?

Your transverse extrusions, how are you going to cut them to length accurately? This impacts how parallel the rails will be.

I have ordered these from China, so I assume they will not be perfect. I plan to take them to either a CNC company, or take them to a local aluminun profile company to cut accurately. Is there any DIY way at home to do this?

Have you left enough room to assemble the saddle? Have you thought through the assembly logic of the Z axis, saddle and gantry? Your design traps the Z axis assembly so it cant be assembled on the bench checked for square them slide onto the gantry. The saddle cars have to be on the gantry prior to all of that, so access to bolt heads and even the length of the bolts have to be checked. On my first machine I had to pull it down and add holes to plates so I could get access to these things. They do come loose and you don't want to have to pull the whole thing down to get to a carriage bolt just to tighten it! Keep at it. Peter

Yes correct. See attached pictures. I plan to:
Step 1: Assembly Y-Rails on Gantry on bench first. Once perfect, add on the connecting plate (I call it YZ plate). This will already have the bearings on it as shown in the picture. Easy access to the bolt heads also. The most outer bolts for the Z side bearings will be harder to access, but my M6 allen key will fit up between the 8080 profile and YZ plate fine. In this step you can connect to the Y- Balscrew also.
Step 2: Pre-assemble Z-axis on the bench (rails, ball screw), and slide onto the bearings.
Step 3: Screw from the rear to connect the YZ plate to the Z-bearing.

It's not perfect, but it seems all bolts, lengths of bolts can be accessed. Do you think there any risk with this method?

Fully agree with peteeng. Also adding that you will have problem with establishing Xaxis ballscrew bearings with a ballscrew nut and Xaxis rails, because you place each other in a different plane.

Hi Pipik - sorry can you help explain what you mean by "Establishing x-axis ballscrew bearnigs"? Both x-axis ballscrews are mirrored between the left and right side. Exactly the same location and setup.

And for god sake - why you have a hole between X-rails?? How do you want to establish parallel between them? Do you realise that it need to be at very good precission (catalogue 0,02mm i guess)?? If your gantry is one solid plate, then its a piece of cake to do that, but two independent pieces...?

Sorry can you also explain where the hole between x-rails is? Do you mean the gap between the two gantry 8080 Aluminun profiles?
The x-axis rails/profile (80120) is joined together with 4 pieces of heavy 8080 (8kg/meter). There is also a laser cut 12mm steel plate on both ends holding these together.

Also - I dont think that screwing a Y-axis ballscrew nuts like that is a good solution. You must consider that there will be a lot of vibrations and force, so I dont think that four screws holding that sidepiece with nuthousing is a solid and firm solution.

This is a fair point. What do you recommend to do? I could move the ball screw over further, but then it would be hanging off the 8080 profile edge.
See "Side View" Image. This gap is filled with a 40x40x62mm block. Other options I can think about:
1. Total redesign putting the Y-Axis rails on the top and bottom of the 8080 profile to bring the distance in (but losing Z-height)
2. Move the bearings further accross, and ensure beefy 20mm plate holding the bearing blocks
3. Make a fully custom part that attaches to the ball screw and the YZ plate directly (not using the off-shelf bearing block, but a custom CNCed version).
Any other ideas?

What is the advantage of placing spindleholder in a distance of lower edge of the Zaxis plate? Otherwise than that you want to make some obstruct with that Zaxis plate? And it will obstruct, you can bet ;-)

See picture attached. I have put a hole pattern so it allows me to move up/down the spindle position depending on what works. If the bit is very short, yes I would plan to mount towards the bottom. If it's very long, then where it is now (50mm length) there is no obstruction. Is this what you mean?

[peteeng]

Hi Redding,
1) The construction extrusion has slots so as long as the rear is clear you bolt it on
2) They need to be squared on a mill in one go. If the piece gets moved or over ended like on a saw at an extrusion company then I think its out of spec
3) At home need a machine big enough and square enough to DIY
4) That's about it from me... Keep at it Peter

[redding]

Hi Redding,
1) The construction extrusion has slots so as long as the rear is clear you bolt it on
2) They need to be squared on a mill in one go. If the piece gets moved or over ended like on a saw at an extrusion company then I think its out of spec
3) At home need a machine big enough and square enough to DIY
4) That's about it from me... Keep at it Peter

Thanks so much for all your help Peteeng. Just curious, what did you think about the assembly method.

Did you also understand the comments from Pipik and can share more light?

[peteeng]

Hi Redding - Hard to say without looking at actual geometry. If your clear in your own head it works then that's good. Just bringing things to your attention. Pipik is concerned about how you are going to accurately position the motion parts when they are on different structural parts. This is a two fold problem. Firstly you have to correctly place the structural parts and if they are out then you have no chance correctly placing the motion parts. But people do it. It will come together, just may not be at the accuracy you expect (the CAD model is perfect the world is imperfect). You will need good straightedges, good levels (machinist level preferred) dial gauges & patience. Keep at it. Peter