[ardenum]

Hi Ard - Its looking like laser cut and fabricated parts to achieve this design. So there will have to be a minimum run to make the $$$ reasonable. Until its designed and fully spec'ed can't estimate cost. From prior projects I expect to make 3 or 4 sets of parts to meet minimum orders at cutters. But I haven't eliminated casting yet. If it boils down to fabricated steel I think I'd revisit the gantry design, this cures the column torsion but creates other issues..... Seems we may have gone full circle...Peter

I feeI I'm more on a möbius strip then anything. Someone on my build thread over at mycncuk suggested my switch to double column lifting gantry. Here's the video they posted.

[peteeng]

Hi All - Heres an interesting guy with good work. Peter

https://wrbl.tumblr.com/mill-build

[peteeng]

Hi Ard - The only issue I see with a lifting column is say you have a 200mm high object and you want to mill 100mm down into it. Means you still need a very long tool (100mm+) and the gantry will be high. But I think it has merit. Deans machine very nice but it has a high speed spindle. Peter

[peteeng]

Hi All and Sundry - Its wet here with a flood warning... So when in Rome do as the Roman do. Or when in steel universe use steel when in aluminium universe etc. That's how I feel about these things anyway. So I built a steel machine base from 100x100x9 SHS and its stiff. Current stats in N/um are X 21 Y 13 and Z 24. Y would be improved with an extra spinal SHS. So I think this ends here. The steel stats are column 16mm thick x 650mm high arm is 12mm steel and saddle is 40mm steel or 40mm aluminium not much delta on that. Envelope is X600xY250 & Z450. I'd have stainless steel laser cut and I'd weld or braze it together, finish machine then have it electropolished. Don't like paint as it mucks up fits... The assembly is 280kg with the base being 105kg. So thats OK...

Since I'm keen on CSA I'll revisit the gantry. Yes the Mobius is alive and well here... Look up Moebius the French artist wonderful stuff.... Peter

[peteeng]

Hi All on sunday - I revisited Milli No2. At No2 CSA didn't exist and the direction wasn't as firm and she was big, 1200mm at the front. So I cleaned it up reduced its footprint to 1000x1000 and loaded it up. It scored X 60, Y36 and Z32. The model had some connection issues and the top beam was not across the columns. So I'll tidy this up and start here. Milli-G off we go! Peter

[ardenum]

Are those columns something you make yourself or bought?

[peteeng]

Hi Ard - I shall be casting the columns and bridge in CSA or fibreglass . The base & gantry will be same (CSA). I looked back through my early models and I was restraining them using the feet of the base. This made them much stiffer then they should have been. My current method of the post on the bed is much closer to reality. Peter

[peteeng]

Hi All & Others - Here's a quick summary of the past. There were some side trips. Peter

[ardenum]

Would you mind sharing the material properties for CSA you're using for FEA? How much N are you applying?

[peteeng]

Hi Ard - The CSA supplier has given me copies of test sheets for the material. With no aggregate it is 26Gpa with blue metal aggregate it is 40GPa. The applied load does not matter as you should normalise it to a N/um or Lbs/in. Most machine parts stiffness are expressed as N/um so I stay with that. I use 1000N (100kgf) as the design load. This means in Millis case I'm looking for a deflection less then 50um as the target static machine stiffness is 20N/um. This is in each axis. Peter

[peteeng]

Hi Ard - CSA concrete is similiar to UHPC .



Peter

[peteeng]

Evening All ! I have tidied Milli-G and ran the model. I added the bridge and corrected the connections. Stats in N/Um are X 18 Y 36 and Z 20. Creeping up to the mark. I'm now wondering whether to use a fixed and lifting gantry vs the moving/lifting gantry. Milli-G is 100kg heavier then Milli23. But I can par the base down a bit. The fixed columns are attractive but that puts the table drive down in the muck.... but it makes sense.... Peter

[peteeng]

Morning all - I looked at No2 today and decided to drop the columns onto the bed to see what that looked like. Very good. The first version had the very hollow gantry and it lozenged. Since I'm casting it I can control the internal shape so I made it thicker. So now the stats are X 20 Y 28 and Z 37 so were on our way. So now rebuild from scratch using all the new knowledge. Only issue now is that parts being concrete means I can't machine them flat or true. Seems I'll have to make a wet machine to do this... maybe the stonemason down the road can help.... Peter

[Mud_]

Morning all - I looked at No2 today and decided to drop the columns onto the bed to see what that looked like. Very good. The first version had the very hollow gantry and it lozenged. Since I'm casting it I can control the internal shape so I made it thicker. So now the stats are X 20 Y 28 and Z 37 so were on our way. So now rebuild from scratch using all the new knowledge. Only issue now is that parts being concrete means I can't machine them flat or true. Seems I'll have to make a wet machine to do this... maybe the stonemason down the road can help.... Peter

Why not put the spindle on the other side of the rising gantry? Shared plane between the x and z bearing top faces seems good for manufacture/accuracy. The carriage gets stuck between the pillars, but it's a small loss of x travel.

[peteeng]

Hi Mud - Good thought. I have looked at that before and it hasn't quite worked due to loss of travel . But worth a new look. Casting allows shapes that std tubing does not so maybe new opportunities Also will put gantry rails on the diagonal as other threads have shown that to be stiffer,,,, Lots of things to explore...Thanks Peter

The question now is - Is the lifting gantry worthwhile? Do I go back to a Z axis? ??? more Q's to ponder

[peteeng]

Hi Mud - Like this? But allowing the gantry to lift. Peter

[peteeng]

Hi All- Interesting German company making UHPC machine parts. Peter

https://www.rampf-group.com/en-us/pr...ds-components/

[ardenum]

Hi All- Interesting German company making UHPC machine parts. Peter

https://www.rampf-group.com/en-us/pr...ds-components/

Didn't know that 3rd option (composite constructions) was viable enough for this type of company to make, I guess that solves the question whether filling steel tubes with EG is a yes or no.

[ihavenofish]

Hi All- Interesting German company making UHPC machine parts. Peter

https://www.rampf-group.com/en-us/pr...ds-components/

They changed their name / merged, or something, used to be called epucrete. Thats the company that made/makes the composite based of DMG machines. the middle one is the DMF travelling column machine i think.
Weirdly, this newly named company is, is 20km from me here in canada.... ! (edit: ahh, they just do carbon fibre tooling here, boooooring)

[ihavenofish]

Didn't know that 3rd option (composite constructions) was viable enough for this type of company to make, I guess that solves the question whether filling steel tubes with EG is a yes or no.

when filling steel fabrications, you need to account for thermal expansion. most of these high end machine bases will have active cooling and live in precision climate controlled buildings, so they never change more than 1-2 degrees in operation. if you dont have that, you could end up with a machine that warps itself.