Designing new Router called Maximus

Hi All - Found a small issue. Will sort it. Peter

fixed - simpler then I initially thought...

Opps another one...

Hi All - Been trolling through Maximus via the drawings. When I find something on a dwg to check I go to the assembly and have a look at all things associated with that part or feature. Currently there are nuts and bolts to add so I do this and I'm finding various interference issues which I then resolve, sometimes I end up way away from the original part!! Just spent some time working through how to tram the spindle. Its held to the back tool plate with a central screw so it can rotate about this. Then 4 outer screws clamp it to the back plate. Its clamped in the spindle mount via a large cross screw and once all this is done its connected to the front plate with 2 or 4 screws. For the tube system to work its got to be well connected at this edge. Slowly getting a pink tick on all the parts!! Peter

Also slowly getting all the nuts and bolts into the assemblies. Usually I don't bother with nuts and many of the bolts as they are a pain, but this time I want an accurate and automatic nut and bolt BOM and I want the N&B's so they run in Simsolid for the final FEA checks when I get there... Plus I'm adjusting the length of the bolts to the correct length so lots of quality bolt time.... I had to make a little CAD hacksaw to cut the long 120mm long bolt to 110mm on the spindle clamp...lucky I already had a CAD vice...Peter

Hi All - One objective with Maximus was to make a 3D printed part. Maybe the spindle clamp will work for this. So next time it's open I'll design a funky webbed version... Peter

Hi All & Sundry - More trolling through the dwgs fixing things and checking things. Finished detailing the welded columns. These in my view will be the trickiest bits to get right but I have confidence in the company that will do it. The slots are overlength so the top tabs can be introduced and then slide down into the base of the column. The tabs are less deep then the basic thickness so when they TIG the slot, the weld will not be proud of the surface. Fingers crossed the slot weld pulls the parts together but does not cup them. I may add a note about preheat with a torch, will be a better result. 200-250degs C would be a good preheat to minimise distortion. Only a few more holes and N&B's to go... I'm going to design the 3D printed spindle clamp now. Peter

Hi All - Here's the 1st attempt at the printed clamp. Its also here as a 3d pdf if you want to look at it. Back to N&B's Peter

Hi All - More trolling thru the dwgs and I noticed that the edge gaps of the columns were all different so started evening them up and realised the top of the buttresses were quite different as well. So made those more even. This is the sort of thing done in the polishing phase but may as well do these things as they come up. I also added 0.1mm to some of the slots to give it a slight wiggle room, last time I did something like this I had to linish some of the edges so they would "click" together nicely. Sorting out the gantry rails and its many part holes now. Peter

Hi pete.
I read an article somewhere a while ago about how new design and testing software is coming up with parts and shapes that are really organic looking completely different to what we are used to but far superior in strength, flex, load etc. Your maximus is reminding of that. Be interesting to see it in real life.

Hi V2 - Those sort of parts are usually built using 3D printing. I'm looking at metal printing for some of Maximus parts but so far too expensive. Fusion 360 has shape optimisation if your keen. Thanks for your interest. I'd like to make these parts more sculptural but holes in laser parts cost $$$ but in 3D printed parts holes make it cheaper. One rule for Maximus is to be as cheap as possible so extra or non-functioning holes are out... Peter

Hi All & Sundry - I was going to do a mini review but not much to say. Looked at the last reviews and we're on track really. Doesn't seem to be a lot of action because its all about holes, fits and drawings done to a manufacturing detail level. Currently I'm working through 14 assemblies and 40 parts dwgs to check everything is correct so I can push the big red "Manufacture" button. The last physical 10% of a project takes 90% of the time... details details and more details. But I thought I'd mention some stuff about CAD systems to those out there that are new to parametric modelling such as Fusion 360, Solidworks. Alibre etc. These systems are based on relational database processes. So parts are connected to assemblies, drawings are connected to parts and assemblies and BOMs are connected as well. This means that there is a large chunk of the software dedicated to tracking all of these relationships, plus Windows is not a robust database environment due to its DOS DNA and corporate regular change policy. Mac systems are very robust due to their Unix DNA which has extra systems in place to "address" data vs having to go and find it. All that said even though you can put parts, dwgs and whatever you are doing anywhere in your file system its best to have a single folder that you put everything into for that project. So have a Project X folder and everything you do in CAD in project X goes into that folder. No other folders in there just the CAD stuff. This means that it can be copied and moved easily. It also means that as other parts of Windows change due to updates, failures, operator error all of those relationships stay within that folder and will always work. Been in companies that have data distributed over various folders (dwgs in one, parts in another for instance) and they have major issues with CAD models and data integrity when something goes wrong. So guys and girls keep it very simple otherwise it may take you a month to rebuild a complex model at some point in your future. At corporate level best plan is to use a "vault" but that's another story.

Oh yes I use a register for my projects in excel all documents within a project are entered into the register so I can find them easily without going hunting like Windows does. So every doc is serialised in the register then it is hyperlinked to the doc (part, assm, BOM, calculation whatever it is) . The only issue I've had is because Microsoft likes changing their software over time so for various reasons the hyperlinks fail over time. For the last 10 years I have been using open office and none of my Open office registers have failed over that time. Highly recommend using Open Office..

I was doing a job at a company once and set up a drawing register for the company on a particular machine. Spent 3 hours setting it up and getting the linkages right etc etc. Then I sent that to the lady across the office who was going to work with it and she opening it and as she had a different version of excel it decided to change things, then she saved it back across my version and it all fell over and self destructed. Could not happen on a Mac or in open office... Cheers Peter

Hi All - Another milestone passed. All dwgs have pink ticks, which means on face value they are ready to go. Now I have to do cross checks on all mating parts or significant parts and have a think through some assembly fits for manufacture and its balloon and confetti time!! Are we still having fun out there? :) Peter

edit - started cross checking and I've sent some dwgs to the bender for comment. Hopefully no changes needed but I think the tool holder may be a challenge... we shall see, it's good to start on the manufacturing phase.

Hi All & Sundry - Some feedback is back and I have to adjust some flange lengths. My holes are too close to the bend and will flare. They have sent me their tooling geometry so I can check other parts. Some things are easy and won't change things much but I can see others now that will make the parts much bigger. May have to drill holes after bending... more thought bubbles needed...Tuning phase begins always pushing some boundary...

Their 8mm tool is 80mm wide and they don't want holes in the throat of the tool. But their 6mm tool is only 30mm wide which means the parts can be much narrower. So some of the 8mm parts will go to 6mm thick. Peter

Hi All - So far so good. Changed most of the parts that where 8mm to 6mm thick. Hasn't changed operation but stiffness will suffer. Will do some more simulation before next review to see how its travelling. The long rail sleds will be the worst affected and will be quite wide I think. But until I get there not sure. Peter

GOOD MORNING - CNCers... I've adjusted the 8mm parts to 6mm, the last ones to change are the tool tubes. These have many holes along the edge and currently will flare when bent. I'll look at increasing the flange width to see what happens. I think I may go back to a shoebox connection to save width. But this still has a wide flange but I can use this to contribute to stiffness. Stay tuned nearly ready for quoting and polishing. This model is now called P4.

As a technical note - there are many mirrored parts here but mirrors don't automatically update when they are buried in a sub assm. So I don't mirror them until the design is finalised. Then I do the mirrors at the top level last thing which makes updating them easier. That's why some things are missing from the model, but they will be there eventually... Peter

Collisions!! As I make the flanges wider various things collide. These need to be fixed!! Working thru the parts, readjusting dims..... Peter

Hi All - Been working thru the parts adjusting flange lengths and hole edge allowances. Lost a bit of machine width then regained it but not quite enough to suit the large plywood sheet. So looks like the Tool tube will become a shoebox arrangement. While working on the bearing beams and gantry (the gantry has gained some height due to hole allowances ) I put in a bumper on the LHS. I'm now about happy to start reworking the tool tube, have to squeeze the Z drive width down a little more then it's nearly onto the finish line. Peter

Hi Lurkers - P4 shoebox looks quite a bit different to P3. Now I have to make it a bit narrower and add all the bolts and holes. This will also entail having threaded holes vs nuts so a little more costly. Peter

Edit -while adjusting parts I looked at the cutting envelope. Its currently Z270mm, gantry 1216mm and table 2550mm. The gantry width needs to be made 60mm wider (one hole on the rail) and it will be a generous machine.

Back to ground zero - Hi All, P4 has the flanges and hole clearance allowances that I think are acceptable. So update the dwgs, do final checks and get the package to the metal shops for quote. Yippee. Peter

GOOD MORNING CNCers - Back to some real work. Maximus has changed structurally while it's being detailed for manufacture. So I did a quick simulation to check we had not lost too much stiffness. Turns out with 1000kgf applied to the tool holder it deflects 0.8mm which is under the 1mm target for 10N/um static stiffness. Very good. I also looked at it minus the internal gantry brace and it's a bit softer without it so it seems it needs it. I'm also going to look at the extra bearing. The Z axis moves slightly sideways which you can see with the colour contours. But at normal tool loads this will be nearly zero. Next steps is to continue updating the parts files and get them ready for RFQ stage. It's good to know that the stiffness target is still on track. Plus I want to run the model with full bolts and friction joints asap. Feel like we may be on the home straight :) Peter

Edit - added a buttress to the small bearing sled to stiffen it on that side and keep the aesthetic going. Back to detailing... and some commercial work....

Why is the Y axis carriage two separate pieces? (Blue bits in image Max P4-3)

Hi Pippin - This feature was discussed in prior posts if you drill back. But to summarise. It was one piece initially but it flexed across the top two bearings. So the plate needed a vertical stiffener up the middle somehow. So I split it into two so the vertical edges of the "big" side could be flanged to provide the stiffness. This area of the machine in a traditional layout has three structural loops around the bearings: a) the gantry b) the bearing plate c) the Z axis plate. Perhaps having three load paths means the bearing plate is redundant to some degree. So far the FEA has shown no issue with the design. To help the side drive I have also included a support bearing near the drive nut. ie the gantry has 3 bearings on top rail and 2 on bottom. I'm about to look at if the 3rd bearing is needed now I have resolved the design a little better. I've attached an image of a very exaggerated deflection of what the two plates are doing under a "push" load on the tool. My current interest is to try to keep the motor mount and the end bearing mount more parallel under load. They currently are about 0.15mm different but this is with 1000kgf applied so probably under normal loads not relevant but this seems to be the biggest twist in the two plates. I could move the bottom bearing over a bit to help reduce its applied moment, a little investigation needed.

To step back a bit - One objective of the design is too minimise the tool cantilever to the gantry. In this design the tool centre is only 110mm from the gantry face which is very short compared to other machines. This is why the Z drive does not go down the middle in between the gantry face and the Z axis. Cheers Peter S

Opps - in the image attached I have tried to "move" the wrong bearing it should be the one below it. Can't move the one circled, sideways!!

to explain that a little further see nut image. When the Z axis is pushed backwards or pulled fwds the drive nut does the same. The drive system is on a cantilevered plate on the side of the Z axis mechanics. Worst case is when its fully extended as shown. The nut pushes the plate down a little. In the mid travel I have the support bearing so it can;t push it down. Right at the top it has little cantilever so will be minimal effect there as well. So I could either put another support bearing at the bottom, or move the inboard bearing out a little. I think I'll move it a little outward.