[ihavenofish]

Fun inspiration

What you see with all of these, vs yours, is the proportions. You keep trying to build a spindly machine with big travels. Yes, you have a weight restriction, and size to a degree, and want as much travel as you can, but you can definitely give yours beefier proportions. Start with the potato chip base. make it 3x thicker then hollow it out strategically. DMG uses a lot of lattice box elements. Complex molds to make, but you make up for it with lighter and stiffer castings.

Also notice they seem to have dropped mineral castings from every model. every base seem to be an iron casting, Probably ADI for the moving parts.

Seems the "new" things is to concentrate on thermal stability. Linear rails and ball screws are all water cooled, even on the cheapest cmx models. (not that any dmg is cheap).

[peteeng]

Hi Ifish - Epoxy is 3-4GPa stiffness by itself. Granite is 40-60GPa by itself. 80% granite plus 20% epoxy has to be stiffer then 4GPa as quoted in this article. The articles top figure is 6GPa. I think they missed something in the calcs. They say 4GPa is very stiff!! I think the Prof that wrote it needs to review...

see the "cast yourself" article their EG starts at 25GPa...

Re- proportions - The current project is finding out the configuration and proportions for a benchtop machine. Its easy to design a massive machine if there is space. The hurdle is to design a small machine with high stiffness. The process so far has been very useful.

Re - mineral casting - In europe with the demand for bigger machines and tighter environmental rules and carbon footprint rules and more complex machines, its going the other way. Making huge cast iron objects becomes more and more difficult and costly. Casting in UHPC is easier, can put in integrated cooling and wiring passageways easier. Companies like DMG have momentum to keep the status quo but ultimately as various things mature they will change. The same happens in the auto industry aluminium and plastic push into various areas then get removed, then they reappear everywhere once the new tech is sorted and costed. In America they don't have foundries that can cast the big machines they have to come from China which is now a great big issue with reshoring so they will have to go composite beds... a simple thing like slag can shut down a foundry if the EPA says you can't dump it here anymore. This is what happened in europe, they moved their "dirty" processes to china now China is saying we don't want slag here now as well. So alternatives have to be found. China are now reprocessing slag and sending it back to europe as concrete additive!! Lifes complex...

[ihavenofish]

bah, cant post more than one vid per post.

v=z6wjhHIuCtY
v=E-GlLSfKi5k
v=19qYM_Zjd0g
v=m8qJpUIvxjI
v=sDHgGbs74_A
v=iXgTIt66jAs
v=TepFt5CYL5g

[ihavenofish]

Hi Ifish - Epoxy is 3-4GPa stiffness by itself. Granite is 40-60GPa by itself. 80% granite plus 20% epoxy has to be stiffer then 4GPa as quoted in this article. The articles top figure is 6GPa. I think they missed something in the calcs. They say 4GPa is very stiff!! I think the Prof that wrote it needs to review... Peter

or not.

show epoxy as a baseline of 2gpa. 4.5gpa at 80% or so aggregate. seems right to me a first glance.

real granite is WAY stiffer than eg.

[ihavenofish]

schneeberger is rating theirs at 40gpa.

i should say, the common these among everyone who has tried to diy cast EG was that it was rather flexy.

[peteeng]

Hi IFish - Here's a simple calc that shows EG should be around 30GPa. I have used an E=50GPa for the granite but there are much higher modulus aggregates to use (eg Alumina is 300GPa+) and this is how some of the companies get higher E's for their material. You can push the Vf up to 70% but over that its a bit tough. UHPC starts at 28GPa by itself and by adding high modulus aggregates can get to aluminium's stiffness of 70GPa... Alumina is white grit blasting media...

If you use alumina at the same ratios you get the modulus of steel at 217GPa stiffer then steel, just. This is effectively a grinding wheel.... I have done this sort of thing with silicon carbide in the past...

A lot of this is covered in the EG threads... So we can make very stiff composite materials if we want to.

Re thermal stability - I have been working with "Toolmakers" for over 30 years and they always have had air conditioned workshops and stabilised machines vs general engineering shops in ambient conditions its the only way to make something the same size today then the same size tomorrow... plus your materials have to be conditioned as well. Keep making Peter

[ihavenofish]

on the eg mix... on paper. actually mixing it to that isnt easy.

on the thermal stability, its more complex than that. a high speed linear rail and ball screw gets.. HOT. like, my router Z ball screw going 600ipm gets up to at least 50-60C. rails get warm too. this makes them expand. your EG doesnt. machine starts to warp. etc.

not that your machine is going to be as fast as a dmg or need 2 micron precision, but it's things to keep in mind.

[peteeng]

Hi IHNF - Nothings easy in machine design or building. This is Maker grade stuff not a 2um machine Gotter go fix a leak in my car window!! Its at least 1200um wide...Peter

[catahoula]

the DMG mori media team has my number. Can't get enough of those videos.

Agree re: proportion. No material substitute for form stiffness. I had my machine even more bulky than it is, and just before ordering steel decided it was overkill, so backed off on tube proportions by about 25%. of course in the building process I've been adding material back, again and again, at this point it would have been slightly cheaper and certainly drastically less fabrication time to just go with the larger sizes, heaviest wall thicknesses, and fewer parts. After running a bunch of FEA, I'm thinking any highly loaded sections shouldn't be much more than a 2.5:1 ratio in cross-section

[peteeng]

Hi All - Dr Mori is a bright bunny. Took a small family business and turned it into an International Monster... hats of to venerable Dr of Eng & President - Masahiko Mori. Would be good to have him on the team....

Now in regard to sections I want to cast these as solid as possible to minimise mould complexity. So yes form over material is the key here. Now I have decided on X450xY225 and 25mm rails I'll rejig the model around that, then I can start adding screw gutters and drive parts... I think this maybe my first servomotor machine.. Peter

Hi Cat a machine can never be too stiff...

I did a little chart for the base rigidity. Horizontal lines are equal rigidity and verticals are equal thickness. The individual lines are material stiffness. Top line is high modulus UHPC, red is aluminium bottom line is ordinary CSA. Not really looking at epoxy due to cost and low stiffness.... I have a report here for CSA shrinkage. For 28days its 553 microstrain, for 900 days its 550us and they say it looks to be the same at 1800 days (4.93 years) so they are saying its stable. So my 700mm base will shrink 700*550e-6=0.39mm UHPC has similar shrinkage figures 500-750ustrain.

[catahoula]

a machine can never be too stiff...

Truly, it's the stiffness:budget ratio we're all really worried about...

I found this: https://www.researchgate.net/publica...e_Patch_Repair
which indicates that incorporating short steel, carbon, and basalt fibers into CSA can reduce unrestrained shrinkage to below 300 microstrains. (300 millionths, so .0003 unit/unit)

Comparing to EG, the big EG index thread says essentially "no" shrinkage when filled with aggregate, but,
I found this: https://www.mdpi.com/2076-3417/10/4/1336/pdf
Which shows epoxy polymer concrete shrinkage at 500-1200 microstrains depending on hardener. The mix there is 13% epoxy, 19% fine filler (fly ash etc), 68% silica aggregate. Those ratios would be tough for a DIY'er to hit, so real world may be worse.

So while the 5 year stability figure you provided is short in context of machine life, it's probably stable a bit longer if the rate is so low, and it seems like enough to warrant DIY/hobby use at the very least.

I just might fill my steel tubes with CSA. The only question is: if these numbers are correct, why aren't machine builders using it already?

Did your source provide shrinkage values before 28 days? i.e. how long should one wait after casting before the final machining

(and going back a few pages in discussion, the J Yeon article also shows thermal expansion around 20-30 E-6 m/m/degree Celsius, twice or more that of cementitious concrete)

[peteeng]

Hi Cat - one thing at a time

1) I intend to add steel fibres to the CSA not for shrinkage control but to improve the modulus. I use the same fibres that are used in brake pad compounds
2) EG does shrink. Low shrink epoxies typically shrink <0.5% and when you use it in granite or mineral fillings (which did shrink some millions of years ago but now are stable) it still shrinks but the volume ratio is low so the total shrinkage is low
3) The test chart I have shows at 14 days the CSA was at 442 and at 21 days it was at 534 then no change to 56 then 800 then 1800 days so around 2 weeks and your stable. If you can heat it to 60deg it will do this in 1/3 the time. I intend to post cure the parts as well. Epoxy should be post cured as well
4) They have been using UHPC for over 30 years in europe which is similiar to CSA.
5) epoxy has a high CTE around 55-60 so this contributes to a high CTE in EG
6) The CSA I'll use is blucem HE80 from Bluey in Australia. If your overseas you will have totalk to cemnet suppliers and figure out equivalent

Metals shrink on solidification and grow and shrink as the metallurgy develops. Much more then 500ustrain its in the % range... Peter

https://www.rampf-group.com/en/news/...ief-history-1/ history of concrete machines no cast iron after WW! so concrete stepped in...

[peteeng]

Hi Cat- since you are filling steel tubes I think you need an expanding cement/grout. These are used for repair of holes so they expand into the hole and grip it. I'm sure your local hardware has something like this. Peter

[catahoula]

Good stuff. If I had gone down this particular rabbit hole a few months back I'd probably be pouring a CSA base around a steel internal skeleton right now. Looking forward to your results here. Not to derail your thread here, but in the context of filling tubes, the bonding of CSA to external steel walls is questionable, and with even small amounts of shrinkage it'll likely pull away, so I'll probably stick with EG for tubes and use CSA to fill the webbed base with some steel angle welded in to make sure it's secure. I'll use another tube down the middle of the tubes to minimize volume requirements. I'm satisfied on the CTE of EG vs CSA and other materials. My guess is that due to the low modulus (1/6 of steel), EG inside steel isn't a huge real-world problem with thermal warping, as long as the steel structure is robust.

Seems the main brand of CSA here in the US is Rapid Set. It's also getting terribly late here in the US, so goodnight!

[peteeng]

Hi Cat - did we go here before?

https://www.ctscement.com/datasheet/...=Professionals

cheers Peter

SDS says its CSA cement excellent....

[peteeng]

Evening all - All this discussion about materials is good. if I were making a one off machine I think I would be having steel laser cut parts made and then braze together then fill with CSA. This seems to be a good path. But since I intend to make many parts and its intended to be a production machine I will be making moulds. I'm thinking about making the moulds in CSA as well.... so I start with MDF parts cut on YaG assembled into a machine to sort out wiring and anything needed to be added to the mould such as mounts or wiring passageways etc, then pull it down update mdf parts then use those parts to cast the mould. Boxing up a part and backfilling is a much easier way to make a mould then laminating a fibreglass one. All that's down the track, have to get the design done yet!! Lots of pressure to get commercial stuff done by Xmas so may neglect this project until next year... Peter

[catahoula]

Hi Cat - did we go here before?

https://www.ctscement.com/datasheet/...=Professionals

cheers Peter

SDS says its CSA cement excellent....

Brilliant. Tailored for the purpose. I'll be giving them a call soon to see if there's any tips about interlayers for steel bonding and what kind of fibrous fillers to use if any. Cheers.

[peteeng]

Hi Cat - They may have a primer for steel. Peter

[spumco]

As soon as my basalt fibers arrive I'll be filling my castings with Rapidset; slow shipping and all that. Assuming testing goes well, I'll be using the locally-available Rapidset Cement-all and as much of the "Flow" plasticizer as the instructions permit. Castings are webbed and extremely rough inside so I'm not concerned with the CSA decoupling from the casting.

Still have no idea on the fiber loading I should use.

It will be a little while before I can provide meaningful stiffness increase data - still have to put the thing back together and get it all aligned.

[peteeng]

Hi Spumco - what length basalt fibres (fibers in US) did you get? Peter