[catahoula]

Great stuff Peter, thanks for sharing. I think it looks great as is. Interested to see your modulus measurements when they happen.

I think I'd be scared of warming the mix for fear of adding any thermal expansion stresses to the mold or mix while it's curing. But I've been wrong before

[peteeng]

Morning all, you scruffy lot - By warming I mean to about 40C. In summer here its 30C plus so 40 is not a stretch. The epoxy flows very very well at 40C (but the gel time drops from 1hr to 30mins)... Now back to mixes. Here's the result of the alox/al mix. The grey colour is the Al as it has segregated. I tried mixing it a few ways but found I could not mix it. The alox/al mix has clear resin whereas the RHS has white resin. The LHS cup grey areas are the aluminum powder not the resin colour. This makes me think to keep to one grade not mixed. The solids delta is 58% alox and 62% alox/al so not a great gain. So onward to modulus testing which is the proof of the pudding... These moulds are plastic puddling cups!! Peter

[pippin88]

Very interested to hear the testing results.

How much is ALOX costing you (in current non industrial quantities)?

[peteeng]

Hi Pippin - 25kg bag is $3.96AUD per kg... Peter

[badhabit]

I also did a lot of testing with Al2O3 because of the 9 Mohs rating, my reasoning was pretty much the same as you: It performs better than granite. But i found it way to expensive to be used in practice for the purpose of machinebases. Have you considered that it could be way cheeper just to beef up the construction to get similar performance at a lower cost? Unless you have moving parts that are cast, it shouldn't be a problem, or?

I also had the problem with sourcing a fine enough powder to get proper filling. I ordered "a 25kg bag of each size" from these guys: https://www.clemco.dk/produkter/blae...-sandblaesning (sorry, it's in Danish, but you'll get the idea about sizing options etc). It's used for sandblasting stainless steel amoung other things. The white stuff is the purest form of Al2O3("Korund"=Al2O3, "ædel"=nobel/purest)...but the finest material they had was 0,0173mm which still was kinda like very fine sand and not "flour like"...which is think is needed to get proper filling via fuller...the Silimix282 i ended up with have this extremely fine dust in it, looks like flour and it gets into the air and "stays there" if one is not careful when handling it..

/Thomas

[pippin88]

How quickly does the thin epoxy saturate / flow down?

How deep a section do you think can be dry packed then poured successfully?

[badhabit]

25kg bag is $3.96AUD per kg...

For reference i paid something similar:

I paid 20DKK/kg (4,17AUD) for these sizes:
2.00-2.80mm
0,71-1,00mm
0,15-0,21mm

and the fine one(0,0173mm) was somehow more expensive, 48DKK/kg(10AUD)

I would have loved much smaller sizes aswell as something bigger(like 5-8mm)...

/Thomas

[badhabit]

Just found the manufacturer behind the Al2O3 i bought: https://www.kuhmichel.com/en/product...fused-alumina/ it seems like they have a lot smaller sizes...nothing bigger though....

[peteeng]

Hi Pippin - The alox cup took about 3mins to fill. The mixture took about 7mins with small fills. I'd expect you could fill any size you like to car size. The 12 grit is 40% porous and gravity sucks. The epoxy I use has a 1hr gel time at 25C so you have over an hour for it to seep through.... river table epoxies take hours to gel and days to cure if you use those...

Hi Thomas - I feel the graded aggregate approach is just not enough gain for the effort. Plus in terms of cost the UHPC is half the price of the epoxy approach. But the epoxy process is so easy. Will make concrete and epoxy test coupons soonest. Peter

[badhabit]

I feel the graded aggregate approach is just not enough gain for the effort. Plus in terms of cost the UHPC is half the price of the epoxy approach. But the epoxy process is so easy. Will make concrete and epoxy test coupons soonest.

An Al2O3 mixture with the same packing as Silimix282 would be a killer though...

[peteeng]

Hi Thomas - Yes I suppose that's true. 80% packing would give 120GPa vs 91GPa at 60%... something to play with into the future. But 60% pack with csa gives 102GPa at half the cost of epoxy process so I'm happy with that for now. Peter

[peteeng]

Evening all - I've been infusing things for over 22 years and there's always something new to learn. One area of interest is how fluid flows in, on and around things. The 12 grit cast flowed really well under gravity. The 12G+420G didn't flow as well and I started to think about just the 420G AL to see what it does. For flow in this case to occur you need gravity flow, capillary flow, stokes flow, saturated flow and unsaturated flow. Very complex it seems. Then you need the liquid to wet the surface otherwise it just sits on top. So I set up a cup of aluminium powder that is 420G which is about a particle size of 0.035mm. I set the cup up so it seeped down one side. It started well but about 15min in it slowed down. So I placed it in my vacuum chamber and pulled it down to 4000pa abs... It degassed but didn't really seep much more. So I took it out and added more resin to increase the head on the powder. Then I placed it back under vacuum. Still no real flow, so I stirred it and degassed again and it wet out. Probably has too much resin in it now. So the outcome is that at a certain particle size the voids are too small and saturated flow stops (Look up darcys law) . I think if I infused this it may work but clearly a bigger aggregate size is best. Once I get thru the 12 I'll go to a smaller grit to see. By the weights the AL is only 41% solid so I added too much epoxy. But its an interesting exercise.... If I can get large grit aluminium I can use this in areas that I want to machine vs using solid aluminium (maybe) We shall see what this cup looks like tomorrow.... Peter

[peteeng]

Hi Thomas - I wouldn't go smaller like the 0.017mm I'd go bigger. The selected grading should give you the same solids content ratio with the biggest aggregate possible. This makes the smallest voids large enough to stay away from the weird or zero flow regimes. Zero flow isa zero wet out.

Theres little value in having a high theoretical solids ratio then having large amounts of air in there diluting the properties. I was surprised how much came out of the mix I did. Modulus test is needed to prove material.

The issue with large agg is that if you have a 50mm gap say and 20mm aggregate it may not pack into it as well as designed. A dry pack test would confirm/deny this. I like it simple so a small singular grit is good for me. The CSA grout I use has a version with 3mm basalt in it and its 34GPa out of the bag. I'm thinking of using that and the alox at max pourable ratio as the basic material for big casts and for intricate parts the alox epoxy. Onward to modulus tests. Peter

[peteeng]

Morning all - Released the Al cast this morning and checked its weight weight etc. Added too much resin. Its 41% al by volume. Maybe do another but calculate correct resin addition this time. But it won't dry fill so would need to be stirred... surface is very good as expected.... Peter

I looked up the supplies spec for the 420A powder and the average particle size is 0.23mm not 0.035mm as stated earlier. Their grit spec is not the same as abrasive grit spec...

[peteeng]

Hi all you Arenophiles out there - (I just found that word looking up sand properties). I'm a little stumped and intrigued on a few things in the EG mix. I've just tried to mix the correct ratio in aluminium powder and found it too stiff so added more resin until it liquified. It stuck me I can get a guage on this by the height of the fill in the mould. When 100g of powder is in the mould its dry surface is 25mm below the top and this should be ~60% volume solid as tested many times before with big grit. The first batch at 40% Val was 20mm below and the second batch I just did was 23mm below. So my impression is that at a certain void size the viscosity of the resin won't allow it to pass (this is called the capillary size and means there is zero saturated flow). This occurs in my experience with glass fibres vs carbon fibres. Glass is about 20-22um diameter and CF is <9um. Even though the void ratios are the same it takes 2 or 3x longer to fill a CF laminate then a glass one. Some CF won't fill at all. So this tells me to use as big an aggregate as possible, grade it if you like but make sure the resin you use will flow into it small scale. Then you can dry stack the mould and pour the resin in until filled and you will know you are at correct ratios. This also minimises air entrainment due to mixing. Otherwise I'll infuse it but use bigger powder then 0.4mm for the inserts.... The metal powder supplier I use has 1mm grit may try that....Peter

The photo of the basalt is my driveway, its 20mm aggregate. That's what all this stuff looks like except smaller...

[peteeng]

I took some close ups of the different mixes-- Peter

[peteeng]

Good Morning all - I haven't done any machine development for a while, been obsessed with the material. So I looked at a few of the priors and decided maybe there's an option of casting the entire frame as one piece. So belted out a suitable design to see how it goes. It falls short on X axis but its close so will widen it and try again. The only issue I see is how to machine the table lands under the bridge... I looked back thru the history and many of the designs got to the 20um/N target. The later models were all about reducing the base weight. This designs frame is 200kg so is a one lift thing... Peter

[peteeng]

Good Morning all you Lurkers - I have made the base a little wider and deeper and it now hits the target. 22/26/41um/N. So Milli-Mono-2 is go. I'll digest the design a little more then maybe?? break it into two. I think I'll make it smaller and I need to decide what size that is.... Then I can get serious with motion parts...Peter The frame is 234kg at the moment so 9 bags of grout

[peteeng]

Evening all - The cast part weighs 225kg at the moment. My plan for the bed and Z axis is to make an aluminium rail/motor sub plate that is then screwed and glued to the frame once aligned. The gantry rails will have inserts in the part. The current gantry design is not suitable for a subplate. The frame is 1000mm high 600mm wide and 700mm long. Peter

[peteeng]

Hi all you CNC addicts out there - I came across this UHPC that gives you 45GPa out of the bag. Available most places USA. Peter