Re: Milli a new composite mill kit
Quote:
Originally Posted by
peteeng
Hi Thomas - With composites I don't think it works that way. Even with metals their dampness ratios are "similar" vs their stiffness. So I'm not concerned that the E80+ material will be less damp then the E35 material I think they will be about the same. Peter
https://durcrete.de/technical-data-nanodur/?lang=en
Ahh, that datasheet shows a rather significant lower damping on E80 vs E45...so it is not entirely false as i see it... the real question is then how big an impact it will have on the real machine performance.... but it shows that pure e-modulus is not the only factor one have to consider...otherwise one could just as well build the machine in ordinary steel(a big oversimplified and exaggerated, i know)...
Maybe a test to determin the difference in damping on a regular EG mix, pure CSA mix and a CSA+Alox mix?
Re: Milli a new composite mill kit
Hi Thomas - To do a damping test is quite expensive. Plus depending on the test done gives different outcomes. So its hard to compare apples to apples in some docs. Lets just say that "concrete" type materials are significantly damper then metals and That's one reason to use them. Peter
Re: Milli a new composite mill kit
Quote:
Originally Posted by
peteeng
Hi Thomas - To do a damping test is quite expensive. Plus depending on the test done gives different outcomes. So its hard to compare apples to apples in some docs. Lets just say that "concrete" type materials are significantly damper then metals and That's one reason to use them. Peter
Piezo crystal + osciloscope attached to a test sample and "hit it with a hammer"....do the same ting to all samples....done:-) One might not be able to get absolute values for the materials, but at least you'll get relative values...apples to apples...
My thought was more on your process of choosing a material for your project. You seem to study the CSA quite a lot and are now playing with AlO2 in the mix aswell. But i feel like you are only seeing half the picture, if you only consider the e-modulus when choosing your mixture. You also made me question my choice on EG by showing your CSA results(but i do appreciate your questions though, that is the only way to improve!)... but that was why i tried to "provoke you" by suggesting to just make it in plain steel because thats a lot stronger, but doesn't damping as well...
In the end, i do get that one have to make some choices at some point in the process. And filter out some options. Otherwise material science is a rabbit hole you are never going to get out of... i choose EG some time back for my project and have to stick with that now, nomatter what your CSA results show.... but that is ok, now i'm better equiped to make a choice next time around :-) (Not saying that i'd choose differently right now, but i have a better foundation for the decision process than last time i was in the situation)
/Thomas
Re: Milli a new composite mill kit
Quote:
Originally Posted by
peteeng
Hi Thomas - When I get them set up I'll post some pics. But they are a melamine board top and bottom with some 6mm pine keeping them apart. The mould will make a plank 80x6x1100mm long. I think I'll do an epoxy/ALOX one first then the alox/cement one. The grout supplier has dropped their price from $2.90 AUD /kg to $1.55AUD per kg for E34 out of the bag... They call it high strength grout...Very happy with that...Peter
I was reading through some test results for the grout I'm using. Its self deairing and one test puts air content at 1.3% by volume. That's really low, in the composite world you have to do a lot of work to get to that level in a laminate...
Is that the CSA stuff from Bluey? HE80AG?
What volume required for that price?
Re: Milli a new composite mill kit
Hi Pippin - Its HE80 or HE80AG same price. Its the bag price (20kg) 1 bag is OK to buy.
Hi Thomas - Can do it by ear as well... My driver for casting parts is to be able to make any shape I want & to consolidate parts. I have made routers with pressed metal and machined parts and want to move on from the limitations those processes have. You say "stronger" but these sort of machines strength is not an issue. Stress in parts rarely gets above 5MPa so low strength materials are OK. Its all about stiffness. There are lots of fabricated steel mills out there that work fine so material dampness is not the only factor in design....
Dampness is a secondary property or concern. Stiffness and rigidity are primary properties. If its stiff enough dampness does not matter (broadly speaking). Making machine parts in steel implies machining every part. This is costly and limited. People get carried away with the damping thing in cast iron. The primary reason to use cast iron is its castability (Its not called Cast Iron for nothing :) ) . Its machining and damping are secondary features. If your going to make many identical parts casting is the process of choice (although in the next 10 years printing will become the norm). So I don't have a hot foundry in my back yard but cold casting or infusion is an easy process for me.
So in regards to Milli materials in review:
1) Its a production machine so the requirement for making many parts drives various decisions. If I was making a one off like Makers and DIY do I wouldn't labour various points I'd just jump at the "best" presented path and move along. I'd go CF for small parts as I've been doing that for 30 years or so...
2) My material target stiffness has always been the same as aluminium E70. Then the fallback is to machine the parts in aluminium. But the theory told me 70+ possible, then the E80 German material confirmed its clearly possible.
3) The higher the stiffness the less material is needed or the stiffer you can make the machine in the same space. I can easily make E80 carbon fibre but the cost is prohibitive. Again if it was a one off for myself I have enough CF scrape to make some parts but this is no good for a production solution
4) I picked 20N/um as the basic static machine stiffness based on a 50% efficiency resulting in a real machine stiffness of +10N/um... Turns out that's a hard thing to do in a small machine no matter what the material. Geometry is King. I have designed some of the versions in steel and they still wouldn't get to target in the required envelope...
5) Originally I thought I could make infused fibreglass skins then backfill with some stiff material. Infused FG is about 30GPa and is machinable. Oh yes another rule was that the material was to be machinable so near nett casts could be done and finish machined (I want the lot :). Turns out 30GPa is not stiff enough for small members
6) The alox is the key and I found this well before I found the German stuff but the machining rule struck it out. Then once I struggled enough with steel, aluminium (fibres and grit By the way steel grit is still on the backburner as its machinable. Trouble with steel grit is its very hard, so hard to machine looking for a soft steel grit ), fibreglass, blue metal, basalt fibres and nano stuff I came back to alox. This means using inserts, so be it.
7) There's a lot of heated discussion about packing factors and at the end of the day all of the DIY and Makers that make EG, mineral cast, polymer concrete etc etc make something then don't check what they actually have. Even if you have 85% volume pack there are other factors that will dilute this at the part scale, so who knows what resulting modulus they get. At the end of the day its the real material modulus that matters not the theoretical packing factor. That's why I'm testing stuff. Plus I have to design with a number so I need to know its modulus...
So onward to defining the material better... Peter
1 Attachment(s)
Re: Milli a new composite mill kit
Hi All - To add a bit to the last post with casting I can do generative design in which part development will look like the image attached... Peter
1 Attachment(s)
Re: Milli a new composite mill kit
Morning All - Been reading up on Inventors shape generator (generative design) and it optimises stiffness. I was under the impression it optimises on stress but that's great to know. You set the weight reduction target say 50% and set the mesh then wait... Then you can use that shape as a template to build the next part version. Maybe create a Milli block and see what Inventor does.... Peter
5 Attachment(s)
Re: Milli a new composite mill kit
Again Good Morning - I have been playing with the shape generator. I booleaned Milli together, defeatured a couple of things like cars and rails as their details are too small for the shape mesher to cope with then ran them in shape generator. The grey images maybe hard to see. Trends are:
1) gantry and carriage are quite good. In the X case it hollows the gantry which is easy to do
2) It makes the columns triangular or slotted. Easy to make them triangular
3) It removes most of the base every time... The base has to be there because the table moves... Would love to get rid of the table... Better start the working day... Peter
In all cases it removed 50% of the weight... I'll have to pout the equal stiffness to the test.
Re: Milli a new composite mill kit
Re: Milli a new composite mill kit
Quote:
Originally Posted by
peteeng
Hi All - To add a bit to the last post with casting I can do generative design in which part development will look like the image attached... Peter
in fusion? they charge $100 per result exported, and, from my experience with the trial they had, its not giving you anything useful for a machine frame.
Re: Milli a new composite mill kit
Hi IHNF - I have Inventor so maybe different to F360? Peter
Re: Milli a new composite mill kit
I will give the Bantam guys a lot of credit for a design that's easy to mass produce, with the round rails and machined plates to locate them. But once workholding is taken into account, a 7x9x3.5" work envelope is just tiny. The small envelope is obviously necessary to make the round rail design rigid enough. And tellingly, they quote repeatability, not accuracy, which is just the step angle of the motors and the pitch of the ballscrews.
I'm with IHNF on generative design for machine tools, as mass is a good thing and GD is meant to minimize it. But I know you're trying to keep the shipping & handling weight down. Seems like an L-shaped column cross section would help optimize a bit. (or triangular as I just noticed you mentioned)
Re: Milli a new composite mill kit
Hi Cat - I think the Bantams saving grace is that the spindle is in-line with the Z bearings, removing torque from the deflection and it has a very small envelope. Well executed though and it uses a 250W spindle which I thought 400W min would work so its a neat little package. Plus it gets good reviews so for $4000USD is it good?? ... seems the "desktop" people will pay a premium as they don't have a garage or a bench? Peter
2 Attachment(s)
Re: Milli a new composite mill kit
HI IHNF & Other lurkers - Topological "optimisation" is pretty much standard on common CAD systems so this is what is being used more and more to design all sorts of parts. Weight costs money to buy and handle and create. So its not just weight of the working part per se that's important. The weight is great lobby is outdated, lighter machines at same stiffness are faster, suffer less vibration and are cheaper to make so I'm sure manufacturers are well into this sort of stuff.... here's an advocate for lighter machines, a bit maths heavy but is readable to get the gist.. Peter
Re: Milli a new composite mill kit
Definitely not arguing against optimizing shape- that's one of the fun parts about design in general. A given mass should be used most optimally when possible. You could have it more rigid at a given mass, or use less material (and therefore less $) for a given rigidity. But the edits suggested by your recent example of GD are pretty minor, which makes sense as you've already spent a lot of time optimizing manually with FEA.
If I've learned anything about product design for manufacturing, it's that the ease or automation of manufacturing is #1. Yes, a molded part offers some options in shapes that aren't feasible when manufacturing from stock material, but simple shapes mean simpler molds, which cost less, or can be made more precisely at the same cost, parts are easier to demold, finishing operations are quicker, etc, etc. To make a margin on producing parts, you have to minimize # of materials, minimize # of operations, minimize hand finishing, minimize potential for errors, and then just produce and produce and produce. A project for a projects sake is one thing, but for actual production, especially for a small outfit, every step has to be as easy as possible. It feels tyrannical sometimes.....
Re: Milli a new composite mill kit
re: bantam, for $4k USD I'd much rather have this: https://www.toolots.com/r4040-cnc-router.html
Though I'm sure it's a lot fussier to set up.
Re: Milli a new composite mill kit
Hi Cat - Yes product design sometimes feels like the tails wagging the dog. We shall see in a year or two if Bantam cuts the mustard, same with Milli. The price has already gone up from $3500 to $3999USD in a short time so they are sorting out their nitch... Peter
Re: Milli a new composite mill kit
Hi All - How's the flotsam and jetsam today? Been watching the Americas Cup races in NZ. Some amazing races. Most races lost not won... Todays a very big day for NZ. Peter
1 Attachment(s)
Re: Milli a new composite mill kit
Evening All - NZ has won the Amercia's Cup again. Amazing how a small country with small funding can win against the big players.. Anyhow I have organised more mould bits today. I thicknessed the sides and here's how it will look. I'll do the Alox/Ep first. I'll seal the strips with acrylic sealer and give everything a good wax. It will be screwed together allowing a top fill. Its 80x6mm and 1100mm long.. It will need a hole for filing at the top, to be figured out as I go along.... Peter
Re: Milli a new composite mill kit
Quote:
Originally Posted by
peteeng
Hi Cat - Yes product design sometimes feels like the tails wagging the dog. We shall see in a year or two if Bantam cuts the mustard, same with Milli. The price has already gone up from $3500 to $3999USD in a short time so they are sorting out their nitch... Peter
Using round rails and a wimpy spindle for $4000 and a 20 week wait, just another one to the mix, small start up machines like this are every where, it's a $1000 to $1500 machine at best compare it to the Cheap Chinese and the Chinese machine wins