512,348 active members
2,338 visitors online
IndustryArena Forum > Mechanical Engineering > Mechanical Calculations/Engineering Design > Moving table (fixed gantry) with two ballscrew

Thread: Moving table (fixed gantry) with two ballscrew

1. Re: Moving table (fixed gantry) with two ballscrew

Morning Ivan - Your doing fine with your english and the tests. When you quote a ratio or % always state whether its by weight or by volume so others know what you mean. Also as a note you will need to dry your test sand before you use it!! In an oven is best. Fit it in with the sunday roast... If your achieving 70% volume fractions I think that's as good as you will get. Well done. I think this is the conclusion of the concrete companies as well. They used the high pack theories for years thinking they where getting 90%+ but in the end this is not achievable. 70% Vf which stands for Volume fraction vs Wr which stands for Weight ratio is a good number.

Infusion epoxy is very thin and yes it will stick even the smallest particles. It will take 1hr for the resin to gel (if you get 1hr plus hardener) and in that time capillary action will permeate everything as long as the sand is dry. I have glass tissue 0.2mm thick and epoxy will flow through this quite well under vacuum. So I expect if the sand is dry and you are not living somewhere where temp is below 15C then its fine.

Your resins density I estimate as 1095kg/m3, from the net quartz is 2650kg/m3 if you achieve 70%Vf then it goes like this if we work in litres:

700 litres quartz weighs 700*2.65= 1855kg
300 litres resin weighs 300*1.095= 328.5kg total = 2183.5 kg

sand is 1855/2185.5 = 0.85 by weight
resin is 328.5/2185.5= 0.15 by weight or 15% resin by weight. This is a weight fraction not a weight addition!!

The weight addition is 328.5/1855= 18% weight addition based on sand weight. Weight additions are usually used for pigments and modifiers, not the functional parts these are usually expressed as weight ratios. Check the maths and check your ratios on the day!

Not sure what you mean by air bubble inside? Can you see large air voids or lots of small bubbles? This sort of open process will always have small air bubbles unless you use a vacuum process. Air is 1.2kg/m3 so it takes a lot of air to change the numbers. As more info you can add a small drop of detergent to the water to improve its wetting of the sand and its bubble release. There are surface tension improvers for epoxy as well and I have played with these. They are expensive and I have not found them to be useful. If the air is in the epoxy resin (the hardener is very thin and it seems to self release) I have found heating it to 40-50 degs releases a lot of air as it thins out. But then allow it to return to ambient temp before use!! If you use a fast mixer its inevitable that air will be sucked in, that's how you whip cream and make pavlova.

When you use epoxy 1) make sure sand and mould is dry (water is your enemy, especially the stuff you can't see) 2) you could warm it slightly if air temp 20degsC or less but if 25C up then alls good. 3) when mixing do it slow with a paddle or spade vs fast with a mixer 4) use a long gel time hardener. This takes the pressure off getting it done, it reduces the exotherm peak and stresses the mould less.

If you have a scale you can weigh the test as well and this can then cross check your figures.

Onward to making some hard stuff.. Peter

2. Re: Moving table (fixed gantry) with two ballscrew

Hi Ivan - some updates on the quartz stiffness now some groundwork has been done. I have always thought quartz was E=70GPa but seems being a crystal it is anisotropic (always a fly in the ointment). So it varies 80-100GPa. Attached doc confirms density at 2650kg/m3 which is great. So reading attached, it seems using 90-100GPa is fair as this is the tested quasi isotropic figure. So your sand mix 90GPa x 0.7Vf x 0.5 strain efficiency = 31.5Gpa and zanites quoted figure of 31GPa is spot on. So I expect you get better then 30GPa so back to the FE work and finalise that design!!! Using 100GPa E=35GPa so range E= 31-35GPa not bad? Half that of aluminium

Oh yes E of epoxy =3.5GPa so full value using rule of mixtures (ROM)

(0.7x90x0.5) +(0.3x3.5)= 32.55 GPa estimated modulus or using 100GPa, E mixture=36GPa are you the optimist or the pessimist? 33 or 36 close enough for FE work.... Peter

I've just had some aluminium powder arrive so I'll be making a block of that in the next couple of days 0.4mm and 0.2mm

Looking at your Biresin, its thinner then the one I use which is great, I'd go with the CH83-6 hardener so the peak exotherm was down and it gives you over 2 hrs gel time. Doesn't quote the mixed density so

100g + 30g = 136g vol = weight/density kg/(kg/m3)
0.1/1.14=0.0877 m3
0.030/0.94=0.0319m3 total =0.1196m3

density = weight/volume = .136/0.1196= 1137 kg/m3 so the prior figures need to be corrected as I used 1095kg/m3 (1137/1095 = 104% missed by 4%)

cheers Peter

3. Re: Moving table (fixed gantry) with two ballscrew

added the resin data sheets - Peter

4. Re: Moving table (fixed gantry) with two ballscrew

Morning Peter. Thanks for the comprehensible calculations and explanations
Everywhere I mean % of volume or ratio of volumes. I corrected 2 places where I did not mention this.
Today I'm going to dry in a oven the fractions for the tests. Yes, I have a digital scale, but I'm not sure of its accuracy (absolute values). But as ratio (relative values), it can be used.
Other tests - afternoon.

5. Re: Moving table (fixed gantry) with two ballscrew

I bought a fraction of 3-5mm. I'm drying the sand now.
In test 2, the ratio of the two fractions is 50% of the volume of the test volume (1000 ml).
This is absolutely wrong. Also the ratio of 70% Vf (1-2mm) to 30% Vf(0.10-0.25mm) is wrong. We cannot thinking up arbitrary ratios. Even with calculation at 10 times the diameter difference between the small and large fraction - the volume ratio is 1000 times.

It is most appropriate to determine the ratio when mixing multiple fractions experimentally. We want to get the highest coefficent of packing.
So, for me, the correct method is the following:

The test volume is 1000 ml.
1. Fill the container with 1000 ml of the coarsest fraction.
2. We start adding small portions from the next finer fraction (recording their volumes) and stirring, mixing, vibrating etc. The fine fraction particles are inserted between the voids of the coarse particles. The volume of 1000 ml should remain unchanged (constant). We fill and vibrate until we can no longer keep the volume of 1000ml constant and it starts to increase.
This means that there is no more space (voids) between the coarse particles. We calculate the recorded quantities (volumes) of the fine fraction inserted and thus calculate the volume ratio.
3. If there is a third fraction - proceed as in point 2. Keep the total (mixed volume) constant - 1000 ml.

6. Re: Moving table (fixed gantry) with two ballscrew

Morning Ivan - So what are you aiming at? Industry information tells us that 70%Vf is about the top figure and you have achieved that. Are you chasing 90%? Industry has settled on 70%Vf for a reason. One of these is scaling. You maybe able to work hard and get your 1000ml sample up higher but then, will it scale and get you the result with a 150kg batch? You can't mix big batches as well as small batches unless you have big equipment. The Basetek company have videos of hugh mixers and graders making very large batches. They would happily make EG at higher Vf as the gravel is cheaper then the resin. The other thing is that by back calculating from their data they are under filling the resin. Perhaps that's the trade off. They are happy to make a sponge vs a solid to save \$\$\$ on the resin. From my composites experience using stitched cloth and UD fibres which are already in a state of high order and using 6 atms pressure in an autoclave we don't get past 60%Vf easily. So why not be happy with 70%Vf and move forward? Back to scaling. Boats have to be surveyed and they use test panels for the engineering certification. They are 600x600mm usually. I have been making these and helping boat builders make these for 30 years. Its always a concern that we make a 20m plus boat using tonnes of material based on a one off small panel. Over the years I have figured out how to make panels very well. I can instruct builders to lay cloth a particular way, use vacuum a particular way, build the bag like this not like that etc. This results in exceptional good figures. But I don't like doing that because it does not scale to the boat. You can't scale all of those hints into a big boat build easily. Maybe same with your samples. So what are you aiming at in terms of Vf and your E?

Did you see the Koehler article I attached to the EG thread? may help in your research... Peter

7. Re: Moving table (fixed gantry) with two ballscrew

Hi Peter.
I don't know how to answer you. Just looking for the maximum with affordable funds. I didn't know the maximum was about 70%. If so - well.
Just from reading a lot of EG threads where there are posts for 90% aggregates and 8-10% resin - I decided to strive for that. Now I see that this is impossible.
I thing that It's not a bad to experiment and find the best. I believe that these tests will not only be useful to me. So they are worth the time lost

I didn't make it in the afternoon . It took me quite a few hours. And half the tests - I repeated.
What I wrote in my previous post was not confirmed. Even the opposite. This in the particle schematic picture is true only in theory. In practice, with small amounts of fine particles, they do not get trapped in the holes between coarse particles. The fine particles just fall to the bottom. No uniformity is obtained.
Test 5 shows this - a bad result. My scale is not very accurate. But this has no relation to the % of quartz volume fraction (Vf).
That is for now.

8. Re: Moving table (fixed gantry) with two ballscrew

Hi Ivan - Yes segregation is a big problem when aiming at high packing factors. More reading this morning and most concrete mixes in reality are about 70% Vf maybe 75% in the lab. The theories that get to 90%Vf are based on spheres with no friction and that's the problem. Real particles are not spheres and they have sharp edges that stop them packing down. So what's your next step? Now you know the practical limits? regards Peter

9. Re: Moving table (fixed gantry) with two ballscrew

Morning Peter.
First I will complete the design with all the details (energy chains, inductive sensors, etc. Then I'll do a real resin test (Test 11). I hope the real 70% Vf (no 75%).
About the large casting volumes. I'm a builder engineer by profession. I have different big concrete mixers (from 100 to 300kg). No problem for mixing.

10. Re: Moving table (fixed gantry) with two ballscrew

Afternoon here Ivan Good keep us up to date. I too am re starting a moulded cnc mill design so thanks for your help so far. I'm trying to get the biresin here in OZ but seems sika don't stock it here. The usual will be fine...Peter

11. Re: Moving table (fixed gantry) with two ballscrew

Hi Peter. Try contact here. They may agree to send to Australia.
https://www.compositeshop.de/xoshop/...d-returns.html

I can buy you cr83 here in Bulgaria and send it to you. But I suspect the shipping will be very expensive.

12. Re: Moving table (fixed gantry) with two ballscrew

Hi Ivan -Thanks for offer but yes its expensive Europe to Oz. I found some milled carbon fibre in france that I wanted. 20 euro/kg for the sample and 180 euro to get it here. I'll have to wait until I can build that into a project. I'm waiting for the local sika people to see if they can get it in. Otherwise my usual epoxy will be fine. Peter

Page 4 of 5 2345