[Herbertkabi]

Dear Friend

I appriciated your machine

With respect to your material, would you pls advice me

1. what's compressive strength of your epxoy granite?
2. what's bending strengh of your epoxy granite?
3. what's e-modulus of your epoxy granite/
4. is there any creep of your epoxy granite for a long time?
5. is there any air bubble of your epoxy granite?

Density kg/mc 2283.0 5x5x1 cm EN 14617-1
Water Absorption % 0.03 5x5x1 cm EN 14617-1
Bending Strength Mpa 48 6x5x1 cm EN 14617-2
Flexural Strength N 3371 6x5x1 cm EN 14617-2
Compression Strength N/mmq 159 7x7x1 cm EN 14617-15
Deep Abrasive
Strength
mm 26.6 15x15x1 cm EN 14617-4
Surface Hardness MOHS 6 20x20x1 cm UNI EN 101
Linear Thermal
Expansion Coefficiency
10-6 °C-1 27.4 2.5x0.5x0.5 cm EN 14617-11
Dimensional Stability mm 0 30x30x1 cm EN 14617-12
Resistence to
Slipperiness
SRV Asciutta 68 20x20x1 cm
EN 14231
SRV Bagnata 16 20x20x1 cm
Impact Strength Joule 1.9 20x20x1 cm EN 14617-9
Fire Resistence Class 1
Frost Resistence KMf25 1.1 6x5x1 cm EN 14617-5
Chemical Resistence Class C4 7x7x1 cm EN 14617-10
Thermal Shock
Resistence
Damage Suffered None 2cm UNI EN 104
Colour Resistence
to Light
Notable Varation of colour Slight DIN 510994
Slipperiness
Rubber on Dry Surface
Rubber on Wet Surface
Leather on Dry Surface
(Honed Surface)
μ
μμ
0.88
0.41
0.45
30x30x2 cm B C R
Emission of lead (Pb) and
cadmium (Cd)
0
OKITE® Test Results
Pb C d

http://www.okite.com/pdf/OKITE_Tech_Manual.pdf

[RobotFixer]

Greetings all,

just found this thread a few days ago.
heroic contributions by very insightful folks.

i think i have read all the posts but i have not seen strength data (msds) for the us composites 635 poxy. except for post 1082 which refers to it as 19000psi epoxy. is that compressive strength?

i was hoping to compare the 635 with this
http://www.aircraftspruce.com/catalo.../polyepoxy.php
and other high strength offerings. its double the price, but i only need 2 gallons or so.

i work on machine tools and used to have some samples of commercial e/g cores, they were ROCKS.

i keep thinking that the 635 isn't strong/hard enough for this application.
could be real wrong though..


Chris

[ckelloug]

Chris,

I'll look up the relevant data and provide my opinion when I get a chance. Anybody else want to chime in?

Regards all,

Cameron

[ckelloug]

Chris,

Sorry for the long delay in responding. Too many projects, too little time.

I have taken a look at the aircraft spruce epoxy. The tensile strength of 8.8ksi and modulus of 500ksi are good as is the working time but it worries me that they do not give a viscosity. In general, it is very similar to the numbers given for most epoxy. It's better sounding than the U.S. Composites 635. In fact it could be something similar to the Hexion 813/IPDA since they make the comment about 2 different cure mechanisms.

For comparison, from a post I made on pg 277, the 635 combo has a tensile strength of 10ksi and flexural modulus of 375ksi. I can't figure out where I got this data from as my notebook is in the lab right now across town.

In general, the 635 combo e.g. Reichhold 37-127/37-606 is suboptimal but it is cheap and easy to get. I personally want to use Hexion 813 with Evonik-Degussa Isophorone Diamine.

The Epoxy in the us composites 635 low viscosity system is diluted with C12-C14 glycidyl ethers which have a moderate negative impact on strength and modulus as reactive dliutants go. It's much better to use something like the Hexion 813 which is diluted with cresyl glycidyl ether.

In general with quartz aggregate, the difference between the 635 and the aircraft spruce epoxies is likely to make about 10% difference on modulus and a negligible difference in strength according to the nomograph I made and posted back in http://www.cnczone.com/forums/showpo...postcount=3119.

In short, the aggregate is the important part. I've had limited success with the aggregate modeling effort I started and I haven't found a lot of time to finish this effort. The best result I have published here that I believe is reasonable is equal parts by solid volume of Agsco #6 Al2O3, #4 Quartz, #2 Quartz, #2/0 Quartz, G800 Zeospheres and G200 Zeospheres. I predict a fill percentage of 87% plus or minus 3% and perhaps some segregation under vibration. Given 635 epoxy, this should have a modulus of about 4000ksi. Given the aircraft spruce epoxy, this should have a modulus of about 4500ksi.

We really want an aggregate mixture that is closer to 92% fill rate which would produce a modulus of closer to 6000ksi with the aircraft spruce epoxy.

At any rate as a rule of thumb, if you design a solid machine, strength won't be a problem as holding deflections small enough for machine work usually ensures the parts are strong enough to carry their load.

I'd be glad to do flexural strength and modulus tests on a sample if it would help with your design. You'd need to make a plate 1/2 inch thick, 5 inches long and 7.5 inches wide for me to test it. I can handle the sawing that plate into the 1/2 thick, 3/8 wide by 7.5 long pieces required for the D790 test.

Regards, Cameron

[jberg]

Hi all,

I have look info about "epoxy granite" and similar products from web and find this thread. I just read trough this whole chain and get a lot of info so thanks for all you who have share you info.

Have anyone notice this pages http://www.silicafume.org/ku-mix.html where is some program to made aggregate optimization? Could this program be usefull for yours purpose?

Do you know any other shareware programs what could be usefull?


I will do some epoxy+fillers stuff (maybe using aluminiumoxcide) but not for CNC machine frame. For my pupose epoxy granite is to britle but with some carbon etc fibers it could also be usefull.

I made some test patch where I add some glass fibers but so far probles has been that those don't mix well (if those are longer that larges aggregate particle). Those fibers will stick together so have you any trick how to avoid that?

I have tried short one also and if those are short enougt (few mm long so more like a sand) to mix with other aggregate then those don't give as much stifnes as longer one could give.


- Jberg

[ckelloug]

Hi jberg,

I just took a look at KU Mix. None of the methods it employs are particularly bad but if you believe <u>Concrete Mixture Proportioning, a Scientific Approach</u> which is the book I have been working from, the methods KU mix uses are suboptimal. It also seems very specific to concrete.

Fibers affect the aggregate percentages required and they do affect viscosity a lot. Siloxanes and titanates could dramatically reduce the issues with fibers, especially short ones, on the viscosity of the mixture.

E/G is useful in situations where what is needed is compressive strength and the tensile loads are relatively minimal. If you have an application where there is even a question of whether E/G is too brittle, you're probably better off with a more conventional carbon fiber or kevlar composite.

I don't know of any freeware or shareware programs that help with E/G mix design although it may be possible to download a trial version of Beton lab Pro, the software written by Francois de Larrard, the author of the book I am following. I've written my own software to implement the models in the book but I haven't completed the optimizer.

This is a long thread and E/G is a complicated subject.

Good Luck,

Cameron

[RobotFixer]

some mindless and only roughly researched ideas follow...

i was thinking of a simple way to test epoxy durometer with a ghetto shore scleroscope type tool. drop a ball bearing (or better shapes) from a fixed height(maybe in a glass tube) and measure the rebound height.
maybe capture the drop with a webcam, measure in pixels (ad nauseum)... still looking for a good grocery store drop shape that would let everybody easily test with similar results. m&m's maybe... ha.

with the capability to pour all the weird shapes that are possible with e/g, any recommendations from our engineer friends on how to roughly structurally simulate these new options?

i typically model with Rhino, Qcad, or Blender, and use Beamboy and spreadsheets to get rough deflection data.
that leaves a big gap at the end of the tool chain though, so i started looking into FEA or some other types of software to analyze 3d designs.

i am just starting the research, but, the required knowledge of all the fea solvers and methods of implementing the elements seems to put it beyond all but the most tenacious of hobbyists. fea is a fairly esoteric discipline, very little for non-engineers to get up to speed. tutorials, modeling for analysis, etc...

a python script for Blender that could do some types of analysis would be a monster for hobbyist's. but i dont know the math to write the code beyond Hooke's law or so..

is it possible for mere mortals without years of maths to make use of the free fea like tools for e/g creations?
or is there some middle ground software tools between something like Beamboy(single beams) and Lisa(FEA) that we are ignoring?


Chris



Resources:

Book:
Building Better Products with Finite Element Analysis
seems to have practical implementations without tons of math overhead.
$118 though

Statics:
http://physics.uwstout.edu/Statstr/i...0OF%20MATERIAL

FEA
i have been picking around here:
http://www.freebyte.com/cad/fea.htm
seems like a fair compilation.

Z88 - seems good and free, but i am still lost.
moving to...
Lisa - limited to 1200 nodes, but only $50 to register. recommended for beginners.
i can do the simple beam example from the manual, haven't been able to import from cad yet.

[galaxyman7]

I was thinking that another, cheaper plastic could be used instead of epoxy. You could use HDPE resin mixed with sand to achieve a stronger more rigid HDPE. Or you could try mixing sand with starch plastic, or "bioplastic". Mixing with HDPE would be hard since it does not become fully liquid, but I'm sure it can be mixed in with resin and then melted. I will try this, as I have much excess HDPE powder from my milling machine.

[ckelloug]

robotfixer,

FEA is difficult. I'm an engineer but not an engineer that's good enough at solving PDE's on the back of a napkin to write my own FEA code. The best course on writing FEA code I've found is at http://www.colorado.edu/engineering/...rses.d/IFEM.d/

I gave up and bought a copy of NEI Fusion as I don't have the time and skills to write my own FEA software in a reasonable timespan. Unfortunately, it was a $4k expense justifiable to me only because I am doing a lot of other things. Fusion contains the modeler from solidworks and NASTRAN based linear analysis software.

The worst problem I see right now, and I am ashamed to admit it, is we still don't have a good working material design as hard as many of us have worked. I have tested a few samples from folks on this thread. None of these were heavily engineered materials and it showed with too much epoxy. The results were approximately 2000ksi in flexural modulus and about 2ksi in flexural strength. Commercial material is about 2.25 times as stiff and 2 times as strong. One could still engineer acceptable machines from mediocre material but one would have to ensure with FEA that the stresses match what the material can take. For a lot of home shop style machines, I don't think this would be a problem but I haven't done the analysis.

I started writing a quick dirty GUI this weekend for my packing simulation from <u>Concrete Mixture Proportioning, a Scientific Approach</u> by Francois de Larrard in hopes of managing to hand optimize an aggregate formula from the AGSCO materials discussed over and over here. Most of the automated optimizations strategies I derived from the book only work well for optimal aggregates so It's time to break out the intuition.

galaxyman7,

Adding mineral admixtures to thermoplatics is a well known technique in the plastics industry. As far as I know however, it is necessary to process HDPE with high pressure injection molders to get it to mold well. I can't see any of us getting access to high pressure injection molders without some good luck on surplus acquisition.

Epoxy is nice because it is liquid at room temperature. It also has better thermal expansion characteristics and is less likely to have creep problems than HDPE. You have a good idea, but I don't know how I could implement it with the tools, resources, and preconceived notions I have.

Regards all,

Cameron

[galaxyman7]

I have created a very simple system for pressing recycled plastic into blocks that can be milled. All you have to do is make a square hole in a piece of mdf, then make a square block that fits into that square hole out of mdf. You screw an aluminum plate to the piece with the hole, putting the screws all the way around the hole. You use a barbeque or hot pad to get the aluminum hot, then you put your plastic scraps onto the aluminum inside the square hole. Once the HDPE is totally melted, you take the square block and press the plastic down with it until it fills all the spaces, making a square block of hdpe that you can use to mill. After milling, I take the leftovers and dump them back onto the aluminum and melt them again. What I am saying is that you could grind the hdpe into small pieces, then mix it with sand. You pour that mixture onto the aluminum and melt it, making a sand/ hdpe mix. When you press the block down, the hdpe will form around the sand, making your super strong plastic. I will try this when I get some propane for my bbq.

[brunog]

galaxyman7,
the problem with a HDPE sand mix is that HDPE will not bond to the sand particules. Wthout proper bonding tensile strength will be affected, However a HDPE/sand mix can surely make a decent abrasive resistant product.

Best regards,

Bruno

[jonixllc]

We make nanometer positioning equipment
Epoxy Granite (EP)
1. It's expensive - consider temporary mold costs, then production mold costs - mold changes (you always have some) - fixturing to set inserts or plates to bolt things to - grinding of cast-in inserts or plates (if necessary)

2. There is debate over material - some use granite - some use only quartz - some use diabase

3. Curing - there are some studies that suggest that thick sections act similarly to concrete - that is that some curing is not complete when you need it to be - measurements of slide straightness and dynamic errors (roll, pitch and yaw) will change as the material cures under your metrology equipment

4. Having said all that, it's good stuff
5. Impala Black is my choice for a machine base - no mica, therefore no grain, therefore more stiffness. It can be finely finished and is predictable during the finishing process

[galaxyman7]

Is there any kind of aggregate that will bond to HDPE? If so, it needs to be fairly inexpensive.

[Zathras]

A bit of Googling for sand & plastic came up with this interesting roof tile product;

http://www.yunona.in.ua/index.php?op...mid=41&lang=en

[jberg]

Hi

I find this study and it seems that they get similar strenght EG than comersial one (hergrate, zanite etc) using just two different size of aggregate.

http://www.scielo.br/scielo.php?pid=...pt=sci_arttext

I don't know how this large amount of epoxy affect to damping but compression strenght looks fine and they get that quite easy as far I understand right (no any micro stuff, vacuum etc tricks).

Then:

I made one table part just to see how good surface I can made. I use diabase and quartz sand and some epoxy (from local hadrware shop so they sell it under own brand and I don't know what that stuf actually is).

I use 5 different size of aggregate and about same amount each size (bye volume). I vibrate this cast so that I press sanding machine to table where that mold was fixed (table part is for PC keyboad size and I made class fibre mold using old table part) and this vibration seems to be enough because I need to add fine sand to mold becasue epoxy rise to top of mix.

Part surface is not perfect (have some small holes) but part is more rigid that origin hdf one and this looks as fine as granite table use to look.

BR
Jberg

[ckelloug]

Hi jberg,

It sounds like you've got a good experiment going. Congratulations on making a test part. I'm a few thousand miles form my lab right now so all I can do is post and write code.

In terms of your reported results, In general, the papers I have read suggest that more epoxy will mean marginally better damping at the cost of stiffness. Keeping the epoxy to a minimum is what the models suggest is needed to get the stiffest material possible.

Vacuum degassing your epoxy, the addition of a deairing agent such as BYK A525, the addition of an epoxy or amine siloxane such as DOW corning Z6040 or Z6020, or the addition of a suitable titanate could all help with your tiny bubbles problems.

The equal parts mixture you said you are using is the simplest one and should generally perform better for stiffness than Fuller's mixture which was being used by Thomas Zietz's group for machine tools.


I've taken a look at your Brazillian reference. I think it is interesting but in my mind flawed. That matter is arguable, but I see a few problems with the applicability of their work:

<ol>
<li>They optimize compressive strength rather than compressive modulus which is the ultimate factor in E/G usage since stiffness generally defines cross section rather than strength.
<li>They divide aggregate into 500-106um and 106-45um. The wide variety of particles present in these two fractions could mean any number of actual particle distributions and certainly doesn't distill to a true 2 component mixture.
<li>They vary components in 10% increments. I think this means they could miss the peak performance which will likely be in smaller increments.
<li>They do everything by mass percent which is not as physically meaningful as volume percentage in the models I have seen. See the Hashin Shtrikman Model of Stiffness for an example of why to use volume.
<li>They assume that 15% and 20% by mass are the only epoxy ratios to use whereas 100% minus the volume fraction of the aggregate is likely the more reasonable answer.
<li>They make no mention of titanate or silane additives and merely complain about microstructure while first principles and other papers explain why these additives can fix microstructure problems.
</ol>

In my opinion, these researchers are trying to use what they think is an easy problem to demonstrate their prowess in Design of Experiments techniques. A Design of Experiments based approach was suggested on this thread by sigma_relief on pg 257 see <A href=http://www.cnczone.com/forums/showpost.php?p=437776&postcount=3074> post 3074</A> 8 months before this paper was published. I wonder if that's where they got the idea?

I like the bibliography of their paper and think we may find some interesting references there. Their paper also does provide a nice quick overview of how to apply DOE principles to E/G: if you ignore their data and reparameterize the optimization with meaningful variables and optimization parameters. I don't think the authors are particularly serious about epoxy granite. In my view, they have shown that it is possible to use a very powerful design of experiments technique to optimize the wrong parameter in terms of the wrong variables to get a suboptimal result. They did some of the right things but in my view failed to investigate the context into which their result should fit.

[samara samara]

hello...i'm new here...i'm looking for a scimatic daigram for control board for forklift still type 7030..7034..7042
plz help...

[tootalew]

Just a bit off topic, but it is a machine and it has granite table. I walked in Home Depot today, and there new table saw has a granite top. I am guessing that it is composite quartz, just guessing.

Here is a link.

http://www.ridgid.com/Tools/R4511-Table-Saw/

[Brenck]

Man... This thread is going crazy, if u google epoxy granite 99&#37; of the findings point it to cnczone.

Does anyone knows how % of epoxy walter used on his castings? I could find the agregate recipe but no epoxy/agregate ratio.

Did he finished his machine?

[Quad2T]

Just a bit off topic, but it is a machine and it has granite table. I walked in Home Depot today, and there new table saw has a granite top. I am guessing that it is composite quartz, just guessing.

Here is a link.

http://www.ridgid.com/Tools/R4511-Table-Saw/

Nah, it's probably a granite slab right from China. Ones that size usually run about $100 depending on the level of post grind/machining work.