[Zumba]

Thanks, Martin for the epoxy/steel shear strength figure. 2900 psi should be a more than adequate bond, considering the massive surface area compared to even a full length weld bead.

[lgalla]

Hey Martin
At 95%fill I think we would call it solid granite.The properties of E/G are the result of two dissimilar materials acting against one another.Two different resonate frequencys result in cancelation of resonance or vibration.In previous posts it has been stated E/G does not have the modulus or stiffness of steel.Perhaps a few layers of glass or carbon would increase the bending strength.Any thoughts?
Zumba
If you are worried about the shear strength of epoxy,I would not get on an airplane.
Atenman1
I beleive you posted a link to a concrete vibrator.I cannot find it.Please post again.Thanks Larry

[Zumba]

In regards to the stiffness, keep the formulas for beam deflection in mind. The important one here is the area moment of inertia of the structure.

Area moment of inertia of a solid rectangle cross section I = bh³ / 12

The important constant here is E, modulus of elasticity, which is 5000ksi, 1/6 that of steel.

If you want to get a rectangular beam of E/G to be equal in stiffness to any given rectangular bar of steel, you'll have to either increase "b" (base, aka thickness) by six times, or increase h (height) by the cube root of 6 -> 6^(1/3) = 1.82.

As for reinforcement, fiberglass has a modulus of elasticity of 2600ksi, inferior to E/G. Carbon fiber has a modulus of 77,000ksi, which is very good. But a thin little bit of CF will still flex more than a thick pour of E/G, and EG is way cheaper.

I know that for my machines, I'll be use a steel/EG combination. I want the flat surface for grinding.

[martinw]

.In previous posts it has been stated E/G does not have the modulus or stiffness of steel.Perhaps a few layers of glass or carbon would increase the bending strength.Any thoughts?

Dear Larry,

My guess is that the best way to increase stiffness would be to add steel reinforcing bars (and/or square mesh) near the top and bottom surfaces, just like a conventional RC beam. However, an increase in stiffness may reduce the vibration damping properties that people want. I would have thought that with moderate steel reinforcement, the "E/G/S" would still retain damping properties. That's just a hunch though.

Best wishes

Martin

[Geof]

....In previous posts it has been stated E/G does not have the modulus or stiffness of steel.Perhaps a few layers of glass or carbon would increase the bending strength.Any thoughts?..

I think the concern about E/G being less stiff than steel or cast iron may be misplaced. Deflection under load depends on both the elastic (or volume) modulus and on the area under load. The area under load is going to be very large and deflection of the main structure will most likely be much less than the deflection of things attached to it such as rails.

[brunog]

Larry, Martin, Geof,
IMHO, i will be going with rectangular steel bar inserts coming out of the E/G approx 1/4" with bolted anchors to support the linear rails. I will also use steel plates (again anchored in the E/G) at all the fastening points with other E/G components. I will use inserts at other fastening points to the E/G components. I personally believe it's the best way to go so far. I also believe that it's much easier to grind, drill and tap steel than E/G and that's the reason I will be doing all of the above.

It only takes carefull planning and minutia in making the molds to get great results, I don't and will not leave any place for any surprises eventhough I am pretty sure I will get some surprises anyway. I prefer measuring twice before cutting (or maybe 6 times if I had 1 beer too many). I plan to make my machine out of multiple E/G components and building it one piece at a time, one mould at a time.

Bruno

[lgalla]

It just hit me on the head.(chair)
I should not worry about E/G stiffness.
What is stiffer?4"of E/G or1/2"cast iron?
Larry

[walter]

I plan to make my machine out of multiple E/G components and building it one piece at a time, one mould at a time.

Machine frame in the form of bolt-on E/G components.. Sounds like a good idea.

One could build a 3000 lbs monster and still be able to relocate..

[martinw]

Dear all,

This assumes spherical aggregates which pack in a random fashion rather than an ideal one.

My understanding is that different sized balls pack in a different way randomly compared to ideally. My post #390. The link suggests four sizes of balls. If you ignore the tiniest of the balls and replace them with epoxy, you have three balls (!)

If

(1) you choose 3 ball sizes...and..

(2) the density of the balls is about double that of the epoxy..and..

(3) the balls have the same density..

Here is the mix by weight

Epoxy 9%

Small balls 10%

Medium balls 22%

Large balls 59%


As regards size of the balls..

Whatever the diameter of the Big Ball.....

The Small Ball is 2% of it

The Medium Ball is 12% of it

BTW The 9% epoxy weight was chosen somewhere in the mid-range of what others use.

I guess the real problem is that it is hard to do the same calculations for non-spherical aggregates,

Best wishes

Martin

[martinw]

Please read post#380 not #390.

Fat fingers

Martin

[brunog]

Dear all,

This assumes spherical aggregates which pack in a random fashion rather than an ideal one.

My understanding is that different sized balls pack in a different way randomly compared to ideally. My post #390. The link suggests four sizes of balls. If you ignore the tiniest of the balls and replace them with epoxy, you have three balls (!)

If

(1) you choose 3 ball sizes...and..

(2) the density of the balls is about double that of the epoxy..and..

(3) the balls have the same density..

Here is the mix by weight

Epoxy 9%

Small balls 10%

Medium balls 22%

Large balls 59%


As regards size of the balls..

Whatever the diameter of the Big Ball.....

The Small Ball is 2% of it

The Medium Ball is 12% of it

BTW The 9% epoxy weight was chosen somewhere in the mid-range of what others use.

I guess the real problem is that it is hard to do the same calculations for non-spherical aggregates,

Best wishes

Martin

Martin,
the idea on non-spherical aggregate is to use an average size per size range. For example if an aggregate size is .4 - .5 inches use .45 inches etc..

Based on the 3 sizes you gave, I would use 10% small, 23% medium and 63% large proportions in weight or volume as you can assume that aggregates will have the same density.
However the 63% of large aggregate should fill in 100% of the volume to be casted, as the smaller aggregates with fill in the gaps between the large aggregate. The added epoxy will also fill the in more gaps as the aggregate will be soked with it.

Bruno

[martinw]

Based on the 3 sizes you gave, I would use 10% small, 23% medium and 63% large proportions in weight or volume as you can assume that aggregates will have the same density.
However the 63% of large aggregate should fill in 100% of the volume to be casted, as the smaller aggregates with fill in the gaps between the large aggregate.

Bruno

Bruno,

You may well be right. Given that the aggregates will not actually be balls,
the difference between your mix and mine probably will not make a blind bit of difference.


Best wishes

Martin

[brunog]

Martin,
The 63% should be 67%, I knew there was a bigger difference for the large aggregate (chair)

[martinw]

Martin,
The 63% should be 67%, I knew there was a bigger difference for the large aggregate (chair)

Bruno,

Again, you could be right.

Just for the hell of it, I "googled" random close packing of spheres again.

Someone has done some work on chocolate...

Random close packing of spheres in three dimensions gives a packing density of only (Jaeger and Nagel 1992), significantly smaller than the optimal packing density for cubic or hexagonal close packing of 0.74048.

Donev et al. (2004) showed that a maximally random jammed state of M&Ms chocolate candies has a packing density of about 68%, or 4% greater than spheres.

This is the link....

http://mathworld.wolfram.com/RandomClosePacking.html


Now, I very much doubt if "E/C" will take off as an alternative to "E/G", but I would stake my life on the guess that you could experiment more easily at home.

Best wishes,

Martin

[Geof]

Now, I very much doubt if "E/C" will take off as an alternative to "E/G", but I would stake my life on the guess that you could experiment more easily at home....Best wishes, Martin

But what about C/G. If it didn't work you could break it up into VERY hard candy.

[martinw]

But what about C/G. If it didn't work you could break it up into VERY hard candy.

Dear Geof,

I couldn't afford the dentist's bills.

BW

Martin

[jsage]

I knew there had to be a practical app to those questions about how many M&M's would fit in a given jar ; )

Math was my least favorite subject. I have become more interested in it as I go along. I have been looking at that site a bit myself (fractals).

On the EG I'd go with larger open reinforcing if you use it. Any tight weave will stratify the layers, create suboptimal aggregate matrix and reduce strength.

[lgalla]

Anyone done costing?From all the posts on aggregates,sizings ratios etc,I came up with a cost of $7.00Per sq ft.1"thick.It is also important to calculate your weight of your project as you may use large quantities of E/G and steel.If your mini mill weights 1000lbs is that accetable?Calculate your weight first and decide if it is acceptable or movable or won't crash through your floor.Thick base plates or filling of weldments may be much heavier than anticiapated.Pre calcaluations are important as you do not want to be short of epoxy and wait a week for delivery.Mutiple pours should not exceed 24hrs for secondary bonding.
Concrete vibrators appear to be a good choice for compactation as ratios are similar to E/G.Harbor freight has one for $79.00 and there are others on Ebay cheap.Correct vibration frequency will release air and void the vacuum de gassing process.
Larry

[walter]

It just hit me on the head.
I should not worry about E/G stiffness.
What is stiffer? 4" of E/G or 1/2" cast iron?
Larry

Larry, can I ask you a question...

How would you stiffen the 16' beam?

[Zumba]

Larry, can I ask you a question...

How would you stiffen the 16' beam?

Make it taller to combat vertical loads. Make it wider to combat horizontal loads.

Like I said in a previous post, the area moment of inertia is bh³/12.

Double the height and your beam will be 2³ = EIGHT TIMES stiffer vertically (but only twice as stiff horizontally). Vise versa if you widen the beam.

However, you must keep in mind that the beam will deflect under its own weight. Hollow centers are a good idea.