[ckelloug]

Fyffe,

Are we sure that the compressive and tensile moduli are the same for E/G. I'm not experienced enough with composites to not want to double check this. At any rate, I think that the accures number for Young's Modulus is probably wrong: For thinking numbers, I'm thinking in terms of the numbers for Zanite available from http://www.zanite.com/Zanite_brochure.pdf. 4.5e6 is the value they give for E at the Zanite site while 4.5 e5 is the number they give over for E at accures. It engineering: What's an order of magnitude right?

[fyffe555]

Fyffe,

Are we sure that the compressive and tensile moduli are the same for E/G.

Heck no! , or rather there's definetly a question of what the numbers actually are. They are an order or more less than steel certainly and, No, comp and tensile are not the same for E/G, but they're not the modulus for the material either. The modulus sounds wrong as I tried to suggest but I've since found reference to the same 4.5e5 so who knows...

I'm not experienced enough with composites to not want to double check this. At any rate, I think that the accures number for Young's Modulus is probably wrong:

Agreed, 4.5e6 is more likely but there's such a wide range possible with the material, just look at the experiments in this thread. A carbon or glass laminate composite which I was once more comfortable with 15 years ago can increase by an order if the laminate is heat cured after layup for example.

If it IS 4.5e6 the same calcs I tried to show still apply with E/G at 1/6.4 that of steel. It still will not make much difference in Larry's examples. Not to say E/G isn't an excellent material in larger cross sections.

I do have a question about the Zanite pdf as it seems to say E/G is 45 times better than steel in suppressing vibration, because E/G has higher 'dynamic stiffness' whatever that's supposed to mean. Someone's playing loose with the dictionary I think, generally stiffer materials transfer vibrations better....

[Geof]

....I do have a question about the Zanite pdf as it seems to say E/G is 45 times better than steel in suppressing vibration, because E/G has higher 'dynamic stiffness' whatever that's supposed to mean. Someone's playing loose with the dictionary I think, generally stiffer materials transfer vibrations better....

I think Zanite plays loose with a lot of things;I think they are the ones giving numbers like +/-0.0001" precision off the mold in a page I found on their site.

I think the claim for 45 times better damping is more or less believable but I think the dynamic stiffness, whatever that means, has nothing to do with it. I think the E/G, because it is composed of two materials with radically different individual moduli, simply acts as a constrained layer system.

[lgalla]

:drowning: Quote from Andrew
"I do have a question about the Zanite pdf as it seems to say E/G is 45 times better than steel in suppressing vibration, because E/G has higher 'dynamic stiffness' whatever that's supposed to mean. Someone's playing loose with the dictionary I think, generally stiffer materials transfer vibrations better...."
Being a non engineer type and bad at math I wonder about the claims also.Go to Home Depot and knock a granite tile.I will ring like a bell or a steel tube.To exadurate,their claims may be due to a machine base 4'thick.I assume any E/G structure relies on mass or thickness???
Please forgive my ignorance.
Larry

[martinw]

Twisting forces are somewhat different from a simple deflection.

To answer the question; adding E/G doesn't appear to resist torsion or twist by any more than the tube alone would.

Andrew

Dear Andrew,


I hope I have not misunderstood.

I can't back this hunch up with calculations , (wish I could) , but IMVVHO, adding E/G to fill a hollow steel section will almost certainly increase its resistance to twisting.


Suppose you have a hollow steel box section. Suppose it is vertical, and you are looking down into the box. If you twist it, it will not deform into a perfect parallelogram, but some parts of the walls of the box will be closer together than before, and some will be further apart.

If you repeat the process with the box filled with a material which is strong in compression, surely it has to increase the section's ability to resist twist, even if the filler is weak in tension?

My 2 cents.

Best wishes

Martin

[fyffe555]

Dear Andrew,


I hope I have not misunderstood.

I can't back this hunch up with calculations , (wish I could) , but IMVVHO, adding E/G to fill a hollow steel section will almost certainly increase its resistance to twisting.


Suppose you have a hollow steel box section. Suppose it is vertical, and you are looking down into the box. If you twist it, it will not deform into a perfect parallelogram, but some parts of the walls of the box will be closer together than before, and some will be further apart.

If you repeat the process with the box filled with a material which is strong in compression, surely it has to increase the section's ability to resist twist, even if the filler is weak in tension?

My 2 cents.

Best wishes

Martin

Mostly correct - but the E/G will only have an effect if the box does deform. If you take larry's example, adding E/G will have an effect only if compression of the core occurs - and even then it's not as strong as steel. If you have a tube of sufficient strength to keep defelction within the requirements of a machine, ie in the thousanths then the E/g filler will not get compressed and so add minimum strength.

The twist should reduce the cross section area and so compress any core, not put it in tension., which is good since E/g is much stronger in compression. But the E/G is still weaker than the steel around it an dthe peripheral tube will carry the majority of the load.

You could use a weaker tube and rely on compression of the core or greater core strength to contribute to the total strength - but why would you? Steel is a $1 or even 0.50c a pound. E/G is more expensive.

E/G has it's place but you can't expect to do everything with one material.

[martinw]

[QUOTE=fyffe555;287890]Mostly correct - but the E/G will only have an effect if the box does deform. QUOTE]

Dear Andy,

Maybe I'm being stupid again.

If the idea of filling the box is to prevent deformation, and it has that effect....

I'm lost.

Best wishes


Martin

[lgalla]

I think I got it.
You would not design a machine that relied on the modulus and compressive strength of E/G.Our tube is strong enough on its own and our E/G does what it does best,vibration damping in this case.
Larry

[ckelloug]

http://www.engissol.com/Article2.htm

Still Working on Calculations.

[brunog]

Fyffe,

Are we sure that the compressive and tensile moduli are the same for E/G. ?

Based Report #6361 from the NIST the specs (page 3)are the following:

Density: 1950-2400 kg/m3

Linear shrinkage: .003-.05%

Compresive strength: 50-150 Mpa

Flexural Strength: 15-50 MPa

Tensile Strength: 8-25 Mpa

Modulus of elasticity in compression: 20-40GPa

Modulus of elasticity in tension: 12-15 GPa

Brinell hardness: 240-400 MPa

Grindability: .10-.30 cm

Poisson's ratio: .30

Linear thermal expansion 10-35 (1/K)*10^6

Adhesion to steel: 5-14 MPa

Adhesion to concrete: 4-6 MPa

Water absorption: .02-1 %

Best regards

Bruno

[lgalla]

Finally nice epoxy curing weather.No posts for 2 days.Walter and the other experimental guys are taking their epoxy blocks outside to enjoy the good spring weather and speed up the cure.Progress reports are fourth coming.
Larry

[eloid]

have a quick question, sorry if allready asked
the thread is to big to all of it,


what resin can be mixed with water and concrete?

[ckelloug]

Eloid,

I'm the thread's newest participant but the idea is to use epoxy as the matrix to bind granite particles. It's called concrete because it has a matrix with embedded stones but in the discussion on this tread the stones are embedded in epoxy, not portland cement.

[lgalla]

Yes the thread is to big or long to expect someone to read it all.I have been onboard since the start and find it difficult to remember everything.
Your reference to resin and water would be portland cement which we have discounted as not suitable for the purpose due to shrinkage and possibly rusting.Stick with epoxy.
Larry

[lgalla]

Thank you for contacting us about your casting application. If you can
live
with the long set time, our CLR 1190/ CLH 6230 would meet your
requirements.
The data sheet is attached. Please let me know if it looks promising.

Do you know how much you might need for your application?

I'll look forward to your reply.

Best regards,

Barbara Zymiak


Technical Sales Support
Crosslink Technology Inc.
6380 Viscount Road
Mississauga, Ontario
Canada L4V 1H3
Tel: 905-673-0510
Fax: 905-673-0519
Toll-Free: 1-800-563-3769
www.crosslinktech.com



-----Original Message-----
From: [email protected]0 [mailto:[email protected]]
Sent: April 18, 2007 11:42 PM
To: [email protected]
Subject: General Inquiry

************************************************** **************************
***
ProdType: Epoxy
ProductNo:
FirstName: larry
Do you have an epoxy for casting quartz sand/granite aggregates
mix.Pours
would be 1"to4" thick at a time.

The viscosity should be under 600cps and have no reactive dilutents.The
purpose is to damp resonances on machine tools

Thanks

Larry



I have not been sleeping on E/G but large corps E mailed have not responded.
I Emailed company's about "air release additives but had no reply.I emailed US composites about our project and received no reply.Crosslink Technology was the only responce.The big guys don't care about the small guys.
Larry

[ckelloug]

The other night, I pulled Materials Science and Engineering by William D. Callister Jr. off my shelf and did a touch of reading. Callister makes the interesting point that particle reinforced composites are divided into large particle composites and dispersion strengthened composites. The two have separate mechanisms of action.

In large particle composites, the matrix carries load between the large particles which have the majority of the strength. An example of this would be ordinary concrete with aggregate.

In dispersion strengthened composites a material with a grain size on the order of carbon black is added to the matrix to pin dislocations in the crystal lattice of the matrix. Callister cites synthetic rubber in tires dispersion hardened with carbon black as an example of this type.

The upshot of this is :idea: Walter and martin, the carbon black actually dispersion strengthens the epoxy and should probably be considered an essential ingredient in our recipe for E/G not for looks but for strength


For Igalla,

I worked out that to hold deflection of .0001in on a 4ft long beam supported at the ends with a 200lb load in the middle one would need a 6.6 inch square solid steel bar weighing in at 595 lbs (steel density .285 lb/in^3) 123$/ft of 3.5 cold rolled from online metals*4ft*3.44*cost of 3.5 inch bar for 6.6 inch bar=$1692.

To hold the same load with a square Epoxy Granite beam, the beam would need to be 12.5 inches across and weight 525 lbs (density .07 lbs/in^3) at a cost of $260 (60/cf). The E/G calculation assumes 2300ksi (16GPa) for the Elastic Modulus of the E/G which is 1ksi higher than the pure tensile strength since I haven't worked out the exact load capacity due to the fact that the compressive modulus is much greater than the tensile modulus. The 12.5 inch cross section beam is an probably upper limit on the necessary size for a .0001 E/G beam until a better calculation can be done.

Note that these calculations ignore the weight of the members in the deflection calculations.

I'm currently planning to write a java application to do beam calculations for us as it will be portable. I'm working on adding the effects of shear. Shear is generally considered negligible for members 10 times their diameter in length and thus my books are a bit sparse on how to do it.

The bottom line is that the lower density of the E/G appears to allow square E/G members sized for a given deflection to weight less than solid steel members sized for the same deflection. Due to the vibration tests already reported, it would be guaranteed that the E/G member would vibrate much less than the equivalent steel member.


Conclusions: E/G is definitely cheaper than steel for solid members. It's entirely possible to build an all E/G gantry though it will be larger physically than a steel one. Even though E/G has a lower tensile strength than steel, it has a better strength to weight ratio by 1.13 and a much (6.5 times) better strength to cost ratio. Carbon black should substantially strengthen any E/G recipe provided that there is enough epoxy to wet it and all of the aggregate. I suspect that the carbon black will suck up lots of epoxy: Walter?

[walter]

Great stuff Cameron, much appreciated.

Seems that I interrupted the engineering session, when I posted that article earlier. I apologize. The role of engineers may be crucial here so please continue. I'll step aside for a bit.

Re: Carbon Black

I used the Cabot Carbon Black substitute. It was somewhat difficult to mix, so basically I've overdone it. This seems to be the reason for failed sample. I'll work on it the next time, your theory sounds interesting.

Beam calculation application would be awesome. Thanks for doing it.



Larry,
The big guys do not respond. The technology is simple and fairly easy to use - they may not want to let the details out in the open. But it's too late now, I think I got it.
_

[brunog]

Cameroun,
do you have any idea what ratio per weight or per volume of carbon black should be used in the E/G mix

Larry,
does carbon black have any effect on the epoxy viscosity?

Best regards

Bruno

[ckelloug]

Walter,

Fear not about interrupting. I thought it was cool that you thought of carbon black for a pigment. I would not have thought of it at all. I figured you were right but when I pulled Callister's book off my shelf I saw that comment about dispersion hardening and I wasn't even looking for it. Anything with particles the same size as carbon black ought to exhibit the effect of carbon black. Thinking back to something I saw recently, Silica Fume is another product whose addition would provide a similar effect. It has the added benefit on this thread of being almost Granite too The key need for dispersion hardening is particle size as much less than 1 micron as possible.

[ckelloug]

Brunog,

The Materials Engineering book I looked at was not specific. I've been looking at this document on using silica fume to reinforce concrete and it gives some sample non-polymer concrete recipes with silica fume which is similar in effect. http://www.silicafume.org/pdf/silica...ers-manual.pdf

It's general consensus is that the right amount of dispersion hardener be it silica fume or in our case perhaps carbon black, is somewhat trial and error.

Anybody have a friend in their local highway department who could sneak some samples in for compressive strength tests? Otherwise, can anybody come up with a quantitative compressive strength test that can be done on samples at home?