[badhabit]

Hi Thomas - I think testing in opposite directions is a good idea. What is the biggest size of aggregate? Add stress to the spreadsheet so you have an estimate of breaking. Peter

Specific Beam Loading Case: Simply Supported: Center Load (efunda.com)

The biggest size is 4mm in the Silimix282...if this works out fine, i think i'll cast a couple more, but with some larger aggregates in...

I'm not sure if i can calculate anything that will tell me the breaking point of the EG?

[peteeng]

Hi Thomas - You will be able to calculate the stress in the beam at the applied load. From concrete data you will find a typical tensile stress of the material. Here's a grout with a flexural strength of 10MPa. So load to 5MPa.... Peter

[badhabit]

So...the results are in...short answer: 22-23Gpa...for a 12% by-weight Silimix282 with R&G ResinL + GL2 Hardner. Cured for 2 weeks at 30c, no postcure treatment.

I've attached an excel spreadsheet with my measurements and the calculations. There is a lot of averaging, but feel free to comment if you think i've made a mistake in either calculations or method...

So the test-samples have plenty of holes and trapped air, just like my mill has. It can be hard to see, but if you look carefully there is a huge amount of small airpockets just below a "film" of epoxy:
Attachment 505142 Attachment 505144

The setup for testing was simple. A sturdy table. Two sturdy pieces of iron with as little overhang has possible which the sample rests on. Measure the deflection in the middle and load the sample with a variable load, in this case we just used a bucket of water. We weighed each "liter" we poured in and calculated the applied force the the total mass in the bucket.
Attachment 505134 Attachment 505136 Attachment 505138 Attachment 505140
(The wire doesn't touch the probe, even though it looks like it in the pictures)

We tried rotating one of the samples 90deg and repeated the loadtest and saw identical numbers, so we ended up concluding that the samples yield the same results nomatter the load-direction. So we only recorded the numbers for 1 load-direction per sample.

I'm not sure if i'm excited or surprised about the results or not. It is kinda what i expected. My previous calculations was using 20Gpa, so my guess wasn't far of at least... Could have hoped for 30, but feared 10....so i guess 22-23Gpa is a "good solid expected number"? At least there is no more guessing, thats a good thing, yes?

Let me know if i missed something or if i should try to measure something else with the samples...

PS: I got a free sample of degassing agent from r&g.de after talking to them about my issues with trapped air. I might try and cast a sample with this and see if i can get more air out and see if it makes any difference....

/Thomas

[peteeng]

Hi Thomas - Good news. The numbers are in. Good to see they are the same order as mine. I've had a restless night thinking through the epoxy granite stuff actually. Come to the conclusion its orthotropic and visco-elastic and low modulus. Makes engineering with it difficult. Well at least for me trying to do optimisation. I think the main issue is its tensile modulus and compressive modulus are very different hence the low mean modulus in a flexure test. But then it will have large hysteresis so will be damp.... there's always an up side Next step I suppose is - Are you going to break the sample? Peter

I have used degassing agents and they sort of work. They work by having a bipolar molecule that one end likes air the other end likes the resin. So they stick up thru the surface and reduce the surface tension allowing the surface bubbles to burst easier. But this does not help the internal bubbles. So the mix has to be agitated to get the bubbles to the surface. This is OK in a resin mix but in a "solid" mix its tough to get the resin circulating. I made a vacuum chamber with a gland that a mixer went through and mixed under vacuum then added the solids. This helped... Peter

[badhabit]

Hi Thomas - Good news. The numbers are in. Good to see they are the same order as mine. I've had a restless night thinking through the epoxy granite stuff actually. Come to the conclusion its orthotropic and visco-elastic and low modulus. Makes engineering with it difficult. Well at least for me trying to do optimisation. I think the main issue is its tensile modulus and compressive modulus are very different hence the low mean modulus in a flexure test. But then it will have large hysteresis so will be damp.... there's always an up side Next step I suppose is - Are you going to break the sample?

Yeah, i could, but not sure if there could be anything interesting to figure out in the process?

PS: I've been thinking that i should try adding larger aggregates to see what it changes. The largest aggregate size in Silimix 282 is 4mm, was planning on adding some rough 16mm granite gravel. So i might try to create 2 new castings(since i still have both forms)...and now i have a "baseline" in the two existing castings which performed very similiar... then i could make two new ones, one which i would try to degas(still doubtfull, but i'll give it a shot) and then one with the larger aggregate. I don't expect the degassing to be succesfully or affect the modulus significantly, but it would be nice to confirm it and remove the guesswork, nomatter the results. I expect the larger aggregate to increase the stiffness some, but i'm not sure by how much or if it significant at all.... thoughts?

/Thomas

[peteeng]

Hi Thomas - The concrete chemist I worked with said that the bigger the aggregate the higher the modulus. We use a stone called blue metal here, mainly basalt. I don't understand why that would be. But a test proves it. What is the aggregate you have? Do you have any grout or concrete to test? Peter

[badhabit]

Hi Thomas - The concrete chemist I worked with said that the bigger the aggregate the higher the modulus. We use a stone called blue metal here, mainly basalt. I don't understand why that would be. But a test proves it. What is the aggregate you have? Do you have any grout or concrete to test? Peter

Yeah, when i see cutouts of "professional machinebases" they usually have very large aggregates embedded(guess: size=50-80mm!) within them. But it also makes sence, less flexible epoxy=stiffer. The tradeoff will be a less dampning effect i guess...since there is fewer "interfaces" in the mix... I think the gravel i have on hand is "granite", but i'll have a look at the bag later on and see what it is...and take a picture...

I was actually thinking about trying some grout... i have access to this:
https://www.moertelshop.eu/buy-nanodur-cement-cheaply_7 (which is apparently this: https://www.dyckerhoff.com/documents...3-e205050e524d )
....but it doesn't seem to be "ready mixed", i need to buy aggregates, plastizer and so on along with it...

/Thomas

[peteeng]

Hi Thomas - One of their mixes claims E=80GPa. Give that a go. I expect in flexure it will be lower than the compression value. That would be interesting. I tried to get some shipped here some years ago but they would not sell it outside EU. Some of the gouts and cretes I see here now have the compression modulus and the acoustic or "dynamic" modulus quoted. The acoustic modulus is usually close to the compression modulus. Unfortunately, the acoustic is a compression process so they are similar. Need a pure tension test to really see the delta. Peter

larger aggregate is cheaper less processing and fills large gaps.

[badhabit]

One of their mixes claims E=80GPa. Give that a go. .... Need a pure tension test to really see the delta.

Yeah, i'll see if i can order the stuff for a test af that...it will take a couple of weeks since i have a lot of traveling in the upcoming weeks...

Maybe i should figure out a (simple) way to measure dampningen aswell...as i feel that it would be the major tradeoff one is "sacrificing" for better modulus... otherwise i would just grap af piece of steel with 10x modulus of EG and say that "steel is much better"...i just haven't found a good way to test this "at home"....it would also be interesting to compare this between EG and grout...

/Thomas

[peteeng]

Hi Thomas - If you have a high speed camera you can image the cantilever damping. So support the beam on your nice steel table deflect a set distance and image the vibration / damping. Count how many cycles the beam goes to come to rest. If you have a similiar size steel or aluminium beam do the same to try to show a difference. Peter

[badhabit]

If you have a high speed camera you can image the cantilever damping. So support the beam on your nice steel table deflect a set distance and image the vibration / damping. Count how many cycles the beam goes to come to rest. If you have a similiar size steel or aluminium beam do the same to try to show a difference.

Yeah, well, i have a lot of stuff, but high speed cameras is sadly not on the inventorylist :-( Apart from this, the deflection will be very small with these sample-sizes...so i don't think this is practical...

I have a quite good osciloscope however, thought about using a microphone or straingauge somehow to measure the oscilations, then i could measure the "half time"(like in radiation) and give an arbitary, but relatively comparable, value...but i need to excite the sample in a repeatable way for this to work...and i don't have to good way of doing that....yet...

I also feel that it might be hard to repeat this in an reliable way between different samples which might vary in length... i might need to somehow saw the samples to a fixed length...which has the sound of "diamonds" to it :-/ I also need a predefined and repeatable fixture of sorts...

/Thomas

[badhabit]

I have a quite good osciloscope however, thought about using a microphone or straingauge somehow to measure the oscilations, then i could measure the "half time"(like in radiation) and give an arbitary, but relatively comparable, value...but i need to excite the sample in a repeatable way for this to work...and i don't have to good way of doing that....yet...

I also feel that it might be hard to repeat this in an reliable way between different samples which might vary in length... i might need to somehow saw the samples to a fixed length...which has the sound of "diamonds" to it :-/ I also need a predefined and repeatable fixture of sorts...

Maybe an acceleration sensor, fix the sample in one end and "hit it with a hammer" in the other? I'm unsure if the damping is frequency dependent or not? If frequency can be ignored then length and amplitude can be ignored and the only data left is the logarithmic "decay" of the vibration, which can be read from the osciloscope....i think i'll need to find a suiteable sensor and try it out and see what happens....

/Thomas

[badhabit]

Maybe an acceleration sensor, fix the sample in one end and "hit it with a hammer" in the other?

A lot of talking to my self here, but i just ordered a couple of accelerometers. I'm going to see if something can be concluded from that...

/Thomas