How to make the best epoxy concrete
Вставка
- Опубліковано 22 кві 2024
- There are quite a few videos on how to make or strengthen machine parts with epoxy concrete. But there is little evidence to substantiate which concrete gives the best results. In fact, people often just try something. In this video I try to get a better grip on this matter.
Again, no music. Too much to tell. - Наука та технологія
Lots of comments here regarding my mixing method. I would be doing that completely wrong, someone even called it shocking. I could count on a below average result with my odd way of mixing.
So I did another small test, and did what a number of you recommended.
Instead of mixing sand and epoxy for nearly 3 minutes, to which the hardener is then added and mixed again for nearly 3 minutes, as in the video, I mixed the epoxy and hardener for 5+ minutes, then added the sand and mixed again for 5+ minutes.
After curing for 48 hours the Young's Modulus of this rod was 13.5 GPa, so in fact it was slightly weaker than the rod with sand in this video (16.6 GPa).
But this bar was much smaller (ran out of epoxy) and the measurement was not that thorough. So let's call it a draw. 😉
Of course: Porous filler should never be mixed as shown. The epoxy then penetrates into the material without hardener. But non porous fillers can be mixed as I did without any negative consequences as far as I can tell.
A few tips I forgot to mention:
- Make sure the pebbles and sand are bone dry (a few hours in the oven at 150 degrees).
- Be very accurate when measuring epoxy and hardener. Follow exactly the recommended mixing ratio. 5 percent too much hardener means that 5% of the epoxy concrete is soft as butter, and 5 percent too much hardener leads to the same result because it contains liquid hardener. Please don't ask me how I know that.
- Mix very thoroughly. I mixed 2x3 minutes, but is was only a small amount of epoxy and filler. The concrete for the column of the mill was mixed mechanically for in total 10 minutes.
I have never seen a better example of some performing live science in my life. VERY well done!
Wow! Thank you ☺
A few things I have learned about epoxy granite.
The highest fill factor and best strength comes from:
1. A mix of medium and small granite, and quartz sand.
2. Ramming the material during filling for compaction, followed by vibrating (an orbital sander works for small jobs).
3. 12-16% epoxy by volume. The more surface area your particles the more epoxy it takes to coat them.
The strength comes from the granite and quartz sand. The more different sizes the higher your fill factor. Epoxy is compressible so you only want that to hold the filler in place. This is why it is important to compact it fully to get as much particle contact as possible.
There are some advertising materials from professional epoxy granite machines being cast on youtube, and their mix seems even higher epoxy content, since it looks quite liquid.
Depending on use case it can also be better to use gravel not pebbles. The jagged shape allows it to interlock, helping with tensile strength and deflection. Pebbles are better for compressive strength.
Excellent tips! Would a vacuum chamber help with small objects?
If the epoxy is thin and slow curing, you can definitely try that. Make sure that you do not apply this low pressure for too long, because if the epoxy hardens in the meantime, you may be left with some enormous vacuum bubbles in the epoxy. And I would reduce the pressure slowly so that the epoxy doesn't bubble or even overflow.
@@Michel-Uphoff there's a fantastically long thread on the CNC zone forum regarding epoxy granite. The CEO of one of the synthetic machine casting companies kind of broke it down in one of the posts, but the right mix of particle sizes is key. I also seem to remember that hot curing and degassing were one of the things that would add more, but was likely out of the hobbyists range
I have been contemplating doing this to my mini mill. It’s gratifying to see that the sand epoxy mix is confirmed as the superior choice. Also thank you for suggesting slow cure epoxy
Your welcome 🙂
A few things. First, for something like this you’ll want a fuller curve mix. This maximizes the filler and minimizes the epoxy. The benefit of a many sizes is that you get layers upon layers (in microscopic scales) of reinforcement. So optimally you would want like 10% X
Using a vibration device to help get the air to move helps alot, some have even went as far as it create a vacume on the ends to draw the air bubbles out.
MetalMusings did tests also that you might fine interesting also.
I like the testing methods used tho, you gained a new sub :D
look forward to more of your work.
Thanks for subbing
Thanks for taking the time to explore this. I'm way too Attention Deficit to even consider keeping good records! 👍🏻
Outstanding video. Thank you very much for preparing and presenting such a rigorous analysis.
Thank you very much!
Love the science lab you present to us here. Very well though out examples, variations of the rig, excellent methodology.
Thank you for sharing. 👍💪✌️
Thanks!
I read somewhere that it is necessary to use the most heterogeneous(from stone dust to noticeable rocks) filling, this allows you to avoid vibrational harmonics in the concrete frames of the machines. I hope Google translated it clearly))
Geweldig weer Michel. Dankjewel voor de test.. Zeer leerzaam !!!!!!
It’s was interesting seeing you work through the experiment
Great experiment. Very interesting
Tech Ingredients did a couple of videos where they created their own epoxy-based thermal compound. On one of the shows, he demonstrated how adding graphite powder to epoxy made it many times more rigid. I strongly recommend you watch them all. The host is very informative and explains everything in a scientific, yet approachable way. The channel is a treasure trove of technical information for DIYers. I'm sure you'll thank me once you've seen them.
😂 how sure?
I know them (of course)
@@Michel-Uphoff What did you think about adding graphite powder to the epoxy sand mix?
@@bobweiram6321As a dye for an outside layer? Or to increase the strength? I'm not convinced at all about the latter. Graphite is a form of carbon and often occurs in the form of thin, easily separated layers. So at least in one direction it is very, very weak. That's why it is used as a lubricant. However, carbon fibers are a completely different story, they are very strong and could be used as reinforcement.
Superb! Nice attention to the experimental design.
Thank you. 🙂
Thank you, always good, always thorough . Good work!
Thanks!
Thank you Sir for your extensive testing and sharing the results with ua.
It would be very interesting to understand how, to end of the day, the filling is impacting rigidity of the column?
crushed gravel aggregate with a wide range in sizes would make a difference, but your measurement method isn't really appropriate for the material. concrete is really only strong in compression, while this method has half the bar in tension. you'll have more meaningful results if you try to crush the bars lengthwise. that said, for your use case, you don't care about tensile or compressive strength, this is meant to damp resonance at the frequencies of your mill. rigging up frequency response test in the shop might be a bit of a struggle, but if you have a strain gauge, speaker and oscilloscope, you could get some interesting data
Cement concrete can hardly bear a tensile load (without reinforcement), but that certainly does not apply to epoxy concrete. My goal is not just to dampen vibrations, my goal is primarily to increase rigidity, especially resistance to deflection. That is why measurements are taken this way here.
@@Michel-Uphoff damping can be done mostly by converting vibration energy to heat. Heat near main source of vibration will be most, and localized, make local bend due thermal expatsion, adding positional errors. Maybe better will be use solid metal rigid base, but add counterweight with elastic material between (polyurethane known to be one of top of damping feature, there are also Cu-Mg and Al alloys). It will separate vibration source and dumping making whole structure more stable. Gray cast iron still top of the hill damping feature ofc.
Great work! Thank you for sharing!
Thank you!
Fantastic work!
Thank you!
Good work! :) -but remember that the increase in static or long term rigidity is probably the smallest part of the gains you will see with mass loading and epoxy granite :)
-static rigidity: deflection decrease is only determined by stiffness increase
-anti-mode rigidity (periodic but non-resonant movement): deflection decrease is determined by the combined product of stiffness increase AND mass increase
-modal rigidity (system resonance movement): deflection decrease is determined by the product of stiffness increase, mass increase AND mechanical loss increase (damping)
So for any transient load system (like a machine tool...) the decrease in the average max deflection under load is typically much, much larger than what the stiffness increase in it self would suggest.
My recommendation: concrete grain size curve. Study of this will cost a few minutes (maybe longer) but it can save a lot of time and effort in the future. BTW your results are good, this properties of the composites was measured a few hundred years ago. Anyway, good work.
Did they had epoxy a few centuries ago? 🤔
@@Michel-Uphoff Sorry, this is not about cement or epoxy or another bond agent. It is about how filling change a properties of composite. Try look to these a few graphs and try (in mind) replace word cement to another word, for example some glue (for example wood glue). It is necessary some detachment of focus just to specific bonding agent, in your case epoxy.
@@mirosolto8078 Ah. You mean that a clever distribution of very fine, fine, medium and coarse material gives a higher packing density. If so, yes I know. See the sketch in the video at ua-cam.com/video/sBn0Um6XFX0/v-deo.html
There you can see that the ridiculously coarse filling of only two round pebbles at best leads to a filling of 66.66%, and also that with pebbles of the same size (and as small as you want them to be) you never reach more than 74.05% density. But with a clever mixture (like that in ready made cement concrete filler, the sketch in the center) the density goes up. In my case to around 80 vol%, so I needed only 20 vol% (8 weight%) epoxy and filler. Possibly I could have achieved slightly better results with even more variation in grain size and very thorough compaction of the filler.
@@Michel-Uphoff Thank you, my message was received. Base on my (expensive) experience distribution of size of grain is important, the best curve is B. Next important parameter is shape of grain. The best is irregular, coarse, round is bad. It is reasonable, people in the past found than desert sand (round) is very bad building material (they must use diged sand) and for example grain under railroad as foundation must be coarse, irregular, never round.
I used crusher dust and epoxy to fill the stand for my lathe. Before doing, during tests I found that I only needed 10% epoxy to wet it out, possibly even less. It made my lathe behave like it weighs 5 times more.
I thought cruster dust is pretty much known to compact well so I figured the grain sizes are already figured out. I did try adding fine sand to it and it lost rigity due to the extra pockets so ended up just using it straight.
I would love to see the results if you used "crushed stone" that is specifically made for the base when laying pavers. It is a manufactured product. It compacts to a hard surface on it own. It's the combination of crushed stone & stone dust that gives far greater compaction than anything else.
Your videos are awesome, but on a lengthy side. Please consider adding chapters for individual operations, such that time-deprived folks can strategically navigate the videos.
Very interesting! Have you considered using pebbles which are rough? (Not round like washed from a river) they would interlock better. Same for the sand, there are sands which have round grains and sands wich are made to have a rough surface which interlocks better.
Great video 👍
Thank you 🙂
Looks like the granite table is 2 layers. If the 2 layers are bonded together the stiffness will increase from 2 x the stiffness of one layer to 8 times. Stiffness increases as the cube of the height.
great work, though it makes me weirdly nervous to see you not mix the epoxy and hardener before you pour it into the particle mix.
Also, I'm pretty sure there's a guy on CNCzone that had high hopes for EG (epoxy granite) and did a lot of testing, in the end he actually concluded the stiffness was disappointly low. So low, that you have to make very big structures to make up for it which then becomes quite costly. These days/years he is looking into cement /grout but as others here have mentioned, shrinkage is a real problem - even for the types that state it doesn't shrink.
Why should I mix the epoxy and hardener first? Now I have all the time I want to let air bubbles escape first from the 'premix' if I just take care to mix the hardener thoroughly enough, which I did. And yes, I agree with you (see the calculated Youngs Modulus and compare that to steel) that one shouldn't expect too much from this kind of concrete. That's why I warned about it in my video.
Well yes, EG is used as a substitute for large, very laborious and expensive iron castings. If the member is small enough for EG stiffness to matter, iron or steel should probably be used to boost stiffness and mass. Anything less than 40 kilos and 250m in cross section is probably not worth the trouble.
@@Michel-Uphoff Mixing the two parts of the epoxy first will make sure that it is thoroughly mixed. A common tip is to mix the hardener in for 5 minutes when you think it is done, then mix some more. In the same way the air bubbles were caught in the medium, the same can happen with the two parts of epoxy. Epoxy that has not fully cured can alter your results significantly.
I have seen elsewhere, that a mixture of different sized medium ranging from ~10mm down to fine sand will fill the most voids to make the best use of the epoxy binder. Making sure that all air is removed also seems to be important too.
This was a great test, and I've been contemplating using EG for a while now, so thank you for showing your methodology and results.
Interesting and thanks. Liked that the detailed efforts were explained. Very well-thought-out and applied. The bottom line might be that composites made of sand and epoxies are less deformable than epoxies alone? An observation made with other composites too, like steels.
Yes for sure. Sand is silicon oxide, (SI02). That material has an Youngs Modulus of roughly 75 GPa. This resistance against deformation of epoxy is around 4 GPa. So the concrete mixture sand & epoxy is considerably less elastic than epoxy alone. In fact, I measured roughly 15 GPa at best.
I know it sounds unusual but if you look at the fines, medium and course grading of regular ready mix concrete then mixing it with epoxy you get about the perfect ratio (of course you don't use any water to hydrate the cement, your just using the cement as 'extra fines' in the epoxy matrix)
In fact, that's where my pebbles and sand come from. I sifted ready-made concrete mix.
@@Michel-Uphoff I worked as a millwright for decades, and filled in a lot of equipment bases and voids in equipment. The engineers specified Redimix / Sackrete for the aggregate, occasionally we added carbon black to the top layers for UV resistance if it was outside exposed to the sun (like pump or compressor bases). The cement powder soaks up the epoxy quite well.
@@bobjoatmon1993 Sackrete is sand & cement powder isn't it?
@@Michel-Uphoff cement powder, sand and pea gravel or similar sized crushed stone. Two names but similar product
That agreeys with what i have seen elsewhere where the strength has been measured. I wonder though if that is the whole story, as you also add mass which will reduce the frequencey of any resonance (and may or may not be better) and vibration damping, which will make things some degree of 'better' If greater rigidity is required, the Tech Ingredients channel found adding a small ammount of graphene powder to epoxy increased the strength significently.
Yes, there is a noticeable improvement (see the previous video) although it was a mere 40% gain in rigidity in that case. Of course the base and the connection between column and base are not changed, so that accounts for a part (the most?) of the remaining flexibility. At high rpm's there is also less chatter in my Proxxon mill now I stuffed the entire column with epoxy concrete. But one shouldn't really expect too much from it.
If you decide to explore this further, mixing the epoxy then adding the filler should produce better performance and repeatability.
Could you elaborate on this? Why would that be the case?
It’s not best practice in industry to do so, reason is one part of the epoxy can soak in to a surface. Said surface will not cure as it’s loaded with only one part of the epoxy mixture.
@@machinewrangler4682 Epoxy doesn't soak in pebbles nor in sand. Tested that in the previous video where I reinforced the Proxxon column.
@@Michel-Uphoff As you noted getting the epoxy mix right is crucial. Mixing it in a separate step allows you to scrape down the mixing cup sides and bottom getting a complete mix. It's recommended to dump the mix in another container and finish the mixing. Most people skip that belt and suspenders 2nd cup. Then mix peanut butter filleting or fairing compound or start your laminating. The problem is either component hits a surface and wets to it, the other epoxy component then mixes with the bulk resin but can't easily mix with the film wet to the surface. Adding the components to the filler maximizes surface area wet by one component and physically interferes with mechanical mixing. This can leave areas that won't properly cure or bond.
@@anullhandleSo you state that mixing the epoxy and hardener first and then adding and mixing the filler will lead to a measurably stronger compound than mixing the filler and epoxy first and then adding and mixing the hardener. I'll test that hypothesis soon.
Awesome tests!
A bit worried about the cylinder with epoxy end though. It rests on the alu stand so you measure more of the deformability of the pure epoxy on the end I think.
That's true, but it cannot be much. The touching surface totals to more than 200 mm². So roughly 1 N/mm² max. I couldn't measure the indentation of epoxy at that kind of pressure.
Pour avoir un meilleur comportement en flexion du cylindre, il est préférable que celui-ci soit appuyé aux extrémités sur un cylindre plutôt que sur un bloc rectangulaire.
view ideas to make the measurements more consistent:
1. epoxy in Machined steal rings as support surface and measurement surfaces
(maybe uses (old) ball bearing races for this)
2. Separate the point of applied load and measurement a view mm and measurements next to but not at the point of load.
like an electrical "4 point measurements"
I doubt those rings wil aid to a better measurement. They have a very high Youngs modulus and would influence the measurements considerably. As you can see in the video I separated those two points in the final setup.
I'm not really convinced that the results for the sample test at 23:16 would be reliable, as one end of the sample bar was pure epoxy (no aggregate), and it was resting on the support. Basically, that small section of pure epoxy under load would have a different compression value compared to the compression of the epoxy concrete blend supported at the other end of the sample. This could give a false deflection value.
In practice that didn't happen. The pressure at the edges is around 1 N/mm² and that doesn't deform the epoxy measurably. But to be sure I tested all bars off camera at 15 cm length, so the 19.5 cm rods where sticking out on both sides more than 10 mm. There was no significant difference.
@@Michel-Uphoff Fair enough, and thank you for your reply. It sounds as though you were very thorough, and I applaud that.
This might be an ignorant idea, but I’ll ask anyway…hydraulic cement does not shrink, but expands, could a mix of hydraulic and normal cement create a non shrinking or very low expanding form of concrete?
I don't know, maybe.. There could be a problem with different curing durations I think. Perhaps the chemical reactions interfere with each other? Anyway it will not adhere to steel, so it is probably useless for reinforcement of machines.
Interesting. I stiffened 1½” pipes on a build using regular sand gravel concrete mix surrounding a 5/8” rebar down the center of the pipe.
my guess was the bigger pebles create a better bond with the epoxy , they be harder to fit inside sumting like a columb for a minimill though
having broken up quarts might be better than smooth pebbles too as it gives the epoxy more to bite intoo
the fine is better...... hmmm think it has sumting to do with the fact that the stone is less compressible / harder than the resin the smaller size makes them fit closer together
it could be that cement or sumting fine like talcum mixed with epoxy might be even better though at a certain point might stop the resin and material bonding together
Regular old sand and oil works really well for stuff like that
Have you ruled out the square tube not being stiff enough? As I understand it, stiffness increases as a cube /3rd power of the square/circle width/diameter, so a doubling of width on the arm might 8x the arm stiffness
No I didn't. Yes, double the length and you weaken by 8, double the diameter and you strengthen by 8 (the mass of the object not taken in account).
Great research. I hope you didn't throw these samples away. And they were not subjected to the test of destruction.
It would be extremely interesting content for UA-cam if you laminated these samples with carbon fiber. And repeated the hardness measurement.
If your concern is vibration, why are you testing just structural strength?
Didn't mention vibration was my main concern.
@@Michel-Uphoff Apologies.
Maybe if you were to make multiple tubes/specimens of the same mixture setup each time than you can get a better average between the mixtures instead of one example to make a suggestion or conclusion. It would also help eliminate tubes that had some open pockets in them. More specimens always is a better scenario in any experiment to seek a defined results...
Yes, that's right. I had a limited amount of epoxy. Nevertheless the results are clear I think. And I wouldn't rule out tubes because of air bubbles, because that will happen in practice.
Metrology is hard 🙂
Surely mixing concrete would be much cheaper and stronger.
You mean just cement concrete? That is cheaper, but it is of no use in these circumstances. It doesn't adhere and it shrinks. You will be left with a rattling stone in your machine.
Epoxy Granite has a bunch of other benefits specifically for machine tools, primarily in vibration damping compared to most other materials. Concrete or cement wouldn't really work well at all.
Couldn't you have just taken off of the rod instead of the test sample
Sorry, I don't know what you're referring to. Could you elaborate?
@@Michel-Uphoff sorry totally forgot a word >measurements<
Couldn't you have just take the nmeasurements off of the rod instead of the test samples, to avoid error on the dial indicator due to concentricity deviance in your test samples
@@mikecunningham3423 No, because rod pushes into sample denting it a little more in place of contact than whole sample deflects. If samples were softer, you would see that as a mark where sample was pressed. In second scenario with bands, you would just measure stretching of the band itself, it would probably dominate deflection results.
@@mikecunningham3423 I tried that, but the surface finish was insufficient, especially with the mix of pebbles and sand. That small pad provided a major improvement in repeatability and accuracy.
@@Michel-Uphoff I gotcha I misunderstood the video thanks for clarification
Which mixture had the highest density?
High mass will help reduce magnitude of vibration.
That's the mixture with the least epoxy, so the mix of pebbles and sand. See the photo of the bars. Their weight is noted on it.
Yes the higher the mass, the less vibrations. So tungsten powder would be a great (but very expensive) choice.
@@Michel-Uphoffso could one cast lead in the extrusion?
William Fairbairn, a pioneer structural engineer encountered your loading means issue when testing tubular bridges in the 1830’s round tube failed before rectangular of comparable section. It is thought he had not taken into consideration the local deformation due to how the test load was applied. As a result there was a rectangular tube bridge for the railway to Anglesea till the recent fire. It’s fun how cyclical history can be.
@@johnsherborne3245that was my first thought, but while lead would dampen any vibration pretty well short of using tungsten powder, but it is way too soft in my opinion to be if any use in resisting deflection under any meaningful load.
@@johnsherborne3245 If the goal is to dampen vibrations, that could be a good choice. However you shouldn't expect a considerable rise in rigidity.
why not use regular concrete/cement instead of a resin/plastic?
I think that regular concrete is strong in compression, but not as good in tension (including bending and twisting). That is why rebar is used in concrete.
The epoxy provides strength in tension but is not as good in compression without the harder materials added.
Someone else may have a better answer. I keep thinking about other materials (granite, garnet, steel swarf) but know I would never get around making a thorough comparison like this.
Cement concrete shrinks, and doesn't adhere to metal. You run the risk that if you pour it in a machine for reinforcement, you will be left with a rattling stone that does not improve anything.
@@Michel-Uphoff I know a guy that uses palsticizers and some other expensive ingredient and has concrete strength tested to 15kpi. How does your test compare? His was tested at a local lab but I didn't see the full report. I work with epoxies too. Which type of epoxy did you use? 1:1, 2:1, 3:1, 4:1, 5:1?
@@firstmkb The aggregate is stronger than the binder. Fibers are often used to increase tensile strength. Glass fiber to Carbon fiber. Also the bonding is enhanced by the porosity of the aggregate. So extremely smooth material yields weaker bonds. As an example I was mounting studs to Black Granite, but it required a hammer drill to create a rough hole for the epoxy to sufficiently bond. The other drawback to resins is creep, and humidity. You really have to do your homework in some projects. There was a tunnel where the overhead concrete "tiles" fell out when the epoxy failed. Bad day...
@@Michel-Uphoff simpler/better answer than mine!