[brunog]

Walter,
After looking and the sample photos, looks llike the aggregate % per size looks like 40% large ,20% medium and 40% small. It makes sense to me.

One thing i am curious about though, is the hardness of the cured epoxy of these samples. I know you to not have the equipment to do Shore D testing, However if you have and automatic centerpunch you can easily compare epoxy hardness by punching one of your samples and one of the sample you got from overseas. The sample that the dent is smallest is the one that has the hardest epoxy.

I have a hunch you will find out that the sample from France has a harder epoxy than the resin you are using.

I believe that using a resin that has a cured shore D hardness of 85 for instance with not additives will out perform any resin with a 65 with additives.

it is a known fact that Hardness of steel has a direct relatioship with it's tensile strength, I am sure it is the same for epoxy resins, this is a gut feeling as i have not put my hands on any studies that proves it for epoxy resins or plastics in general.

Just to give you an idea;a Shore d hardness of 65 is a bit softer than Polypropylene and a shore d hardness of 85 if stiffer than nylon and has a similar hardness to brass and aluminium.

Larry, Cameron I need your 2 cents on this.

Thansk, best regards

Bruno

[walter]

I have a hunch you will find out that the sample from France has a harder epoxy than the resin you are using.

There is no doubt in my mind. It's hard as ROCK - my stuff is more like Elmer's glue on the wooden chair (too much epoxy in the mix). This thing is very dense, crumbles when pinched, and doesn't have anything glueish about it. Like concrete... I'm going to risk it and say that it is NOT epoxy based. Some other 'epoxy', acrylic, polyester, plastic, whatever. People who know the system can use plastic and get the results, believe me.
Did I mention density? My biggest problem is lack of density. And since I'm not using vacuum.. What can I say.


Article:

"Machine tool bases have traditionally been made of cast iron, but recently inroads have been made by another material which developers claim reduces the effects of temperature variation and vibration.

The material combines granite and either a resin or acrylic binder cast into a machine tool base, which, developers say, costs the same or less to make than a conventional base.

Although new to many machine tool builders, especially those in the United States, the composite base has been available in Europe for several years. One developer, Fritz Studer AG of Switzerland, first came up with its Granitan S-100 technology in the early 1970s. But, according to Hans Schenk, vice president and general manager at Studer Inc., Brookfield, Conn., the company's U.S. subsidiary, it had been kept proprietary until about four years ago.

The granitan base is a mixture of reactable epoxy-resin binder developed by Ciba-Geigy Ltd., Switzerland, and granite or gravel. Granitan produced with sand and gravel was developed specifically for the requirements of machine structure manufacturing, Studer officials have said.

There are other "synthetic granite" products on the market, but they are newer than Studer's technology. Helgard Koblischk GmbH of Germany has developed Motema-AC, which uses an acrylic plastic binding agent. Also, Plastibeton Inc., Montreal, a subsidiary of Lone Star Industries Inc., Greenwich, Conn., has developed Quazite, which uses a polyester acrylic binder.

Another company, Diamant Metallplastics GmbH of Germany, has also developed a process which uses an epoxy acrylic binder, but after about four years of production has left the market. Rather, according to the company's U.S. agent joseph Devitt of Devitt Machinery Co. Inc., Media, Pa., Diamant is doing the fitting work to connect synthetic bases with machine tools.

Despite their differences, all of the synthetic granite developers claim similar advantages over cast iron and steel bases. Vibration dampening is one of the key advantages. In technical publications, Studer officials have explained that machine parts, and particularly machine tool beds, are subject to vibration by the machining process. Effective dampening by the structural material is therefore important. The dampening improves machining quality and steps up production rates, especially where metal is being cut, they said."

_

[ckelloug]

Bruno,

From what I was able to find, the relationship between shore hardness and tensile strength for steel is well established. For other materials it doesn't look like it.

I can find an <u>opposite</u> relationship in the case of the two epoxies listed here between shore hardness and ultimate compressive strength: http://www.glueit.com/epoxies.html

I finally found a chart on the subject in the following article
http://www.calce.umd.edu/general/Fac...d_.htm#rf15%22

but quoting from the article:
<blockquote>
A correlation may be established between hardness and some other material property such as tensile strength. Then the other property (such as strength) may be estimated based on hardness test results, which are much simpler to obtain. This correlation depends upon specific test data and cannot be extrapolated to include other materials not tested.
</blockquote>

I'm going to guess that without a bunch of lab work, we can't use shore hardness as a guide for what is the best epoxy to use. I agree that it seems like higher hardness "ought" to just work better but I can't find data to support or more than cursorily refute the conclusion.

In general, hardness is a combination of unknown portions of compressive strength, flexural strength and fracture toughness. We often get these in the material datasheets, so I'd wager that maximizing these three will ultimately set us on the right road.

I am suspicious that epoxy mixing procedures may be the culprit behind epoxy that is too soft. It also looks like post curing might be required for the US Composites 635 epoxy.

Several of the vendors published a technique called double cupping in which the epoxy is first mixed in one cup and then poured into another cup and mixed again to avoid the unmixed epoxy that sticks to the surface of the first cup.

Keep up the good work,

--Cameron

[lgalla]

Cameron,Bruno,I don't have a concrete answer,but I guess bond strength or shear strength would be the determinating parameter.A lower shore hardness may not be a bad thing as stress or an impact may absorb the energy rather than than fracture from being too hard.The main problem is a void free matrix.
On Walters core samples,if you torch the sample and it supports combustion,It may be polyester as epoxy should not support combustion.
I have stated many times before on mixing epoxy which is critical.Use calibrated plastic containers from Home Depot or a body shop supply.E/G use a one litre container,fill with A to one half or ratio desired and pour in B to the desired ratio and mix.After mixing and adding aggregates restrain from scraping every last drop from the mix container.You will get poorly mixed epoxy from the sides of the mix vessel.
My application is focused on damping weldments or tubes which lowers the strength requirements of E/G.Puffed wheat epoxy would probably work for my purpose.Total E/G machine is beyond my needs or expectations.Anyone doing samples must give accurate procedures as to their technique to "anal ize "the results.
My $2.00 US {inflation}

[ckelloug]

Here's another one (before you buy the aggregate):

High grade quartz contributes to great wear resistance but also lowers elastic modulus. In another words, buy stuff other than quartz to get the best Young's modulus.
_

Gotta hand it to ya Walter! You tend to call more stuff right by common sense and research than I do by esoteric theory.

During my post on factors influencing aggregate, I started researching fracture toughness which is the determining factor in the strength of brittle aggregate. I found this paper http://www.minsocam.org/ammin/AM67/AM67_1065.pdf on a method for analyzing fracture toughness by smashing a sample of material and examining the sizes of the pieces it breaks into. They analyze quartz and several other things.

One interesting point is that there are two classes of quartz, a weak one and a strong one. Synthetic fused quartz has a crystal structure that is much less strong than some types of natural quartz. Also, they tabulate the fracture toughness values of several materials. It looks like alumina (Al2O3) is about 6 times the fracture toughness which means it will likely perform substantially better than quartz.

This is borne out by a data sheet from Coors Tech, a spinoff from the Coors beer company that makes technical ceramics. (Beer used to be packged in ceramic jars in the late 1800's. . .) The data sheets point out using alumina as an epoxy filler. http://www.coorstek.com/products/grindingmedia.asp

It's probably on the insanely expensive side but Zirconia has over 10 times the fracture toughness of the low strength quartz variant.

Of course all of this is moot if we are failing in the epoxy or in the bond between the aggregate and the epoxy although walter's last test with the large home depot special aggregate showed fracture in the aggregate.

Finally, this post is mainly academic until we determine whether our epoxy is sticking as well as it ought to be but I thought others might be interested.

--Cameron

[Joe Crumley]

A year ago I was doing the searches as previously mentionedand with frustration. It was a slogggg. Fortunately I've settled that down and will be glad to share my findings and experience. We are now in production and glad to share info..

We've been in the sign business for twenty years. Our product has been 3D wood signs from cedar and redwood. It became necessary to branch out with the lack of redwood available. With the addition of a CNC I saw the possability of topcoating our signs with acrylic stucco, sometimes called polymere concrete. This is a concrete look and will last a long time. I'll try to post a photo of one of our projects.

There are many different concrete formula's available. I've found Fossil Creeks "Vertabond" what we needed. The base for all of these is concrete. This is laced with nylor fibers, for added stregnth, along with a host of other compounds. The essential one being a an acrylic polymere which makes it stick to just about anything. I believe this stuff will even stick to glass.

Many contemporary structures, such as TacoBell's are stucco laid over a lath then the concrete is applied. The first coat is called the Brown Coat" Once dry the final coat is either trowelled or sprayed on. The product name for them is Drivit. Forget it, they only sell to approved installers.

Our process, simular ot Drivit, is to apply the liquid polymere, with a roller, to the structrue. This stuff looks like milk. Let it set up, we either trowell or spray the vertical concrete. This is a very forgiving process. It's not very hard on hands or equipment either. Also the surface can be sanded for several hours after application to correct flaws or give as smoother affect. You can build this up to any thickness desired. It takes several days to reach a good hardness.

We hoppergun on about 1/2" and let it set. When allmost dry we sand to a nice texture. Then coat with latex paint. Yes you can mix in color, but I've found it expensive and not very precise. Besides, if we chip a piece during installation, paint is the easiest to repair.

So in conclusion: You need the Polymere and concrete mix. The 40lb bags cost about 25 bucks but the polymere, from my supply, in five gallons, is about $155. That's diluted 4 to 1. Goes a long way.

My supply is www.thestampstore.com Look under vertical concrete mix.
Also there is www.demandproducts.com I think they are cheaper.


Joe
www.normansignco.com

[walter]

These are beautiful signs Joe!

[walter]

Could you post a dimension so that i can give you something that is similar to what you have been testing,,,,, in case you would like to standerize your units?

My recommendation would be 1.5" height, 2" width, length optional but 4" minimum. I PM'ed the same dimensions to Davo. I'll send the address, please enable your PM box.

[walter]

Walter, I hope you will continue to hang out on this thread even once you succeed at making a usable machine. You've had a lot of good insights into making the ultimate E/G and I hope you're still having fun here. Most of the useful results from the theoretical approach will come long after you are done with your machine at the rate things are going. If there is anything I can do specifically for your machine, please don't hesitate to ask as I'm trying to study making E/G thoroughly before doing much building and haven't posted much that's been directly helpful to a machine builder lately.

Yes I'm still having fun and appreciate your kind words.
I've been thinking (always dangerous) and decided to give Greybeard's modular idea a try. I'm going to pour 6 different pieces and bolt them together.

Could you advice me on the bolting hardware? (similar to the one used in kitchen cabinets or whatever). What is the toughest and meanest hardware that can be used here? Not specifically for E/G embedment but for metal to metal etc. What would you use on this 200lbs machine ↓
Thanks.

[millminer]

Polymer concrete frame?

I have seen 5 foot by 4 foot x 10 inch deep surface tables in factories that are made of poly/granite/....

The surface is like glass and exceedingly true.

Some large lathes have filled bases.

[ckelloug]

Walter,

If I understand your picture correctly, you are trying to bolt the top beam vertically all of the way through the column supports to and through the table and you are trying to bolt several laminations together on both the top beam and the column supports with the bolts running horizontally in both cases.

Is this correct?

My thoughts based on this understanding are these:

The column supports for the beam are loaded in compression so almost any of your E/G formulas ought to work reasonably. As for bolts, I would suggest that home depot threaded rod and nuts is probably adequate for holding the beam down to the columns and the table. I'd also think that a bit of epoxy in the joints before the bolts are tightened and after you've made sure all of it is square would be good for this application.

If you're planning to cast the beam in short pieces and bolt them together then there are probably a couple ways to do it. I don't know how to predict the ultimate strength of any of them.

The most promising method I can think of is to cast 12 x 4 x 1 inch plates with 4 holes an inch from the edge in all directions. Here is a lousy ascii art of one of the plates:

<code>
******
* *** *
******
* *** *
******
</code>

I'd then slather the ends of the pieces in epoxy, abut them end to end and run 4 threaded rods down the entire length through the holes and tighten the nuts on the end. I'd also put some quarter inch thick steel plates drilled the same way on the ends for the bolts to screw down against. This will put the E/G under compression from torquing down the threaded rods and even if there were no epoxy on the ends, friction would hold the pieces together. It ensures that the E/G is loaded only in compression where even bad batches are at least a bit strong and also that the moment and tension loads are carried mainly by the steel rather than the E/G. This mainly uses the E/G for weight, damping and to hold the threaded rods far enough apart to carry the moment loads in the beam.

As for bolts in the general sense, they come in grades 1, 2, 3, 5, 5.1, 5.2, 7, 8, 8.1, 8.2 according to my Marks Handbook. They range in strength from 36 kips for grade 1 to 160kips for grade 8.2. Some of the higher grades are available from home depot. Grade 5 bolts with a strength of 120 kips can be identified by a pattern that looks like Y embossed on the head of the bolt.

Generally speaking, I can't see a need to be too picky on the bolts for this application. I'd probably use 1/2 inch mild steel threaded rod for all of the threaded rod but I haven't calculated it out.

I haven't thought this suggestion of bolting plates together through very thoroughly yet and some part of it could potentially be stupid but hopefully it will spur some discussion on the thread.

Regards,

Cameron

[lgalla]

Sum Summary of previous posts.
635 slow:
HARDENER>>

With 2:1 Ratio Slow Epoxy Hardener (Description)
NOTE: Our slow hardener should only be used at temperatures above 80degrees unless an extremely long curing time is desired. Temperatures below 70F during the cure can result in a 2-3 day drying time.
The pot life at 80f is 35 to 40 minutes.At 80F or below you should "sweat" the mix of A&B.Leave it for 20 or 30 minutes to start crosslinking and producing heat of cure.Adding aggregates after this time will cool the mix and extend pot life.
Heavy,thick mixes are very difficult to de-air.Multiple thin pours will entrapp less air.Sweating will lower the viscosity to 200 to 300cps.A mega sucking pump is not gunna suck air out of a bread dough thickness batch,4" thick.Vacuum while mixing should work as vacuating the mix vessel results in no air to entrap.Remember the saying"Don't sweat it"does not apply to E/G.
Walter says the core samples shatter easily.I assume they are from a damping machine tool base.Any gantry E/G's I have seen appear to be inspection type machines.I would not attempt a gantry beam to support a 150lb,10HP spindle.E/G filled tubing for me and epoxy surface plate to square the gantry.
NEW IDEAS FOR ME!For a large format router why build a table?Surface plate the shop floor and mount the rails to the floor.This would be a low profile router and eliminate the table and cost.It is common practice on wood routers to mill the spoilboard to be flat to the gantry.Any thoughts?
Larry

[lgalla]

Steel frame bolted together with epoxy grout has been used by Madvac and other Zoners to acheive squareness.The problem is when the epoxy heats and crosslinks it flows out.Kevlar pulp or merge makes epoxy extremely thixotrop and it will not flow out of the joint.Kevlar pulp has no use in E/G as it is an extreme thixotrop and would hinder high epoxy aggregate ratios.Bolted joints reduce vibration and kevlar has high vibration loss.
Larry

[walter]

Walter, If I understand your picture correctly, you are trying to bolt the top beam vertically all of the way through the column supports to and through the table and you are trying to bolt several laminations together on both the top beam and the column supports with the bolts running horizontally in both cases. Is this correct?

Well, not exactly. (This is not my machine, just an example).
I'm trying to find a commercial (off the shelf) high tech way of assembling E/G parts, all kinds of parts. Some type of blind joint hardware or whatever. Maybe some inserts that can be bolted to a metal part and then encapsulated. Like in furniture industry, you assemble the cabinet, insert the big black screws and the whole thing locks together. This application calls for some serious clamping force and I need to be able to disassemble the structure- so no epoxy.

Just a shot in the dark. Don't worry about it.
I'm still in a design phase...

[brunog]

Well, not exactly. (This is not my machine, just an example).
I'm trying to find a commercial (off the shelf) high tech way of assembling E/G parts, all kinds of parts. Some type of blind joint hardware or whatever. Maybe some inserts that can be bolted to a metal part and then encapsulated. Like in furniture industry, you assemble the cabinet, insert the big black screws and the whole thing locks together. This application calls for some serious clamping force and I need to be able to disassemble the structure- so no epoxy.

Just a shot in the dark. Don't worry about it.
I'm still in a design phase...

Walter,
I am actually working on threaded inserts made from Allthread couplings.
Actually this could work straight off the box. All you need for each insert is 1 allthread coupling, 2 flat washers with ID an extra size bigger and a short carriage bolt; make sure the carriage bolt thread screws in at least least nominal size length.

Example: Take a 3/8-16 UNC coupling, 3/8 carriage bolt x 3/4", and 2 flat washer 7/16" ID. you assemble them together and cast them in the E/G.

What I am planning to do is turning about 3/8" long of the outside of the coupling to have a nicer finish on the surface.

I will send photos of a sample when I am done.

Best regards

Bruno


Note: Do not use bolt grade less than Grade 5 !!! All grades below are not marked and are not consistant in strength.

[greybeard]

Gosh, with Walter thinking about my epoxy lego idea, and Larry considering my "why not build it on the floor ?"(watch out for dogs and kids), perhaps the old grey cells are still ok.

How about combining the two ideas ?

John

[Zumba]

[walter]

I think these are all good ideas. Let's keep the creative juices flowing!

[ckelloug]

Hi all,

I've been a bit busy lately with house roof repairs, shop construction and sprinklers so I've done less E/G work the last week or so.

I did want to say that while I think the lego brick construction is an excellent idea for the table that I'd hesitate to do it for the beam unless metal reinforcements are carrying most of the load.

I think Bruno and Larry have the right idea of using a metal beam filled with E/G since our current E/G recipes are in all likelihood not up to carrying the moments associated with the loads on the beam.

Most industrial metal distributors will cut a piece of metal for you so a piece of steel rectangular box channel at least 12x4 in cross section is not unfabricatable even for the "I don't have a metal shop" crowd.

Also, epoxy grout is a good idea for holding things square after they're assembled but Larry pointed out that epoxy and not epoxy grout or epoxy with a thixotrope will tend to squeeze out and not help.

These are my $.02 while I am on the run.

--Cameron

[DAK3333]

Here in the lab (back in double door land) the house vac is a water ventury backed vane system. We get about 26 mmHg at the bench. This has been sufficient to degass my samples. I would guess that near 28mmHg for about 15 min would do it. Yes it will expand to great volumes (kinda neat). I have used a refridgeration compressor at home with good results in other ventures.

For the Shore hardness. These are good questions. If someone could share again the epoxy being used and a data sheet, I would like to see the chain length to crosslink position ratio. It may be a simple matter of the wrong epoxy chemistry. A good adhesive may not be a good binder and vise versa.