Thread: Epoxy-Granite machine bases (was Polymer concrete frame?)

**MaX-MoD** · 10-13-2009, 05:50 PM

Mmmh, just one tip: you may prefer to separate each air springs pneumatic inlets. I guess that one air spring could empty into the others in such a way that you may have lower and higher air springs... just like cylinders linked in parallel.

good luck for the rest and... don't forget the pictures

**Eric R.** · 10-13-2009, 06:32 PM

Originally Posted by MaX-MoD

Mmmh, just one tip: you may prefer to separate each air springs pneumatic inlets. I guess that one air spring could empty into the others in such a way that you may have lower and higher air springs... just like cylinders linked in parallel.

good luck for the rest and... don't forget the pictures

I agree and that is exactly what I am planning to do.

**romihs** · 10-13-2009, 07:52 PM

Eric,

Great, thanks.

Just one question regarding synchronizing the two motors. How do you intend to synchronize, or rather shall I say, control the angle of two motors relative to each other?

For the table to function correctly, the two motors need to be 180o out of phase in the horizontal plane so as to cancel each other out and thus producing only vertical vibration/motion.

Sorry if I am ragging you a bit about these details....

Regards

Sandi

**Eric R.** · 10-14-2009, 10:07 AM

Originally Posted by romihs

Just one question regarding synchronizing the two motors. How do you intend to synchronize, or rather shall I say, control the angle of two motors relative to each other?

For the table to function correctly, the two motors need to be 180o out of phase in the horizontal plane so as to cancel each other out and thus producing only vertical vibration/motion.
ut these details....

That is not too difficult. It can be achieved in several ways, but I will do it by connecting the two motors parallel to the same frequency inverter, so they are being fed with the same phases. Then I will either:

mount one motor 180° rotated around its vertical axis (just like you would turn your car to drive the other way) compared to the other motor
or I may have the motors mounted identically and just wire motor 1 as L1, L2, L3 and motor 2 as L2, L1, L3 (so switching two phases, which makes the direction of the field rotate the opposite way).

Before powering up the motors, I will remove the protection caps from the balance weights and make sure that:

the weights are placed in such a way that they create most force (the balance weights are adjustable to create the desired force)
rotate the shafts to ensure both motors are in sync
Then switch on the frequency inverter and let the motors rotate with a very low speed, to ensure they are in sync. If not, switch off the inverter and resync them and power-up again.

If the motors are still slightly out of sync, I may have to adjust the position of the weights of one motor versus its shaft to match the other motor. But as these motors are brand new, I assume the weights have been mounted in a standard position. After you are satisfied, do not forget to place back the protection caps. If anything goes wrong (i.e. if a weight breaks or gets disconnected somehow) these weights will launched like projectiles.

That should do it.

Please be aware that connecting motors parallel to one shared frequency inverter only works well if you have two identical motors. Otherwise one motor may draw more current than the other (and/or force a lower voltage if the frequency inverter is current limited during soft-starting), which may give start-up problems for the other motor. No need to mention that the frequency inverter should be able to provide sufficient current, voltage and frequency to feed both motors (then why do I mention it

). In my case I will be using a programmable inverter (2.2kVA), that allows me to define max. values for voltage and frequency, as well as start and stop profiles, for soft starting and stopping, prolonging the life of the motors.

**andrebenschop** · 10-14-2009, 10:19 PM

Eric,

I do not think you need to sync them very accurately, as one motor will sync the other and vice versa. The horizonal components of each motor forces the rotor of the other motor to (eventually) be offset by 180 degrees. Even if they start at a random angle, all it takes to sync them is time (or patience

).
I'm on this side of the pond (maybe even closer than you may think) and will contact you (if that's OK) in 3 weeks or so.
Here's another guy anxiously awaiting those pictures.

Good luck building your table !

Andre

**Eric R.** · 10-14-2009, 11:15 PM

Hi Andre,

Sounds like you are experienced. I will put your advice to the test once the table is completed. Feel free to contact me at any time. Or just drop by, as I see you are also based in the Netherlands.

Thanks,

Eric

**HBFixedGear** · 10-15-2009, 03:33 PM

ontGrowAutofit/> </w:Compatibility> <w:BrowserLevel>MicrosoftInternetExplorer4</w:BrowserLevel> </w:WordDocument> </xml><![endif]--><style>  </style> Please do not think I am lazy or not willing to do my due diligence because this is not the case. The information I need is certainly buried in this thread but digging out the details isn’t an option on this one. I simply have a small window of opportunity to act and need to cut my learning curve to zero.

<o

> </o

>
The problem:
I have a small gang tool lathe that shakes like a three legged washer on spin every time I make a G0 move. It is a CNC retrofit on a Hardinge HSL-5C speed lathe sitting on the standard Hardinge U shaped sheet metal base. This was never intended to be a CNC base and it shows.
<o

> </o

>
The solution:
It is simply to light and I am going to fill the base.
<o

> </o

>
The question:
Who sells material prepackaged that can be mixed pored and cured with a minimum of blood shed.
<o

> </o

>
My expatiation is to make this machine perform better. It is never going to be a high precision machine so the solution to my problem needs to be appropriately scaled. As I stated above, there is a very small window of opportunity for me to act so finding a vendor that can sell me the agg, & resin ect as a kit or at least from a single vendor is the only way this will fly.
Reinventing the wheel isn’t an option here and with 127 pages in this thread, someone out there must have done this before.
<o

> </o

>
Thanks in advance for your help.
<o

> </o

>
Bob J

**HBFixedGear** · 10-15-2009, 03:36 PM

Sorry I have no clue where all the little smiley faces came from. I must have had one of the buttons pushed.

Bob J

**ckelloug** · 10-15-2009, 03:51 PM

Bob,

If all you're trying to do is fill the base of a machine that is too light, get yourself some play sand at lowes and some of the low viscosity epoxy with slow hardener from uscomposites.com. The sand will make the base heavy and sand is cheaper than epoxy. The epoxy required will be something like 30 percent of the solids volume of the sand. Divide the weight of the sand by the density to get the solids volume.

Damping will be as good or better than an optimized E/G mixture. Of course, you won't get the stiffness of an optimized mixture but you won't need it so it won't matter. This is a slam dunk easy thing to do.

Bob Warfield used one of our earlier E/G mixtures to fill the base of his Industrial Hobbies Mill and reported excellent results. see http://www.thewarfields.com/cnccookb...lEpoxyFill.htm

Best of luck!

Regards all,

Cameron

**ckelloug** · 10-15-2009, 04:15 PM

Eric R,

Do you have references about 4500 rpm being optimal? The thesis of Castillio German (in French) suggests that 50 hz e.g. 3000 rpm is fine. I'm also curious if you have a reference on why to use only vertical vibration. I can see why it would be good to use only vertical vibration but I'm not sure I've seen a reference on it.

At any rate, congratulations on starting experimenting. I'd be happy to reiterate my offer to test the flexural modulus and strength of your material and post it here. I'd need 5 samples 7 1/2 inches long 1/2 inch wide and 3/8 inch thick. I can cut these from a single 1/2 inch thick plate. I haven't been able to build my vibrating table yet so I don't have experimental data.

Brenck,

From my research into packing theory and de Larrard's model, 95% is likely not achievable with any mixture using only vibration. It's possible it could be achieved with substantial external pressure and vibration but it still might be a stretch. The models suggest 93% is fairly close to the upper limit of the achievable solids percentage with vibration.

Regards all,

Cameron

**tootalew** · 10-15-2009, 05:08 PM

Here is some good reading from American machinist.

http://www.americanmachinist.com/304...se/80845/Issue

**jberg** · 10-15-2009, 06:58 PM

Hi

I just wonder that is it really needed to use epoxy. I understood that epoxy is used because it don't shrink. But If packing density is "perfect" how that stuff could shrik anymore (if used some other that epoxy e.g. vinylester)?

In theory if we made gube like 10*10*10cm using different resins/epoxy is there any shirinking if aggregate are packed alredy as tight as possible?

I think that there may be some internal tensions if used something else that epoxy but is that a problem?

Just wondering....

**Eric R.** · 10-15-2009, 07:11 PM

Hi Cameron,

A while ago I have had some interesting discussions with a representative of a German company that builds vibration tables for both concrete and mineral casting purposes. He advised me to use frequencies in the mentioned range of 4500-6000rpm and vertical vibration only, as that would give the best results. Some time ago I have also read similar statements regarding vibration direction in some documents in French, German and Dutch, but unfortunately I did not bookmarked them. I will see if I can find them again.

Anyways, as I will be able to adjust the different vibration parameters, such as frequency, amplitude, direction (vertical only, rotational, etc.) I will see if I can produce some samples under different vibration conditions and see which gives the best results. When I have found the best method, I will post my findings and may kindly accept your offer to test my best samples.

I have one more question regarding the composition of the aggregate that you are calculating/simulating. Do you have the sizes of these components in inches or mm? The Agsco sizes that you are referring to are not entirely clear to me and on their website I did not find a clear specification either.

As I highly value your input, please let me know if you have any additional suggestions or concerns.

Thanks,

Eric

**Brenck** · 10-15-2009, 07:32 PM

Cameron,

Thanks for the research, its good to know. It's kinda amazing that some dudes from the german forum got a mixture of 93/7 then, since is the limit.

HBFixedgear,

If you want something that would be even cheaper you could fill with portland cement...

**Brenck** · 10-15-2009, 07:54 PM

Eric,

That American Machinist link says that the molds are vibrated 70 cycles per second, thats 4200 rpm.

Indeed vertical vibration is the best, at least is what the catalogs says

But all the vibromotors catlogs I've looked looks like the more rpm the best is the compactation, I'm not really sure if you have a limit on that, I was wondering what would happen if I used an old router 21000RPM I have here with an offset weight, well I will just wonder because I won't be close from it to see... I think too much vibration might start making the opposite and separating the epoxy from aggregate.

Looks like they heat the mold to 133º F, and heat more then that have adverse effect.

I remember I one of Walters post that he tried to heat the cast with heat lamps he got a bad casting, I wonder if that was the cause.

**Eric R.** · 10-15-2009, 08:20 PM

Hi Brenck,

Yes, I just read the article. That information is in line with what I have been told and what I have learned from other sources.
I don't think they were heating the mold. What they refer to is the result of the exothermic reaction >> curing epoxy causes heat. They state that if the product becomes too hot (over 113º F), it may cause stress when the cured product cools down.

I was wondering what would happen if I used an old router 21000RPM I have here with an offset weight

My strong advice: don't try this. A router motor is simply not constructed for moving around unbalanced weights. Most probably the bearings will die or worse... something may break and the weigth may be launched, which can cause dangerous situations. Please check posts 3687 by Roy and Cameron's reaction in post 3691 (both on page 308), discussing similar issues.

**ckelloug** · 10-15-2009, 10:20 PM

Erik,

Thanks for the info on vibration. If you find the other references, I would love to see them!

That new American Machinist article that totalew just posted is great. The 133F they're suggesting is the temp they get from exotherm is surprising but understandable since I have only handled lab sized batches. The isophorone diamine hardener suggested by roach which I have tried is so slow that I never noticed my small samples get above ambient. I have in fact treated the raw epoxy samples in a curing oven as IPDA cured epoxy can behave differently depending on the cure cycle.

In terms of vibration, the info I can remember suggests that an acceleration between 2.5g and 4.0g is very important and different authors seemed to differ on frequency. Getting this much vibration on a part of non-zero size is hard for somebody not going to the lengths Erik is going to to build a real vibration table.

De Larrard's book states that some of his samples were vibrated at 3 different frequencies because different sized aggregates resonate at different frequencies.

As for the agsco aggregate, it is graded by sieve size. There is also a chart on their site to convert sieve sizes to millimeters or inches. I converted all of the sizes in the aggregates I modeled to millimeters. I generally assumed the average size to be the size of the sieve. It seems like I had to do a bit of guesstimating to assign the pan fraction to the smaller fractions I was already tracking.

I separated the aggregate into the following fractions in the model where numbers preceded with a pound sign are U.S. Standard Sieve numbers.

#4, #6, #12 , #14 , #16, #18, #20 , #25, #30, #35, #40, #50, #60, #70, #80 , #100 , #120 , #140, #170, #200, #230, 53um, 18um , 12um, 9um, 4um, 2um, 1um

Assuming I haven't fouled up anything posting, the de Larrard Beta coefficients I used and the millimeter sizes that the sieves they translated to via the agsco chart for average size passing the sieve are:

.64,3.070
.66,1.68
.66,1.41
.66,1.19
.66,1.00
.66,.84
.66,.71
.66,.59
.66,.50
.66,.42
.66,.297
.66,.250
.66,.210
.66,.177
.66,.149
.66,.125
.66,.105
.66,.088
.66,.074
.66,.063
.66,.053
.71,.018
.71,.012
.71,.009
.71,.004
.71,.002
.71,.001

I've also attached a chunk of an octave sheet representing the percentages of each size for each fraction I assigned in the model. It's really wide and a pain to view.

Brenck,

The German forum guys in the past seem to have been measuring by mass ratio where 93/7 is not as impressive as it sounds. They were using an aggregate mix based on Fuller's formula last time they posted here which guarantees that their volume percentage is likely only in the high 80's. If you can verify the way they took their measurement, it would be great. I don't read german, only English and French so I haven't been following their excellent work lately.

My calculations are by volume percentage of solids since that is the parameter which can be used to successfully predict both packing density and stiffness. Making a 93% solids volume mixture is substantially more difficult that making one that is 93% by weight. It is only difficult to get a density around 93% by volume not near-impossible. Getting there simply means that throwing some random sand in a mold won't work.

Based on the models of stiffness in Kinloch's book, I suspect that the article totalew just posted is wrong that it is a 93% mass percentage used. Using the stiffness numbers from one of the commercial vendors in Kinloch's model, their stiffness is predicted to occur at about 93% solids volume.

I suspect that 21,000 rpms would not be good at all for vibrating E/G not to mention the danger of flying parts. I think that it's way too far above the resonance of any of what you're trying to move to get any energy transfer. Thus I agree with Erik that it's probably not going to work well.

Regards all,

Cameron

**Brenck** · 10-15-2009, 10:42 PM

I am talking to some companies here that make vibration tables for concrete. I've notice that all the table sizes that they have, the ones that use 2 motors are square shaped, like 500 x 500 or 1000 x 1000 and so on...

Does anyone know why?

Will a rectangle shaped table change anything on vibration effect? Due to motor placement or something like that?

**Eric R.** · 10-15-2009, 11:56 PM

Cameron,

Thanks for your extensive explanation. It is highly appreciated. This weekend I will compare my own aggregate composition (based on the Hexion mineral casting document) with the sizes you have selected.
Regarding the exotherm reaction: I once mixed some resin and hardener, resulting in a complete melt-down! It was during the summer and I was building an ultra-light electric guitar, covering the entire body with a layer of carbon fibre. I had just mixed the epoxy (the volume was less than 300ml), and put it aside for a few seconds to grab the carbon fibre. Unaware as I was, I had placed the epoxy in direct sunlight. Shouldn't have done that! Next moment I looked, it was literally boiling and the pot it was in started melting.

Brenck,

Originally Posted by Brenck

I am talking to some companies here that make vibration tables for concrete. I've notice that all the table sizes that they have, the ones that use 2 motors are square shaped, like 500 x 500 or 1000 x 1000 and so on...

Does anyone know why?

Will a rectangle shaped table change anything on vibration effect? Due to motor placement or something like that?

I think that they just like squares

No, really, it does not make any sense. Seemingly that is what they needed for their purposes. You can make vibration tables in any size and shape you like, as long as you make sure that (1) the table is rugged, (2) the force meets the load and (3) the motors are somewhat equally distributed over the length/surface of the table, which contributes to the rigidity of the table. As you can imagine, placing all motors in one spot may not be the best approach

Eric

**Brenck** · 10-16-2009, 01:29 AM

I think that if you make it square you can use just 4 springs while when you make it rectangular u might need to use 6 or I might just be dumb and my thinking doesnt make any sense :P

Thread: Epoxy-Granite machine bases (was Polymer concrete frame?)

Thread Tools

Search Thread

Display

Filling a machine base

epoxy vs vinyl etc resins

Similar Threads

Epoxy Granite In Practice (Mineral Casting, Polymer Concrete)

Voice Your Opinion On "POLYMER CONCRETE FRAME" Thread!

Epoxy granite (Mineral casting, Polymer concrete)---Particle size distribution

Index to "Epoxy-Granite machine bases" thread

Epoxy-Rice Machine Bases (was Polymer rice frame?)

Tags for this Thread

Posting Permissions

My IndustryArena Login

My IndustryArena Login

Thread: Epoxy-Granite machine bases (was Polymer concrete frame?)

Filling a machine base

epoxy vs vinyl etc resins

Similar Threads

Tags for this Thread

Posting Permissions

This website uses cookies