PDA

View Full Version : You mill drill guys....



torker
03-20-2008, 11:30 PM
I've been asked many times to show the 3" post I cemented into the column on this mill...and to show the jacking bolt setup I use to aid in tramming this thing. The bolts also add stability to the lower base. Without them the base can actually flex. This whole affair ties it all together. It's ugly.. I did it in a hurry and grabbed whatever fit at the time. The big bolt in the bottom of the shaft goes right through and jacks against the base toward the front. You can figure the rest out I guess. Some of the stuff is welded on afterwards...way inside. More bracing to steady things.
Add this to the "Million Ways to Improve A Mill Drill" deal that everyone hates :D
http://img.photobucket.com/albums/v210/torker/DSC00002-33.jpg
Russ

BobWarfield
03-21-2008, 01:31 AM
Good stuff torker!

I am engaged in filling my mill's base and column with epoxy granite to dampen vibration and increase its dynamic strength. At least that's my story, and I'm sticking to it!

The fill material is a mixture of granite gravel, sand, and epoxy. It molds real nicely:

http://www.cnccookbook.com/img/MillStuff/CNC/EpoxyGraniteFill/P1010301.JPG

Filling the base was very straightforward:

http://www.cnccookbook.com/img/MillStuff/CNC/EpoxyGraniteFill/P1010290.JPG

http://www.cnccookbook.com/img/MillStuff/CNC/EpoxyGraniteFill/P1010379.JPG

What you see there is all dry and solid as a rock. It increased the weight of the base from about 100 lbs to nearly 200 lbs. I will be partially filling the column as well as increasing the strength of the connection between column and base with a monster big bolt to add to the existing bolts.

The nice thing about the epoxy is it bonds well to the cast iron, it adds no moisture so does not promote corrosion, and it is pretty neutral in terms of changing size as it sets up.

I like your jacking screw arrangement for tramming your column.

Lastly, I am adding a one shot oiling system I'm fabricating.

You can do quite a lot to improve these machines.

Best,

BW

Doc Nickel
03-21-2008, 04:51 AM
That's an interesting idea. It had been suggested to me many years ago to fill the column of my JET mill-drill with concrete, to help stiffen it and damp vibration, but as noted, I was concerned with moisture and corrosion.

The column of my mill is cast iron, thinner walled that I'd have expected (especially given the weight of the head) and extremely rough inside.

I eventually made a heavy table with supports on either side of the regular column, and that did help some, but I have been gathering ideas to optimize it more so it's more useful.

Is this epoxy granite a commercial item, or does one just mix fancy fishtank gravel with some JB weld and call it good. :D

Doc.

Doc Nickel
03-21-2008, 05:07 AM
Here's a photo of the stand:

http://www.docsmachine.com/machineshop/jet2.jpg

It's all heavywall pipe and plate, fully welded. The perforated legs are "gun casing"- oil well casing that's dropped down the well full of a seies of small shaped explosive charges. The charges perforate the casing, crack the rock, and allow greater oil production.

But I digress. :D

I put a lot of miles on this poor mill before I picked up the Bridgy clone- you can see the "shadow" on the wall, notably behind the leftmost upright just above the table, of the "spray" from cutting oils, chips and other swarf.

The one thing I didn't do, though, is build in any option to tram the mill- I'd assumed that the fixed column couldn't be trammed, so I fitted the upper bracket as closely as I could and called it good.

Doc.

KiloBravo
03-21-2008, 05:26 AM
That's an interesting idea.
Is this epoxy granite a commercial item, or does one just mix fancy fishtank gravel with some JB weld and call it good. :D

Doc.


You can read all about it on his website/blog

http://www.cnccookbook.com/CCMillEpoxyFill.htm

Pete H
03-21-2008, 09:22 AM
Very cool, and some tips for working with epoxy that I'd never come across elsewhere - like using a little torch for popping bubbles.

One point - I've had epoxy "cook off" on me - get hot enough to smoke, when it was in large volumes. Could that be a problem in the column? Or is the volume of epoxy low enough that it doesn't really get very hot?

WEST makes a "high-temperature" hardener for use in the Summer (that is, it's slower than the regular hardener), and that would slow the reaction down.

Thanks for a very good, and easily-followed, description.

BobWarfield
03-21-2008, 12:25 PM
I've had epoxy "kick" and get real hot too, but have not seen that for this project. It could be the West epoxy I'm using, or it could be that I mix it up in relatively small batches added one after the other, but I have hardly felt it getting warm to the touch, let alone hot.

FWIW, there is a cheaper epoxy supplier than West Marine that is mentioned on the original CNCZone E/G thread. I don't have the link handy, but you can find it. My West epoxy cost about the same because I got some "scratch and dent".

As to how precise it all has to be. There are thousands of posts over on CNCZone about minute details. Those guys were ordering precision aggregates in 5 sizes, "Zeospheres", wetting compounds, deairing compounds, lamp black, you name it. It's not clear to me that they were getting dramatically better results from all that, and I wanted to get on with a project using locally available materials.

Filling existing castings is a much less demanding task than casting up entire machines from this stuff, which is what the CNCZone guys were after. I think this simplified approach will work very well for filling castings. Since I am lucky enough to have 2 identical mills, I intend to try to cook up some tests of some kind to quantify how much this has helped. Any ideas would be appreciated.

Meanwhile, DocNickel, I like your column brace and as mentioned I really like Torker's jacking arrangement too.

One thought if you were to fill a column entirely. You may want some kind of vibrating arrangement to help the air out of the mixture inside the column. Some of the CNCZone guys were using concrete vibrators from Harbor Freight for the purpose.

Cheers,

BW

lazlo
03-21-2008, 01:47 PM
You can read all about it on his website/blog

http://www.cnccookbook.com/CCMillEpoxyFill.htm

I'm surprised that Bob's Blog never mentions that a great deal of the hard work on the CNCzone epoxy granite thread, including the optimal aggregate forumula that Bob quotes, the sources for the aggregates and the West Epoxy, came from Cameron ("ckelloug") -- an active member here.

Cameron's "optimal" formula, which Bob quotes on his blog, was derived by some heavy-duty mathematical analysis of the de Larrard Packing Models and then computer simulation of the classic 3D packing problem: how to fill the greatest percentage of the 3D volume with the smallest spacing between the aggregate components.

The reason the packing density is important (and the reason the commerical epoxy granite mixtures are trade secrets) is that the tighter you pack the aggregate, the higher the dynamic strength of the resulting pour.

Cameron's optimal mix achieves someone on the order of 93% packing, which is amazing:

20% Agsco #6 Aluminum Oxide
20% Agsco #4 Quartz
20% Agsco #2 Quartz
20% Agsco #2/0 Quartz
20% 3M #800 Zeeospheres

...and here's what Cameron's formula looks like when you pour it. That entire block is epoxy granite:

http://www.cnczone.com/forums/attachment.php?attachmentid=41711&d=1186248447
http://img69.imageshack.us/img69/7480/twogranitedeviceshb9.jpg

Cameron wrote a paper on the analysis and derivation of the formula here:

http://www.cnczone.com/forums/attachment.php?attachmentid=53869&d=1203602933

He posted a general synopsis of the problem and his solution here:

http://www.cnczone.com/forums/showpost.php?p=320419&postcount=1824

Great job Cameron!!!

speedsport
03-21-2008, 01:59 PM
why not just fill it with lead?, prolly cheaper.

lazlo
03-21-2008, 02:40 PM
Lead would work exceptionally well to damp vibration Speedsport, but it wouldn't do anything to improve the dynamic stiffness.

speedsport
03-21-2008, 03:10 PM
lazlo, you never heard of "having lead in your pencil" & dynamic stiffness?

Willy
03-21-2008, 04:01 PM
Thanks for the info Russ, I know you work the bajeezus out of that mill/drill, so if it's working for you it has definitely passed the acid test.
I'll want give that a try as soon as I unpaint my way out of the corner I'm in now.

I'm not sure what you used for a concrete mix but I do remember that you used a rod mounted on a air chisel as a vibrator for help in removing entrained air and improving density. Correct me if I'm wrong.

I'm also curious if anybody has ever used a product such as Blok Hard (http://www.hardblok.com/info.html) .

For those not familiar with this product, it is a product commonly added to the water jacket area in high horsepower drag race engines in order to increase block stiffness and cylinder integrity. I've seen it work magic in engines not originally not designed for such high power levels. Just thought that this application would be a natural for such a product. Not cheap though, as i think it works out to about $70-$80 for twenty eight lbs.

ckelloug
03-21-2008, 04:09 PM
Thanks for the kudos, Lazlo. There are lots of folks that deserve credit for the success of the E/G thread on CNCZONE! They're all over the world including Canada, England, and France.

The supplier of cheap epoxy that Bob mentions is www.uscomposites.com

The photo of the cnc router gantry top quoted in Lazlo's post is by Walter Jarog in Chicago posted on CNCZone. Walter was the original instigator of the Epoxy/granite thread. I just happened to be interested and worked on it further providing walter some formulating suggestions.

In general, if the machine that you are trying to fill is already structurally sound, the E/G mixture used to fill it won't really matter: you're basically just adding mass and mass that has high damping at that. The point of optimizing an E/G formula is for building all composite machines.

Lead in epoxy would also be a good damping addition although it has very low strength. It must be remembered that with epoxy granite, a lot of the vibration damping is from the fact that E/G is like a giant bucket of masses, springs and dampers rather than something with uniform properties like metal.

According to the rule of mixtures and the Hashin-Shtrikman Bounds, the stiffness and ultimate strength of the material can be extremely sensitive to the proportions of the stronger and weaker components. The sensitivity and the gain to be had by a tiny improvement in density are largest once you get over about 90% aggregate. With aggregate densities of 50% and 80% it may hardly make a difference. Between 80% and 95% aggregate there is the potential for a doubling of the modulus.

The torch for popping bubbles is good for cosmetic surface bubbles but less effective at the internal bubbles that lower the density of the material overall. The torch was suggested by lgalla on CNCZone many times. In that air bubbles only occur in the epoxy phase, the less epoxy you use, the fewer the bubbles is the attitude I've been taking.

At any rate, after a year of modeling and advising the adventurous, I am just about to start running modulus and strength tests on my admet materials test machine. Those results will get posted to the epoxy granite thread on CNCZone when I get done.

Finally, it should be noted that Walter added an entry for Epoxy/Granite to the wikipedia which is a good short read http://en.wikipedia.org/wiki/Epoxy-granite

I'm glad to see the interest here and happy to see credit go to all those deserving souls on CNCZONE who have done the hard work (including lazlo, myself, and BobWarfield whom I know are on both boards).

Cheers all,

Cameron

lazlo
03-21-2008, 04:26 PM
I'm also curious if anybody has ever used a product such as Blok Hard (http://www.hardblok.com/info.html).

Blok Hard came up in one of the discussions on the Yahoo Mill/Drill group about filling the column with concrete. But it just looks like iron-filled hydraulic cement?

At $2.75/lb for block hard, you may as well go for DuraBar or VersaBar, which is $1.70/lb (but not pourable, obviously)

Willy
03-21-2008, 06:00 PM
Lazlo I'm sure Torker will fill in the gaps here but I think he used the steel shaft encased in concrete. That is why I mentioned the air chisel vibrator.

If I remember correctly he used rags to plug the bottom of his column and had the shaft project out of the bottom.

JRouche
03-21-2008, 09:58 PM
Great thread, thanks Russ. And Bob, you have put alot of consideration into your method.. I used high pressure concrete for mine not knowing anything else to do, it was five years ago..

I am curious if there are any ideas on conductivity. Meaning mechanical conductivity. For example. If a large piece of metal were to be welded to another the mechanical conduciveness for vibrations would seem to be high. Mechanical bolting would also seem to be high.

For the other end of the spectrum, just to get my point across. If section of metal were bolted to another with a sheet of rubber between the two the conductivity would be greatly reduced.

Any idea as to the conductivity (as expressed here) of the various materials. Such as concrete, epoxy, lead pour, steel shot fill, lead shot fill and so on..

I think of epoxy as a type of plastic, I dont know if it acts as a plastic, just my seat of the pants feel. A plastic seems like it would be some what of and insulator to carry mechanical vibrations.

And without a carry through of the vibrations the load (epoxy load) would in essence isolate the vibrations to the casting instead of absorbing it, not dampening it..

Very interesting idea though adding mass. I imagine one could place a mechanical device for measuring vibrations on the center of the mass (epoxy) and set up a vibration on the casting and see how much of it is actually felt at the center of the epoxy. A graph showing a trailing off of the vibrations would be a good indicator of the mechanical connection. And a known "control" could also be tested, say a fill of molten steel, or just a solid block of steel.

Ok, now Im going way off here...

So, if you or I should say me.. LOL Made up these various test "beds" They would be a steel box of a small size. Say a foot by a foot by half a foot deep.. Fill the beds with various dampening materials. From epoxy, lead, sand, steel and what ever else comes to mind.. Test them all for mechanical conductivity!!!!!

Oh wow, sorry, strayed abit...

I was just curious if the is even that much of a difference between the fills of is there is a huge difference... Just for dampening purposes, not structural strength. That would be a whole nuther discussion LOL JR

torker
03-21-2008, 10:48 PM
JR... Holy Smoke! Yer a deep thinker!
Not so here. Willy is right...I stuffed rags around the bottom of the column to stop the concrete from dripping through. And I DID use an air chisel to vibrate the concrete into the base. Used a big flat face punch(welded on a washer) with a rubber furniture foot pad slipped over it.
Kept the air pressure at 40 psi and rattled it for a loooong time between "fillings" Gawd... I shoulda been a dentist.
Mechanical attachements adding to vibration...maybe. But the concrete and the big geezuz piece of shaft added a lot of mass to this mill.
Why did I do all this chit???
I was trying to tram the mill. Someone had mentioned putting a DTI on the column of a mill drill to test for flex.
AFTER I put the concrete in I did that. Wow! I was depressed! You could lean hard on the top of the column and see shift on the DI. I wanted to know why.
I put the DI all over the base and came up with the reason it was flexing so much. The walls of the base at the back where bowing out slightly when I pressed on the top of the column.
What I did cured that.
You all go ahead and laff but this lil' mill is not your ordinary mill drill anymore.
It's a stiff little sukker. A 5" post of solid steel and concrete...held very tight by 4 1/2" bolts at the bottom. Compare that to other "bigger" vertical mills.
Mass and rigidity are what it's all about. This has it now.
It has 2 hp and I can use far more of it than I could before I did the one millionth modification to the mill drill. This thing gets a steady diet of a 2" indexable end mill at the most DOC it can stand. That cutter is almost always in the quill.
Russ

ckelloug
03-22-2008, 12:06 AM
Damping is a tricky beast. Nasty math (for the average HSM) comes into play. I can't say I'm too great at differential equations either but. . . Not only is material a factor but also mass, cross section, and vibration frequency.

The relative damping effectiveness normalized to aluminum is given below:

T6 6063 aluminum 1.0
mild steel 1.8
cast iron 2.4
epoxy granite 14.0
lead 16.0
Portland cement 48.0
sand 398.0
(Normalized to aluminum by me from table 7.3.2 in precision machine design by Alexander Slocum)

Generally, vibrations have an easier time traveling through a material of uniform composition than one with many phases. Epoxy Granite is a good damper because epoxy is relatively soft while the granite is very stiff. As a result, the vibrations bounce off of the interface between every tiny stone in the material and the epoxy surrounding it. Every time the vibration bounces, it loses energy. Since each stone has a bit of compliance in the material, not all of the stones in the material are moving the same direction at the same time and so it's very hard to get the whole piece to move in the same direction at the same time.

Sand is excellent because there is air between the grains and because the grains are not constrained, there is energy absorbed by the friction between grains.

So, in summary, aluminum is reasonably strong but crappy vibrationally. Steel and iron are very strong but mediocre vibrationally. Lead is good at damping but has little tensile, flexural or compressive strength. Portland cement has incredible damping but is not dimensionally stable and has very little to no tensile strength. Lastly we have sand which has great damping but no tensile strength, no flexural strength and behaves like a liquid rendering it useless for structural components.

So, throwing out the low damping materials and the high damping materials that are useless; that leaves Epoxy granite as the major practical solid vibration damping material: Epoxy granite has extraordinary compressive strength near cast iron, passable tensile strength, and passable flexural strength. It's also one of the few things an HSM can cast easily and it also has the benefit of low shrinkage and the ability to cast precision parts to tolerance.

Portland cement concrete or loose sand are likely better materials than E/G for base filling would be what I would conclude from the data in Slocum I processed and presented above for easy comparison.

Regards all,

Cameron

interiorpainter
03-22-2008, 07:00 AM
I would like to see a before and aft machining sample before going after all this trouble.
At the risk of sounding a bit like a broken record, how about a refrigirator pump to get the air out. Could be a bit to coors to get the air running.

Dynamic dampening properties of plastic is very high. That the reason for dynamic failure in plastics because of heat buildup.

QSIMDO
03-22-2008, 10:17 AM
I follow along when you discuss fabricating a machine structure out of E/G.

But when filling a cast iron machine base/component, is it just a case of adding mass to a structure?

I understand how the E/G would prevent the mill base from ringing like a bell but couldn't that be accomplished by adding stiffeners?

BobWarfield
03-22-2008, 12:06 PM
QSIMDO and interiorpainter, there is more going on here than meets the eye.

First, I agree that some experiments need to be done to validate the efficacy. You have one such in Torker's view that his mill is capable of a lot more than used to be the case. I will also be performing some experiments both in the area of surface finish/performance at various depths of cut, and in the area of measuring deflection and looking for increases in stiffness.

Second, there is a tendency to conclude it's all about mass. This leads one to want to use lead and other materials, or even just sand. But, there is more to it than that. Too much mass will lower the natural frequency of the system, despite the damping, which is considered bad all other things being equal. Reading literature like Bamberg's MIT thesis on machine design, he suggests avoiding sand and lead for that reasons. A higher natural frequency, preferably one much higher, is the desirable outcome.

Third, it's easier to just do it than talk about it. It takes very little effort to do one of these fill jobs. Most of the time involved is waiting for things to set up or planning how to avoid getting the epoxy where it isn't wanted. This was actually one of the few machine shop projects to date where my kids got involved, mixing up Dad's "mudpies", LOL.

The practice, BTW, is common in the commecial machine tool world. You can Google, for example "Hardinge Harcrete" to read about their approach.

Best,

BW

Mcgyver
03-22-2008, 12:26 PM
interesting subject, in measuring dampening, what exactly are you measuring? Vibration is a wave, and dampening is the decay of the wave right? So it has an amplitude, frequency and decay rate. What is being measure when comparing dampening; the time to go to zero, the amount of energy absorbed over a period of time and how does frequency and amplitude affect it - is it linear for both? Cameron, I've read those sorts of numbers (maybe at cnczone) but I've also in a book or company web site (I'm trying to remember the source of the info) that epoxy granite has about 2x the dampening ability of cast iron.....so I'm wonder where the difference is, ie could the dampening vary a lot with frequency, amplitude or application?

As I understand it, one of the ways better damping is achieve is when the wave has go from one material (or density) to another and that this transfer absorbs energy. This is why (i think) the EG (epoxy granite) is so effective and why just adding stiffeners won't do the same. However it does have a lower modulus of elasticity than steel or cast iron (yeah, I know, a well hashed cnc discussion). I think, and I'm in good company (hardinge et al), that using it to fill machine structures is an excellent application - it adds great vibrational dampening and some stiffness.

imo concrete is a bad idea is it can take years to cure and moves around - these problems are why EG's used instead.

it'll be interesting to see what sort of success you get Bob, I think you are on the right track with it

QSIMDO
03-22-2008, 01:08 PM
QSIMDO and interiorpainter, there is more going on here than meets the eye.
BW

Oh, no argument there!
I'm just trying to get my head around the mechanics of this having a positive affect on the cutter/work interface.

Hardinge, et al start at ground zero (not to take anything away from those who've labored long in the hobby mode on this thesis) and I'm just wondering if a retro fit with this fascinating material is worth more than, say, bolting the machine down to a granite surface plate.

I'm following with great interest and Kudos to anyone/everyone involved, but I'll wait for the movie!
I have a rush order of swarf to get out!
;)

lazlo
03-22-2008, 01:24 PM
I think of epoxy as a type of plastic, I dont know if it acts as a plastic, just my seat of the pants feel. A plastic seems like it would be some what of and insulator to carry mechanical vibrations.

That's right -- the epoxy is the weak-link in the epoxy granite. Although is does have some vibration damping characteristics on it's own, the purpose of the epoxy is just to hold the aggregate together. The majority of the vibration damping, and almost all the dynamic strength, comes from the aggregate.

This is why road surfacing aggregates, as well as Cameron's formula, go through great lengths to minimize the binder volume/weight, and maximize the overall content of aggregate -- hence the 3D Packing Problem.

The biggest challenge that Cameron and Walter had with mixing the optimal formulas was that they're so dry (so little epoxy) that they're incredibly hard to mix.

lazlo
03-22-2008, 01:38 PM
Vibration is a wave, and dampening is the decay of the wave right? So it has an amplitude, frequency and decay rate.

As I understand it, one of the ways better damping is achieve is when the wave has go from one material (or density) to another and that this transfer absorbs energy. This is why (i think) the EG (epoxy granite) is so effective and why just adding stiffeners won't do the same.

That's my understanding as well: by mixing a variety of aggregate sizes, the epoxy granite is encapsulating a large assortment of mass dampers, where each aggregate component has a different natural frequency, and damps a different resonant frequency. So the finest aggregate (the Zeosphere's in Cameron's forumula) damp high frequency vibration components, and the coarse quartz aggregate damps low frequency components*.

So for maximum vibration damping, you want a large variety of aggregate sizes in the epoxy granite.


I've read those sorts of numbers (maybe at cnczone) but I've also in a book or company web site ... that epoxy granite has about 2x the dampening ability of cast iron.....so I'm wonder where the difference is

I've also read the various machinery manufacturers quoting that steel has twice the vibration damping of aluminum, grey cast iron twice better than steel, and epoxy granite twice that of grey cast iron.

I'd imagine the difference is in the implementation: you have to have a very tight coupling of the epoxy granite with the structure to get the best damping properties.


However it does have a lower modulus of elasticity than steel or cast iron (yeah, I know, a well hashed cnc discussion). I think, and I'm in good company (hardinge et al), that using it to fill machine structures is an excellent application - it adds great vibrational dampening and some stiffness.


Like you say, we're summarizing the looonnnngg discusion on CNCZone, but epoxy granite, at it's best, is pretty weak compared to grey cast iron, so you give up a lot of dynamic stiffness in return for the vibration damping.

That's why Harcrete, for example, is used as a filler, and not as the primary structural material for the machine tool.




* Another advantage of Cameron's epoxy granite formula is that the Zeopheres look like black graphite, which is why Walter's epoxy granite linear rail in the picture I posted in the first page looks so pretty (like polished ebony) :D

speedsport
03-22-2008, 02:48 PM
McGyver, its DAMPING, not dampening, sheesh, just a pet peeve of mine.

Paul Alciatore
03-22-2008, 03:00 PM
This is a most interesting thread. Who would have thunk of epoxy concrete?

I wonder if adding some lead shot to the mix would add to the properties. Gravel, then lead shot to fill the voids, then sand to fill the smaller voids. It would add more mass than just sand and another set of properties for the vibrations to bounce around in. Or perhaps small pellets of some other metal.

lazlo
03-22-2008, 03:54 PM
That's an interesting idea Paul.

There was an old HSM article that I need to get a copy of, where the guy made a lead-shot filled boring bar -- similar to what you're describing.

ckelloug
03-22-2008, 04:56 PM
One way of measuring damping is taking a square rod of the material, clamping one end in a vise displacing the other end, letting go and measuring the decreasing displacement as it rings.

I haven't gotten a hold of a copy but According to Precision Machine Design by Alexander Slocum, an authoritative text on the subject of damping is B.J. Lazan Damping of Materials and Members in Structural Mechanics Pergamon Press, London, 1968.

Mcgyver,
I've always seen a a 7x number between cast iron and epoxy granite. The first source I've seen was in a Machine Design article by Terry Capuano of Accures casting.

The numbers in my post are zeta (damping coefficient in a second order system) values from Slocum's text Precision Machine Design although I normalized the presented numbers to aluminum in the post to give everybody more intuitive relative numbers. The ratio between cast iron and E/G also comes up to about 7 in these numbers.

QSIMODO,

The problem with simply making something stiffer is that stiffness alone does not make vibrations go away. It just makes them smaller. If you could make (which of course you can't) a system which had no damping, it would continue to vibrate forever. The higher the damping, the quicker the piece will stop vibrating.

If you add mass to something without increasing it's stiffness, then the resonant frequency will in fact go down. If you manage theoretically to add so much mass to a system without enough stiffness, it could bring the resonance down to a harmonic of your spindle speed. If that happens then you are boned because the system will vibrate horribly.

Of course, it's generally impossible to add stiffness without mass in a column filling operation regardless of the material. In theory however, the idea is to make sure that the mass, stiffness, and damping are such that the vibrations while operating are either substantially below the resonance peak or substantially above it.

This vibration problem is all tempered however by the fact that in addition to having to deal with vibrations, machines also have to resist cutting forces and deflection under their own weight. Thus they have to be stiff enough to resist all these forces without deflecting significantly.

Without modeling it, I'd say that the mass added from filling the column on a mill drill slightly lowers the resonant frequency but adds damping. The filling also should increase the stiffness of the machine which has a corresponding effect of raising the resonant frequency. Which effect wins out is a function of the geometry of the parts and without getting specific information about the machine in question it's impossible to predict accurately. Intuition makes me thing the effects on resonant frequency are probably a wash and the net effect is a stiffer machine with higher damping and maybe a slightly lowered resonant frequency.

As long as you manage not to gum up leadscrews or ways with what you're filling the machine with, it's almost sure to be of some benefit.

Just to make this discussion go somewhere other than epoxy granite:

I Read Bamberg's thesis as suggested by BobWarfield see http://www.mech.utah.edu/~bamberg/research/PrinciplesOfRapidMachineDesign/Principles%20of%20Rapid%20Machine%20Design.pdf

It looks to me like a rather incredible constrained layer damping system could be made by inserting something like a weather balloon into the column and filling if up with concrete laced with aluminum chips which turn to hydrogen in the caustic concrete causing it to expand. Can't imagine where an HSM might find something as exotic as aluminum chips.

In conclusion, I think filling machine bases with E/G or some other material with good damping is likely to increase both static stiffness and dynamic stiffness without making the resonance point come down too much. I'm a theoretician here however as my Bridgeport is too heavy to experiment with as is.

Regards all,

Cameron

lazlo
03-22-2008, 07:01 PM
Cameron, this is Slocum's table from his book, and his vibration damping constants are consistent with what I quoted for aluminum, steel and cast iron. Note that he quotes polymer concrete, but not epoxy granite:


I've also read the various machinery manufacturers quoting that steel has twice the vibration damping of aluminum, grey cast iron twice better than steel, and epoxy granite twice that of grey cast iron.

http://i164.photobucket.com/albums/u15/rtgeorge_album/Slocum.gif

lazlo
03-22-2008, 07:15 PM
Actually Cameron, I bet I know what you're thinking about with the 7x number: Slocum designed a prototype cylindrical grinder for the Weldon Company that used viscous epoxy shear dampers as an integral part of the machine's base design. Rather than rigidly bolt or weld joints, he bonded alternate layers of viscous and structural material to the machine surfaces -- IIRC the shear damper bars were layers some kind of exotic rubber and polymer concrete filling square cross-section cast iron tubing. He even had coolant for the grinder running through the center of the cast iron tubing.

I seem to remember that he quoted some exceptionally high vibration damping on that machine, but that was a highly tuned shear damping system, that's not saying that polymer concrete or epoxy granite has 7x the vibration damping of cast iron.

There's an article about the grinder Slocum built for Weldon somewhere on the Machine Design web page -- someone usually finds the link for it...

Edit: I found the link, on Design News. Slocum used Vibradamp, a commercial epoxy granite from Philadelphia Resins. He also patented and trademarked the hollow, concentric shear damper bars, which he calls "ShearTube." Note that he claims that the shear damper bars can decrease amplification at resonance of a metal beam from 500 to 20. But that's the shear damper tube with all the exotic elements, including VibraDamp, that's not saying that epoxy granite is 20 - 500 times better vibration damping than "a metal beam."

http://www.designnews.com/article/CA151212.html


Weldon engineers and Professor Alexander Slocum of the Massachusetts Institute of Technology employed an epoxy replicant called VibradampTM from Philadelphia Resins. The composite is injected in and around viscoelastic covered structural tubes, designed by Slocum, to improve damping on the precision grinder.

The damped structural tubes, winners of a 1994 R&D award for one of the 100 most technologically significant new products, have doubled the accuracy of Weldon's precision grinder, according to Weldon engineer Jim Flinchbaugh.

The design, used in the base of the grinder, enables engineers to highly damp a structure without imposing strict limits on the structure's geometry or materials of construction. Tests have shown that the ShearTubeTM damper can decrease amplification at resonance of a metal beam from 500 to 20.

According to Slocum, the tube design uses a thin film of viscoelastic material that dissipates energy at a rate orders of magnitude higher than most designs. And coolant can be circulated through the hollow tubes providing thermal stability for added precision.

JRouche
03-22-2008, 07:16 PM
McGyver, its DAMPING, not dampening, sheesh, just a pet peeve of mine.

LOL And yer so right, I miss spelled it three times myself. I should know better.. Used to have to deal with damping on synchro/servo systems years ago.. Me brain cells are failing me now :) JR

Evan
03-22-2008, 07:48 PM
There are ways besides adding mass to obtain good damping, even from aluminum. I was aware of the problem when designing my all aluminum mill and drawing on my experience with aircraft construction I used the same sort of solution that prevents the wings from fluttering and breaking off aluminum aircraft. By building up an aluminum structure with layers of material held together with fasteners it is possible to produce a highly damped structure.

In the case of my mill each of the two columns are built up from three individual components. Two side by side 2"x 3" columns are bolted together every 5 inches or so to provide damping in the X axis. To provide damping in the Y axis a face plate is then bolted to the front of both columns. The very slight scuffing that occurs between the bolted components under stress is extremely effective in absorbing energy and can provide damping as good as materials that are inherently highly damped.

This is the structure of the vertical columns, left is front.

http://vts.bc.ca/pics3/millcolumn.gif

I used this same technique elsewhere in the mill. The base of the machine is built up from two layers of 3/4" aluminum tooling plate with additional stringers below and on the sides. The same applies to the carriage that carries the Y axis ways as well as the attachment system of the vertical columns to the base.

Partly for these reasons the mill behaves much as a similar cast iron machine would and doesn't exhibit the chatter that one might expect from an aluminum machine.

ckelloug
03-22-2008, 07:51 PM
Lazlo,

From reading Slocum and Bamberg, Engineered Shear Tube dampers are substantially better than ordinary epoxy granite beams but require a reasonable amount engineering work. I believe it is actually true about the material damping of E/G being an order of magnitude better than cast iron however.

This is the article from machine design magazine I mentioned. It is from Accures castings' president Terry Capuano. It cites research that the material damping factor for epoxy granite really is better than cast iron by an order of magnitude.

http://machinedesign.com/ContentItem/5839/PolymerCastingstakeonmetals.aspx

Table 7.3.2 as cited by Lazlo lists the actual zeta coefficients which I just normalized to aluminum in my post. I used the exact zeta1 values from the table but divided them by the zeta1 values for 6063 T6 aluminum to normalize them for easy comparison.

I considered the entry in the table for polymer concrete to be epoxy granite since polymer concrete is usually epoxy granite or epoxy quartz in the machine tool sense. In the general industrial sense however polymer concrete is often used to mean a mineral polyester composite rather than a mineral epoxy composite. We started using "epoxy granite" over on the thread on CNCZONE to cut down on the people talking about polyester countertop material. Reichhold's polymer concrete apps engineer told me that he does polyester, not epoxy and that I'd have to talk to the epoxy group. I also googled about for Gandalf Inc. where Slocum got his "Polymer Concrete" zeta value but they have gone out of business.

Lazlo, I think we agree mostly but have some different references, and slightly different interpretations. I think the material damping for E/G really is an order of magnitude greater than cast iron however.

Thanks for posting the scan of the table. I don't have a working scanner right now.

--Cameron

lazlo
03-22-2008, 10:18 PM
Lazlo, I think we agree mostly but have some different references, and slightly different interpretations. I think the material damping for E/G really is an order of magnitude greater than cast iron however.

Yep, I think we're in complete agreement Cameron, I just wasn't sure about the actual magnitude of the improvement of EG over cast iron.

The Zeta Slocum is quoting in that chart I posted from Precision Machine Design is the natural log of the amplitude of the vibration, and I just noticed that he's using 6 x 10 -4 for "cast iron", which is the vibration damping of common maleable or ductile iron. Fine-grained flake grey cast iron (as used in machine tools) damps about 5 times better than "common" iron (which is precisely why it's used in machine tools).

So Epoxy Granite damps 10x better than common iron, and around 2x better than fine grey cast iron, which explains both our recollections:

http://i164.photobucket.com/albums/u15/rtgeorge_album/vibration.gif

By the way, here's a pictorial showing the amplitude of the vibration of steel, "common" cast iron, and grey cast iron (aluminum isn't shown because it would be off the charts):

http://i164.photobucket.com/albums/u15/rtgeorge_album/vibration3a.gif

Now compare those damping charts with EG, on the far left (from the Machine Design article). Epoxy granite really is amazing stuff, but you won't get that kind of damping unless you have really precisely chosen aggregate, like Cameron/Walter have done (the Accures EG measured in the far left chart has 92% solids):

http://www.accurescasting.com/Machine2_files/MecRG04-00.gif

ckelloug
03-22-2008, 10:40 PM
Now compare those damping charts with EG, on the far left (from the Machine Design article). Epoxy granite really is amazing stuff, but you won't get that kind of damping unless you have really precisely chosen aggregate, like Cameron/Walter have done (the Accures EG measured in the far left chart has 92% solids):



Do you have some more data on this statement lazlo?

I personally have never been able to find anything in the theory that says that the damping is specifically improved by improving the aggregate to epoxy ratio. What I can say with certainty is that the modulus (stiffness for the non-engineers) is non-linearly proportional to the aggregate to epoxy ratio and that the last few percent of aggregate content can theoretically make up to a factor of two difference in stiffness.

I would like to do some research on the modulus and the strength with the equipment I've got. I'd also like to eventually be able to obtain damping measurements with respect to aggregate percentage but that's a long way off for me.


Also,

I'm a bit fuzzy but it appears the data you posted from the iron handbook appears to be what slocum calls the log decrement (Slocum's equation 7.3.8) where as zeta is the second order system damping factor which is a function of the log decrement (Slocum's equation 7.3.9). Have I got this wrong?

Regards all,

Cameron

Mcgyver
03-24-2008, 10:40 AM
McGyver, its DAMPING, not dampening, sheesh, just a pet peeve of mine.

damping is the better, or rather perfect word, frankly, didn't know it existed so thanks for the erudition. however dampening was not wrong to use; "to deaden, restrain, or depress", whatever, now i know the right one.

Lazlo, that's the stuff I'd read that made me think 2x!

There's application practicalities though that make it less important exactly how much better than cast iron it is......for example it might be practical to fill a void in a milling machine cast column with say a cross sectional area of 1.5 sq feet - even if the EG was only 2x, the resultant structure might have 20x better damping properties just by the shear volume of EG added. Similarly, while it has a much lower Young's modulus, I'd expect adding (in this example) many many times the cross sectional area of EG compared to the cast iron that was there to start with would add a lot of rigidity.


I personally have never been able to find anything in the theory that says that the damping is specifically improved by improving the aggregate to epoxy ratio.

Cameron, I'm a bit in imagination land on this because i don't really know.....but....as i understand it, its not the epoxy or the granite or the mass that creates the superior damping, its the combination of different materials; that energy is absorbed in the boundary as the wave passes from epoxy to granite. Therefor, wouldn't it be reasonable to expect dampening to increase as you increase the surface area between granite and epoxy for a given volume?

torker
03-24-2008, 11:13 AM
Guys.. A lot of interesting theory etc. However...
I started this whole thing on the "advice" from someone who had never owned a mill drill.
He suggested the column was weak and was the problem...said filling it with concrete would cure the "flex" problem.
I checked the head flex with a DTI and sure enough..just like he said...the column was indead flexing... or so I thought.
Then I went further..thinking that the concrete alone wouldn't be as strong as if I had a big steel post in there. I never meant to use the jacking bolts...the post was that long so I used it as is.
After all the fooling around, I put the DTI back in the same location. Pushed hard on the head and was very dismayed to see the SAME amount of flex...even with a bunch of extra strength and weight.
The guy didn't know what he was talking about. The flex comes from the open backed base.
The sides actually flex in and out depending on how you push on the head.
They don't flex much.. I could barely get a reading on the DT...but add up the bit of flex then tranfer that to the top of the machine and you have the .005 flex I was getting.
That's how the jacking bolt thing came to pass. I thought I could stabilize the thing with these.
It still didn't really work til I tied the base sides together.
The concrete fill??? The post??? Yes it added mass but I don't think the column was a REAL weak point anyway.
The fact that the bolts work for tramming the head in was also a by product of the whole thing..it works but it was a fluke. I "stumbled" upon that.
And I can guarantee that the whole works made the mill a lot better. I used to have to take .020 max cuts with the 2" endmill(the mill would actually start to shake). Now I can take .050 easily with half worn inserts.
With new inserts it'll take a .075 cut but the positive rake inserts don't last very long with heavy cuts so i keep it around .050.
Russ

ckelloug
03-24-2008, 01:53 PM
Cameron, I'm a bit in imagination land on this because i don't really know.....but....as i understand it, its not the epoxy or the granite or the mass that creates the superior damping, its the combination of different materials; that energy is absorbed in the boundary as the wave passes from epoxy to granite. Therefor, wouldn't it be reasonable to expect dampening to increase as you increase the surface area between granite and epoxy for a given volume?

Mcgyver, I think you are probably right that increased surface area plays a role. Size of the particles and the epoxy filled space between the particles also plays a role. I suspect that the minimum epoxy to bind the mixture effectively is probably not the optimum amount for damping. If the data on the right ratio exists, it is proprietary and not available from the literature. I intend to work on this once I am set up for lab work.

I just checked out a copy of B.J. Lazan's hard-to-find textbook on structural damping from my local university library. I plan to delve into this more if I get time during my travel this week.

--Cameron

lazlo
03-24-2008, 02:09 PM
Do you have some more data on this statement lazlo?

I personally have never been able to find anything in the theory that says that the damping is specifically improved by improving the aggregate to epoxy ratio.

I thought I read it either in Slocum's book, or his student's PhD thesis (the one that BobW quotes all the time with the steel castings filled with rubber bags filled with portland cement). I'll see if I can find it.


What I can say with certainty is that the modulus is non-linearly proportional to the aggregate to epoxy ratio and that the last few percent of aggregate content can theoretically make up to a factor of two difference in stiffness.

Right, the epoxy itself is very weak compared to the aggregate, so it's there only to hold the aggregate together, and to the walls of the machine.


I would like to do some research on the modulus and the strength with the equipment I've got. I'd also like to eventually be able to obtain damping measurements with respect to aggregate percentage but that's a long way off for me.

You're going to need a strain gage or accelerometer and a spectrum analyzer. Basically, 90% of a dynamic balancing machine. :)


I'm a bit fuzzy but it appears the data you posted from the iron handbook appears to be what slocum calls the log decrement where as zeta is the second order system damping factor which is a function of the log decrement.

I'm at work. I'll check Slocum's book when I get home...

ckelloug
03-24-2008, 02:23 PM
torker,

Wow! That's a good analysis of your situation. You have demonstrated conclusively that the largest problem with mill drills is not lack of damping or lack of weight but lack of stiffness. Bolts are a lot cheaper than epoxy granite and extremely effective at adding stiffness since the back of the column is expanding. There is next to nothing that just adding high damping materials will do for you if the column flexes .005 under a load with no vibration.

I suspect however that filling the column of the mill drill with epoxy granite might have helped your mill as much or more than the bolts did. This is not because it is a damping material but because it has some strength in tension. Combine the strength in tension with the large cross section of a column and it should translate into a large stiffness gain. This doesn't work for concrete however as it doesn't tend to stick to things and behaves very poorly in tension.

Sorry to get into the damping discussion and hijacking your thread. I guess I should slink back to the E/G thread on CNCZONE now;)

Regards all,

Cameron

ptjw7uk
03-24-2008, 03:15 PM
Could you stress the column with some rods inside and give it some stiffness that way or would it be better with external struts.
Just a thought.
Peter

BobWarfield
03-24-2008, 10:15 PM
that bamberg thesis mentioned earlier is worth a read for several of these posts.

First, Evans sliding plate idea is spot on. In fact it works exactly as h describes with vibration damped via sliding friction. For that reason I works even better with a thin layer of elastic dampener like epoxy between the plates. Thin is less than 1mm.

Second it can be a good idea to add reinforcing rod. Bamberg got great results adding rebar to his concrete fills.

Btw, he encased his fills in a rubber bag to protect against any corrosion and to increase that rubbing damping effect.

Anyway, it'll be fun to test. I hope I'll see as much improvement as torker!

Best,

BE

torker
03-24-2008, 11:39 PM
Cameron..LOL! No worries about the hijack. I learned a lot from you guys during this thread.
About the concrete inside this column. Someone else mentioned how rough the inside of theirs was. Well that'd be an understatement with this one.
When I looked down inside of it I thought my eyes were decieving me.
I've never seen a rougher casting than this. There are lumps and bumps that vary over 1/4" through out the whole thing.
I had absolutely no qualms about putting concrete in there and worrying about it ever letting loose.
Never thought about the corrosion issue. Should I worry...nah..I've seen bumper posts in outdoor parking lots that are filled with concrete. A couple I remember from 30 years ago. They've been hit repeatedly and are still fine.
They survive in a far harsher world than my mill.
If I was to do it all over again.. I actually would do the same thing but I would also fill the base as Bob is.
That 3" post in the column does add a lot of weight and it certainly does add stiffness if there was ever a question there.
The added mass certainly doesn't hurt as was discussed.
Another thought... they sell these mill drills with 2 hp. Even with the added things I've done and giving it a good bite with a 2" indexable end mill...the machine has never even grunted. The only limiting factor is the real lack of stiffness and to some degree the R-8 spindle. I almost think they only need 1hp motors.
BTW...the head slipping under a heavy cut has never been a problem with this mill. The only slip was when I forgot to tighten it.
If you overtighten the clamp that this Taiwan built mill uses, it will clamp down so hard that you can't loosen (the clamp) or the head without major problems. You have to remove the bolt...strap a long snipe on the mill head, jerk like crazy and tap a long punch in between the clamp blocks to hopefully loosen them.
How do I know this??? Again.. I read (on chaski) that one should always get a torque wrench and tighten the snot out of the bolt each time you move the head. I did that ONCE. It took me two hours to loosen the head.
Anyway...carry on with your talk about epoxy/rocks :D
Russ

lazlo
03-25-2008, 12:02 AM
If I was to do it all over again.. I actually would do the same thing but I would also fill the base as Bob is.

That 3" post in the column does add a lot of weight and it certainly does add stiffness if there was ever a question there.

I don't see how filling the base would do anything? Filling the column with concrete is an old, established mill/drill trick, and the posts I've seen on the Yahoo groups were as positive as your's Torker.

I replaced the column bolts on my mill/drill with large grade-8 bolts, and that made a huge difference on the column flex. I never got around to filling the column before I sold it...

torker
03-25-2008, 01:27 AM
Robert.. I'm totaly sold on mass. My old Ohio...six inch shell mill... .150 doc, absolutely no vibration...a cut so smooth you can see your nose hairs in the reflection.
It's not a huge mill...about the size of a BP but it weighs over 3500 pounds.
The knee weighs almost as much as my mill drill.
I'm betting another 100 pounds or so wouldn't hurt a mill drill either.
It'd just hurt your back when you tried to move it.

lazlo
03-25-2008, 11:45 AM
Sure Russ, filling the base certainly wouldn't hurt, but I don't think it's going to do much for rigidity.
Now filling the column, that's where the action is (pun intended)!

By the way, it's critical for improving the dynamic stiffness of the column, that the cement (or epoxy granite) have a tight bond with the mill's column. Normal Portland Cement is hydrophilic, so it shrinks as it cures, and will pull away from the mill's column.

Cameron mentioned adding aluminum chips to the Portland Cement, but you don't need to go through all that -- "hydraulic cement" expands slightly as it cures. I know that sounds like a rare and expensive industrial compound, but common grout, available at your local Home Depot or Lowes, is hydraulic cement. Even better, many grouts have quartz aggregate, which is getting very close to Hardinge's Harcrete mixture :)

ckelloug
03-25-2008, 07:13 PM
People Keep citing Erhard Bamberg's PHD thesis without understanding it fully as near as I can tell.

Bamberg's paper is not about filling machines with concrete for damping and expecting good results. Bamberg's paper is about viscoeleastic constrained layer damping.

Reduced to layman's terms, the rubber between the concrete and the machine is providing the damping and the concrete is there for mass and cheapness. The damping is caused by the behavior of the thin layer of rubber when subjected to vibrations. The attempts of the layer of concrete inside the rubber to move with respect the the machine base squeezing the rubber do the damping. The expanding concrete (created by the addition of a Sika product) forces the rubber tightly against the inside of the column of the machine.

Bamberg tested with cylindrical columns so until the column of the mill drill becomes as rigid a a cylindrical column, Bamberg's technique is unlikely to help out weakly performing mill drills.

Bamberg specifically states that his model does not account for the intrinsic material damping of concrete. He says in the paper that his models are off by a factor of 6 for the concrete constrained layer damper and that the error likely lays in not accounting for the damping of the concrete.

Mass is a good thing to add because when added properly, it increases stiffness. Stiffness is the most lacking point in light machine tools. If you can build a light machine tool that is stiff then adding damping will greatly improve the vibration performance. If you make a tool that is not stiff enough, specifically adding damping will not produce a particularly pronounced effect.

I conclude from torker's data that mill drill guys should listen to torker: Work on improving the stiffness of mill drill columns is paramount to seeing improved performance.

Regards all,

Cameron

BobWarfield
03-25-2008, 10:27 PM
actually, the paper is about building machines more cheaply with a variety of techniques. One chapter is damping and the results of multiple methods are compared. Rebar made a big difference and the rubber bag was unchanged over many of the cases.

That the model didn't match experimental results is a good advertisement for not getting lost in the math and other details rather than getting on with it. Torker got a great result by just getting on with it.

You do have to track down the weak links as torker did. Looking for flex with a dti is a good approach. He also made a great case on his mill for how the base and the columns mount to it mattered as much or more than the column itself.

bigalbert
01-16-2015, 01:35 AM
Does anyone know what size the 2 step drive pulley on Docs Mill is?