[greybeard]

Thanks Walter, but I make no claims for the idea. I found the info through a patent search, and constructed my own version to see if, or how well it would fit our requirements.
I must admit that yesterday I was disappointed that I coudn't report an amazing result.
As they say - "No pain, no gain" - so I'm still thinking about it, both possible ways to improve the design, and looking at much simpler methods.

One that occurs to me is to drop a fine stream of sand down between to sheets of glass, separated by only a few mms, then have a horizontal jet of air from a compressor knocking them sideways.

It's the same principle as my attempt above, but much simpler to build.

John

PS As they say over here - "Massive, man."

[walter]

Well there you go, it's Sunday 9 pm and the site does not load; took me 20min to open this window. I have it opened and now I can't use it- another 20 minutes..
This blank screen is really getting into me so I'm going to make this my final update. I'll post some more when I finish the machine, probably some time next year.
Test base will be done next Saturday.


Beam test: 300lbs in the center; deflection less than 0.0001"



_

[romihs]

Walter,

I have not had much time in the last few weeks/months to check out the progress on this forum. And a LOT of water has flown under this bridge since I was last on it.

I am amazed at what you have achieved! It looks great!
Good work! Especially with the deflection you posted there, well done!

Do have a separate thread about your build?

Regards

Sandi

[jhudler]

Jesus! Even with gantry splay that's better then I expected! What am I saying, what gantry splay!!
Was the Y axis bolted or hard coupled to legs or just sitting on top?
If it was sitting on top, then I would test it again, bolted, only get a better indicator!

BTW what are the dimensions of the gantry? H/W/T of the Y axis beam? Weight of each beam and legs?

Wow! Very nice!
Jack

[walter]

Good to see you Sandi!

I don't have the separate thread, stuff will show up somewhere on the internet at some point. I appreciate all you did for us, your data helped a lot. You're welcome to stop by more often!


Jack,

dimensions:

Y axis 42.5" x 6" x 9.5" 150lbs, unsupported length 32.5"
legs 5.5" x 19" squares 150lbs each

Y axis just sitting on top. I can't use better indicator- the machine wasn't even meant for any kind of precision. All I need is basic rigidity for wood carving.

Having said that, it would be totally easy to to build a decent metal machine tool. This is the minimum speed I consider for hobby E/G machines: http://www.youtube.com/watch?v=80va0W0qk48 I mean that in steel. This particular machine is not even semi professional, just bunch of welded tubing, rolled screws.. gimme a break. 20 bags of sand will give you better performance- just go for single piece designs:





btw, I redid the test. Bolted everything and placed the weight on 123 blocks



I set up video camera and did the acrobatics. 564lbs total, with 0.0001" deflection after I claimed up there;
I did my best. However these results are not very reliable, you know how these things are.
_

[romihs]

Walter,

Thanks!

I'll be around... I am thinking about this all the time....
I am trying to obtain 3M G800 and G850 here, in Slovenia, at the moment..


What is the recipe you based your mix on?
I know it is somewhere in the last 500 postsor so......

Regards

Sandi

[ckelloug]

John,

It's a regular Kit-atostrophe that your particle separator didn't meet specs. It is, in it's glory, pretty than some people's CNC machines however. I got your data package in the mail today. I have no idea how long the postal trolls take to deliver mail to the UK. . .

I also broke down and ordered an Admet eXpert 5601 flexural test system today. I'll soon be able to produce data on what is good vs. what is bad for both modulus and strength.

Walter,

That machine of yours is looking absolutely fabulous. I am blown away by the beauty of it. Congratulations on excellent work. The deflection numbers are also excellent. I suspect the modulus will be much higher than my sample calculations were made for.



Regards all,

Cameron

[jhudler]

"Admet eXpert 5601" that's serious! What model?

Jack

[brunog]

Having said that, it would be totally easy to to build a decent metal machine tool. This is the minimum speed I consider for hobby E/G machines: http://www.youtube.com/watch?v=80va0W0qk48 I mean that in steel. This particular machine is not even semi professional, just bunch of welded tubing, rolled screws.. gimme a break. 20 bags of sand will give you better performance- just go for single piece designs:


_

Walter,
High speed machining does not need big heavy machines like regular milling, all what you need is a high speed spindle and carbide bits. the whole idea behind high speed milling is making narrow passes that will increase the temperature of the chips close to melting point, that is why you will often see sparks when machining carbon steel. And because the passes are narrow the needed forces to pull out chips is minimal.

In am positive the machine you are building would be suitable not only for wood but also machining steel.

By the way, the gantry and posts look great!

best regards

Bruno

[ckelloug]

The description of the materials test machine I have ordered is here:

http://www.admet.com/assets/eXpert560x.pdf

It is a model 5601 with 225lbs force maximum. I've ordered it with a flexural test fixture for breaking samples of up to 3/8 inch by 1/2 inch square and 7.5 inches long with 6 inches between supports.

Regards all,

Cameron

[dfro]

I have almost everything I need to start making some castings. I drove to Chicago and got the stuff on Walter's list from Agsco and The Cary Company. The people at both companies were very friendly. The US
Composites epoxy arrived today, also. All I need is the de-airing agent from BYK, which they wrote is on the way.

I am still trying to reach somone at Cabot to get a carbon black sample for dying the leveling epoxy seal coat. (After the seal coat I will pour a final clear leveling coat, btw). I found a company that supplies lamp black powder:

http://www.naturalpigments.com/detai...umnBestSellers

I think I will try this, since it is only $5. Four ounces will go very far. It says on the webpage that it is amorphous carbon, with some traces of other combustion chemicals. That sound close enough for me.

I do not like going to huge companies like Cabot and asking them for samples. IMHO, I think we should find retail sources for everything that is needed in the e/g casting process. Agsco and The Cary Company, while obviously used to handling much larger orders, were happy to sell to me.

Instead of using carbon black to dye the epoxy seal coat, I was thinking the smallest size of zeeosphere could be used, instead. I like the greyish-black color of Walter's gantry. Walter, I imagine that the zeeospheres are mainly resposible for the blackish color of you parts. True?

That is truly inspiring, beautiful work!

I am sure you have answered this earlier somewhere, but:

After all of the aggregate has been mixed into the epoxy, do you put your mixture in the vacuum chamber to try and de-air it a little more, or do you just rely on vibration at this point to work bubbles out of the casting?

Thanks,
Dave

[jhudler]

Walter,
In am positive the machine you are building would be suitable not only for wood but also machining steel.
Bruno

I totally concur!!

[jhudler]

[QUOTE=greybeard;343489]I found the info through a patent search/QUOTE]

What's the patent number you looked at?

Jack

[greybeard]

A quick look at the pdf I kept has the number 4213852, Bernard Etkin, July 22 1980 on it.

If you have no luck tracking it down, let me know and I'll repeat my search.
Regards
John

Edit It's at freepatentsonline.com with that number.

[jhudler]

I believe, your gut feeling about "turbulence in the air stream" is correct.

The real important part of this separator is creating an air stream that's relatively turbulence-free and laminar.
Etkin must have never seen a wind tunnel, because in essence that's what he's created.
Which is pretty amazing, considering all the guys before him (way back to 1909) seem to understand albeit intuitively, the need for laminar flow.

First thing I would do is toss the screens (28) and replace them with small plastic drinking straws (~5 mm) and about 50-100 mm in length. The longer the better however, you'll start to pay a penalty in form of air resistance and require more SCFM from the fan. They usually come in a box that you can cut in half while the straws are still in the box.

Also if you can, roll the inside corners of box, it would help reduce the corner eddy currents thus reducing air resistance in the system. A simple piece of cardboard or sheet metal would do the trick.

In fairness to Etkin, the screens they used were probably carefully aligned and polished brass or stainless, points they judiciously left out of the patent. If he didn't... he got lucky, wasn't that picky, or just wanted a patent.

Jack

I just remembered another point about fluid dynamics; http://en.wikipedia.org/wiki/Reynolds_number. You would have to gauge the screen based on it's wire diameter, which I'm not going to dive into at this late hour!
I bet it's small enough to make very tiny eddies that would decay rapidly, not disturbing the laminar flow.

[greybeard]

I believe, your gut feeling about "turbulence in the air stream" is correct.

The real important part of this separator is creating an air stream that's relatively turbulence-free and laminar.
Etkin must have never seen a wind tunnel, because in essence that's what he's created.
Which is pretty amazing, considering all the guys before him (way back to 1909) seem to understand albeit intuitively, the need for laminar flow.

First thing I would do is toss the screens (28) and replace them with small plastic drinking straws (~5 mm) and about 50-100 mm in length. The longer the better however, you'll start to pay a penalty in form of air resistance and require more SCFM from the fan. They usually come in a box that you can cut in half while the straws are still in the box.

.

Hi Jack.
If you look at the 1st picture in 2099#, at the left hand end of the "tunnel", you'll see the ends of the multicolored straws that I used in my version of Etkin's idea. Three boxes of straws neatly filled the cross-section, rather than trying to cut them without crimping the ends. There are also descriptions and drawing of my previous versions in earlier postings. (chair)

Wind tunnel design was in the back of my mind when I was thinking about the cheapest way of doing it. The wind source is a couple of ex-photocopier fans I have which just happen to rotate in opposite directions while driving the air in the same direction. Thus neatly giving me a more efficient source, with lower losses.
Good point about the corners. If I go back to this one (I've resisted the temptation to take it to bits), I'll do that.
My latest idea is look at the possibilities of dropping a small sample of mixed sizes of sand between two glass sheets, with a narrow gap between them.
I can see that there will be collisions between the particles and the glass, but a crosswise laminar flow of air might allow the separation I'm after.
As this is only to provide size distribution of a very small sample, I can afford to use a very slow feed to minimize particle interference.

If that doesn't work, the next idea is to try vibrating the v.small sample, about 1 gram, down between two microscope slides held to form a tapering cell. Then use a film scanner to do a particle count , or not.

Then, a single jet of air at a low angle onto a single sheet of glass inclined at an angle. The air will stick to the surface, and the sand falling onto the air flow should be deflected across the glass in an arc as it falls, giving a size distribution.

And last, for the moment, dropping a stream of sand into the junction of two air jets which are meeting at a slight angle. The two jets should coalesce, trapping the sand into the middle, fastest stream, and take a narrow path before falling into a series of collectors.

Regards
John

[jhudler]

LOL! I thought that was some sort of morie or optical illusion!

[romihs]

Cameron,

What is your opinion of silica fume in E/G (E/Q) compositions?

Has anyone experimented with it in the E/Q yet?

Regards

Sandi

[walter]

Bruno,

Thank you for your comments. Anything you can contribute on the topic will be appreciated. IE, what would be the spindle requirements for such machine?


Dfro,

Zeeospheres are responsible for dark color. Formula calls for G200 and G800- I used G800 and G850 (pictured below). G200 will give you much better fill.
I did not use vacuum chamber, but you definitely should. There's still ton of air in the mix; besides, we desperately need someone to vacuum test it. Would you be able to do a sample?


Sandi,

The formula calls for G200 zeeospheres, not G850. Slight change in color and much better fill. Higher price, too. Recipe quick links are in this post http://www.cnczone.com/forums/showpo...postcount=2062
Silica fume- I did some tests with M5 Cabosil thickening silica fume- not a good choice, because of "thickening" part; aka ketchup effect, aka thixotropy. It did however, add 10% in terms of strength. Beats me..
O and by the way, have you ever seen a 10lb bag that's the size of a minivan? LOL. Apparently, 98% of that stuff is air..
(So why am I paying $90 for it?)
_

[jhudler]

I did not use vacuum chamber, but you definitely should. There's still ton of air in the mix; besides, we desperately need someone to vacuum test it.

Walter,
After seeing your sample and grinding off the outside layer, I'm going to reverse myself and recommend using vacuum. That is unless we can get higher frequency vibrations for a longer period.

E/G grinds/sands very well with the exception of the large aggregate "#6 Agsco Brown Aluminum Oxide". You'll need to remove this to prevent gouging the surface grinder if you want to use this option.

The exposed sanded sides do absorb moisture however, and I don't have a sensitive enough scale to qualitatively determine the amount. It's not very deep as it dries quickly under compressed air. Oil is another matter, once in, it's there! But again, it's not deep.
Note: the unsanded side does not allow water to penetrate; it beads up.

From this; if you plan to surface grind E/G then use small aggregate, and reseal with a very thin seal coat. Note: USE A RESPIRATOR!!!!!

Density is 0.08175 lb/in³ or 2.252 g/cc which is less dense than Aluminum at 2.7 g/cc.

Air entrapment:

Size of bubbles in one square inch:

>.03 2
>.025 5
>.016 16
>.008 108
<.008 ~140

Note: the top part had a much larger set of bubbles, which was to be expected.

Cameron: Can you extrapolate the volume gas sphere's per cu/in?

I'm on to CTE test next. I need to finish a simple thermal chamber. I'll also test to see if moisture causes any swelling... which I doubt but it's an easy enough test.

Here are some pictures after and initial sanding on surface plate using wet dry sandpaper 180/220/350/600/1500

Warning these are large 1200 DPI greyscale scans.
http://www.hudler.org/pub/eg/Face%20after%20sanding.png
http://www.hudler.org/pub/eg/Edge%20after%20sanding.png
http://www.hudler.org/pub/eg/Face%201%20Inch%20sq.png (air entrapment count)

Again Walter thank you for the sample!

Jack