[ckelloug]

Jack,

Not to harp but the biggest problem for E/G in terms of making a vacuum chamber is the fact that E/G's normal failure mode is brittle failure through crack propagation. While it is true that metals can also fail by cracking, they usually fail in a ductile mode when fatigue is not an issue according to my understanding. Also, the exact position of mesh reinforcement in the part will have enormous consequences.

Keep in mind that the flexural strength of E/G is only about 3ksi whereas aluminum is at least 30 depending on the alloy and heat treatment. Also keep in mind that because the material is created by a random process that it doesn't necessarily have perfectly known or consistent material properties. B.W. Staynes in his 1975 paper suggested that an overall safety factor of at least 2 be used for load carrying designs and given the brittle nature of the material my handbooks suggest that 4 is probably the appropriate safety factor.


Nothing is impossible but I'd want to do some pretty rigorous engineering before being within the event horizon of an E/G vacuum chamber.

Sorry I haven't had a chance to do more calculations on fracture properties and thanks to new poster dilbert for the estimate of voids.

Regards all,

Cameron

[jhudler]

Cameron,
Harp taken...
Think I will back off the vacuum chamber for now, and perhaps return to my previous position of no vacuum needed but can't hurt.

Jack

[greybeard]

Jack - any chance of a lead on your use of PShop re my post 2133# ?

Regards
John

[jhudler]

Jack - any chance of a lead on your use of PShop re my post 2133# ?

Well I generally play around with levels sometimes... you know I would have to do it again in order to know. It's just something I do.
What would work best would be to paint the block with white paint, then scan it. Or scan it in color and use red paint, as the field is pretty monochromatic the red would be easily filtered.

Jack

[jhudler]

The Sample is in the CTE test. Right now it's not in a chamber it;s sitting on the bench.

The act of walking up to it adds 3 to 5 nanometers just from the body heat it absorbs!

Jack

[jhudler]

Well it looks like the CTE is 34.16253 µin/in-°F or .00003416253 in/in-°F or 6.138E-05 m/m-K.

For comparison: aluminum is 13.3 µin/in-°F

So we are ~2 1/2 times more expansive than aluminum.

I need to do a few more runs to average them out, but that will have to wait until next week.


Jack

[greybeard]

Well I generally play around with levels sometimes... you know I would have to do it again in order to know. It's just something I do.
What would work best would be to paint the block with white paint, then scan it. Or scan it in color and use red paint, as the field is pretty monochromatic the red would be easily filtered.

Jack

Thanks Jack. By this I guess you are counting by eye.
I had jumped to the conclusion that you had a method/tool in PS that turned all the "holes" into separate objects, and the software told you how many there were !

Thinking aloud here, but I wonder if I then did a "trace" and saved the outlines as a vector, I could then get the number that way ?
By doing a series of "reduce outline" steps on the original scan, I could also get at the size distribution through a set of traces on the series.
Hmmmm. :idea:

Regards
John

[jhudler]

John,

You'd still have a lot of noise to contend with without a contrast medium.

I think we have stumbled upon a test that we hadn't really considered. How does one know thay have the correct process down (i.e., vibratory compaction, vacuum, correct packing density, wetting, etc.)

And anyone with a scanner can do it.

Jack


Jack

[greybeard]

John,

You'd still have a lot of noise to contend with without a contrast medium.

Sorry Jack, I should have explained my original request a little more.
I'm now looking at the idea of using software for a different problem.
My primary interest at the moment is analysing any commercialy available aggregate mix into a particle size distribution (seived <1mm).
With this info, I would know what size particles, and in what quantity, need to be added, in order to acheive the de Larrard mix that Cameron proposes.

Regards
John

[jhudler]

John,
I've seen a device that uses a CCD line scanner (a CCD 2048 x 1 pixels) for material separation however; it only sorted 5 different sizes, and only because there were only 5 sizes in the mixed.

It worked like this:

A stream of material was blown through the aperture of the CCD (don't remember the speed... it was fast though). These were small items on the order of 1mm to 4mm, screws, nuts, bolts, flat washers, and lock washers.

CCD was illuminated quite well, too bright to look at. Think there was some polarizing filters used.
If I remember the CCD was taking a scan every 1/2500 of a second.
This was fed into a fast microprocessor that maintained a water fall type of image that was only 256 lines. There was an algorithm that detected speed and width of the material.
Then perfectly timed puff of air diverted the item from the stream (quite a few air jets in a manifold). Only one item was sorted at a time. Any item that was not diverted was simply recycled back through until all were sorted or it gave up (sometimes there were item or parts than had interlocked and the vibratory bowl wasn’t able to break them up, like screws with nuts, lock washers hooked in flat washers).

You can do this with aggregate; how hard depends on your electronics, programming and mechanical skills.
You could do this a vibratory bowl that took aggregate and vibrated it around to flatten out the aggregate and then drop it through the CCD scanner. At least its slower than blowing it.
The vibratory bowl could even roughly sort the aggregate thus easing the detection.

Sounds like a fun project; only by the time I was finished I could have bought the aggregate!

Jack

Search ebay for: "Vibratory (feeder, bowl, controller)" remove the quotes.

[greybeard]

Good grief, you've started me off in yet another direction.
The story of my life.
But I'll certainly think about the possibilities, so thanks anyway.

regards
John

[BILLFUNK29]

I have not cast a standalone base, however, I have used polymer concrete to fill welded tube machine frames. In some cases I didn't have the time to fill the machine right away. When I later filled the tubes, it made a huge difference in vibration.

I used Philidelphia Resins material. And, yeh it is expensive. Try to design for minimum volume. I sometimes put foam in the center of the tube prior to filling. For example I put a 2" round inside a 4" square tube. Use welding rod stubs to keep it in the center. A PVC pipe works if you want to run wires through. Make sure you are done welding first. (voice of experience)

Bonel makes a cool vibration stress relief gizmo for the welding.

Good luck.

[jhudler]

Bonel makes a cool vibration stress relief gizmo for the welding.

Bill,
Nice to have you aboard!

Can you supply some more information on Bonel?

Thanks,
Jack

[walter]

Yes.. and that Philidelphia Resins material, too.

Thanks in advance.

[jhudler]

Probably Chockfast Red... about 1 1/2 cubic feet cost about $250 if I remember and comes with 4 (40lbs) bags of aggregate about 2 gallons of resin and some hardener.

Once used this to fill a column of a mill, and saw it used to mount a huge deburring machine (vibrating stones). Not much in the way of tensile strength, but compression was 2 million or something like that.

Here is a link; probably find more info here. http://www.chockfastgrout.com/itw_techmsdsCF.html

Jack

[walter]

Thanks!

I visited ITW Philadelphia Resins page when reading on Zanite E/G composite, they seem to be using it in their 4ksi castings. It's probably not the chockfast foundation system though, it's got silly name and anything "foundation" related means lots of shrinkage. There has to be another "highly specialized engineered epoxy system" in there somewhere..

[speed33317]

I wonder if this post would qualify for the longest lasting forum post?

Can't believe its been going on 3/4 of a year....

Though it seems to be slowing down a bit.

Anybody else build anyhting yet? Besides Walter.

[jhudler]

I'm in collection mode at the moment.

Collecting this bit of aggregate and that... scored about 10 lbs each of titanium diboride and zirconium oxide. Curious about there high density and non uniform shapes and how they'll work in an epoxy matrix.

Working on Tungsten Carbide next (density 15 to 18 g/cm3!!!! Bloody ell).

Jack

[jhudler]

Oh and looking at different epoxies with low CTE... but that property doesn't seem to coexist with viscosity!

[lgalla]

Speed,
Just building dreams for the moment.
Slowed down for the moment.All this good info takes time to absorb.
Walter will probably do some more amazing casts and "Light the fire again"
Larry