We all like shiny stuff, don't we?

Thanks for the reply, i get it now. Though, precisely what is abrasive sand? Sand used for air blasting or?

Very cool, I like it.

Joe

Thanks...


uhhh, it's a white bucket.


Thanks. I forgot to mention that BucketHorse can be used to mix the concrete for the base, so you might want to build it first.


"Blasting sand", widely replaced with "slag" these days since silicosis has become more understood. (Don't ever breathe blasting sand dust...)


Thanks Joe. Today I'll experiment with pumice stone, gravel of several types, and mixes of sand and stone. I'm looking for a fast combination that still leaves a nice finish.

Is the bucket impervious to the abrasive sand, or does the bucket fill with white powder? You know, from the bucket being sanded away...

Ah, the great thing about softer materials is that they are relatively immune to wear in the presence of abrasives. An example; tonight while experimenting with the tumbler abrasives, I added an alum test piece with a piece of duct tape on one side. You may think the duct tape goes quickly, but it doesn't...the rest of the piece "finishes" with the duct tape largely intact. This is easily understood by the "bounce" the softer material has.

That said; if the bucket wears out, it's a bucket.

Tumbling: Ran several experiments today with a variety of materials for abrasives. I'm looking for a combination of materials that can cut quick but leave a good finish. I don't want my tumblers tied up for the industry-standard 24-72 hours. I am leaning heavily toward ceramic and slag. "Bead size" plays an important role, as I have learned. If the bead or "drop" becomes too large, the piece will become marred. My goal is to cut the finishing process to 4 hours total, 1/2 the work day. It is an optimistic goal, yet I feel I've nearly accomplished the first two hour phase after running 6 separate experiments today. If what I intuit is correct, tomorrow will see the completion of the rough phase tumbler, and the near completion of the finishing tumbler.

Some starting materials including alum oxide, foamed ceramic, pumice etc...:

Click to view attachment

After two hours the pieces come out de-burred, yet needing bead polishing. The piece on the right is a commercially prepared piece. I think I can get that in 2 more hours of prep...

Click to view attachment

In order to finish the pieces, I plan to vibrate them in powdered glass. To do so, I have to make a "vibratory tumbler", which is a simple (yet expensive non-DIY) device that can vibrate the work in abrasive at high frequencies, leaving a very smooth finish. You can see the advantage over bead-blasting, since the unit is self contained and requires little notice. I don't see why these machines cost so much, so I began building one. I have lots of surplus motors and such just laying around, so the combination of this PMDC motor and a piece of RoGR scrap was waiting to happen:

Click to view attachment

About an hour's work later, I had the circle drilled off center for a vibratory weight, and an 8mm heavy duty set screw embedded in the piece to clamp onto the motor's shaft-flat. Here I am using my mini-lathe to make easy work of drilling the set screw hole:

Click to view attachment

4 machine ops are need to finish the circular weight: drill for shaft, drill for set screw (8mm), drill for screw head, tap set screw hole. The finished piece:

Click to view attachment

The weighted motor is simply mounted to a sheet metal alum platform about 18" diameter. The weight hangs below and is off-center with the motor axis, providing the vibration. ("Vibrators" work the same way, albeit mercifully are much less powerful! ) Upon the platform is mounted the "tub", and the platform is mounted to medium-stiff springs (shown below with the weight, where you can clearly see the weight's offset):

Click to view attachment

I got a video of me holding the weight whilst Robin hooks up the leads. The weight is too fast for my cam, so not much action (unless you are holding it!):

Hmmm interesting...

I grabbed a few other pics of interest from the "lab" tonight. Every executive needs a desk, here is mine:

Click to view attachment

A magnetized aluminum "chip flower" formed during deep milling of 2" alum block (magnetized alum is freaky):

Click to view attachment

Loving what I'm seeing BC, I have been wanting to build a CNC router for a long time. As I have limited space and time. will there be a full kit available at some stage and would it make it to Australia?

Greetings to Oz Brett. Yep, a whole kit is soon available, and like always, I will export my enabling technologies everywhere I am allowed.

FWIW, RoGR is much more than a "CNC router" (though it does that quite well too). I challenge all of you to rethink the tools that can be used here; from "Sharpies" to cake icing, you can see the potential...

yep point taken, I do think out side the box, and can see when i have one that it will let me explore new things, I did work on a CNC Trumph punch press for a few years and the things that you could do with that was cool enough. I would just like to get started. as there are many things that I would like to make and see this as a way of doing heaps of things very accurately...... Cant wait,

WTF? Magnetized aluminum?

I was thinking it would be cool to stick a marker in the machine to label CDs (if you're in a band or something). The CDs could be clearly labeled with a marker, but still accurate and uniform. You could have the machine "sign" each CD.
Better yet, use it to forge celebrities' signatures and sell them on eBay.
...
Think of all the stuff you could counterfeit!

I see you're using air cooling, how are your bits holding up?

Two possible explanations: The aluminium is contaminated ferrous metals. Or this is a result of static charge. If you have ever used an electric plane on MDF, the fine dust becomes statically charged and lines up with the magnetic field of the motor housing.

DJ

EDIT:
And a third possible. The swarf just so happened to mechanically line up in that pattern.

Yep, answer #2 is correct. You can even get the shavings to stick to a screwdriver, appearing for all intents to be "magnetized". I've heard talk of it before in machine shops. The phenomenon is truly one of surface charge, not of ferrous magnetism. The surface charge comes from the rubbing of the bit on the metal, allowing the alum to respond to magnetic fields.

Air cooling works great for lots of stuff, though these last pics are using a solid carbide bit, which needs no cooling at all. The coolant line is there whether it is used or not. As far as I know, the carbide never wears out...these bits usually break instead, typically due to resonance. Since they are expensive, it is best to avoid breaking them.

Got 9 hours more work done on RoGR:

The 2" thick by 4" wide X-ways shaft and motor supports are finished. These massive pieces of alum were milled with a trim router, at high precision. The tool marks are from the lack of a finishing cut, since the pieces get tumbled.

Click to view attachment

The back of the block, showing a facing op for the ways support, and the bore through for the motor coupling:

Click to view attachment

A great way to start your day:

Click to view attachment

While RoGR was milling, and in between shipping DIY PJ parts, I had some time to work on the tumblers some more. The rotational tumbler got a smaller stirrer. The smaller stirrer doesn't "drop" the parts as far, which only became critical for the really heavy block.

Click to view attachment

Good old BucketHorse has run for more than 24 hours without a hiccup, I think we have a winner!



In between RoGR jobs, I had Robin cut me this disk for the vibratory tumbler I started yesterday. The disk is 1/16 alum with a "bump" machined for the motor mount. This job took about 5 minutes on a RoGR...once you have a gantry robot, your possibilities expand exponentially...

Click to view attachment

The disk has the motor/weight mounted at the rear with the weight facing down. The tumbler (RuMBLER, for Dk ) simply "rides" on 3 springs, and the elliptical weight provides the good times. A pic from underneath the vibratory tumbler:

Click to view attachment

A (bad) pic from the top:

Click to view attachment

There are two identical tubs. One tub is attached via machine screws and locknuts to the alum base, the other simply sits inside for easy removal. Here is a video of the first try:



Ok, my eyes are closing...last vid for tonight. Watch the pieces move around by osmosis; it is mesmeric. Also note the loving care given by the "helper fish" red pumice stones. They know just where the burrs are it seems. It is fascinating to watch...



Alright one more! This one really shows the crazy "helper fish" red pumice stones, with a seemingly mystical attraction to the burrs.



------

Conclusions:

Since the tests were short (4 hours for 6 different experiments) I can't make any recommendations yet, but I can say the RuMBLER runs smooth and quiet and performs excellently based on what I've seen. The question is why these things start at $600 and run up to thou$ands when the RuMBLER was about $10? Will the RuMBLER die tomorrow? I don't think so...the motor is made for RV water pumps, and is heavy duty, rated at 24v. I believe an AC "sewing machine" motor or fan motor in that range (perhaps 1/8hp) will work well too, plus you get to use a fan speed control for motor speed. If you get a long shaft motor, the shaft can run through an abec 7 bearing to bear the brunt of the weight. Mine doesn't work this way though, it's a $5 surplus motor after all, so I'm going to run it until it dies on its own bearing...I have to know after all!

nice tumblers, very good concept for other to follow.

joe

Thanks Joe. It's been fun experimenting with them.

Amazing ingenuity.....again!

Thanks...If necessity is the mother of invention, then poverty must be the father!

Looking good here is a ghetto macro photo set up when you get all the pieces gleaming and really want to show it off!
http://www.strobist.blogspot.com/2006/07/h...oto-studio.html

Appreciate the link!

I've noticed that some people mount their stepper motors on 4 poles sticking off the block about 3 or 4 inches.

Why do they do this? I've got a couple guesses...
- Maybe this reduces vibrations, both being transfered from the rest of the machine to the motors and from the motors to the machine. Stepper motors vibrate more than normal motors right?
- I don't know what they do, but I've seen weird couplers between the motors and screws of some CNC machines. Does the space afforded by the offset mount allow for easy changing/maintenance of those couplers?

It looks like you're mounting your steppers directly to the aluminum blocks brainchild. As it's clear you and RW Turner (is that name right?) know what you're doing, and it looks like each piece is thought out. I'm curious if the direct mount approach has any advantages/disadvantages that you can help me understand.

you do not need to change or maintain the couplers but yes the space is for coupling the motor shaft to the lead screws. I use Lovejoy couplers with a spider between them.

Joe

I'm using a "slit coupling" like this:



The coupling fits through this bore:

Click to view attachment

The bore allows me to mount the motor directly to the block.

RW Turner is "Robin", Lumenlab's draftsman and my all around lab sidekick. Robin is responsible for most of the shop's CAD/CAM, some of the RoGR design and breaking lots of carbide bits, among other things. Since my name is Grayson and his is Robin, we have a Batman theme going, so I make Robin wear a mask and tight pants at work.

Ahh, but does he cry out "Holy broken-bits!" at all?

Yes, POW...BAM....Holy carbide! $%&#$

Hirudin:
The rigid mounting of stepper motors is so that as much energy as possible is transfered through the lead screw. without rigid mounting, it would be possible to loose steps here and there as the motor torsioned in its mount. The connection seen between the motor and leadscrew is often a flexible shaft coupler- this is so that the leadscrew can be driven at slightly "off" angles (the leadscrew its self is not particularly rigid, and can sag, or in other ways develop a slight bend) but not transfer forces that would lead to in-accuracies and quicker wearing of the parts.

My guess as to why you'd see a motor mounted off from the block is this: The designer wanted to use as much or the leadscrew as possible without having to sacrifice some of its work area for a coupler.

Robin

Edit:
For what it's worth, the steppers do create vibration, but this is nothing compared to the vibrations made by the tool (AKA router).

Ahh... that makes sense
Thanks for the explanation!

What is happening with the cast concrete table? are you foaming the concrete with the aluminum waste?

We're going to pour the table in the next few days. The only admixtures I'd considered were fiber and acrylic fortifier. I'm not sure if the alum chip would weaken the structure...I'll see what I can find out.

While I continued improving the tumblers, Robin finished machining the Z carriage. Here it is shown with the Y-way bearing blocks installed:

Click to view attachment

Click to view attachment

Countersunk, flat bottomed holes for 8mm Alan-head cap screws. This kind of stuff is easy with a RoGR:

Click to view attachment

Click to view attachment

On the tumbler side, I added some supports underneath the vibratory tumbler to better transfer the force of the weight into the tub:

Click to view attachment

A mixture of slag and alum oxide seems best for quick cutting:

Click to view attachment

I've read that alum oxide is actually miniature rubys, perhaps that explains the red color:

Click to view attachment

For the polishing stage, I plan to use powdered glass. Here's a vid showing the polishing, definitely going to need a lid for this stage:

I can see this RoBlock project quickly turning into the devils play ground leading to a sintering CNC machine http://www.luciferfurnaces.com/ Aluminum Oxide makes concrete rise like dough i the concrete should be formed and under pressure finished by steam it makes an air entrained light weight mix but probably too fragile, for a table. However I would be curious if it could help dissipate the harmonics from all those motors whirling in opposite directions.

That Aluminum is begging to be reliquified and cast into something useful.

Good luck project looks great

Indeed, deposition printing>sintering is one of the things I'm really aching to try...to make things like this:

After reading, reading and more reading for the past 24 hours - coming into this fascinating world of CNC a total noob - i have to say very very well done.

I love the ingenuity that comes out of lumenlab - and especially how diversified it is. From projectors to CNC machines to Q's - just awesome. I'm finally starting to understand the terms you guys throw around like frisbees, a lot in part to Joes well-documented 3 year old CNC plog (the MDF one).

Unfortunately i still don't understand everything as thoroughly as i would like. There doesn't seem to be a pace that says, "new to CNC? start here." Maybe there is, and a link or book reference would be great.

I'm definitely interesting in picking one of these kits up though. If there's a waiting list, put me on it. Thanks for everything you guys do.

It was ESPECIALLY cool seeing you guys were in NC! If you ever have a shop tour running, let me know...

Thanks again..

D

i was curious about the concrete table, how will you make it perferctly square/flat/true? Would a concrete table be better or worse than an aluminum table?

Hi Dale, greetings to Cary NC. If you're ever in Asheville, be sure to give us a ring. As for the CNC...you are correct about the info being very cryptic and hard to find. I am writing a tutorial that has a basic but thorough explanation of the process, the tools, and the important factors when building these machines. I am committed to keeping this information free and open to all, and I am very committed to the notion that high-technology belongs in the hands of all who can make use of it.



Several things: The form is made from MDF which is dimensionally very flat and stable. The forms can then be machined on a RoGR to a higher tolerance, but since concrete can not be used as sacrifice (hehe), it is only important to be fairly accurate with the slab. Fairly accurate is still within...say 1/64" across the diagonal. A layer of sacrifice is used atop the crete and is milled flat by the RoGR prior to use. The RoGR I'm building now has tiles that can be replaced as worn, and form an integral part of the vacuum table I plan to build.

I will be anxiously awaiting that... thanks for your contribution.

Any idea when that will be available?

I believe I can have it done by the time the bearings are here; about 4 weeks.

Exxxxxxxxxxcellent...

Amazing I love to see works in progress, I get home check the email to see what has been done now.
Great job

Thanks, it's fun to build...

The worse thing is that I keep changing the design! I have always planned to add 2 more axes and I think I have a design which will make it much easier. It doesn't change the original design though...more like a variant on a theme, but hey, that's why it's "modular robotics".

I would like to request a topic to cover in this tutorial. I have always wondered how the final cut to a piece works. I have watched many videos on YouTube, but no one seems to show the final cut. What I am wondering is whether or not the CNC machine actually cuts the piece free from the starting stock.

For example, if I have a piece of plywood that is 2 feet x 2 feet, and I want to cut a circle with a diameter of 1.75 feet, then I can see the CNC machine making multiple passes to slowly cut the circle out of the starting square plywood. However, when the final pass of the CNC machine is made (final meaning the pass that cuts through the wood entirely), what keeps the circle piece in place? Won't the circle piece want to fly off the table once it is cut free and the router bit is still running? Supposing that the square starting piece was anchored to the cutting surface, once the circle is cut out, the circle isn't anchored to the table, so it will want to bounce around and maybe get nicked by the router bit; which would leave an imperfect edge to the circle. From all the pictures I have seen, the final pieces from a CNC machine look perfect, so how is this achieved? And, if the cutting surface is going to be concrete, how is the starting stock piece held in place.

I hope my question is clear enough and that you can include it in your tutorial.

Thanks,
Pirin

I've wondered about this too Pirin... I think double sided tape is used for smaller pieces, but that seems a little... non-mechanical.

Also, I think sometimes "they" leave tabs between the part and the material so the whole thing stays as one piece until the tabs are cut by hand. Sounds fine for wood, but not for... well, much-of-anything else.

Very good question..

Oh and for Brainchild... 2 more axes? is that functionality youre looking to include with the first release of the robloks? That would be insane. What would rotate? I don't know if there's a standard for each additional axis, thats why i ask - sorry if its a dumb question.

Dale

some people use a vacume hold down, some use screws, some use tape (carpet tape). there are all kinds of systems, you chose the best for you.

I use hold down clamps on my stock material, then use tabs left usually 4 of them, and cut them when machining done, and use a flush router to trim the tabs, very easy, on metal people just grind off the tabs if used. Once again depends on what you cut and what system you want to use.

Joe

5-Axis machining example

http://www.rainnea.com/cnc.htm

Joe

Wow. I'm kinda glad my initial cnc project attempt had to go on the back burner 3 years ago. This looks like it will far surpass my original intentions for a lot less money. I can't wait!