Installing the linear actuators

I now have a machine that moves easily by hand... but that is not enough to make a machine.. the next step will be to install the linear actuators. These will allow the axis to move under software control.

I can't find ACME thread anywhere in the country, and at nearly 1000 bucks per meter from South Africa this is way more than I'm willing to pay.... On a previous machine I used 10 mm threaded rod as the actuator, the distance between threads was 1,5 mm which do give great resolution sure, but it's slow as hell... Normally machines using ACME thread has about 10 threads per inch, that is a good mixture between speed and resolution. To compensate for the fact that I cannot get the ACME thread here I will use 20 mm threaded rod which have a thread pitch of 2,5mm that would give me 10 threads in 25mm, very close to 10 TPI on the ACME thread.

The X axis :

I installed 2 bearing holders cut from MDF on the small machine, I couldn't cast these this weekend since it was raining the whole time.. I'll replace the MDF bearing holders later with Aluminium ones. It's quite simple I locked the rod between the 2 bearing holders and welded up some mounting brackets for the stepper motor. the motor is then fixed to the rod using a pice of rubber tubing. It will cause some backlash, but this is a temporary solution till I can find a proper connector.

Here is the x axis with both bearing blocks and stepper motor attached.


one side of the bearing block with the gantry cross member attached to the threaded rod.













Other side of the rod with the stepper motor attached using rubber hose and hose clamps.













y axis using two of the cast aluminium bearing blocks.



with stepper motor and rubber hose attached.







Installed on the gantry.. I removed the stepper for this, didn't want to damage it.



Notice the nut in the center that will be fixed to the carriage to move the Y axis.

Okay there you go that's 2 of the linear actuators installed.. I'm still planning on how to do the Z axis.. will keep you updated.

Buildup of the Big CNC machine.

Right since I cast some parts for my big CNC machine this weekend it's only fair I make a post about the buildup as well I suppose.

01 : The Frame

The frame is just a few bits of angle iron welded up to make a nice sturdy frame. I welded to pieces of square tubing to the frame which will be the X-Axis rails.






This picture already has the whole gantry assembly attached.. I didn't take any pictures of welding up the frame since that was around 3 years ago.

Just note the angle iron frame and the square tubing rails.


02 : The X-axis bearing runners.
This is all still very experimental, as in I hope it works... Well here is what I did.

First I took a 200mm piece of stock square bar and drilled the holes for the bearings to attach to.
These holes where then tapped and the bearings attached using a lock nut as a spacer between the bar and the bearing.









Hope you get the idea of what's going on here.


This will make for a very sturdy 90 degree rolling surface with the square tube rails I have on the frame.

Here is the runner on a angle iron to show how it will run on the rail.









Okay I didn't take any pictures of how I fixed that setup to the frame so I hope with these you'll be able to see what I did.















So there you have it my take on a linear bearing .. homemade. I hope it's as precise as It feels when I move it by hand.


The first picture in this post shows the complete x axis linear bearing system with the gantry attached.

03 : The Y and Z axis runner (linear bearing)

The Y and Z axis bearings are tied together and is just the same as the X axis bearings except that they both run on a single piece of square tubing instead of 2 like the x axis. there is a total of 8 bearings for each of these axis so my calculations say that they will hold up under the stress since the load are quite evenly spread.

Here is how I made the linear bearings.




The Y axis running on 40 mm square tube rail.

















The Z axis running on 40 mm square tube rail.













Here is a picture of the whole bearing assembly for both the Y and Z axis.













04 : The Complete deal

Here is the complete linear moving stage of the gantry router so far.


Everything runs smoothly there is no noticeable racking. I will have to reinforce the gantry uprights though. they are causing a slight wobble when the axis move and come to a stop. the inertia in the z axis still wants to keep going. That will be sorted out at a later stage.













well that is what I have so far... in this post we've completed the linear movement.. in the next post we'll start with the linear actuation.

keep checking in....

First real usable parts

OK, here we go again...

After last weekend's failures I realized the biggest problem I'm facing is the runners, there is a whole section of science dedicated to that topic. So in my quest to see what works and what doesn't I made some simple brackets to be used on the big CNC machine I'm building. Difference is; this time I'm cutting the parts out with the runner already attached, then once the casting is done I can simply cut them off. This has 2 advantages.. one I don't have to glue odd pieces onto the casting, since it's styrofoam no glue with any sort of solvent works, so the glues that do work isn't really all that good. 2nd I don't have to cut the damn things out after the part is cut out.. I hate styrofoam!!! it gets everywhere and on everything.

so this is the parts after I cut them out. I only need 2 of the top ones and 1 of the bottom ones, but hey at least that way I can try a few times even if I do have failures.















This is the parts after I dipped them in the plaster... now the wait for them to dry out.















Well about 24 hours later it was time to start up that furnace again. I'm getting to a point now where I can start pouring in around 20 minutes...

Crucible with molten aluminium.. From here it gets poured into the foamy which is completly covered in sand.











Damn I love these pictures showing the aluminium flowing out of the crucible... I'm still kinda amazed that with a pile of junk I'm turning a once properly messed up rim into things that will go onto my cnc mill.


The pour...












And then again that agonizing wait to see if the part came out successfully... being too hasty will warp the part since the aluminium has not set yet, and being over cautious will end in a heart attack... coz really what the hell is going on in that sand!!!!


I pour the sand into a stainless steel trough that I have around, that way new sand don't have to be sifted every time.










The moment of truth!!!!


Happy days!!! It seems to have worked!!













And voila!! here is the two different parts I cast, this is after a rinsing under the tap to get rid of the plaster.


This one will be used on the Z axis of the mill.

















I cast 2 of these they will by used on the Y axis.















During this week I will start praparing the parts I cast to be put on the machine will keep you guys posted.

See you all later.

Steep learning curve

Well as promised I have some updates today...

Lets start at the beginning of my casting process. Remember those lathe chucks I cut out of foam in a previous update well I dunked those into a very watery mixture of "polyfilla" It's used to repair cracks in brick walls. I don't know what it would be called where you're from, but it seems to me to be nothing more than plaster of paris.

Oh and don't forget to add a runner as well that is the "tunnel" into which the aluminium will be poured then run into the mold.


This is the "foamy" covered in polyfilla.
















Here is the backside of the foamy, you can see the runner sticking out, it's just a piece of styrofoam cut and then glued onto the base. this was done before covering it, so inside that hard crust will be a foam part exactly the shape I want the aluminium to be.











Next I covered the whole part in sand with just the top of the runner sticking out.. I also welded four "feet" to a piece of 120mm pipe, the feet is just to increase the surface area so the pipe cannot fall through the sand into the mold when the styrofoam has been burned out. This is where I will pour the aluminium into, It should stop any spilling from occuring, that will destroy the plastic bucket I'm using. Another benefit is that it will give a bit of fluid pressure to force the aluminium into all the little nooks and crannies I want it to go.



plaster foamy covered in sand with pouring spout

If at this point you're wondering why everything I make seems to be rusted, well the answer is simple .. it's scrap metal... I have a pile of it in my backyard and I use scrap exclusively as a construction material... You simply can NOT beat the price.





The Money Shot!
Okay the next step is one I've shown a few times before, start the furnace and melt some aluminium!! I poured the molten aluminium right into the funnel I made and I was amazed at how little flames the burning styrofoam actually makes.


This is a very exhilirating experience for one you're carrying a very hot steel bucket filled with molten aluminium around, and the anticipation of what might be if this comes out good.. A very odd sensation indeed.



See at least I got some gloves and a leather apron now.





......The wait!
This is an agonising 20 minutes... after pouring you have to wait for everything to settle before messing with it... this 20 minutes felt like an eternity. then finally I figure it's about time I see what is going on in that sand.. I dumped the bucket out to reveal the part.


I saw a heavy lump in the sand and hey it looks like the foam I put in there ... it's just a LOT heavier now... It's still very very hot, but the aluminium has definately set now. The plaster leaves a bit of a crust around it but it comes off really easy with some water.








.... the let down!!!
Okay off to to the tap for rinsing off all the plaster.. and the I see it ... it was a failure :( ...After all this effort and anticipation I have just created a very interestingly shaped ingot to be melted down later.


It looks like it came out good ... just the runner that needs to be cut off...















But NO... the aluminium made it through the cracks in the plasterand since the sand didn't fill those cavaties completely it messed up the casting.







*Look at the detail one can clearly see the beads of the styrofoam on the casting.. In the top picture you can see the cut levels from the cnc process.


So hence the title of this post... The learning process..
I should have used 4 runners with the slots pointing upwards. Well it was a great learning exercise...

Oh and since you may think this is the one and only failure ... you're wrong! Before I welded the feet onto the pouring funnel the first attempt was a complete failure the funnel fell through the sand and into the casting cavity, I was also too hasty to pull it out of the sand and I pulled the whole runner out and messed it up good... at least I knew it was a failure before since the funnel fell through.


A really good mess I made with this one..


But hey it takes 8 minutes to cut another one from foam ... so well worth the time.









So apart from these 2 castings most of my time was used cutting down that rim and making ingots.

Got some new aluminium

Well It's been almost a full week with no progress. I was looking for some good metal to use in the melt, and finally a good Samaritan helped me out with an aluminium rim. I cut out 2 pieces to test if it's actually aluminium or again some weird concoction.

















I started up the Furnace and after some adjusting it ran beautifully.


If you look carefully you'll see the chunks of rim in the crucible.












5 minutes later.......


I turned down the flames and voila! molten aluminium.














The Aluminium in the Ingot mold right after the pour. I should have used more metal to fill up the whole thing, but the idea was just to test whether it actually was usable aluminium in that rim.










...Nice and shiny




After about 20 minutes cooldown. I now have a piece that will be perfect to remelt later ...
















I really want to start casting something useful this weekend, Hopefully on Monday there will be some big updates again. The worst hold up has been the rain. It is very very dangerous to do this when it's wet outside.. Steam is around 1600 times "bigger" than the water used to make it. so one cubic millimeter of water will be 1.6 cubic meters when it flash boils .. that difference makes it possible for concrete to explode!!! ... Yeah EXPLODE!!! ... so taking care is very important...

The CNC process

Well Since I don't have much to post after this weekend disaster, I'll just take you through all the steps involved in CNCing (the machine that cuts out the foam blanks)

1st

Design your part in 3d... this is the lathe chuck designed in Rhino3d



















2nd
Look at the part and visualize how the steps will be performed on the machine.
Then it's on to generating the "toolpaths" this is how the physical cutter will move through the material.. It is different for nearly every CAM (Computer Aided Manufacturing) software Package there is... For this Chuck design I've used the add-on to Rhino3D called RhinoCAM.

This the toolpaths calculated for the face of the chuck.

This one will cut the outline of the chuck ...



Note the Blue lines is where the centre of the 3mm flat cutter will move. This package automatically calculates the tool cutter offset.. That means it will move half the diameter of the cutter away from the part and the resultant cut will be exactly to the measurement of the design.







This cut will "pocket" out the centre cross slide grooves.















3rd
Next thing will be to translate this tool paths into a language that the machine interpreter will understand.. The language is called G-Code, owing to the fact that all operations start with a G-code word..

This translation is called post processing.. so the blue cutting lines in the software package will look like this when processed..

N1 T2 M6
N2 S4851 M03
N3 G00 X-6. Y83.5
N4 Z5. D0
N5 G01 Z-10. F 500.
N6 Y77.5 F 800.
N7 X-5.8496 Y76.1649
N8 X-5.4058 Y74.8967
N9 X-4.691 Y73.7591
N10 X-3.7409 Y72.809
N11 X-2.6033 Y72.0942
N12 X-1.3351 Y71.6504
N13 X0. Y71.5
N14 X4.1574 Y71.379
N15 X8.3006 Y71.0165
N16 X12.4158 Y70.4138
N17 X16.489 Y69.5727
N18 X20.5064 Y68.4963
N19 X24.4544 Y67.188
N20 X28.3197 Y65.6525
N21 X32.0891 Y63.8947
N22 X35.75 Y61.9208
N23 X39.2899 Y59.7374
N24 X42.6968 Y57.3518
N25 X45.9593 Y54.7722
N26 X49.0663 Y52.0072
N27 X52.0072 Y49.0663
N28 X54.7722 Y45.9593
N29 X57.3518 Y42.6968
N30 X59.7374 Y39.2899
N31 X61.9208 Y35.75
N32 X63.8947 Y32.0891
N33 X65.6525 Y28.3197
N34 X67.188 Y24.4544
N35 X68.4963 Y20.5064
N36 X69.5727 Y16.489
N37 X70.4138 Y12.4158
N38 X71.0165 Y8.3006
N39 X71.379 Y4.1574
N40 X71.5 Y0.
N41 X71.379 Y-4.1574

NB. This is just a very small portion of the file, the actual file is about 100 kB of pure text... it's a LOT of lines.

If you look at the code you'll see it's nothing but a list of Cartesian coordinates.. A cnc machine is actually nothing but a Cartesian robot with a cutter attached. The same principles are used for robotic assembly lines, pick and place robots that assemble computer motherboards etc. etc. etc.

4th
Load the G-code file in the machine interpreter software, I use an old DOS based program called TurboCNC.. it looks like this with the file loaded.











Then it's time to load the material on the machine. Make sure that it's positioned correctly and then "jog" the machine into the correct position to start. This is usually coordinates X,Y,Z(0;0;0)

This software will send electronic pulses via the printer port of the pc to the electronic drivers which in turn moves the motors to the correct position.



this is the electronic drivers made... The motors used is also not ordinary motors that spins continuously these are stepper motors which turns exactly 1.8 degrees every time power is supplied in a specific sequence. The software outputs signals to step the motor and the electronic drivers then set up the correct sequence to step the motor in the right direction.






And the result of all these things coming together is this..





























... And there you go That is a short run down of how to go from a 3D design to a real life model using the easiest of grade 6 trigonometry.

It's not only simple shapes that can be cut out. Here is another project of mine.. a Concept model of a peugeot... It's still a work in progress though.












Hope you enjoyed this .. and maybe it inspires you to start your own Cartesian robot one day...

First serious failure

well as much as I hoped to post a nice shiny lathe chuck in aluminium on today's post I have to admit I have nothing... Absolutely nothing..

I thought I'd use an old Volkswagen beetle engine as the donor scrap for this project.. but it turns out that beetle engines were made from an alloy of magnesium and aluminium. It did make some really great fireworks but nothing usable, so I'll have to get some other scrap to try.

This is the magnesium just oxidizing, bear in mind this in broad day light... the camera tries to auto adjust the levels and that is why it looks dark.




The light created by the magnesium oxidizing is just the whitest brightest light you can imagine. Although it was a massive failure It did entertain.



























The aftermath....

Well after a good 20 minutes the flaming magnesium burned itself out, and the resultant magnesium "rust" was nothing but a pile of white powder... which stuck to the inside of the crucible so that will have to be chipped away... I tried a ingot cast but that came out wrong... though I must say this ingot does have some really cool properties It weighs next to nothing and it looks powdery, but it's quite strong.




Messed up ingot...




















My poor crucible... I hope I get all this out with a chipping hammer.. it's really stuck.