Milling boxes in fixtures

As Koka Nikoladze had the need to machine some holes in some electronic enclosures, we produced some jigs to repetedly mounting and machining of these boxes.

For Bernt Isak's project, The Cosmos, we made a similar jig, also for cutting out por't for the Arduino's interface.

3D milling aluminium and stainless steel

This weekend i held courses in Fusion 360 at Bitraf. We started of with a course aimed for people without any 3D modelling experience on friday, then went trough some of the more advanced functions on saturday, before moving on to the CAM function and 3D printing on sunday.

On saturday i lectured about assembly in fusion, and the difference between bodies and components, and how to structure the file when building an assembly with moving parts. The case for this weekend was this machinist vise, as we then could use the model when we were milling on sunday.

After modelling the vise, we modelled a test case, 60x60x100 with some random topografy for testing out Fusion CAM's adaptive clearing. As we now had a 3D model of the vise, we could now easly activate collision detection, aswell as grabbing the necesary geometry for mounting the vise to our machine.

As this weekends goal was to experiment with 3D milling, we also tried some different finishing strategies. For aluminium, what worked best where a 6mm singleflute ballmill. Above is a rendered image of the vice, with the finishing toolpath projected on to it using gimp.

The machining process leaves a mess, but since i where using adaptive clearing, it's a uniform mess. All the chips are uniform, since the tool cut the material with a constant load.

I did several tests with different stepdown and stepover. For each test, i let one side be finished, while the other side still had the surface from the roughing pass.

I also machined a 50x50x50 block of stainless steel, this was the result:

This is the toolpath for the steel block. The model was produced by warping a hexagon along an the center axis. As the material was stainless steel, the stepdown was kept low at 0.2mm, but the feedrate high at 1600 mm/min.

The three samples sitting at my desk after a long weekend. i have now aquired the knowledge needed for producing my first injection mold.

See through wooden sign

 Today, a client told me he wanted signs to guide his customers through his bar. The signs where to be subtle, but still clearly show the direction. They must also be able to blend in with the rest of the interior when not in use.

After a day of experimenting, this was the result, a wooden slate milled so thin that light pass through from the inside, leaving people puzzled when it turns of, only leaving a clean, untouched surface.



 

3D printed cookie cutter

Got a sudden craving for christmas cookies, so i decided to 3D print a cookie cutter to produce the cakes. Christina Lewis designed the cat graphics, which was used for the cutter.I extended the features creating the imprints of the eyes and other facial structures of the cat. This gave me a problem, as the dough needed to be uniform, or else, the markings would not appear correctly. 

I solved this problem by milling two rims for the rolling pin, ensuring uniform thickness of the dough, as well as cutting a perfectly parallel slab of the dough. How convenient.

Progressing with the screwdriver case

Here is the latest progression on the screwdriver set. Now that i had made one functional prototype of the drawer, the time had come to produce five of the new improved versions.

After generating toolpaths, the parts where milled from plywood and aluminium. The plywood parts where to be milled from both sides, so three wooden dowels where used to ensure that the material lined up from both sides.

Then, once i had five aluminium frames, and five wooden trays, i had to begin drilling the holes for the small brass screws that hold the frame in place. As much as i enjoy operating the CNC machines, there are something special about forming material by hand. Therefore, grabbed my toolbox, and began the process of marking, drilling and countersinking the holes.

After ensuring that the screws did infact fit in the aluminium frames, i mounted the frames to the wood. Using one of my favourite tools, a brass hammer i made my self some time ago, i hammered the frames onto the wooden trays.

Then using another of my selfmade tools, the small parallel clamp, i secured the two pieces together while i predrilled the hole for the screws. Then, i mounted the piece on a small board with a welding clamp, and inserted the threee screws.

As i had not used Bitrafs lasercutter for quite a while, i decided to produce a temporary rack to store the drawers in when i am not working on them. After some thinking, i came up with a quick design in Fusion 360. I took roughly half an hour to cut.

Thats it for now, the next thing i will have to do to finish this project, is to lasercut the wooden inserts for the aluminium frames. Then i will need to cut the final wood cabinet in wich the drawers will sit.

Wooden case for wrench

As i have been working hard the last week, i needed to take some time off, and relax my own projects. So, today i made a wooden case for one of my most used tools, the wrench to change the tools in oour CNC machine.

First, i began scetching an concept, in Autodesk Fusion 360.

Then, the cutting commenced:

After Some sanding and oiling, i assembled the case, glued on the velvet, and suddently, i had this sitting on my desk:

This was a fun quick project, all together, it took less than three hours.

Assembling giant vacuum wall

Months ago i agreed on designing and manufacture a 2100x2500 vacuum wall for Norwegian artist pushwagner´s photo studio.

After producing the technical drawings, the contract was signed, and the product taken into production at Fellesverkstedet

Below are som photos from the production of the vacuum wall.

The vacuumwall must be adjustable. Therefore, mechanical valves where added in form of PE slides.

As a consequence of this, the vall became very thick, and several hundred screws had to be used to keep it from imploding from the underpressure.

This is one of the two halves making up the vacuum wall under assembly.

Here, the two halves of the vacuum table stands ready for delivery and mounting at the customers photo studio:

Autodesk Fusion 360

Yesterday, i and Christian Anker from bitraf spent all day produsing parts using Autodesk Fusion 360's CAM module. We produced the toolpaths on my macbook air, then seamlesly beamed them over to the CNC machine using Dropbox.

The more we learn about production technology, the more we can reduce the turnaround time from concept to reality. For the case of the two copper pieces below, that time ended up at 43 minutes.

Before the copper parts where made, prototypes where made in aluminium. The 3D milling where done at 25mm/sek, with a 1/8 inch ballnose mill.

This assembly is the hammer parts of Koka nicoladze´s new instrument; the solenoid drummer.

As we did further adventures into CNC milling with the shopbot, Christian anker decided he wanted to produce the build platform for his huge part for his 3D printer.

As it must be perfectly flat, it had to be facemilled. We decided to do this with this 32 mm milling bit.

This is the finished part rigth after milling:

Since teaching Christian to mill aluminium with the shopbot CNC machine, he has produced a number of parts for his 3D printer:

Holddown

I prefer to use lasercut wood washers to mount the material on the CNC, aswell as keeping the piece from moving under cutting.

This part were cut from 5mm polyethylene, a very soft material. Therefor, it had to be secured well before cutting.

The washer are cut to spec, so that they fit exactly in the slot.

This part needed to be chamfered around the outer perimeter. To allow this, it had to be secured with a number of washers and screws. Itwas not possible to avoid collision with the washers on the thin parts, so these where hold down with wood washers, while the rest of the part where secured with steel washers.

In this example, i was making a table, and the legs where to be chamfered from both sides. To acomplish this, i cut a jig for the part from some scrap wood. As this jig where cut by the machine itself, i can now position the part perfectly in the machine each time.

Testing out 3D milling in Fusion 360 once more

Today i did some more testing of Fusion 360`s CAM module.

I made a sample 100 X 100 X 50 body, then made a  T-spline plane ontop of it. I then gave it some random topography to work with.

After boundary filling , the part is ready for Toolpath generation!

I intended to run the first files in foam, so i ramped up the feedrate to 3000mm/min @ 15000 RPM.

I used 10mm two flute HSS tools from GARR for this job, a ballnose for the topography, then a flat end for cutting out the part.

The results where good! At 2,5 mm stepover, the toolpaths are clearly visible.

I continued to use the same file for testing in wood. This was made using only adaptive clearing with a HSS 1FL6 from norswiss.

Rigid FDM delta printer

Altough we have a few FDM printers at Bitraf, i now see the need for having my own at my office. i started to develop a small scale delta printer last week.

The printer is constructed from a plates milled from sheet of 10mm 6082 t651 aluminium.

Movement is performed by NEMA17 steppermotors, and HIWIN linear bearing rails.

This is what i have modeled so far:

As the ultimakers, this design also equips the captured nut concept, in this case M6 nuts, as seen in this cutaway:

In this cutaway, the belt tenison system can been seen in the lower left corner. Notice the stiffening bracket in the middle of the linear rail assembly, and the milled weigth reducing pockets.

These are the main components of the printer, here laid out, ready for toolpath generation.

I decided to use this project as a testcase for Fusion 360`CAM strategies, in this case adaptive clearing.

The milling bits normaly break When peak loads is achived when running into a corner. Therefore, the feedrate must be reduced for the entirety of the toolpath when using normal CAM strategies for avoiding milling bit failure:

What adaptive clearing does, is to slowly dig the milling bit into the material, before performing a series of deep cuts at maximum depth, with optimal chipload. In this way, the tool will not experience peeks in loads, as the tool will have constand load. This reduces the runtime with a suposed 40%

I will spend the next weeks experimenting further with Fusion CAM. This is my first result with adaptive clearing:

Dust extraction for the shopbot

Today me and and Jens Dyvik assembled the new dust extractor for one of the shopbots.

The parts were previously milled out by Jens, using his CAM plugin for rhino, called barkbeetle.

Barkbetle can stream to our projector, so that we can preview files before we run them.

Seen here is the finished parts, along with the prewiew.

Milling aluminium with

When milling difficult parts, installing them in a jig is the way to go. Normaly i would use valchromat, but we had problems with it cracking when holding small aluminium parts. Therefore, Christian Anker decided to use PP plastic, which is a nicely machinable material, with more hold than valchromat.

The CNC machine produces a smaller CNC

I produced the parts for a friends smaller CNC, on Bitraf's huge CNC. How cool is that?

The cutting has begun!

The baby CNC! The banana is for scale. NOTE: This is an imperial banana, not a metric one!

Here, the brackets is installed. 

This is the machine with the bridge installed, waiting for the spindle.

The tools used for the roughing and finnishing. The bigger 8mm bit has two flutes, and provides exelent chip removal rates, while the smaller singleflute 6mm leaves a great finish, at the cost of efficience.

For more info on the CNC, follow this link.