Community challenge #10: Milling plastics (closed)

Probably silly, but how about surfacing then stacking/glueing several 3x5x1.5" thick blocks? I have no idea whether the glue layers/marks would show after milling. Wait, maybe that’s a…challenge!

I can imagine how a 4"x4"x4" form factor would be really uncommon and hard to find unless placing a custom order (but then the price will not be low)

1 Like

Would round be workable?

a 4" diameter Delrin rod 4" long is $39.03 at McMaster-Carr:

Other plastics are less, but available in differing lengths.

Also check Fastenal and Grainger.

4 Likes

They have the 4"dia HDPE by the foot for $US37.04/ft, as well. https://www.mcmaster.com/hdpe/moisture-resistant-polyethylene-hdpe-rods-and-discs/.

You can get 4" thick sheet from them, too. At these sizes, expect issues with material movement due to trapped strains.

Note that the price goes up out of proportion to weight for most larger sizes, except for a few special ones.

3 Likes

If you’re fortunate enough to have a plastics supplier nearby, they almost always have a scrap bin for odd sized leftovers and cutoffs that are attractively priced in my experience.

4 Likes

https://www.tapplastics.com/product/plastics/cut_to_size_plastic

My go-to for HPDE.

3 Likes

Submission (old project, but why not get the ball rolling) part 1:

Description: Several years ago, I CNC’d my Atlas lathe, just because. This was a very large project, as I rolled my own controller. I dabbled at it for several years, so by the time I heard of GRBL, or got my Nomad, I was better than half way.

Last year, I made a few changes based on my experience, and two of those are presented here:

The ones I will show here are a mount used for stepper drivers to control cooling (Big Easy Driver replaced the original for the leadscrew) and a decent mount for the leadscrew stepper to replace the cranked up sheet metal mount I threw together for development, and left in place since it worked. Noisy like crazy, though, and limited quality of work on the machine. No damping, poor stiffness, and lots of surface area is all bad choices for for mounting steppers.

The main panel has the operator controls for manual use (feed and threading) and holds the electronics.

This is the interior when I was part way through refitting last year. There are a number of parts visible that came out of the Nomad, including a couple circuit boards (the rotary quad encoder control board is not visible in the housing) and most of the parts for the rotary encoder (rough cast on a $US200 3D printer, then machined on the Nomad to tolerance).

First, the airflow guide:

You can see the fan on the left, underneath the leads to the terminal block.

HDPE, milled on the Nomad. The CAD design shows what it looks like better:

Note the form of the channel. The driver is cooled by the copper mass and vias on the circuit board, so the goal was to maintain good flow over the board surface. There are a couple copper pins I added that reach out into the airflow under there. This took the operating temp from at the limit to about 20C above ambient running full current.

The work was done in Inventor, though Fusion360 will do everything I did here, but slower. The machining steps:

An adaptive clearing for the majority of the material removal with a 1/8" two flute, 10KRPM, 1000mm/min feed



A ball end was used for the fancy surfaces. Morphed for the slope, which makes any scalloping be in line with the airflow, followed by Scallop for the more vertical sides and fillet blend:


Then the drilling. The drill operation here was to locate the holes, and they were finished on another machine for practical reasons. A Nomad is not a drill press. The torque isn’t really there for deep drilling, so if it can be finished on a more suitable machine, why torture the poor beast?


(Part two with the stepper mount to follow)

4 Likes

The stepper mount is a different animal. For a number of reasons, I wanted to make it easy to disengage the leadscrew drive mechanically. I did this using a face-pinned coupling (think a finger coupling) so the stepper could be slid off easily. The mount facilitates this.

The upper portion is slotted and can slide axially on the lower riser. The upper part (the mount) is 6.3mm acrylic cemented together and the riser is HDPE.

The mount was modeled as several parts but was cut as a single assembly:

Machining it started by drilling the holes that would be threaded for the stepper mount screws.

followed by windowing the parts

1/8" two flute square end tool, 10KRPM, 1200mm/min. Really should’ve gone a bit faster, but I used a vacuum to flow the chips out so it was fine. The tool here was TiCN coated. Nice and slippery.

Then the majority of the isolation cutting was done with an adaptive. Slotting is annoying in acrylic. Use a controlled engagement strategy if you don’t want to destroy tools and work.

This was done with 0.5mm axial leave, and 0.1mm radial. The axial holds the parts together, and the radial allows for a nice final finish pass. Glue-up is easier if the surface has a clean finish. THe adaptive strategy leaves a lot of entry points as it gnaws away.

Then the separation cut, with tabs to control breakout. The material was taped down to avoid the cuts, so there was less hold that desirable. Doing it this way keeps good support until the end, where these are very light cuts.

This was going to be the test part, followed by a final one with alignment tabs for assembly correcting any errors, but it turned out fine. I had to cut back the tops of the gussets a little due to interference with the stepper, but that was it. Yes, the stepper is up in a flexible region, but there is no load or vibration out of the plane of the sheet.

The riser block worked out really well:

Three guides for the slots, drilled for the hold-down screws, and counterbored holes to mount to the bench.The height here is critical, as it provides one dimension of axial alignment. This got fixture mounted to a bed-of-holes plate, and rerun to get the height dead on.

First the perimeter and top were roughed:


Then the drilling started. All holes were piloted with 0.85mm drill first. Fast peck cycle to get the chips out.

Then a larger bit to clear room for the tool to do the counter bore. Not truly needed, but makes chip control a lot easier. Note that the holes to this point only go part way through. The bits are short and small, so no desire to push it

The counterbores

There clear drill the rest of the way. The bits are long enough to make it the rest of the way and there is room for the shank


and bore out to size for the mounting bolts

The three holes for the hold-down screws on top will be tapped

Last is the finish sizing of the locating nubs and the counterbores.

I don’t have any shots of machining these, unfortunately.

The Inventor models, for anyone interested: LATHE-parts.zip (1.6 MB)

7 Likes

Epic write up, thank you !

1 Like

So I will start this contest entry with the following quote and i think it will set the scene for my entry.

"A person who never made a mistake has never tried something new" - Albert Einstein

If this is not a valid entry happy to have it moved :slight_smile: and will take no offence as it is not complete. However i wanted to show my sons, that even in failure we can still achieve allot !! and that our failures need to be celebrated as equally as my successes.

My design was simple in my mind at least, i wanted to make something for my office that was useful, so i liked the idea of lit LED Plexi signs, and ended up on a small Calendar with removable months and days that i could swap out.

These would then sit in a wooden base that has a hollow bottom for the electronics and slots for the LED light to shine up and illuminate the tombs.

These are the two tool paths to hollow out and then to chamfer both the top and bottom to give the base a nice uniform (almost oval) appearance

so far so good!

Now things go badly wrong i purchased 2 x Plexiglas (one sheet as backup) 3mm thick. Worked with the feeds and speeks and looked at at Winston’s C3D video on suggested cutting speeds and came up with

Started the cuts and things looked really good ( i must admit i gave a little fist pump)

as you can see bottom left is nice and clean, top left we have a problem, this boils down to the cut building up heat i believe.

then after a few more seconds of running started to to encounter molten issues


and molten issues on my bit

tried new Feeds and speeds after some research, in an attempt to clean up the cuts but this only ended in further disaster lol


followed by “OH SNAP!”

So i got a new bit and second sheet but things just turned out the same again - Heat is defo my problem but i cant figure it out, below shows the cuts i am getting. Though interestingly the contour cuts are amazingly clear, and looked good whilst cutting, but towards the end of the cutting job it was starting to get a buildup on the bit again!! argh!!!


Close up of the edge - really happy with that for sure, but same F&S as the rest so confused why different results

Another issue was cutting integrity something i have not really encountered, by this i mean i obviously did not leave a big enough gap between the tombs to allow the sheet to retain rigidity

It started to pull up the pieces when cutting, i used a bit of scrap wood to hold it down and shift while cutting. (really i should have stopped but was worried that the Plexiglas might shatter) .

So i took some of the best pieces and tried a simple test to see if my design principles at least would work and to be honest i was very happy with the results - if not the finish. the best was number 3!

so in short - i am glad i tried this and stuck with it over the two days, even tough i really struggled with this process. I think if i could get more stock in time i would try again, but will miss the deadline at present, and therefore i humbly submit my entry, even though like my wider journey in CNCing, it is not complete.

Again and for clarity will not take any offence if this is not permitted, i would though ask if it could be moved over to main shapeoko thread.

Thanks and thanks for this challenge it has for me at least pushed me into an area i would not even consider!

Jon

UPDATE 21ST OF JUNE

Used some of my wastage as suggested by @enl_public, I think I have found the spot, it just feels the feeds and speeds are way to fast but below shows second try on left original on right

so try a few more things today.

Even a poor plan executed with ferocity can provide results - my new motto

9 Likes

I was worried I would be the only entry.

This looks like a really good start. Feeds/speeds for acrylic can be a bit squirrely to find. Plan on test material. A summary is: aggressive enough to get the chips out and not heat the tool too much. Engraving is tough because you have no choice but bury the tool, but there are a few threads dealing specifically with engraving acrylic.

Curiosity- feel free to say none of my business- Why can you not get more stock?

4 Likes

I’ll add that:

  • when things do not work out in plastics, feed and plunge faster.
  • single flute endmills are your best friends in plastics
    In your case I would try and plunge twice as fast.
    Thanks a lot @Sherpa, this is exactly the spirit of those contests: learning together

EDIT: to make you feel good, here’s one from my first cuts in plastics:

To elaborate on what I was saying above, I found the hard way that you can mess up a cut in plastics even with good feedrate/RPM if you stick to the usual (very conservative) guidelines of slow plunge rates. And then you end up melting plastic during that first second when the tool enters the material, and after that all bets are off.

Best case scenario, you just get some plastic strings wrapped around the shaft:


which is usually no big deal (unless your job has a lot of retract/plunges, and these strings start to accumulate), and worst case scenario, well c.f. first pic.

Single-flute endmills are cool because they allow for much better chip evacuation, hence fewer chances to have chips accumulating, being recut, getting pissed off at not being able to enforce social distancing, and then melting out of pure rage.

7 Likes

Is there a way one can tell something about the quality of plexi?
I read on another forum (dont know where anymore) that high quality plexi has a better heat resistance whereas low quality (from the hardware store around the corner) tends to melt quickly.

And thanks for that summary! I’ll be really careful if I try plexi in the future!

Not a problem at all, it is because I live in the back of beyond and have been having problems with deliveries in some cases not turning up, but I have ordered more so hopefully get another chance.

3 Likes

Jon,

If you cannot get more stock to finish this, I’ll be happy to be act as a proxy and cut some acrylic for you based on your project files, to finish that entry! I make the rules, so I say this is ok :slight_smile:

4 Likes

Makes sense. I am somewhat spoiled supply-wise where I am. Until recently, I had pickup at McMaster an hour away, or same day courier, welding supply is 10Km away, and a whole pile of others within 20KM, despite living in a smallish rural town. Right on the line between the sticks and suburbia.

Do keep your lossage for working out settings. It is worth the space. Well worth the space.

3 Likes

Dude this made my day lol I love it​:grin::grin:

2 Likes

Do a search in your area for “usable drops”. We’re lucky and have a Alro Steel showroom nearby. They get pallet loads of cutoffs of HMMW, Delrin, HDPE, Acrylic, etc in from their plastics division and they sell them all for $2.65 per pound.

3 Likes

So, what are your guys up to for the challenge?

I cut a new chipguard for my spindle tonight.

Amana single-flute 1/8", 16000RPM, 54ipm, 1mm depth per pass.

I captured a few videos and posted them in the dust shoe videos thread.

2 Likes

Hello everyone I am pleased to share

“A Calendar Desk Creation (LED Backlit) or “ACDC” Version 2.01234

This is my second entry as in my opinion I have changed so much about this project that it warrants it, as well as using this as the end point to my journey.

Files here contest 10.zip (438.4 KB)
Cutrocket link Here

My previous entry ended in failure, burning / melting, poor letter definition, and well folks the list goes on and on for some time and over 45+ design issues ending with “and nothing to hold the tombs with”.

So armed with new knowledge of what not to do I decided to address the issues of design first.

Version one had a wooden base, but I did this if truth be told, because I was afraid of the plastic! – so that had to stop and I then came up with a simple oval design that will sit flush on my desk and have three pockets cut into it to allow the tombs to go in 1 x tomb for month 2 x tomb for day. These had been set to 3mm Deep as I worked out that would give me the best backwards tilt without them falling out or over.

image

This worked really well and can be seen in the picture below

Next I needed a way to get the light to defuse up to the tombs from the bottom base but without looking rubbish.

To that end, I did some reading and watched a few vids on acrylic diffusion, in short rough / sanded edges present the best colours / light diffusion. So armed with this knowledge I then decided to flip the base over, I measured the LED light strip height and width and this came in at a massive 1.5mm deep by 12 mm wide

Now came the scary bit x 2

1st was I have always (no joke here) ALWAYS started a project with a 90Deg start i.e. always on a square bit of stock, so first challenge was to figure out how to get the correct WCS Coords for my oval cut out, turns out this was easier than I thought, Vcarve has all for corners or centre so I choose centre as below and that took care of that., after marking centre on my stock

Next was to simply create a rectangle that spanned the width along the centre of the three pockets. And I would then cut this to correct width length and depth, the LED strip then would sit snugly in there allowing the base to sit flush on the table.

It would also then give a direct path of defused light through the base to the tombs (and it did) pic later on

image

2nd Scary moment though was the realisation that 3mm was defined for the top three pockets + 1.5 for the LED strip depth = 4.5m my stock is only 5.9ish leaving me with a gap between the top and bottom cut outs of 1.4mm.

Now if my machine is off this is going to go bad! – I never doubted you for a min SO3 (well perhaps a little bit :slight_smile: ) but again believe in the SO3 and she will make it work! You can see this is quite close hard to picture

I kept the step over around 40% as I thought this would help defusing the light too.

image

And from above

Now jumped ahead a bit there but never mind, The biggest problem I had was burning / snapping bits, so again after reading and watching Winston’s videos I diagnosed the issue as not going fast enough.

So I did several tests and ended up with these settings, I am sure I can go faster but after 14 tests these seemed on my machine at least, to provide the balance between quality and speed.

For my 1.5mm end mill

image

For my 3.17mm end mill

This was a massive step forward as you can see below on the left new F&S and the Right Original F&S

So armed with my new recipe off I go to the cutting room and start to bang them out. But wait!! in my design i had them so the pocket was at the front but during my testing i had the pocket at the back and this made for a much better light show, so time to flip the numbers and months so the orientation will read correctly when the pockets are at the back, Vcarve again worth its weight in gold two clicks and BAM!

I am not sure if this was a factor or not in the first attempt at cutting, but this time and during all the tests I removed to protective coating from the top side of the Acrylic as I think this was clogging up the 1.5m end mill evacuation.

Looking Good for the numbers

And the names are crisp too

Job complete

and more impressive after 28mins the end mill was Crips and clean but most importantly cool.

Onto the contours, so life is funny been using small end mills allot in this project / tests . so that when i broke out the 3mm one i was like wow this is huge lol.

Now this had me slightly worried as i only have 1 left after well… my science tests so, zeroed out and off we went, but as i started i aborted, i remembered when i did this last time i had a problem with structural integrity of the acrylic sheet once the contours got half way through, causing the sheet to lift, flex, and generally not be nice to me, so i decided to double side tape it like mad :slight_smile:

I did a test shake on the sheet and the workshop moved lol.

3,2,1 x0. .y0 .z0 off we go!

Confidence growing, life is good, wife scared the "%£%$ out of me (side story but worth it) i have my ears on, music going, Cyclone revving and head / eyes focused on the SO3, que my wife to walk in like a ninja to the workshop holding the white sheet from the washing line and decided to come and say hello in my half lit (only my work area) workshop. Well after thinking i had been visited by ghosts, collected myself and heart rate, as well as trying to explain what the hell sort of man scream i gave out was, it was back to business!! BTW a man scream is like a woman’s but much more errr manly!

Lifting the sheet at the end of the run - If there is a more satisfactory feeling that this i would love to experience it, this is to CNCing what peeling your scab off is to kids

yip happy with this they are all super crisp clean and again no dramas!

Time for a little base plate creating same F&S

Moment of truth and first dry fit, again this is coming together nicely, the allowances for the pocket are perfect as you can see below, with a smooth tomb fit and lean

So like a kid at Christmas i could not wait, so tried a little LED test, i think that we can all agree, that whilst this may not be at the @Vince.Fab bend metal to my will level, i have asked Acrylic to form an orderly line! and i think looks pretty cool!

So at this point i have completed 90% of the job, sniped the LED to size, small amount of glue and that is another tick in the ACDC completion column.

Stretch goal time - as stated above i wanted to try and do this all 100% acrylic, so this meant that i needed a nice box to hold them in, OK simple, tap tap tap and

Hang on i hear you say, what how and more info… OK you got it

WHAT - did i use to create the box ?

Well again Vcarve showing me that the kids future fund misappropriate was worth it, (no seriously to begin with i was saying to myself “you just bought £700 worth of software that can only work with a CNC” but it is times like this that i was like “shut up and take my money”

I used the “Box Making Gadget” and with some simple prompts created a perfect box in seconds

Only issue i had was first time i had the allowance at .25mm (this basically means an offset value for the joint allowance (set to 0 and you need a hammer to whack that box together, set it to 3mm it will have loads of slop) .25 is my go to for wood but this ended up being .35 and it is nice and snug (between push fit and small tapping together). 3mm end mill and 8 mins later

All cleaned up

And… it didn’t bloody fit!!! ok so you remember my wife in the white sheet? yip this one is on her lol (that is my story and i am sticking to it) after the heart attack i forgot to change it from .25 to .35 allowance, now i have a real problem stock i needed 160mm x 230mm i had 155mm x 220mm.

Well after a few games of Tetris i thought hang on, again there is something in the Vcarve toolbox i can use, two clicks and the Nesting feature sorted out my space and cut issue, how this nesting works blows my mind by my gosh it is amazing!

so 0 0 0 and away we go again…nope this time it cut well but because i was using wastage i had no protective film on the side i just put double sided sticky tape on, i think this will display my emotion to you effectively…

OMG this stuff is killing me what a mess, all the pieces are impossibly sticky and are causing issues with finish. This is where i feel being older makes me wiser 2 x mistakes oin the trot = step away and re-group, so i did and the following day spent 2 hours trying to clean it up, i had all but given up then i was watching a video on flood milling HASS machine and the machinist said a throw away comment about cleaning Plexiglas up and i decided, meh why not.

weirdest things i have said out to my wife +1 “honey where is the baby oil, i need to bathe my acrylic in it”

Tip top tip - this works like crazy and leave the acrylic baby smooth :), simply soak for 5 mins the remove with a baby cotton wipe

with that time to see that fit

![box dry fit together|666x500]
(upload://w63SnCFqnqcHoXYGHjoxL1ndhnv.jpeg)

The fit on this is amazing well beyond my expectations, and the dogbone corners in the acrylic are hardly noticeable at all and when you do see them they look like a very cool feature!!

And that is it folks this is my submission for contest 10.

Things of note -

I really struggled with this and cutting new stock type - but i have proven to myself that i can and need to be more methodical in my approach to design and testing for F&S etc.

I really wanted to push myself and take on a challenge i think i can say i did that

I really love this community and my SO3 - and i know i say this in 100’s of people posts, but we are only limited to the restrictions of our minds, and that, when we individually or collectively set our minds to something we can achieve greatness.

This was my Appolo13 and i am glad all my end mills have made it back from space safely!

Please vote for me (shameless plug)

Jon

7 Likes

Sure,

And/or you can edit your existing entry as many times as needed before the deadline.

2 Likes