Kevin - A V Carroll Horizontal Mill Rebuild

Firstram, thanks for stopping by. Southport... I've been down that way a few times. You're near my favorite donut shop---Wake and Bake, in Carolina Beach.

Back to the irregularly scheduled show... I made some taper adapters on the 3d-McPrinter. The gray-haired machists don't even bother to cluck at this contrivance. But I'm not a real machinist so it's allowed. Turns out that the printer struggles with a fine taper. It like a low-resolution monitor or an apprentice coil-pot maker. The trouble is the plastic-extruder nozzle diameter. It produces a bead of plastic that is 16 thou in diameter which is a giant step-over in an application like this. Normally a small step-over can be used but the taper is on both sides and the distance from one wall to another is small. A thick part would be able to take up the difference between the outer shells.

IMG-3221.jpgIMG-3222.jpg


Just to get an idea of the bearing, I marked up the collet and spindle. The bearing is 100% on the discontinuity discussed above.
Did the same procedure for the low-profile collet. Similar issues. Also, plastics can grow or shrink a bit. There's a scaling factor in the software. When I was making a grease seal for the spindle nose (a future post) I had to correct for shrinkage of 2.4% when printing with PolyFlex TPU95. I didn't make any adjustments for these parts because these are simply visualization aids and not real parts.

I probably will try the plastic adapters in the mill. I had an amazing experience with 1-1/2" dimple dies in a 20-ga stainless pan. The dies were 3D printed PLA. These are ABS, which is actually not as good as PLA in compression. I think PLA has E = 350 ksi and ABS is half of that. Steel has E = 31,000 ksi. Silly thought it may be, I'm curious how well it will work. There's a video of a kid who made a weird Morse taper collet out of PLA and ran a fly cutter in it. Watch and cringe.
IMG-3223.jpg


Still thinking about how to machine the ID of the adapter in steel. The OD might not be so bad. But if the outside is tapered, there's only a thin flange to hold onto...
 

Attachments

  • IMG-3220.jpg
    IMG-3220.jpg
    1.6 MB · Views: 2
  • IMG-3219.jpg
    IMG-3219.jpg
    846 KB · Views: 1
  • IMG-3218.jpg
    IMG-3218.jpg
    1.2 MB · Views: 1
>I probably will try the plastic adapters in the mill. I had an amazing experience with 1-1/2" dimple dies in a 20-ga stainless pan. The dies were 3D >printed PLA. These are ABS, which is actually not as good as PLA in compression. I think PLA has E = 350 ksi and ABS is half of that. Steel >has E = 31,000 ksi. Silly thought it may be, I'm curious how well it will work. There's a video of a kid who made a weird Morse taper collet out of >PLA and ran a fly cutter in it. Watch and cringe.

31000 ksi is a pretty strong steel. And those are strange units. I had to look it up.
 
You could bore an internal taper as close as you can, than turn the external taper to match the spindle. Part the adapter off, locktite in place and grind the ID in situ.
 
I'd use the flange style. The double taper will want to move back and forth in use, breaking the loctite.

Sent from my SM-G892A using Tapatalk
 
>I probably will try the plastic adapters in the mill. I had an amazing experience with 1-1/2" dimple dies in a 20-ga stainless pan. The dies were 3D >printed PLA. These are ABS, which is actually not as good as PLA in compression. I think PLA has E = 350 ksi and ABS is half of that. Steel >has E = 31,000 ksi. Silly thought it may be, I'm curious how well it will work. There's a video of a kid who made a weird Morse taper collet out of >PLA and ran a fly cutter in it. Watch and cringe.

31000 ksi is a pretty strong steel. And those are strange units. I had to look it up.

Hmm.. I admit I'm out of my league on material properties. I have seen ksi (1000 lb per square inch) in some of the J F Lincoln books and around the web. It is often in tables of material properties, which is why I used it. I'm not religious about it, though, so I'll add GPa for the metric minds out there.

Looking more carefully, I'm seeing 200 GPa (30 ksi) for steel. I found a paper [1] that says that PLA has a Youngs modulus of 3.5 GPa (507 ksi). Those measurements are for tensile stress.

For ultimate strength, mild steel is about the same in tension and compression, so around 200 GPa (30000 ksi). PLA, on the other hand, is much stronger in compression. This study found that PLA failed in compression around 50 GPa (7251 ksi). That's 4 times weaker than steel but it's so damn easy that it's hard to resist. Also, compared to the strength of aluminum, PLA is not bad. Now-a-days there are companies that produce press-brake tooling from PLA, which I take as social proof of the material's suitability for this weird application. It's worth a try, at least.

1. http://2015.igem.org/wiki/images/2/24/CamJIC-Specs-Strength.pdf
 
...

Looking more carefully, I'm seeing 200 GPa (30 ksi) for steel. I found a paper [1] that says that PLA has a Youngs modulus of 3.5 GPa (507 ksi). Those measurements are for tensile stress.
This comparison doesn't look quite right. It is saying that PLA is almost 6 times stronger than steel. Could these be switched around? Also, are you sure the first units are GPa? It seems that MPa would be more reasonable.
 
This comparison doesn't look quite right. It is saying that PLA is almost 6 times stronger than steel. Could these be switched around? Also, are you sure the first units are GPa? It seems that MPa would be more reasonable.

I may not have explained it well. But for clarity, here's a list of E values [1]. The list makes it easy to compare values for ABS, A36 steel, and aluminum. PLA isn't listed but the other study [2] had a reasonable value for E in tension. The most important thing is that showed the compressive strength of PLA, though.

As I was thinking about your comment, I saw that carbon-fiber filled plastic is also on the list. It has E = 150 GPa (21,755 ksi) which is A-MA-ZING. I have carbon-fiber ABS and carbon fiber-nylon filament on the shelf, ready for a project like this. So I'm glad we had this conversation!

Anyway, I don't want to over emphasize this whim/lark/fancy. I'll try the plastic collet adapter and let you know how it performs. But first, I've got the rest of the mill to fix.

1. https://www.engineeringtoolbox.com/young-modulus-d_417.html
2. http://2015.igem.org/wiki/images/2/24/CamJIC-Specs-Strength.pdf
 
CF ABS is much easier to print. CF nylon really wants garolite and glue, or maybe glass and glue.

I'm deep the the glass fiber nylon rabbit hole right now ;-)
910ad17aa680d05f2155cf0effaff527.jpg


Sent from my SM-G892A using Tapatalk
 
CF ABS is much easier to print. CF nylon really wants garolite and glue, or maybe glass and glue.

I'm deep the the glass fiber nylon rabbit hole right now ;-)

Sent from my SM-G892A using Tapatalk

That is a nice looking part. Post a link if you've got a thread going!

I've had good luck with ABS. It will curl up on the corners a bit but a brim seems to solve the problem. I have nylon too but haven't gotten to it yet. Need to install the hardened nozzle and give it a try.
 
Back
Top