Vorpal 180 - A Printed Printer

[sparky06757]

On 6/14/2023 at 1:15 PM, atrushing said:

I lost track of which print was from before and which was with the side panels installed but they both have some pretty bad ringing and surface texture. They were printed at 150mm/s infill and 110mm/s walls and 5000mm/s/s acceleration. Also I haven't printed much with PLA in a long while and forgot how bad it can curl on the overhangs. Maybe I could slow it down a little..

Have you done any testing since this with the changes you made to the frame??

[atrushing]

6 hours ago, sparky06757 said:

Have you done any testing since this with the changes you made to the frame??

With the kids home from school for the summer, it has been a lot harder finding time to work on this.

I finally finished rebuilding the new toolhead with the Vz-RIDGA gears and the new cooling ducts. I re-printed the Benchy while slowing it down a tad for a 40 minute print. Here are the before (blue @35min) and after prints.

It still shows ringing on the smooth faces but I haven't yet tried to test the resonance and apply input shaping.

I am re-printing the astronaut right now and I will print the Benchy again in the matte blue from before but I want to re-slice it so the bow points to the front of the printer and take better advantage of both fans for the overhang.

[atrushing]

Here is a set of comparison shots of the astronaut print. Old print on the left..

The new one looks a little better in person and the new cooling ducts are helping with the overhangs. I really need to hook up an ADXL345 and dial in the input shaping.

[atrushing]

And here is a new Benchy rotated for better overhang cooling. The geometry is not very ideal for toolhead mounted part cooling since it spends so little time at the worst part of the overhang before changing direction. Of course that is designed to be a feature of the test.

The new print on the right took 38 minutes while the old print took 35 minutes.

These prints also show that the ringing in the X direction takes longer to settle out than in the Y direction.

The change in quality might also be connected to the new BMG drive gears. The old one used $5 gears from a random AliExpress seller while the new one uses RIDGA gears from Mellow Fly.

[atrushing]

I have added gear meshing adjustment to the belted Z drive by allowing the stepper to rotate about the top right screw. This gets rid of the backlash that seems to throw off my Z-tilt process. My Trident will almost always succeed in two passes but, for now, the Vorpal always takes at least three if not four passes.

Additionally, when I took apart the original foot assembly I found that the 50T gear shaft was rubbing on the face of the stepper motor. Fixing all of these minor issues should improve the print performance but it can be a slow process..

[atrushing]

I have added hidden handles to the Vorpal 180. This version requires support while printing but I think it looks a lot cleaner with the back closed off.

I could create the same effect while not needing support by making it in two parts but that wouldn't be as strong as this design..

[atrushing]

I was able to remove the backlash at the gears but then I found that the GE5C spherical bearings that I used have up to 0.5mm of vertical play in them. There is wisdom in the saying "buy once, cry once"...

On my Trident I use Igus spherical bearings but I can't seem to find them in stock anywhere locally. I am wondering if I could make a channel in the two mating pieces for a rubber band to maintain tension between the bed and the Z-carriages. This does seem like a sloppy design though and I don't know how well it would hold up to the enclosure temperatures over time.

I would really like to design a flexure joint to replace the need for a bearing here but that could easily become another rabbit hole of trial and error design testing.

[mvdveer]

2 hours ago, atrushing said:

but that could easily become another rabbit hole of trial and error design testing.

The pain of innovation and being a pioneer. I take my hat off to you. (And yeah, I do wear one ) Sure you will find the answer/perfect solution.

[atrushing]

I think I have a working solution!

For the release I plan on adding a fifth flex spoke for greater vertical rigidity but it shouldn't effect the rotational flexibility.

Now I just need to figure out a macro or some gcode using force_move on each z_stepper sequentially so I can have the bed tilt/rotate continuously for at least 30 minutes to see if the flexures loosen up over time.

[atrushing]

I wrote a .gcode file with 6,900 FORCE_MOVE commands to cycle each Z motor up and down by 12mm at 50mm/s and ran the test for over an hour. I wish I could have created a smoother interpolated movement but that is beyond my skill set..

Some of the spokes started separating at the center hub so I will need to modify the geometry for a better transition and to give the spokes more of a straight flex section near the center. I will also need to run the test with the printer enclosed to see how the flexures will handle higher temperatures.

[Killajoedotcom]

What about something like this?

[atrushing]

1 hour ago, Killajoedotcom said:

What about something like this?

I can make one side with that style of flexure and the other with my next iteration design and put them both through another hour of testing! Do you have a link to that model so I can "borrow" the geometry?

[Killajoedotcom]

 

I'll do you one better... Here's the .stp file for your use.Z Screw Arm.step

Edited August 11, 2023 by Killajoedotcom

[Killajoedotcom]

11 hours ago, atrushing said:

I can make one side with that style of flexure and the other with my next iteration design and put them both through another hour of testing! Do you have a link to that model so I can "borrow" the geometry?

 

[atrushing]

On 8/10/2023 at 12:36 PM, Killajoedotcom said:

What about something like this?

I finally got around to rebuilding the bed and testing the new flexure options. These pictures are after two hours of testing (nearly 14,000 moves) and neither side showed any signs of fatigue or cracking at the corners like the first version did.

I feel that the design with spokes mimics the movement of a spherical bearing a little better. I have adjusted the spokes to be more equal in length on the inside and outside curves.

Now it is time to finish the tophat and test this printer with some of my typical working prints including some of it's own replacement parts.

[atrushing]

I tested the flex at the toolhead for each of my printers by measuring the Y-axis deflection while pulling on the nozzle with one finger. The minimal Vorpal gantry doesn't seem to suffer any extra flex compared with my other printers.

It isn't quite a scientifically rigorous test procedure but I pulled several times for each test to get a range of values and recorded the mean value.

V-Minion      0.39mm
V 0.1             1.32mm    -->   0.45mm
Vorpal 180   0.36mm
Trident         0.43mm
Switchwire   0.24mm

My V0.1 has been using the very first Mini Stealth x-carriage I made and was held on with the undersized Ruthex heat-set inserts. The rail is also the stock LDO MGN7H. This gave me a good excuse to design a better x-carriage that should help the MGN9C mod as well. This version does require new XY_Joint_left/right_Upper pieces (which are so very fun to install). I really need to put together a rather large update across my Mini Stealth releases.

[atrushing]

On 4/30/2023 at 9:10 PM, Buurman said:

Looks awesome, even though.. tophat.. I hate this thing.. how often I have broken this thing...

I guess it's not really a surprise, but you were right @Buurman.. This thing is like an annoying puzzle. I will stick with this design for now because I really don't want the boxy look and I'm not going to introduce aluminum profiles to this printer just for the tophat.

I finally got the tophat put together and I will need to re-print all four pieces and re-cut the panels to make it wider and deeper. I also want to change the feet to use 25mm square cork pads instead of the mini rubber feet. At least I can start printing some ABS on this.

And here is a shot of the Vorpal 180 hanging out with its buddies.

I am seriously thinking about selling off the V-Minion and the Switchwire to make room for (and help me afford) a 350mm Trident (which I wouldn't mod at all ).

[mvdveer]

17 minutes ago, atrushing said:

which I wouldn't mod at all ).

Say what!!!

[claudermilk]

2 hours ago, atrushing said:

(which I wouldn't mod at all ).

Blasphemy!!

[Penatr8tor]

6 hours ago, atrushing said:

I am seriously thinking about selling off the V-Minion and the Switchwire to make room for (and help me afford) a 350mm Trident (which I wouldn't mod at all ).

There's really only one kit you don't need to mod at all...

Build a VzBoT 330 

[atrushing]

2 hours ago, Penatr8tor said:

There's really only one kit you don't need to mod at all...

Build a VzBoT 330 

Did I say No mods? I think I meant to say no New mods...

The VzBot looks impressive but a bit pricey. I guess there aren't very many kit suppliers yet.

[Penatr8tor]

41 minutes ago, atrushing said:

The VzBot looks impressive but a bit pricey.

It's ~$1360 on Aliexpress and I said... Hey that's really not too bad, after shipping and tax that number ballooned to ~$1800. So yeah, you're not wrong. I will say that the performance is impressive if nothing else.

[sparky06757]

@atrushing Did I miss the release or have you not released it to the public yet??

[atrushing]

41 minutes ago, sparky06757 said:

@atrushing Did I miss the release or have you not released it to the public yet??

No release yet. Now that I have the tophat installed, it is doing great with ABS which was the last thing I really needed to verify. Also, as I wrote above, I measured the flex at the toolhead and even though it doesn't have an aluminum profile for the gantry, it is right on par with my Trident. The flexure spherical bearings have no problem with the heated chamber so I am happy to push that as part of the final design.

I am printing new feet that will use 6mm cork and then I just need to re-export all of the latest .stl files with the shrinkage scaling applied. Then I plan on creating a folder structure and a readme.md file and will post it on GitHub and share the link here. Then starts the "fun" part, writing up a manual.

------

I forgot how much nicer ABS parts look compared to PLA parts. These were printed with the exact settings that I use when printing on my V0.1. There is a little ringing on the faces but the corners are nice and crisp and the dimensions are spot on.

[atrushing]

I re-printed the new feet to test the cork pads and it looks like they have helped reduce the ringing.

I have also printed a new x-frame piece for mounting an ADXL at the toolhead so I can finally test the resonance frequencies.