LDraw.org Discussion Forums
Part 4210 (hose reel) - Printable Version

+- LDraw.org Discussion Forums (https://forums.ldraw.org)
+-- Forum: Models and Parts (https://forums.ldraw.org/forum-18.html)
+--- Forum: Parts Authoring (https://forums.ldraw.org/forum-19.html)
+--- Thread: Part 4210 (hose reel) (/thread-24141.html)



Part 4210 (hose reel) - N. W. Perry - 2020-07-29

Part 4210 (ex-4208) says it needs teeth. It also needs the holes through the drum spindle.

If anyone has the real part handy:
  • How many teeth should it have? I think 40, but it's hard to tell from photos alone. (The similar part 2585 has 32.)
  • What's the diameter and position of the hole? (In 2585 it's 6 LDU diameter and centered 5.5 from the inner flange.)



RE: Part 4210 (hose reel) - Alex Taylor - 2020-07-30

(2020-07-29, 16:31)N. W. Perry Wrote: Part 4210 (ex-4208) says it needs teeth. It also needs the holes through the drum spindle.

If anyone has the real part handy:
  • How many teeth should it have? I think 40, but it's hard to tell from photos alone. (The similar part 2585 has 32.)
  • What's the diameter and position of the hole? (In 2585 it's 6 LDU diameter and centered 5.5 from the inner flange.)

I've attached the LDD mesh for the part, if that helps.


RE: Part 4210 (hose reel) - Alex Taylor - 2020-07-30

(2020-07-30, 6:56)Alex Taylor Wrote: I've attached the LDD mesh for the part, if that helps.

.dat   4208.dat (Size: 212.51 KB / Downloads: 4)


RE: Part 4210 (hose reel) - Magnus Forsberg - 2020-07-30

(2020-07-30, 6:56)Alex Taylor Wrote: I've attached the LDD mesh for the part, if that helps.

Seems to be accurate. I counted 44, both on the real thing and the ldd mesh.


RE: Part 4210 (hose reel) - N. W. Perry - 2020-07-30

(2020-07-30, 7:12)Magnus Forsberg Wrote: Seems to be accurate. I counted 44, both on the real thing and the ldd mesh.

Thanks, that'll be helpful.

Hmm…44 teeth doesn't adapt quite as neatly to 16/48-prims as 40 would; it means subparting the flange into only 4 slices instead of 8. Is it better to go for the correct number of teeth, or streamline the geometry a bit with greater use of subparts?

For comparison, here's what I came up with for a 40-tooth version (pinhole not finished yet):
   

.zip   4208s01.zip (Size: 2.16 KB / Downloads: 2)


RE: Part 4210 (hose reel) - Magnus Forsberg - 2020-07-30

I think I would make a 1/4 subfile with 11 teeth, or maybe better, a 1/8 with 5,5 teeth.
Attach the inner side to a 4-4ndis around the shaft, and hide the outside under the smaller cone.

btw.
dont use the stud3 primitive like that. I know it was used like that in the old file, but must be corrected now.


RE: Part 4210 (hose reel) - N. W. Perry - 2020-07-30

(2020-07-30, 15:27)Magnus Forsberg Wrote: I think I would make a 1/4 subfile with 11 teeth, or maybe better, a 1/8 with 5,5 teeth.

I'll do the 1/8th then, since it solves both problems (accurate number of teeth, and greater use of subfiles).

Quote:Attach the inner side to a 4-4ndis around the shaft, and hide the outside under the smaller cone.

So the ndis on the inside replaces the disc that's there now? And the cone just the way I have it now, hiding the outside of the subpart?

(By the way, my current version is based on 2585.dat as far as how the prims overlap the teeth subfile. But I can see how the ndisc is an improvement.)

Quote:btw.
dont use the stud3 primitive like that. I know it was used like that in the old file, but must be corrected now.

OK, I'll replace it. What should it be, a cylc?

And for the pinhole, I'm thinking a cyl, a cone and a cylj (subfiled of course)?


RE: Part 4210 (hose reel) - Magnus Forsberg - 2020-07-31

(2020-07-30, 17:35)N. W. Perry Wrote: And the cone just the way I have it now, hiding the outside of the subpart?

No, I would hide it under the smaller cone prim. The bigger cone is better included in the 1/8 subfile.


RE: Part 4210 (hose reel) - N. W. Perry - 2020-07-31

(2020-07-31, 6:54)Magnus Forsberg Wrote: No, I would hide it under the smaller cone prim. The bigger cone is better included in the 1/8 subfile.

OK. What's the advantage that way? My thinking was to hide the overlap between the primitives and the uneven number of teeth (to avoid gaps).


RE: Part 4210 (hose reel) - Magnus Forsberg - 2020-07-31

(2020-07-31, 16:12)N. W. Perry Wrote: OK. What's the advantage that way? My thinking was to hide the overlap between the primitives and the uneven number of teeth (to avoid gaps).

I'm thinking the teeth are part of the sloping cone, so the bigger cone has to also be adapted to the 44 teeth. And the edge line.
If you allow the bigger cone to switch from 16 to 48 is wouldn't fit the 44 teeth.


RE: Part 4210 (hose reel) - N. W. Perry - 2020-07-31

(2020-07-31, 16:48)Magnus Forsberg Wrote: I'm thinking the teeth are part of the sloping cone, so the bigger cone has to also be adapted to the 44 teeth. And the edge line.
If you allow the bigger cone to switch from 16 to 48 is wouldn't fit the 44 teeth.

I think I get it. So the larger cone, instead of a primitive, should be directly modeled, along with the 44 teeth, to match a 48-sided inside edge and ring…and the transition between those and the smaller 16-sided cone and rings in the main part gets hidden behind them?


RE: Part 4210 (hose reel) - Magnus Forsberg - 2020-07-31

   

Something like this....   But with 5,5 teeth of course.
Keep the green cone. Make an adapter surface (light blue), and a 44-sided cone (light pink and yellow)
(Here I inlined 48-sided primitives, but you need to make it "44-sided", matching the 5,5 teeth.)

Make a 1/8 subfile that will attach to the pink 4-4ndis and three vertices in the green cone.


RE: Part 4210 (hose reel) - N. W. Perry - 2020-08-01

(2020-07-31, 21:12)Magnus Forsberg Wrote: Something like this....   But with 5,5 teeth of course.
Keep the green cone. Make an adapter surface (light blue), and a 44-sided cone (light pink and yellow)
(Here I inlined 48-sided primitives, but you need to make it "44-sided", matching the 5,5 teeth.)

Make a 1/8 subfile that will attach to the pink 4-4ndis and three vertices in the green cone.

Brilliant, thanks for the guidance. I was just getting started with the cone as the adapter surface instead of the disc, but I see this way will probably look cleaner.

I'll use PrimGen2 to get the vertices for my 44-sided serrated polygon. :-)


RE: Part 4210 (hose reel) - N. W. Perry - 2020-08-01

A completed draft:
   

.zip   4208.zip (Size: 6.21 KB / Downloads: 1)

I'm not thrilled with how the pinhole blends with the shaft (or doesn't blend) when prim sub is on. I could minimize gaps by using more and skinnier cyl segments, I suppose.

Also, this part would benefit from a cylj2_3x4, if it existed (I had to manually invert cylj3x4).


RE: Part 4210 (hose reel) - Magnus Forsberg - 2020-08-01

(2020-08-01, 7:08)N. W. Perry Wrote: A completed draft

Excellent!
The only thing I don't like is the long centre cylinder, and the short one inside the hole, being divided, into the subfiles.

Don't split things going across a main axis.
1 16 19 0 0 0 -38 0 6 0 0 0 0 6 4-4cyli.dat
1 16 -23.5 2.857 0 2.5 0 0 0 -5.714 0 0 0 2.5 4-4cyli.dat


RE: Part 4210 (hose reel) - N. W. Perry - 2020-08-02

(2020-08-01, 13:46)Magnus Forsberg Wrote: Excellent!
The only thing I don't like is the long centre cylinder, and the short one inside the hole, being divided, into the subfiles.

Don't split things going across a main axis.
1 16 19 0 0 0 -38 0 6 0 0 0 0 6 4-4cyli.dat
1 16 -23.5 2.857 0 2.5 0 0 0 -5.714 0 0 0 2.5 4-4cyli.dat

Great! I've moved those cyls, and the pinhole subfiles, up into the main file.

Glad to try my hand at a part with some actual geometry. :-)


RE: Part 4210 (hose reel) - N. W. Perry - 2020-08-03

Subfiles are now on PT; main file has been submitted, pending admin acceptance uploaded as well:

https://www.ldraw.org/cgi-bin/ptdetail.cgi?f=parts/4208.dat


RE: Part 4210 (hose reel) - N. W. Perry - 2020-10-21

Just one vote away (in the s01 subfile) from moving to the admin review stage.  Wink Thanks to all for your reviews and feedback!