Edge of peg hole in 11458 too thin?

[Jaco van der Molen]

I was building this nice bookcase from a video I found from Tiago Catarino.

In real life LEGO 20482 (brown here) fits perfectly and the peg hole aligns:

  Screenshot 2020-11-06 at 11.40.34.png (Size: 33.46 KB / Downloads: 95)

But, in LDraw see illustrated here (20482 in light grey for better contrast) does not align with the peg hole
   

Isn't the edge of the peg hole of part 11458 too thin? Or is 20482 too thick?

[Jaco van der Molen]

I was building this nice bookcase from a video I found from Tiago Catarino.

In real life LEGO 20482 (brown here) fits perfectly and the peg hole aligns:


But, in LDraw see illustrated here (20482 in light grey for better contrast) does not align with the peg hole


Isn't the edge of the peg hole of part 11458 too thin? Or is 20482 too thick?

I also have the feeling the edge on the side where the peg hole is, is a bit too thick. In real LEGO this is a minimal edge.

[Philippe Hurbain]

They are not the same diameter... 11458 diameter is 7.4mm (radius 9.25 ldu) while nipple tile is 7.85mm (radius 9.8ldu). and models are rounded down to 9 ldu for 11458 and up to 10 ldu for 20482.

...and there is another practical issue: as I tested here, 20482 doesn't hold well on a technic pin!

I also have the feeling the edge on the side where the peg hole is, is a bit too thick. In real LEGO this is a minimal edge.

Not sure what you mean?

[Jaco van der Molen]

They are not the same diameter... 11458 diameter is 7.4mm (radius 9.25 ldu) while nipple tile is 7.85mm (radius 9.8ldu). and models are rounded down to 9 ldu for 11458 and up to 10 ldu for 20482.

...and there is another practical issue: as I tested here, 20482 doesn't hold well on a technic pin!

Not sure what you mean?

OK, I must admit, I have not build it for real, but this suggests otherwise



That is does not really hold is not an issue in the model.

[Jaco van der Molen]

I also have the feeling the edge on the side where the peg hole is, is a bit too thick. In real LEGO this is a minimal edge.

   

[Philippe Hurbain]

Oh yes... same issue, plate part is not really 16mm wide, but is modeled as such, and technic beam is a bit wider IRL than the modeled 9LDU diameter...

[Jaco van der Molen]

Oh yes... same issue, plate part is not really 16mm wide, but is modeled as such, and technic beam is a bit wider IRL than the modeled 9LDU diameter...

OK, then should (can) it be adjusted or is rounding an absolute must?
Maybe I will just scale the tile to make it look more close to the real thing for this particular model.

[Magnus Forsberg]

OK, then should (can) it be adjusted or is rounding an absolute must?
Maybe I will just scale the tile to make it look more close to the real thing for this particular model.

We had the same issue on this part, 85943:



There the beam had to stay at 9ldu, and the difference was hidden with a slope.

[Jaco van der Molen]

These parts suffer from the same consequences then?
   

I personally dislike this very much.
Why don't we make these so that they align?

Part 75535 seems to work just fine. What differs that one from this 62462 one?
   

In RL LEGO these parts are the same diameter, are they not?

[Jaco van der Molen]

They are not the same diameter... 11458 diameter is 7.4mm (radius 9.25 ldu) while nipple tile is 7.85mm (radius 9.8ldu). and models are rounded down to 9 ldu for 11458 and up to 10 ldu for 20482.

How can this be? I now have the parts here in my hands and if you look close you can see a very slight difference, but certainly not 0.45 mm?
Rubbing your nail over it you can feel it. The same goes for the edge I mentioned earlier.
   

Using the same color for 11458 and 20482 viewed from a distance the difference is neglectable IMHO.

However, I have a very strong feeling that 75535 and 62462 are really the same diameter.
I have those now in my hands too, connected with a pin and I can feel nor see no difference.
   

[Orion Pobursky]

I'm with Jaco on this. If things need to adjusted, let's adjust them. Errors on the order of half a mm are noticeable and should be fixed.

[Magnus Forsberg]

I'm with Jaco on this. If things need to adjusted, let's adjust them. Errors on the order of half a mm are noticeable and should be fixed.

Problem is to decide where to draw the line. They can't be fixed to fit both a technic beam width, and a regular brick width.

[Philippe Hurbain]

Problem is to decide where to draw the line. They can't be fixed to fit both a technic beam width, and a regular brick width.

Yes, that's why I made 62462 beam wide. But I checked again, and it is significantly wider than a beam. So I agree it should be fixed.

[Travis Cobbs]

OK, then should (can) it be adjusted or is rounding an absolute must?
Maybe I will just scale the tile to make it look more close to the real thing for this particular model.

Parts are allowed to use up to 3 decimal places, so rounding to the nearest LDU is certainly not required from a spec standpoint. Having said that, I have no idea how easy or difficult it would be to fix it to be correct, and whether doing so would cause other issues with other parts.

Given that this seems visibly wrong due to the two sets of rounding, I personally feel that not rounding (or rounding more accurately) would be advantageous.

[N. W. Perry]

Oh yes... same issue, plate part is not really 16mm wide, but is modeled as such, and technic beam is a bit wider IRL than the modeled 9LDU diameter...

In my imagination, there's a category of shadow data that could be included in a part to account for discrepancies like this, not only due to rounding, but also the difference between real part dimensions and LDraw approximations. Viewers and editors could selectively display this "true dimension" data, just as many now do to approximate gaps between parts, etc.

[Jaco van der Molen]

Have we ever gave this more thought or are we going to "fix" or rather adjust things for this kind of parts?

[Magnus Forsberg]

Have we ever gave this more thought or are we going to "fix" or rather adjust things for this kind of parts?

Tell me how to fix it and I'll start working.

[Magnus Forsberg]

Tell me how to fix it and I'll start working.

Looks like 62462 is fixed. And 11458 is still too thin?

[Magnus Forsberg]

Looks like 62462 is fixed. And 11458 is still too thin?

I've now made the cylinder half 9.5 ldu thick.
https://www.ldraw.org/cgi-bin/ptdetail.c.../11458.dat
As far as I understand it, it should be somewhere in between in thickness.

[Jaco van der Molen]

Tell me how to fix it and I'll start working.

I see the work is being done. Wherefore I thank you all.
I guess in the discussion we determined what was "wrong" with the parts.
They are too small in diameter.
There are certain consequences and compromises to be made from what I understand.