Some edges are missing on that part and 77180 such as the following:
2 4 6 -4 -30 -2.282 -4 -30
2 4 -24.64 40 -170 -26.955 39.985 -165.307
2 4 -6.9384 36 -177.3912 -10 36 -178
2 4 -10 36 -178 -13.0616 36 -177.3912

Confirmed...
Also:
- looks good, but imho much too fine detailed. Back cylinder should be normal resolution!
- one of the pair should use the other mirrored

And... what about this? https://forums.ldraw.org/thread-23790-post-40414.html#pid40414

it's amazing how your and my curve math :-)
https://forums.ldraw.org/thread-23790-post-40645.html#pid40645

why don't you subfile it to have it mirrored? so the opposite parts needs only the mirrored subfile and the stud

Simplify the inside, this does not need so many surfaces

Also the outside could be reduced, where the part is nearly flat. As it connects to the 6091, which has a prim as surface, the 6091 benefits from the smoothing, this cannot be applied at this part, so I was also split between using more surfaces or the standard resolution of the 6091, I finally chose the 6091.

I would strongly advise against using a primitive on the straight rounded section. This will generate holes when the renderer (e.g. Studio) is doing prim substitution

There are a number of stray lines in under the surface around the STUD4:

2 1 -10 35.016 -178 -13.0616 36 -177.3912
2 1 -8.7502 34.4852 -177.7515 -10 35.016 -178
2 1 -8.7502 34.4852 -177.7515 -7.3921 34 -177.4814
2 1 -7.3923 34 -177.4813 -6.9386 33.8378 -177.391
2 1 -6.9386 33.8378 -177.391 -4.3436 32.2902 -175.6574
2 1 -15.6568 24 -155.6568 -17.391 24 -153.0614
2 1 -15.6568 34 -175.6568 -17.391 34 -173.0614
2 1 -15.6568 34 -175.6568 -13.0615 34 -177.391
2 1 -4.343 34 -175.657 -6.9384 34 -177.3912
2 1 -14.2426 34 -174.2426 -15.5432 34 -172.2961
2 1 -10 34 -176.0002 -12.2961 34 -175.5433
2 1 -17.391 24 -153.0614 -18 24 -150
2 1 -17.391 34 -173.0614 -18 34 -170
2 1 -16.0002 32.875 -170 -15.5432 32.875 -172.2961
2 1 -13.0615 24 -157.391 -15.6568 24 -155.6568
2 1 -6.9384 34 -177.3912 -9.9999 34 -178.0002
2 1 -12.2961 34 -175.5433 -14.2426 34 -174.2426
2 1 -13.0615 34 -177.391 -9.9999 34 -178.0002

STray Surface:
3 14 -5.543 36 -182.296 -8 36 -181.1 -4.3432 32.2897 -175.6568

The important things first:

Ouh! .... I didn't notice that someone was already working on the part ... Sorry Gerald !!!

To the object:

*
Resolution:

As far as I can see, there are no adequate 16 and 48 primitives for the unusual surface.
The goal was to strive for a resolution that withstands photo-realistic rendering and optimally meets the shapely curvature.

*
Resolution reference for this:

The longitudinal curve (all line points of the curve of the x-axis brought to zero) comes from the part 45705.dat from Paul Easter.

*
With a resolution of 22.5 degrees on the longitudinal axis with 1-4cyli.dat as a reference, the render image was too faceted.
Correspondingly better in hi-res.

*
Mirror the part:

Yes...Philippe
... had put it on hold for this morning for the time being, under the aspect
also to create the surface as a subpart ...
... as is so often the case, printed versions appear at some point.

thanks Philippe and Vincent

We both wrote comments at the same time, so I was only able to read yours afterwards.
I actually wanted to send you a PM about this

Out of fairness:

If you can use anything from my files .. benefit from it.
Otherwise I have no problem deleting my files and you will create yours for them.

Let me know.

PS: the render results speak for themselves ... neither with studio nor Modo and Cinema 4d are no "holes" or other fragmentations to be seen if you use the hi-res part.
(related to "6091" section)
Whereby the last two render mentioned do not use substitution - neither do they need ... because of their usage structure.
At the front in the "spaghetti" area, however, something is still wrong.
Presumably cond-lines.

Sorry to insist, but clearly resolution is much too high. Eg. back should match 1 16 0 -8 20 0 0 -1 0 1 0 1 0 0 6091.dat. Reduce the number of segment by 3 on X axis, by 2 in z axis and you'll be about right.
As noted by Gerald, it's even worse for inside which could have about 1/20 the number of triangles without any adverse effect.

Addendum:

briefly researched:

*
From February 16, 2021
https://zombinationgebaut.com/lego-10295-porsche-911-im-review-114491/

On February 27th, 2021 I started my work on the part.
So it was impossible for me to see at the time that anyone started doing it.
But still my mistake ... should have noted that under PIWO.

But no matter ... my offer to delete is still there.

corrected link:

https://zusammengebaut.com/lego-10295-porsche-911-im-review-114491/

Keep your file up, no worries!

I tried also a render with Studio, there was no gap as well.

Release 2023-05