The Red Shoulders: Redesigning the Excelsior

Redesigning the Excelsior resulted in modeling not just one ship, but three.

Introducing the Excelsior-class U.S.S. Enterprise:

In-Universe Context

Registry	NCC-1701-B
Class	Excelsior-class Type-B
Type	Command cruiser
Active	2293 to 2329
Decks	26
Length	Approximately 581m
Capacity	1200 humanoids 3500 humanoids maximum limit
Max. Speed	Warp factor 9.2
Max. Cruising	Warp factor 7.4
Armament	Phasers (x12 phaser banks) Photon torpedoes (x6 launchers)

The third Starfleet vessel to bear the name Enterprise, the Enterprise B more than lived up to her storied legacy. Launched in late 2293 shortly after the Khitomer Accords ended hostilities between the Federation and the Klingon Empire, the new Enterprise--designed for combat--spent most of her service at the furthest frontiers of known space, venturing deep into the unknown. After four decades of service under five captains, the Enterprise ultimately disappeared without a trace in 2329, shortly after completing her last mission: escorting a Bajoran refugee flotilla to safe harbor after the Cardassian annexation of their homeworld.

In 2331 the Enterprise B was officially declared lost in action and presumed destroyed. The following year she was succeeded by a new, Ambassador-class Enterprise.

...Also introducing the Excelsior-class U.S.S. Avalon:

In-Universe Context

Registry	NCC-42024
Class	Excelsior-class Type-C
Type	Heavy cruiser
Active	2348 to present
Decks	26
Length	Approximately 572m
Capacity	800 humanoids 3000 humanoids maximum limit
Max. Speed	Warp factor 9.68
Max. Cruising	Warp factor 8
Armament	Phasers (x16 phaser banks) Photon torpedoes (x6 launchers)

Named for the mythical "Island of Apple Trees," the Avalon was launched in 2348 and spent much of her early patrolling the Federation's borders with during the bloody Klingon-Romulan War that ignited after the Khitomer Massacre in 2346. Despite the Federation's neutrality, the conflict frequently spilled into UFP space, and the Avalon participated in multiple combat engagements against both Klingon and Romulan warships.

By the Dominion War, the Avalon had successfully completed 6 multi-year deep space exploration missions, and established first contact with 40 new civilizations. During the Dominion War the Avalon was assigned to the Seventh Fleet, which saw some of the heaviest combat in the opening months of the war. She was one of only 14 ships to survive the disastrous Battle of Tyra, out of a fleet of 112.

...And finally, introducing the Excelsior-class Type-D U.S.S. Chiron:

In-Universe Context

Registry	NCC-42044
Class	Excelsior-class Type-D
Type	Heavy cruiser
Active	2372 to present
Decks	26
Length	Approximately 572m
Capacity	800 humanoids 3000 humanoids maximum limit
Max. Speed	Warp factor 9.7
Max. Cruising	Warp factor 9.4
Armament	Phasers (x13 phaser arrays) Photon torpedoes (x6 launchers)

Following the successful transwarp experiments of the U.S.S. Ardennes, Starfleet designed the Excelsior to be the first of a new wave of starships built to take advantage of the new four fourth generation warp drive (which necessitated the adoption of a new warp scale). The first ship of the Excelsior-class, the U.S.S. Excelsior, was launched in 2285, and it was the first of many. The Excelsior-class rapidly became the "face" of the United Federation of Planets, as it proved to be one of the most capable starship designs in Federation history and was soon a ubiquitous part of the fleet. The Excelsior remained at the forefront of the Federation's exploratory and military efforts throughout the 23rd and 24th centuries, and the Excelsiors combined participated in more first contacts than any other starship class to date.

The U.S.S. Chiron, launched during the third Federation-Klingon War that precipitated the Dominion War in 2372, was the first Type-D Excelsior-class starship built. Assigned to the Federation's seventh fleet, the Chiron was one of only 14 ships to survive a disastrous counterattack operation in the early months of the war. The Chiron's secondary hull was lost in the battle, and the Chiron became one of the first Excelsiors to be refitted into a Centaur-class starship. As a centaur-class starship, the Chiron served the remainder of the war in a support capacity, and eventually restored after the war.

Commentary

Building a model of an Excelsior is a very intimidating project: of all the classic starships designs, the Excelsior is among the most beloved. And for good reason: she is a gorgeous vessel, from every angle. But there are a few problems with the design, especially its scale. You can find an excellent and thorough write-up of all the issues over at Ex Astris Scientia if you're curious. When I construct a model, my #1 concern is keeping the interior and exterior scales consistent. The biggest problem I had to reconcile with the original Excelsior design (as usual) was the placement of the window rows and the decks they represent. The studio model is very problematic in this respect:

Excelsior-class study model in Star Trek: Generations.

The red-highlighted area represents the approximate size of the interior decks if we assume the window rows line-up properly. This results in an Excelsior that is far too big--a saucer four decks thick would result in an Excelsior larger than a Galaxy-class! Given that most Starfleet vessels with circular primary hulls are only two decks thick, and that the Excelsior's scaling relative to the Constitution-class Enterprise and MSD props reflect a saucer only two decks thick, I chose to dismiss the proximity of the windows on the studio model as an error. When I scale a model, I find it helpful to zero in on one or two specific features and extrapolate the rest of the model from there. For the Excelsior, I built my model around a saucer rim approximately two decks thick with enough space to accommodate the raised, third row of windows on the top of the primary hull (green-highlighted area on the above image).

To get the proportions to line-up better with the original design, I had to make the rim of the saucer slightly thicker, and add a bit of a slope going up to the "terraced" deck above. And while I'm talking about the saucer rim, the studio model features three horizontal lines around the saucer that are evenly spaced, but this is only possible because of the close proximity of the two outermost window rows. Because I fixed the spacing on my model, I could not evenly space these lines around the saucer: my solution, as you can see, was to widen the lines and have the top and bottom ones overlap with the window rows.

Scale-wise, my Excelsior-class model has an overall length of approximately 572m. This is perhaps a bit on the large size (that Ex Astris Scientia article, for example, estimates a most-likely length of only 467m) but I find it works well for a ship that bridges the gap between the relatively small vessels of the 23rd century and the relatively large vessels of the 24th century.

My attempt to build an Excelsior model began with the U.S.S. Avalon, the Excelsior-class Type-C, which I intended to represent the "default" Excelsior-class we see "guest-starring" in various TNG and DS9 episodes. Over the course of about a year, the project grew to include two additional Excelsior variants, and led directly to my Centaur and Helena projects, which both of which will be covered in separate posts. My primary goal with the Avalon was keep the overall proportions of the model as accurate as possible, while adjusting the various details to "fix" various issues I had with the original design.

From there, I removed some of my changes to create the Enterprise model (Excelsior-class Type-B, very appropriately), which adheres more closely to the studio model, and then made even more radical changes to create the Chiron model (Excelsior-class Type-D). In order to avoid explicitly "retconning" the original Excelsior design, I justified my changes with a simple headcanon: over the years, Starfleet adopted multiple sub-types of Excelsiors, most notably the Type-A, Type-B, Type-C and Type-D.

The Type-A is the initial model (see U.S.S. Excelsior NCC-2000).
The Type-B is a command variant, designed to accommodate more personnel and a much greater complement of secondary craft (see U.S.S. Enterprise NCC-1701-B).
The Type-C is the most common Excelsior sub-type in the 24th century; a refit of the Type-A with modern technology, it features a ventral shuttlebay on the primary hull and a third impulse drive on the secondary hull (see the U.S.S. Avalon NCC-42024).
The Type-D is the most recent Excelsior variant, designed after Wolf 359 as the "ultimate update" to the Excelsior design. It is more durable and more heavily armed than the Type-C, and can accommodate both saucer separation and recombination.

Excelsior-class Type-D, one of the final variants of the Excelsior line.

As the version of the Excelsior seen in TNG, I imagined the Type-C as a relatively simple update or refit of the original design. As such, I made a few alterations as possible, resulting in only four major changes.

First, I changed the rear shuttlebay to a more conventional circular clam-shell design, similar to the Constitution-class (I could never understand how the shuttlebay doors were supposed to operate in the original design); I added a third impulse engine on the secondary hull, between the two aft torpedo launchers; I moved the second set of forward torpedo launchers from the upper part of the neck to just above the other set on the secondary hull; and I added a third shuttebay on the primary hull, between the bridge and main impulse engines, to allow for the Excelsior to operate secondary craft from the primary hull after an emergency saucer separation.

When I was a kid, I had this fantastic poster of the TOS Enterprise, cut-out to reveal all of the interior decks and mechanics. One detail that always fascinated me was a row of "explosive bolts" at the top of the neck--designed to suddenly, violently and permanently separate the saucer section from the rest of the ship. I preserved this idea in my conception of the Excelsior: the Type-A, Type-B and Type-C would all have similar explosives in the neck, designed to separate the primary and secondary hull, using the force of the blast to propel them in opposite directions.

Obviously this is an extreme and inelegant procedure, which would only ever occur in the most dire of situations. For the Type-C model, sauce separation would result in the total destruction of the neck component, but for the late 24-century Type-D variant, however, I imagined that the neck would consist of an independent module housing a complex docking apparatus more easily allow the two hulls to recombine after separation.

The U.S.S. Avalon's secondary hull after saucer separation.

Incorporating saucer separation provided me with a justification to imagine the secondary hull as an independent starship--further justifying the third impulse engine--but I found that keeping the top of the hull completely flat looked a bit bland. I therefore added a protruding section (2 decks) that would otherwise be hidden underneath the neck.

All of these changes are also present in the Type-D Chiron, along with a few others, but had to be removed for the Type-B.

The Type-B was designed to be more similar to the original studio model, which meant I had to undo those changes. Also, appropriately, the Type-B is designation for the Enterprise B, and at this point I finally thought my skills as a modeler had improved to the point where I could construct my own Enterprise. I tried to keep the Enterprise more faithful to the original studio model--which meant removing the third impulse engine, shifting two torpedoes back up into the neck, simplifying the hull paneling, and adding the various superstructures to the hull. In order to maintain my minimalist aesthetic, I refrained from modeling many of the "greebles" on the hull to present a smoother, simpler vessel.

The most problematic aspect of the Type-B is the pair of hull superstructures on either side of the saucer--what, exactly, are they? On the studio model they tend to glow red, indicating that they may be an additional set of impulse engines--but if that were the case, why design impulse engines to expel gas/waste/heat directly on top of the warp nacelles? Alternatively, the two superstrucutres are sometimes assumed to house an additional pair shuttlebays, a better explanation, but one that suffers from the same problem: the nacelles are in the way.

My solution was to make the shuttlebay doors relatively small, set on the outer ends of the superstructure, and angled outward, so that shuttlecraft could fly in or out w/out needing to navigate around the nacelles.

Moving on to the Type-D, which I designed as the "final evolution" of the venerable Excelsior-class. The Type-D is what Excelsiors built after Wolf 359 look like: same basic design as ever, but with a few modern twists.

Most obviously the Type-D has different nacelle struts, more in line with the aesthetics of other 24th century designs like the Sovereign-class and Nova-class. I tried to keep the back-swept struts simple in order to preserve the overall profile of the Excelsior--because why mess with perfection? Also, of course, I added all of the familiar TNG-era hull elements: phaser arrays instead of banks and visible escape pods.

The other changes to the Type-D are subtle, to the point where I doubt most of them are even noticeable. I smoothed a lot of the edges to better fit with my modeling style, and added a slight curve to the superstructure on the primary hull.

I also adjusted the top of the secondary hull to balance out the larger nacelle struts: the engineering pod has been moved forward slightly, and the square protrusion of hull at the back of the shuttlebay slightly elongated.

The new nacelle struts give the secondary hull a bit more of a smoother, organic flow, which I quite like. Speaking of which, I think that by the time the Type-D rolled out, the whole notion of using explosive bolts to separate two hulls would be a bit anachronistic. But at the same time, I don't want to keep the neck--so for the Type-D, I imagine the neck as a more-or-less independent module w/ its own docking systems, similar to the Galaxy-class. This would make saucer separation a bit of a clumsy affair (effectively resulting in abandoning the neck module), but because the neck module remains intact, it should be easier to recombine the two hulls post-separation.

One of the more appealing aspects of the Excelsior-class, to me, is the art-deco style "ribbing" along the neck, nacelles, and engineering pod. The negative space in the Excelsior's design does a great job at conveying a since of size and grandeur, and the ribbing furthers that by almost making the neck invisible, and "slimming" the nacelles.

For the escape pod hatches, I chose triangular lifeboats similar in size and shape to those found on the Sovereign-class, as it was easier to arrange them in a circular patter. They are slightly smaller than the square lifeboats found on the Galaxy-class, but I think the smaller hatches help add to the sense of scale--they make the Chiron feel a bit larger than it really is. The Excelsior design always felt very "big" and "intimidating" to me, and I wanted to preserve and elaborate on that feeling as much as my abilities would allow.