Craft Gunmaking, No compromises, No corners cut, EVER

Thursday, January 21, 2016

A Dispassionate Look at the Design of American Doubles

Let me preface this post with these statements:  
First, unless you were personally responsible for any of these designs, you have no reason to be offended (or embarrassed).   

Second, there are certain things that I don't allow to cloud or color my thoughts / opinions / evaluations where mechanical devices (especially guns) are concerned.  They are (in no particular order), nationalistic pride, "collector" value, "legendary" status, the opinions of long-gone hack gun writers, rarity or "provenance".  In other words, there are no sacred cows.
Quality stands on its own merit, in design, execution and function.  To me, it is the only measure that matters.   What follows are factual observations, based in actual reality.  If any of this offends you, in any way, perhaps you should wonder if your self-image is too closely tied to a particular "brand".

Third, for those who need things literally spelled out for them, or those whose reading comprehension isn't all it could be, I'll type this v-e-r-y s-l-o-w-l-y:  I am NOT comparing any classic-era American double to any best-quality double, from anywhere. 

Now that the disclaimers are out of the way, let's get on with it.

 It's very tempting to wax romantic at the idea of a white-bearded fellow, wearing wire-rimmed glasses, taking those last few imperceptible swipes with a file or scraper to make the gun perfect and then blowing out the flame of his gas lamp before he goes tottering home for the day, knowing that his day's work was the best that could be done.   

Here's a question: How many guns do you think that the company could have finished and sold if all (or any) of their workers fit the above description?  

The truth is that American doubles were like every American product, a consumer item, designed for maximum production output and to make maximum profit, at minimum cost.  They were (and still are) items whose production was given over to machine-manufacture to the largest extent practicable at the time, with the minimum of hand work.  That is because the guns were made and finished by men who were not gunmakers, but factory workers that specialized in one area of production.  Most operations that would have been too expensive for skilled labor (which then, as now, is always in short supply) were done by machine, things like barrel boring, barrel contouring, frame shaping, stock inletting and shaping.  Add in that many of these men were paid "piece work", meaning that they got paid based upon the number of pieces they finished (not the best pay structure to instill pride in workmanship) and it becomes obvious that a man who fit the imaginary description above would have been told in short order to speed-up or ship-out.   Luckily for him, he exists quite happily in the imaginations of  many collectors and shooters of American doubles.

Now, let's take a look at some of the guns that many believe (sometimes with religious fervor but mistakenly none the less) to be the equal of the best from England or Europe.

L.C. Smith ~ America's best? (spoiler alert, it's not)
There is nothing impressive about any classic-era American doubles, but the L.C. Smith guns stand out as the best example of how NOT to design a shotgun.  The fact that so many design compromises and engineering mistakes could be incorporated into one single product is truly the most impressive thing about these guns.
Evidently, the Smith's being a sidelock is what qualifies it as a "best" gun in the minds of more than a few gun writers.  Clearly these men either had never seen a true best sidelock, or if they had ever seen one, lacked the discernment to know what they actually were seeing.  In any event, all sidelocks are not best guns and all bests are not sidelocks.  No American shotgun of the classic era, of any grade, even remotely came close to equaling an English best, least of all the Smith.   Once one has seen and worked with actual high quality, it becomes difficult to make excuses for guns like the Smith.  Once one disassembles an L.C. Smith for even the first time, the level of Rube-Goldbergian design readily becomes apparent.
To wit,

1 - The toplever spindle that is supported (barely) by the removable triggerplate at the bottom and held in place by a wholly inadequately sized screw.

If (actually, when) this screw loosens, it allows the toplever spindle to move upward, out of engagement with its insufficiently sized bearing in the triggerplate.  The toplever spring forces the now unsupported bottom end of the spindle to one side and causes the entire mechanism to bind.

2 - Wood that the gun simply can not spare is removed to make room for the toplever spindle, leaving two narrow vertical webs to transmit all recoil forces through the stock.  This entire area is solid in any Holland-pattern sidelock.

3 - Both iterations of the double trigger safety mechanism are poorly thought-out and sloppily made.  In truth, everything in these guns is quite sloppily made.

4 - The cocking system is an answer to a question that nobody asked, and it's not a good one.  The cocking levers are extremely short on either side of their pivots and the rear portion of the lever (the cam) is designed to fail by cracking.  The cocking levers are just that: levers.  They are NOT "torsion bars", they are levers, with one end situated at the front of the action bar and the other end situated ahead of the hammer pivot.  The lever's pivot runs axially, rather than transversely, which is why the long pivot shaft that connects each end is necessary.

Some American gun-rag writer of the past once called this system unbreakable, I wonder how he'd square that opinion with the drawer full of broken cocking rods and cams that I've got.

5 - The first-type ejectors (powered by the mainsprings) must have seemed like a clever idea at the time.

6 - The agri-locks, the primitive imitation of a sidelock with one spring powering the tumbler and the sear.  There is no safety sear (interceptor) and the the mainspring simply bears (scrapes) against the hammer directly, no roller, no swivel link, nothing.  The overall level of quality, even in the highest grades is astonishingly poor.  The higher grades do get spotting (perlage), unfortunately, it's done right over top of the machine marks.  The highest grades also got bridles filed up into "fancy" shapes (very crudely).  There are those who will shout from the rooftops about some "high grade" Smiths being the most expensive guns made at the time.  These individuals clearly equate high price with high quality, which, much like wealth and intelligence, don't necessarily correlate.  In the case of the Smith, price and quality certainly do not.

7 - The use of corrosive flux during barrel assembly that invariably leads to the need to strip and relay the ribs.  When you're making more guns than everyone else, ya gotta cut some corners to keep production up.

8 - The late (single spring) hammergun lock is nothing short of an abomination with its single-legged and single-screwed bridle which can (by design) never serve its primary function of stabilizing the tumbler and sear pivots.

9 - The rotary bolt is a seemingly clever design solution to getting a crossbolt into a small space.
In that regard, it may well be.  It's also pretty good at opening itself if not properly fit or if worn, due to the complementary angles of the bolt and the bite in the rib extension.
The rotary bolt, when everything is properly fitted and functioning, is as good as, but definitely no better than, any conventional bolting solution.

Ironically, the best single trigger that was ever available in a classic-era American double came in the Smith.  This was Allan Lard's patented design, first seen in the guns and rifles of Westley Richards, where it always enjoyed the well-deserved reputation of being a reliable design.   Given the overall average level of gunsmithing (in)competence in America, this is unsurprisingly the most maligned feature of the gun.

Regarding the Model 21,
- The dovetailed lump barrel construction adds ZERO strength materially to the barrel assembly because more often than not the actual fit of the dovetails is quite sloppy. The soft solder joint and two small-diameter crosspins do most (if not all) of the work of holding the barrels together.  Take a look at the photos and video below and tell me again about the strength and precision of the dovetail joint.

- The rib and forend lug joints fail with amazing regularity. In 25 years, I've stripped and relaid as many 21s as Smiths. Trust me that's a lot.  Here's what happens when the forend lug joint fails.


- I don't think that any well-designed gun should have ANY parts that are held in place by staking alone.

 - I personally don't consider the 21 to be a "classic" double, it came on the scene very late, is completely coil-spring driven and is made of much more modern materials than the other American doubles (these are good things). It's really a modern-ish double that is a mashup of Ithaca NID cocking mechanism, Ithaca Flues underbolt and, lastly, just like the Flues, there is no mechanism by which the head of the stock is drawn tightly against the frame.

I'm familiar with the legend of the destruction tests of the 21 against the other doubles and all I can say about that is, it's absurd and lacking in any documentary evidence.  In fact, depending upon who is telling the tale, the 21 was either tested against all of the other American doubles, or it was tested against the best of England, or it was tested against both, or it was simply tested by itself.  The catalog photo of a 21 resting on a pile of fired proof shells proves nothing beyond the marketing department's firm grasp of their job.  But, but, advertising was "honest" back then you say?  A company wasn't allowed to make untrue claims, you say?  Well, here is an ad from the model 21's time period.  It's about as true as the supposed "torture test".


Even if it did happen (which is highly debatable), it's comparing apples to oranges, and no one shoots proof loads regularly. The 21 is a gun, made of modern (for its time) materials, designed to shoot modern ammo and the "test" (if it ever really even happened as they say, or at all) proved nothing except that all of the guns purportedly tested were indeed stronger than they needed to be, for the ammunition and use for which they were designed.  But Americans seem to be obsessed with "strength", whether real or imagined, so I'm sure that that also helped to build the model 21 legend.
If you like 21s, that's great, they're a serviceable gun, but to compare them to classic-era (or English) doubles, is an unfair, uninformed comparison.

Just as with my Smith posting, I did not mention aesthetics or handling. These areas are entirely subjective and I am certainly no one to tell another what they should (or should not) like.

H&R - Hammerless:
Strictly from a quality standpoint, the finest shotgun made in America, at the time. This was the only gun actually built under license from (and under the direct supervision of) Westley Richards, rather than designed to get around their patent.  It was an Anson & Deeley boxlock, available in 10 and 12 bore and, but for the name on the gun, could have passed for an English-made shotgun.
Sadly, there are too few of them to be concerned with (about 2000 were made).  High quality items cost more to make then, just as they do now and apparently the American market was not willing to support the manufacture of a truly high-quality and largely handmade shotgun.

Parker - All hammerless:
Another gun referred to by gun writers as America's "best" and "hellishly complex". From the standpoint of quality of fit, finish, materials and workmanship, these guns are second only to the H&R. I believe that the Parker company's making of generally high-quality consumer goods directly influenced the quality of their guns. From a design standpoint, I would never refer to one as "hellishly" complex but they certainly are needlessly complex. I wouldn't be surprised to find that the designer got paid by the part. The early guns used a complex, multi-piece linkage to translate the toplever motion to the single under bolt.
With all of the pivots and contact points in this linkage one would imagine that wear would "stack up" and result in much play in the mechanism. One would be correct. The latest guns and the Repro dispensed with the linkage all together and operated the bolt directly by the toplever spindle.  The cocking system remained unchanged throughout the guns production (and reproduction) run. This is a system in which a spring-loaded hook in the barrel lump rotates a rocker arm in the frame that causes a slide in the bottom of the frame to rotate the hammers via hooks below the pivots, when the barrels are opened.
This design also requires an entirely separate mechanism to disengage the cocking hook when the forend is removed, this so that the barrels may be removed from the frame.  As I said, needlessly complex.
The cocking mechanism (and the need to house it) is the reason that small bore Parkers look out of proportion. Regardless of gauge, the action remains roughly constant in size (from the action flats on down), giving the smallbore guns a very disproportionate "shrunken headed" look.  The photo below of a 28 gauge frame illustrates this quite well.

It is in the ejector mechanism that I would use the "hellishly complex" description. An entire book/manual could be written about this mechanism. I'll spare you. A common problem with ejector Parkers is that even though the gun is tight on face, the joint will be loose when the gun is open. This is caused by the forend iron wearing the forend lug at the point where they come into contact. This area experiences accelerated wear due directly to the pressure of the ejector mainsprings. For all of its complexity, Parkers really do hold up and other than the aforementioned issue with the forend, failures seem rather random. Most of the repair work that I perform on these guns is fixing the mistakes of others.
Lastly, the single trigger. Most of the American makers offered a proprietary single trigger and Parker was no exception. There were two versions, both mechanical-shift with an inertia safety block.
The late style was a fairly simple and reliable design and was used in the Repro guns.
The Parker Reproduction is a materially superior gun to the original.  That is down to modern materials and manufacturing methods and is meant to take nothing away from the originals.   Thanks to their being somewhat shunned by the official "collector community", their prices are very reasonable.

Ithaca - Flues model:
A very simple, crudely executed design that was clearly optimized for mass-production and assembly by unskilled labor, with its round bolt (easier to drill a round hole than cut a square one), coil springs and simple cocking mechanism that sees (basically) the barrel hook directly rotate the tumblers. This system is identical in principal, and similar in execution, to the way a Fox cocks its hammers.   Broken cocking lugs on the hammers are a common failure in this gun.  Unfortunately, accommodating these internal parts didn't leave an excess of material in the frame. Flues models with cracked or broken frames are not a great rarity, with their very thin sidewalls and minimal, almost token radius at the base of the standing breech, especially in earlier versions.  The trigger box, rather than being an integral part of the triggerplate, is a simple U-shaped stamping screwed to the triggerplate, with the wire anti-rattle spring trapped in between, thus ensuring that the screw will loosen over time.  The safety pivot post is also a separate part, screwed into the triggerplate.  The toplever spring is a coil retained by a headless screw threaded into the top tang, in single shear.  The design also lacks any method for drawing the head of the stock up tight to the frame.  The design does not appear to be meant to have a long service life.  It is truly a vile affront to proper engineering.  Like all Ithaca guns (and all American guns), it was available in some high grades with very well done engraving.  Sadly, the engravers' work is all but wasted on this very poorly designed farm implement.

Ithaca - NID model:
The ultimate Ithaca double with a rotary bolt (a la Fox and Smith and sharing the inherent faults of that system), cam and pushrod cocking system (used in whole on the Model 21) and a whole lot of metal everywhere.
Clearly, this is a simple and strong design that should withstand much use and not a little abuse. The factory single trigger that was offered in these guns was actually a Miller trigger, built by Ithaca under license. The Miller design stands out as maybe the simplest single trigger ever devised and they are trouble-free as long as they aren't messed with by incompetent fingers.

A.H. Fox - All
I'll start at the top and work down. This design uses a rotary bolt, same as the NID (in fact, part-for-part, the design is identical) and the Smith. Where Smith decided to support each end of the toplever spindle in separate parts of the gun, Fox (and Ithaca) supported the toplever spindle wholly in the frame of the gun. The result was much more rigid and compact. The Fox cocking system is simplicity itself and while it shares the direct-acting principal with the Flues model, the execution is that used by W.W. Greener in their Facile Princeps boxlock. The hammers are still directly lifted by the barrels but that part of the hammers that actually engages the barrels has been "jogged" inward toward the centerline of the gun. This serves two purposes. First, it makes room for the coil mainsprings and the plungers and struts that drive the hammers.
Second, it allows the action bar to be rounded, again mimicking the Greener. Stepping out of the analytical and into the aesthetic for a moment, this is what I believe gives a Fox its superior "proportions" and shape.
The Fox uses coil springs throughout with the exception of the sear spring which is a leaf spring. Fox used a derivative of the Baker system for its ejectors, very simple and easy to time and repair.
Fox's "house" trigger was the Kautsky. THIS was the trigger that should have been in the Smith. A Rube-Goldberg fantasy of cams and levers that is the single most difficult trigger to regulate that I've ever encountered.
Perhaps it's poetic justice that the Fox and Smith used the triggers that they did. The Fox is not without flaw however. The opening of the barrels is checked by the forward portion of the hammers stopping against the inside of the action flat (I loathe the idiotic term "water table"). When the barrels are allowed to simply fall open, this puts great stress on the frame at that point and often results in cracking of the frame along the barrel hook opening. Obviously this would be considered abuse but it does happen.  The occasional "self-opening" rotary bolt can also be an annoyance.  If you've got a Fox that applies its safety after the first shot, it's probably the bolt that is at fault.

Part Two: