Colt Prototype Self-Ejecting Revolver

RIA’s catalog page for this revolver

Robert Roy was a career Colt employee, who began his work as an engineer in 1963 (including work on the 1971/SSP pistols and the CMG machine gun series) and retired in 1993 as Director of International Sales. One of his side projects appears to have been experimentation into auto-ejecting revolvers. This proof of concept revolver has a gas port added to the barrel and a gas tube which vents gas directly in the 2 o’clock chamber each time the gun is fired. That gas blows right into the previously-fired cartridge case, ejecting it out the back of the cylinder through a spring-loaded aluminum deflector/cover.

In theory, the system seems like it should work just as intended, although I have no information about how successful it was for Roy. The practical problem with such a system, however,r is that it cannot eject the final round, as the cylinder is them empty and there is no additional cartridge to provide the gas to eject the last one. Thus the cylinder must be opened and the ejector rod used to eject the final case – and there is really no difference to the shooter between manually ejecting one case and manually ejecting all six. So the added complexity doesn’t really provide a practical benefit.


    • A partially loaded cylinder could be interesting. Say a shooter went click bang. With no empty case to contain the gas, the shooter might get a better look at the tail end of the gas tube than he likes.

      • “shooter might get a better look at the tail end of the gas tube than he likes.”
        Then so called CUPFIRE system might be interesting starting point for self-ejecting revolver:
        as it would throw spent cases to front, however would need more complicated gas system, possibly with 180° turn to push case in proper direction.

    • This may be a valid idea:
      start with empty closed gun, push in a cartridge in the left side, fire, load, fire, load and repeat and repeat. . .
      essentially the top three spots in the cylinder being in use load-fire-eject
      there looks to be cut out on the left side to facilitate loading in the left side

      essentially a single shot wheelgun

  1. Is there actually a way to NOT eject the 6th round on a freshly loaded cylinder? It seems to me that, if you reload and then fire the 1st round, without being able to cut off the gas impingement, you’d eject your 6th live round anyways.

    • That’s the problem with basically every “automatic-ejecting” revolver that anyone ever designed.

      If you start with a fully-loaded cylinder, the first shot ejects a live round. In the case of a Colt, which rotates the cylinder clockwise as seen from the rear, it will eject the live round in Chamber No. 6 when No. 1 is fired, eject the fired round in No. 1 when No. 2 is fired, and so on.

      The only way to avoid this with a Colt-type revolver would be to follow the standard Peacemaker operating procedure of “Load one, skip one, load four more and lower the hammer onto the empty chamber”. With “five beans in the wheel” the first round being fired will expend its gas on the empty chamber. Of course, that carries with it the risk of the shooter getting the blast of gas in the face, as well.

      With a Smith & Wesson type revolver, which rotates the cylinder counter-clockwise from the shooter’s POV, it’s worse. An ejection at the 2 o’clock position means that when chamber no. 1 is fired, chamber no. 2 is ejected. Pull the trigger twice to get to No.3, and watch it eject the live round from No.4, and so with No.5 and No.6. You end up with six trigger pulls, three shots fired and three live rounds at your feet.

      To work “correctly”, ejection on a S&W type would have to be at 10 o’clock as seen from the rear, and again you’d have to leave the chamber under the hammer empty. Since S&W revolvers swing the cylinder out to the left just like Colts, at least part of the gas-ejection system would have to be built into the cylinder crane. That would be an interesting design issue, and even more interesting in terms of machining.

      Far from simply being Col. Cooper’s “ingenious answer to a nonexistent problem”, it’s more of a “nonsensical answer to a question so stupid nobody with any sense ever bothered to ask it”.



      • With the two open chambers on the left side, you could reload the empty chambers as they came round, and after the first three live rounds, the gun would only eject spent cases, though in a click, bang, click, bang, click, bang, click, click, bang cadence.

  2. Self ejecting revolvers might be acceptable and practical with self cocking and self rotating features like Webley Fosbery.

  3. “Thus the cylinder must be opened and the ejector rod used to eject the final case – and there is really no difference to the shooter between manually ejecting one case and manually ejecting all six.”
    Well, so it is answer to question no one make. It could make more sense with fixed (not-swinging) cylinder and ejector rod in Colt 1873 style, but anyway advantage over then existing swinging cylinder would be dubious at best.
    If anyone desperately needs self-ejecting revolver (lets say it is revolver) DARDICK HANDGUN SERIES 1500:
    giving some real advantages (like capacity) and disadvantages (time of loading from fully empty to fully loaded).
    Anyway for other “self-ejecting revolver” design see BECKER shotgun:

    • If the Dardick had used a detachable box magazine, it would have solved the whole “too much time to reload” problem. Although in reality, the Dardick action would be best suited to a high-RoF weapon like an aircraft cannon, especially if using modern Plastic Cased Telescoped Ammunition.

      Add “Combustible” to that, plus electric ignition, and you might have the ultimate DEFA, ADEN or Gatling/Vulcan type 20 to 30mm cannon.



      • “Add “Combustible” to that”
        Case-less? If you could make such cartridge which would fit into standard revolver, that would be complete solution of need to eject spent case, as there would be none.

        • Not quite; you still have to cope with duds. Some way to cycle them out and eject them would be needed. Also, heat dump from the chamber(s) would be a consideration to avoid cook-off. So some form of “ejection port” would be necessary, if nothing else just to cool the breech area via air flow-through.

          Also, gas sealing would be a consideration. The most likely solution would be a combustible plastic that burned at a slightly slower rate than the actual propellant. Ideally, ignition would be at the front of the propellant charge rather than the back, to ensure a “front-to-rear” burn of both propellant and casing material. With electric ignition, say by a contact “ring” of sintered conductive material moulded into the case, this should be fairly easy to manage.

          Note that “front ignition” is nothing new. In the past there were some rifle cartridges that had a long, narrow axial tube built in to conduct primer flame to the front of the powder charge to ensure that it burnt “backward” rather than “forward”. This feature was mainly seen on proprietary or wildcat rounds intended for benchrest shooting. Electric ignition “up front” would obviate the need for such measures.

          NB that one of the often overlooked advantages of PCTA, combustible or otherwise, with electric ignition is that it would allow the use of liquid propellants with considerably greater energy yield than any propellant powder in use today. Metallic cartridge cases filled with liquids tend to leak;a moulded plastic case, assuming no chemical reaction with the propellant, would be less likely to.

          Storable liquid rocket monopropellants such as the hydrazine family;

          Have been used in the past as experimental liquid gun propellants, notably in artillery pieces. With such liquid propellants, small-arms muzzle velocities could easily exceed 2,000 m/s (6,500 F/S+), reaching the point where atmospheric friction ablation would have to be considered just as it is with modern APFSDS ammunition for tank guns, such as the M829 for the 120mm gun on the M1A1 and M1A2 Abrams MBTs;

          At such velocities, optical sights would not require any trajectory adjustment at ranges under about 600 meters, and “expanding” or etc. bullets would not be required to generate enhanced wound trauma.

          When a slug moving at over 1 mile per second hits, a second shot is rarely necessary.



          • I was thinking about classic revolvers, so it should not heat very much and also dud would have to be removed, but not necessarily instantly.
            Problems with gas-seal holds true, but for which front-loading might be solution, as long it would accept “center-fire hammer strike” as input it should be achievable with cylinder swap, but still there is question how to head-space…

  4. I wonder if the legendary Jeff Cooper would have counted this as “a brilliant solution to a non existent problem” ?

    Arguably, the fundamental design of conventional revolvers is a legacy from muzzle loading days.

    • “Silver and Fletcher mechanism as fitted on Webley RICs in the 1890’s”
      Also, this make more feasible option when compared against background.
      In 1890s there were fixed cylinder design, relatively slow to unload spent cases and load new, top-breaks having, well “broken” frame, which does making it strong and durable more complicated and swing-out cylinder design was newcomer. So improving reload speed of first must look to be logical solution.
      On the other hand we have this… thing, in times when you could have L9 automatic pistol with 13 rounds, in which you can unload empty magazine or load fresh magazine, in time like one second and without quirky ejection. No, this could not be success and in my opinion such idea could be rejected yet at state-of-art review.

      • Context is key, but self-ejecting revolvers are definitely more trouble than they are worth. Another approach to self-ejecting was the recoil-operated Landstad revolver, with a flat rotor in place of the cylinder. It was a miserable failure, but unlike all other revolvers that ejected spent rounds automatically, the Landstad would eject the last casing when empty.

  5. I could see something like this working if a gas or recoil action cocked the ejection mechanism but that the next pull of the trigger released it. That way you could eject the last empty with a pull of the trigger on an empty chamber. It doesn’t change the fact that it’s the loading, not the ejecting, that takes the most time in any case.

  6. I was in a sporting goods store in Denver in the late 60’s and an old guy came in wearing a security guard uniform and carrying (in his hand) a nickel plated .32 of some sort. He said he had only 3 cartridges and wanted to buy 3 more. The clerk refused to sell him just 3 and he left. I have always regretted not asking him how he carried the three. bangbangbangclickclickclick or bangbangclickclickclickbang.There are a couple of other ways to carry 3 shells in a sixshooter but they seem worse although clickclickclickbangbangbang would have some value if the guard was being tried for murder. “Geeze, I gave him every chance I could!”

  7. The Silver and Fletcher mechanism sounds interesting, but I only find vague pictures of it online.

    Strange that the extra, wowie-zowie ejection would be added to a Webley made gun…When they already made an auto ejection much quicker than any side-loading revolver. The RIC guns don’t seem to break open, so I don’t know what S&F’s mechanism was replacing.

  8. Two points, first, it would be handy for people who do not shoot revolvers dry. Much faster and easier than pushing the ejector rod to push up the empties part way then pulling them out one by one. Not everyone shoots everything dry in combat style shooting, especially hunters and casual target shooters.

    But, other point, a fatal flaw is this design could put soot into the ejector star, recoil shield, hand, etc. Those areas need to stay clean. Can you imagine shooting Unique powder (AKA flammable dirt) back in the day in this thing?

  9. Here’s a thought… With some minor changes to manual of arms or doctrinal approach, this could, in its current configuration, be fairly practical. Let me explain.

    Imagine for a moment that this 6-chamber firearm were to be considered a 5-shot weapon in practice, much in the same way that old single-actions were generally carried with the hammer down on an empty chamber for safety. Loading from a 5-shot speed strip is still fairly quick compared to individually loading cartridges into the cylinder, and if you index that empty round to the 12 o’clock position upon loading, this could still eliminate the step of having to eject the empties.

    Would I use one instead of a traditional revolver? Hell, no. But I still see some practicality in it.

  10. I recall seeing online a rising block repeating pistol in a pocket pistol configuration which used the same vent between chambers approach to eject spent cases from the previous shot and it used part of the hammer or hammer spur to deflect the cases.

  11. I believe the missing cylinder latch is because of the milled clearance for loading a fresh round. this is like the Swiss revolver you showed before

  12. I saw a video several years ago of a Ruger single action modified with a gas port to operate the ejection rod to kick spent cases out the opened loading gate. On the plus side this doesn’t blow gas out an empty chamber

  13. Now we need Ian to do a video on self-ejecting revolver mechanisms that dime-novel authors used to love to use.

  14. I have no doubt about Robert Roy was a great engineer, he designed de excellent Colt SSP pistol and CMG LMG, but this revolver is a nonsense and lacks any originality. The idea had already appeared in Spain in the 19th century where Spanish Artillery lieutenant Luis Ibarra Cortazar patented in 1876 a revolver with gas ejection, latter followed by another artillery officer, Clotaldo Piñal Rodríguez. But the drawback was the same, in the first shot they ejected a full round, unless you left an empty chamber. The Spanish Army preferred Orbea copies of S&W Russian or Model 3 DA self-ejecting revolvers that solved the problem of ejection and permitted a faster loading.

Leave a Reply

Your email address will not be published.