Clear all

Reproduction Webley 'Mercury' air pistol

Curator in Chief Admin
Joined: 5 years ago
Posts: 6401
Topic starter  

Reproduction Webley 'Mercury' air pistol 

See also:

Lincoln Jeffries patent pistol 
Haenel Sport Modell 55R 
Frank Clarke twist grip 
Webley Whiting 
Hill patent pistol

Thunderbolt 'Senior' (1947 patent)



This pistol was only made in prototype by Webley and two examples were sold in the Webley Museum sale in June 2006.

See here.

With thanks to John G, who posted the following on the

“The idea of the project was to make a working replica of an unusual Webley spring air pistol design that almost made it to the market place around 1980, but couldn’t compete with the Webley Hurricane, Typhoon and Tempest of the time. This break-barrel pistol had a two-stage cocking system and was patented in 1981 by Webley and their Technical Director, Harold Resuggan. However, the original design dates back to ca. 1960, and was the brainchild of Stanley G. Holder of the Webley Design Department. His original drawings (dated 1960) were adopted for inclusion in the 1981 patent, but curiously no mention is made of his name in the patent. The original prototype was progressed sufficiently in the 1960’s to be given the working name “Mercury” and the only known surviving example Is in fact impressed with that name. By the time the Resuggan version had been developed the name “Mercury” had been abandoned and the only known prototype has no name stamped on it. However, purely for convenience I am going to refer to both versions as the “Mercury” throughout this account. Those of you who have read my earlier posts on scratch-building springer pistols will understand why the “Mercury” pistol was on my to-do list. I am fascinated by any extinct design that was patented but not commercialised, and in the majority of cases making replicas from the patent data is the only way of understanding how these guns functioned and what their advantages and shortcomings were.

Before updating on progress, I thought I should say a bit about the scene of the crime – my workshop, in case you think I have all sorts of fancy machine tooling at my fingertips. I am not an engineer or machinist, so my workshop is pretty primitive and something that anyone could have for themselves. No cnc machining, 3D printing, MIG welding, not even a milling machine. So I tend to have a make-do and mend approach and improvisation is often the name of the game. The workshop measures only about 8 foot by 20 foot, so is pretty cramped as you can see from the first group of pictures.


Taking a 360 degree scan, you can see that I have a lathe, the absolute must-have for any amateur gunsmith. Mine is a 60 year-old Myford Super 10 - which cost me £700 from someone who was emigrating to New Zealand. A bit long in the tooth and worn in the bearings and gears (like me), but when you get used to its peculiarities it can do a passable job. It does have a swivelling vertical slide, another essential in my opinion, which I use for all my milling work. Then there is my work bench with a reclaimed solid mahogany top fitted with a large swivelling vice. (If you are going to get a vice, get a swivelling one. I can’t count the number of awkward situations the swivelling action has got me out of.) The drill press, you can see, is hardly ever used, except for rough household/gardening jobs as the lathe gives you so much more control when drilling. Bit of a waste of money and bench space in my opinion, as I could just as easily use my vice and Black & Decker to do most of the things I use it for. The combined belt and disc sanding machine was a great buy at £90, and I use it a lot - squaring off edges, deburring, contouring, sharpening tools. Although these things are advertised as woodworking machines, I use them for metal without any problem and the discs and belts last for ages (I have been using mine with steel for over a year and still have the original belt and disc), and in any case they are cheap to replace.
As you can see, I use a lot of cupboard and drawer space, and it is all made up from discarded kitchen units etc. scrounged from wherever I could. You will also gather that I like to use rather a lot of chemicals, but I am a chemist by training and old habits die hard. The mini-furnace was a lucky find and has proved incredibly useful for heat treatment, casting etc.

Anyway, getting back to the project itself. As always, I start from the patent drawings and scale them up to life size (Photoshop is great here). Patents never give absolute dimensions, but in this case I was able to work them out both from the barrel calibre of .177 inches and from information from a reliable source that the barrel length of a prototype was 170mm. Both methods gave the same results so I was pretty confident of my final measurements.
(For patent details see , post #22)

However, there was a problem – there are two known prototypes of the Mercury and they are different in some cosmetic respects. One of them closely resembles the patent drawing and quite frankly looks rather weird. The second, later one has sleeker lines, and I decided to go for this one, not only because I preferred its looks, but also because it was going to present me with fewer constructional problems.


Making the basic carcase is the usual starting point and is always the hardest stage and so it is well worth spending some time over its planning . Patent drawings often do not make it clear how the carcase of a gun is put together, and the Mercury patent is no exception. The drawings appear to suggest a single casting for the cylinder and cylinder housing, with no steel insert, and it would presumably have been made from an aluminum alloy like the contemporary Hurricane. However, photographs of the second prototype, which I wanted to make, show a cylinder that is a separate entity (presumably made of steel, which is more practical than alloy) enclosed in a grip-frame (most probably alloy), rather like the wood or plastic grips of the Diana 5 pistol and other similar break-barrel pistols. So this was the route that I was going to take. I decided to make the grip frame that houses the cylinder from three separate parts – an inner steel frame, in which the cocking links and trigger/sear system will be housed, and two outer aluminium cladding panels which will be purely cosmetic and give the frame its proper shape. Aluminium to keep the weight down. The inner steel frame will be attached to the cylinder by screws as shown here:


In a previous post I showed the grip frame that had been crudely cut out from steel plate, and it now looks like the picture on the right. The top has been milled out so that the cylinder sits snugly on it. Still a long way to go though, and more metal will need to be removed from the frame as it still weighs a ton.


The next component to make was the barrel. For this I used a section of an old Relum air rifle barrel (0.177, rifled), and it was turned down on the lathe to give a stepped profile, as showing in the next group of pics. The steel barrel housing block was drilled to accept the turned down section of the barrel as a very tight fit, in fact so tight that it could only be inserted after heating the block with a blow torch. After cooling that barrel wasn’t going anywhere, but just to make sure a small hole was drilled through the block and a steel pin knocked in to keep the barrel permanently in place. The small step that had been turned down on the breech end of the barrel produced the necessary recess for the eventual breech seal. The wider section at the muzzle will later provide the base for the front sight, to copy the appearance of the prototype.


The next thing I wanted to tackle was the cylinder unit. My starting point was a section of precision (seamless) steel tubing, sourced from the Bay, with inside diameter 25mm and outside diameter 30mm. First a slightly oversized plug was turned down from mild steel to make the jaws for the barrel. This was sweated in place at a dull red heat using a blow torch and finally pinned. A slot was then milled in the plugged end of the tube to make the jaws, making sure that the jaw gap gave a sliding fit for the barrel housing, and finally a hole was drilled through the jaws for the barrel pivot pin. The pivot pin was made from silver steel rod.


The barrel unit and cylinder unit were now ready for fitting together, but the mating faces first needed to be milled to an angle of to 8 degrees (as in the patent). This needed to be done fairly precisely in order to give a nice tight fit when closed and, most importantly to end up with the barrel exactly in line with the cylinder.


The next phase of the project was the one I was least looking forward to – the spring detent barrel catch. I have had bad experiences with these and if things can go wrong they usually do. The problem is that there are so many variables - for positive opening and closing action and for rigid closure the two angles on the siding wedge and the angle on the fixed wedge have to be just so; the positioning of both is critical for the barrel to be perfectly in line with the barrel. You can’t always rely on the patent drawings for this and I have come across examples where the illustrated arrangement just does not work. In the present case, the patent drawings gave a good starting point, but I did have to change them a bit, and this proved costly in terms of time. I ended up making three sliding wedges before I got the angles right. The following pictures show the development of the catch. I was quite happy with the final result, as it gave the barrel a crisp break action and in the closed position the barrel was held very firmly in place. Even so, I still don’t like making detents.


With the barrel now firmly located on the cylinder I was able to drill the air transfer port (3.5 mm diameter), after first locating exactly where to drill by inserting a long pointed steel rod down the barrel and punch- marking the end of the cylinder.

The cylinder now had to be fitted with a rear plug, which on the prototype was held in place by a pin rather than a thread. For this I needed some 30 mm steel round bar which I could turn down to 25mm to be a push fit in the cylinder. Luckily, (so I thought), I had a decent length of 30mm bar to use, but as Sod’s Law would have it, it turned out to be half a millimetre undersize. Not enough to worry about? Yes, it would have looked awful to have a rear plug that was not perfectly flush with the cylinder. Unfortunately the only round bar I had that was wider than 30 mm had a massive diameter of 40mm. So I used this, making a series of steps from 40mm to30mm to 25mm to 12mm on the lathe (the last step was to provide a short spring guide). You can’t believe how much swarf this made. When my better- half popped her head round the workshop door and asked what I was doing, I said I was going to make my own Brillo pads as I had heard rumours that people were panic buying and there would be a shortage. She went off happier than if I had tried to explain the truth, but still probably thinking I was mad.


The plug and cylinder were then drilled through to take the fixing pin (4mm silver steel rod). This summarises the whole sequence:

The cylinder barrel assembly was now ready to be attached to the grip-frame. This was done by means of screws (one 2BA at the rear, and two 5BA screws at the front).


The cylinder barrel assembly was now ready to be attached to the grip-frame. This was done by means of screws (one 2BA at the rear, and two 5BA screws at the front).


So the final carcase assembly currently looks like this:

Curator in Chief Admin
Joined: 5 years ago
Posts: 6401
Topic starter  

Reproduction Webley 'Mercury' air pistol part 2 

Thanks to John G.


In Part 1 of this series of threads on scratch-building a copy of the Webley Mercury pistol, I had got as far as making the basic frame, cylinder and barrel components:



Since then I have been carrying on with the project (among other things), during lockdown and thought now would be good time for an update on progress for those of you who expressed an interest in the project.

The next phase after building the basic carcase was to make the working parts of the pistol: the cocking links, piston, trigger and sear. But first I had to decipher the patent, as like most patents it was a bit short on detail. This can be the fun part of a project like this, trying to work out exactly what the drawings represent and how the mechanism actually works. Quite often there are things in the drawings that are not mentioned in the text, and just when you have decided that they are irrelevant and you leave them out of the build ,you find that they are important after all. Sometimes you can see important things in the drawings that are unclear, but there is no helpful description in the text, so you have to work things out for yourself. A good example in this case of this is the shortest of the three cocking links in this part of the drawing, marked as No 21, which the patent calls a “pawl”.(I looked it up and ‘pawl’ is as good a description as any).



This has to press firmly in the notches in the piston during the two cocking strokes (items 22 and 23 in the drawing), but then has to be set free when the barrel is returned home. The patent describes it as “spring-loaded” but there is no indication of the design of the pawl or what sort of spring is used or how it is fitted into the pawl. In the end I took inspiration from the short cocking link used in Webley pistols, such as the Senior and Hurricane, and made it as follows:



Once made and fettled a bit, it seemed to work well, so it must have been along the right lines. It was milled from carbon steel, so that it could then be heat hardened, which is important as it is going to have to take a lot of stress and wear. The long and short cocking links (items 19 and 20 in the above drawing) were also made from carbon steel for strength, using bar of 5mm thickness. This makes for harder work cutting and filing, but it is worth it in the long run for components that are subjected to a lot of stress. These links needed to be a sliding fit in the channel that had already been milled into the steel frame. Then all three components could be pinned together and the whole unit checked for free travel when attached to the barrel.
I should point out that I am making the second version of the Mercury, not the first which is the one in the patent, and there are a few unknowns to deal with. I only have one photograph of the second version to work with (and a knowledge of its barrel length), and it was clear from the start that some of the external dimensions were slightly different from the patent version, and also that there would be some minor differences in the internal layout. This meant I could not just work straight from the patent drawings, and a lot more thought had to go into working out dimensions and the relative positioning of moving parts.
This was particularly true in the case of the cocking link train and the piston. The cocking link train had to be an optimum length to maximise piston movement, and the positions of the two cocking notches and two sear-engaging notches on the piston also had to be precisely spaced for the whole cocking process to work. All this caused me a lot of head scratching but I finally came up with what I thought was a reasonable compromise.
Now for a confession. It’s a sad fact that, whenever I make components from carbon steel for a project, after a lot of hard work and fettling I always end up having to make at least one of the components again, and on occasion even three times. Inevitably something unforeseen crops up later down the line which makes the part unusable – too short (never too long, which would be easy to sort out), pivot hole in the wrong place, wrong shape, you name it. I always try to think ahead and anticipate these snags, but never succeed totally. So when I am sweating away hacksawing or filing a very hard steel piece, putting a lot of thought, time and effort into it, it keeps coming into my mind that this could be the one that ends up in the scrap bin. Very off-putting! And the current project was no exception. When the assembled cocking link was tested with the piston in place, it lined up exactly as shown in the patent drawing shown above. The trouble was, when the gun was cocked and fired, the piston smashed into the spring-loaded pawl, which was protruding into the cylinder (it was being held there under the action of its spring). The patent diagram, which shows the pistol in the uncocked or fired state, was obviously incorrect. Just my luck. It should have shown the pawl sitting slightly ahead of the piston notch, so that it had ridden up against the end of the cylinder slot and had been lifted out of the cylinder cavity. That way, the gun could be fired without damaging the pawl and/or the piston head.
So despite all my best intentions, the cocking link train I had so carefully made needed to be shortened by about 5mm, and as the short and long links had already been drilled and profiled, the only way I could do this without weakening the linkage was to make a new short link, 5mm shorter than the original. So it was back to the hacksaw and file So the cocking linkage pictured in the next group of pictures is actually my second attempt.



You can see from the next pic how the linkage looks when fixed to the barrel and cylinder. When the cylinder is eventually fixed to the grip frame, the linkage will sit in a channel in the frame and the spring-load pawl will be pressed flat against the cylinder.



With the cocking linkage made, attention could then be turned to the piston. I have found that the simplest way to make a piston is to use steel tubing that is a loose sliding fit in the cylinder and then to block one end with a steel plug. This requires a lot less effort than boring out a steel round bar, especially if tubing of the right diameter can be bought. In this case my cylinder had an internal diameter of 25 mm, and 25m tubing suitable for the piston was readily available on the net. The plug for the piston head was turned down from steel bar until it was slightly oversize to fit inside the tubing. The tube was heated to a dull red heat and the plug tapped in place. As a belt-and -braces way of further securing the head in place, three holes were drilled into the head from the side, tight fitting steel pins hammered in, and then the excess cut off and the pins filed flush . The head was then drilled and tapped to allow a piston seal to be fitted.

As a first stab at a piston seal, I decided to use PTFE rod, and to turn one down on the lathe. I have never used a PTFE seal in my repro’ pistols before,as leather was always more appropriate. Anyone any thoughts on PTFE versus leather? Anyway, I will give it a try and if I am not happy with it, I can always go back to leather.

The final parachute-style seal was fitted to the piston with a countersunk 2BA screw and washer. The final piston was a close sliding fit in the cylinder, and when pushed down hard into the cylinder, it held the compressed air well. The sequence of events in making the complete piston is summarised in the following pics. (Note that the actual thickness of the plug used to form the piston head was about one half of that shown in the first frame).



The piston once made then had to be cut with four notches: two to receive the spring-loaded cocking pawl, one for each cocking stroke, and two for catching on the sear after each cocking stroke. It is interesting to see how the double cocking action works, and you can understand this from the following drawings.



The piston notches shown in red and labelled as A1 and A2 are the ones that the cocking pawl locks into. The notches shown in blue, labelled B1 and B2 are those intercepted by the sear.
(1) shows the gun before cocking. (2) shows the first cocking stroke, where the cocking pawl has connected with notch A1 and pushed the piston back until it is held by the sear in notch B1. The barrel is then returned home to give situation (3). The piston is now cocked in the first stage, and if wanted, the gun could be fired for a low-power shot. When the barrel is pulled back again for the second stroke, the pawl engages with notch A2, as shown in (4) and the piston is pushed back again until the sear engages with notch B2, giving situation (5). The barrel is then returned home, as in (6) and the gun is now ready to fire on full power.

As you might imagine, the relative positioning of these notches is critical for the gun to cock properly, and also to take full advantage of the available swing of the barrel to ensure a maximum swept volume. Also the angle of the notches was important to ensure that no slippage might occur during cocking. Needless to say, this stage caused a lot of puzzling out and modelling with cardboard cut-outs before committing the piston to notching. When I was reasonably confident that I had the layout correct, the piston notches were then milled in and it was with huge relief that a trial run showed that all was OK.
The piston also needed a slot milling along its top, exactly opposite the train of notches, so that a screw protruding through the cylinder wall would ride in the slot and so would keep the piston notches always in line with the cocking pawl and the sear. If this feature was omitted, the piston could rotate out of line with continuous use of the gun, and the gun would then fail to cock. The only recourse would then be to disassemble everything and realign the piston, not something you would want to do after every dozen shots or so. The patent makes no mention of this feature in the text, but once I realised it had to be there, I was able to find it the patent drawings, with its guiding screw. The lower picture shows the guide slot.



The next step was to make the trigger and sear pieces. I used the same 5mm carbon steel plate that I used for the cocking links. Although this made hacksawing and filing harder than if I had used mild steel, it did mean I could heat harden them once finished and they would stand up much better to wear and tear. This was particularly important for the sear. This is the sequence in making them. They were first made oversize and only roughly shaped, and with the holes for the pivot pins also drilled, they could then be fitted onto the frame with the pins and then sized and shaped more accurately. The long arm of the sear lever was also drilled with a hole for attaching a spring. The last picture in the sequence shows the trigger and sear in place, with the sear spring fitted.



Now was the moment of truth, as I had all the necessary bits to assemble the working parts of the gun and to see if the gun actually cocked and fired. So using a weak main spring (as I did not want to put the load bearing parts under much stress until they had been hardened) the gun was assembled:



With bated breath I broke the barrel and cocked the gun. Yes, the pawl picked up the first piston notch OK and pushed the piston back until there was the reassuring click of the sear engaging. The gun had successfully achieved the first cocking stage! I then returned the barrel home and re-cocked the gun. Yes, the pawl picked up the second notch and pushed the piston further, but disaster! Everything came to a dead stop and there was no way I could get the sear to engage with the second notch. So the gun was taken apart and checked. The notch distances seemed OK, there was nothing obstructing the cocking links in their channel, and nothing blocking the trigger or sear movement in the second stroke. What the hell could it be? After reassembling and hitting the same problem again, I agonised over the problem, untiI I realised what an idiot I was. When I made the piston, I deliberately made it a bit longer than necessary so that I could trim it to an optimum length later on. I had forgotten this. It turned out that it was just that bit too long and was hitting the end of the cylinder in the second cocking stroke, before the notch could engage with the sear. So out came the hacksaw and with 5 mm trimmed off the end of the piston, the gun worked perfectly, at least under very light spring load. The sear now engaged after both cocking strokes, and the trigger worked fine so that the gun could be fired after the first or second cocking stroke. Testing the action proper with powerful springs would have to wait until the rest of the gun was finished and all the relevant parts heat hardened. This would include case hardening the piston, as it was made of mild steel and the notches would become rounded, and pretty soon useless, if some protective measure wasn’t put in place.

The next aspect of the build I wanted to tackle was the “cladding” that was to go around the grip frame and help make the pistol begin to resemble the original prototype more closely. I had decided that the cladding would consist of two aluminium panels screwed to each side of the frame. Aluminium was chosen, as it would not add too much extra weight to what was already promising to be a heavy pistol, and also it would go some way to representing the aluminium alloy casting that the prototype was (presumed to be) made of.

The following pictures show the processes involved in making and fitting the side plates. Each plate was profiled to give a snug fit against the curvature of the cylinder, but their inner surfaces were milled out in steps so that their weight was reduced as much as possible. The plates were then drilled and screwed into tapped holes in the frame.



So that is where I am up to right now. There is still quite a way to go, including making the grip plates, trigger adjuster, safety catch, and front sight. Fortunately the rear sight used on the original prototype Mercury (second version) was the same as the Webley sight used on the Hurricane pistol and other Webley airguns of the period, and as I happen to have one of these (which I snapped up on this forum recently) that will help save me some time.
Once all the parts have been made, fitted and polished. I will then have to etch the lettering on the gun, blue the steel parts, and anodise and black the aluminium cladding. Only then will I put the gun through its paces, experimenting with different strength springs (the prototype used two concentric springs), piston seals, transfer ports etc. It will be interesting to see how the gun performs and how effective the double cocking action is. All will be reported in Part 3 of this mini-marathon, unless of course the gun completely disintegrates on its first serious outing.

Postscript: I find I have only used up 11 of my 12 pictures allocated to this post, and so I have a spare. I thought I would share with you my all-time favourite piece of workshop kit - a magnetic sewing machine light, costing about £7 off Fleabay. For ages I had struggled with getting adequate lighting for close-up work in my workshop, trying out portable fluorescent tubes, overhead anglepoise etc. but always I would get shadows on the workpiece and could never rely on getting a clear view of close-up work in the vice when hacksawing, filing etc. The same with precise setting up work on the lathe. (You can get purpose lathe lights but they are very expensive and not all that versatile). Then I came across these sewing machine lights. They have a really powerful magnetic base and can be stuck virtually anywhere on the lathe or bench vice, and being flexible you can focus the light in the most tricky situations. I have had mine for a couple of years now and wouldn’t be without it. The best 7 quid I ever spent on a gismo for the workshop.



In late  July 2020, John posted this on the

"It has been more than a month since I last posted, so in case some of you who have been following this thread are wondering what has been happening, here is a quick update.

Everything was going well until it came to milling a slot in the barrel for the front sight. I took meticulous care in appropriately clamping the barrel/barel housing unit in the lathe (difficult due to its awkward profile) and ensuring that the slot would be perfectly aligned with the barrel sight line. I was also very careful to ensure that the slot did not go too deep and penetrate the bore. Unfortunately I was so preoccupied with these details that I didn't spot the elephant in the room. The milling was proceeding slowly and smoothly and I was starting to congratulate myself on a tricky step completed, when I realised that I was milling the slot on the underside of the barrel. I won't bore you with the details of what went through my head, or what I said at the time, but it wasn't "Oh dear me!"

I tried to salvage the barrel housing/detent unit, but I had fixed the barrel in too well and couldn't remove the barrel without damaging the housing drastically. So I had to bite the bullet and make a complete barrel/ barrel housing/ detent all over again. Not something I relished.

I have also had to make a new piston from scratch, as I found that the precision steel tubing I had used was just that little bit too slack in the cylinder. Normally this is not a bad thing in a springer as you only need the piston seal to be a close fit, but in this design you need maximum contact between the cocking link and the piston notch to take the cocking force. Any excessive play and the cocking link can ride out of the slot. I found that with a powerful spring in the cylinder there was a tendency for slippage and deformation of the piston slot.

As I could not buy any steel tubing with a closer fit to my cylinder, or any with an adequate wall thickness that would let me turn it down to the right diameter, I had to make the piston the hard way, by turning down solid bar to the right diameter and then boring it out. Anyway, this has now been done and the new piston works fine, with no signs of slippage. The cocking link has been heat hardened, and the piston itself case hardened, so there should now be little wear with frequent use.

So the gun is more or less finished, and all that remains to do is some checkering to the grips, the lettering to be etched, and the whole thing blued. When it is completely finished and its performance assessed I will put together a final account in Part 3. I must admit, I am getting to like the dual cocking action more than I thought I would."





Curator in Chief Admin
Joined: 5 years ago
Posts: 6401
Topic starter  

Reproduction Webley 'Mercury' air pistol part 3

Thanks to John G.



It has been a while since my last post, but I have now managed to finish my Webley Mercury pistol project, and for those of you who have been following the ups and downs of my efforts, here is a brief account of how things went.
The following pics show the second model prototype gun that I was trying to copy on the left, with my pistol on the right, as it stood at the end of Part 2.
( )


There was obviously the need for front and rear sights, chequered walnut grips and a safety catch, as well as appropriate etched lettering and an overall black finish.

The strategy I chose to make the front sight unit was the simplest I could think of that was within my limited capabilities, and was certainly not the one used for the original pistol. But it gave the right cosmetic effect so I was happy with it. It is summarised in the following set of pictures. I broke it down into two parts - the “ramp” and the hood.


As shown in picture (A), a section of steel plate was used to make the ramp, and it was first grooved along one edge for most of its length; a section of steel tubing for the hood was slotted to mate tightly with this, as shown in (B). The base of the ramp was then milled to give a stepped section, and a recess was milled into the end of the barrel to accept this, as shown in (C).

Having got the ramp to be a nice snug fit in the barrel recess, I then shaped the ramp it and filed a sight blade at its top, as shown in (D). The ramp was then soldered in place (E) and the hood slid in place, as in (F). The hood, which can only slide forwards, is essential for making cocking a pain-free experience, and although it was a very tight fit, it could easily be removed if needed, by a sharp tap with a piece of wood.

The rear sight unit, as shown in the last pictures of the sequence, was an easier project, as the Mercury prototype used a standard Webley Hurricane sight, and I able to purchase one of these off the forum. It needed a slight radius on the underside to make it fit closely to the cylinder, and after drilling and tapping the cylinder it was firmly attached with two 4BA screws. On the original prototype there was a curved block in front of the rear sight assembly, which was there purely for cosmetic reasons.. Pictures (G) and (H) above show how the block was constructed, with a hollow interior to cover the piston guide screw, and picture (I) shows it fixed in place in front of the sight.

After the sights were completed, the next phase of the project was to make the wooden grip plates. On the original prototype, the grips were made of walnut, chequered, and the left-hand plate had a pronounced thumb rest with a recess for a safety catch lever. The sequence of events I followed in making the grip plates is summarised in the next group of pictures.


Two flat sections of walnut were cut from a larger block, as shown in (A). One of the sections was cut thicker to allow for the thumb rest. I always think you get a much better looking result with wood grips if you can arrange for the grain and figuring to follow the rake of the grip, so I made sure of this as best I could. The grip outlines were cut out with a hacksaw. The profiling for the thumb rest I found I could do most easily in the lathe, using an end mill for removing wood and with the grip plate clamped to a rotary table, as shown in (B). Picture (C) shows the left-hand grip with thumb rest after further finishing with a file and emery paper. The grips were drilled for the securing screws. The left grip also needed to have a recess let into it to accommodate the safety catch lever, and this was achieved again by milling. Then came the part I like the least – chequering. It is not something that I am good at, and I have to take it very slowly. The plates were screwed to a heavy block of wood to keep them steady while the chequering was being done, as in (D), and after what seemed a lifetime, the results were as shown in (E). Not particularly professional but adequate I think. The plates were then finished with satin polyurethane varnish, applied in very thin layers with rubbing down with finest steel wool between each coat. The final result is shown in (F), and you can see what I mean about the grain following the angle of the grip.

The safety catch was the last mechanical part of the gun that needed to be tackled, and the construction of this is shown in the next picture sequence. The first version of the Mercury pistol had no safety catch, whereas the Resuggan version clearly did, as indicated by the safety catch lever on the left hand side of the grip. If you study the patent drawings (which depict the first version only) there was no such safety shown. So I had no clue as to how Resuggan’s safety catch worked, but I was able to work out something functional from the location of the catch lever. All it required was an off-set pin on the lever pivot that intercepted the sear lever in the “safe” position and prevented it moving. An anticlockwise turn of about 20 degrees was sufficient to rotate the pin out of engagement with the sear and leave it free to be released by pressure from the trigger. The resultant safety catch unit was made in two parts as shown in the first two pictures, and the third picture shows it in place on the grip. Although very simple in construction, I found that the safety worked remarkably well in practice, so I imagine that this must be very similar to the design that Resuggan used. When the lever is in the down position the gun is ready to fire. In the up, or “safe”, position the gun cannot be cocked. Or if it has already been cocked then it cannot be fired.


With all the functional parts of the gun now complete, I was able to assemble the pistol in its rough state, fit it with a reasonably strong main spring, and then test fire it. To my relief everything worked perfectly – much better than I expected, as usually there is a lot of fettling to do with a new build. The detent broke smoothly and the spring-loaded pawl on the cocking lever made a loud click as it engaged with the first notch in the piston. Pushing against the spring, the barrel could be swung down until another loud click indicated that the sear had engaged with its notch in the cylinder. Returning the barrel home gave another loud click as the pawl engaged briefly with the second piston notch, and once the barrel was closed the gun was then cocked in the low power setting and ready to fire. To get higher power the whole sequence was repeated, needing more effort this time of course, and a similar sequence of clicks could be heard. I found the clicks to be reassuring, as it showed that everything was engaging properly and there was unlikely to be any surprise slippages or premature firing when cocking.

I then shot 30 or so pellets through the gun, at both power settings, with no problems whatsoever, and then suddenly, just as I was feeling smug that I had everything under control, the gun misfired at full cock. The problem persisted intermittently, and to make matters worse, sometimes the cocking pawl would slip its notch during cocking, especially on the second full power stroke, which was not good for the nerves. I had visions of the sear and/or pawl having worn or sheared, or the notches in the piston having chipped off due to faulty case-hardening. Or it could be down to my own poor engineering skills and poor tolerances.

Eventually I found out the cause of the problem. It seemed that as the gun was being fired repeatedly, the rear screw that held the grip frame and cylinder together slowly unscrewed. This enabled the cylinder to lift a minuscule amount from the grip as the gun was being cocked, so that there was reduced contact between the cocking pawl and the sear and their respective notches in the piston. When the tension got to a certain level, slippage occurred.
To solve the problem, the grip-frame fixing screw was fitted with a keeper screw to prevent it turning, as shown in the following picture. Once this had been fitted, I had no more problems, and I have since put several hundred shots through the pistol, even with powerful springs fitted, and the gun has worked perfectly so far (touch wood).


Now that the gun was up and working it was time to think about the final finishing touches – the lettering and the bluing. I wanted the lettering to describe accurately what the pistol was, also making it clear that it was a reproduction. The following picture above shows the final etching result (taken after the gun had been blued). As is usual, I also etched my signature, date and place in an inconspicuous part of the gun, in this case on the heel of the grip.


The next stage was to colour the metalwork of the pistol, and all the relevant steel components were blued by the hot salt method which I have used successfully before on other projects. This went well and a 10 minute immersion in the bluing bath at 140 deg C gave the result shown in (A) of the next picture sequence:


The side plates for the frame were aluminium and so needed a different approach. One option would be to use a black lacquer of some sort, but this somehow did not appeal to me. Also the original Mercury prototypes, which were almost entirely made of aluminium alloy, appeared to have been chemically blacked rather than painted. With aluminium and its alloys the most popular way of achieving a chemically generated coloured layer is by anodising and dyeing, and this is most probably what was used on such iconic air pistols as the Walther LP53 and the Feinwerkbau. The result is a hard, scratch and corrosion resistant layer, which retains much of the character of the underlying metal. I had previously had a go at anodising in my previous project, the Haenel LP55R pistol, but in that case the anodising layer was not dyed and a silver effect was obtained. This time I would need to take the aluminium parts straight out of the anodising bath and into a black dyebath. Once the dye had been absorbed to give the depth of black I wanted, the surface would then be permanently sealed by boiling the parts in water.
Picture (B) shows the plates coming out of the dyebath, and (C) shows how the plates looked after sealing in boiling water. They were an excellent match for the blued steel frame shown in (A).

With all the finishing completed I could now assemble the gun. The following pictures show the final result:



A box was made up with a veneered plywood top and bottom and hardwood sides, avoiding joints by fixing the whole thing together with PVA glue. After a bit of staining and applying two thin coats of polyurethane clear satin varnish, the box looked OK.



The interior was fitted out with green baize and foam rubber. A mint, period tin of Webley pellets, courtesy of Troubledshooter, together with a manual, a few targets and a couple of implements completed the picture.

The rectangular green pellet tin is not true to the time period of the 1980 "Mercury" and will offend the purist Webley collector. In my defence, I can only say that I did not look at my project pistol as part of my Webley collection, and saw it as just another project to investigate obsolete British air pistol mechanisms. After all, my pistol differs from the originals in many ways, both from a materials point of view and how it is constructed. The case and fittings are for safe storage and just a bit of window dressing, and to me the rectangular tin is evocative of the late 1960's-1970 period when the design was first proposed, and I just like the look of the rectangular tin more than the round. So apologies, but I will be keeping the rectangular tin even though Troubledshooter has kindly offer to supply me with a more period round one.



So the pistol was now complete, and it was as close enough in outer appearance and mechanical operation to the original Mercury to make me happy, but the question remained, how would it perform? The first, and most important thing to decide was the best spring, or combination of springs, to use as the power source. The patent drawings show that a double spring system was used. Presumably the springs were pre-compressed slightly in the cylinder, but there was no way of knowing to what extent.
I had available three suitable springs of different sizes and strength as shown here.


The larger spring had an outside diameter of 19 mm, and the other two springs fitted easily inside it without any binding. The wire diameters of the springs, which give a rough indication spring strength, were 2.5, 2.0 and 1.5 mm.

The muzzle velocities and energies were measured by chrono’ for different spring combinations, using unsized .177 waisted pellets (Diana flat heads), each weighing 0.505 g. Measurements were averaged over a string of 10 shots, and the consistency was reasonable, at about plus or minus 10-15 fps.
The first set of tests were the lowest in power, the gun being fitted with the wide spring (A) only. At the single-stroke setting the gun registered 180 fps, and this increased to 315 fps for two strokes.

The spring power was then upped by using a combination of springs (A) and (C). There was a noticeable increase in the effort needed to cock the gun, but even the second stroke was still very easy to do because the double cocking action uses a shorter cocking stroke than conventional break-barrel pistols. With this spring combination, the lower power setting gave an average muzzle velocity of 233 fps and the higher setting a very respectable value of 385 fps. The lower figure compares with a value of 287 fps claimed by Webley for their Junior pistol, and the value of 385 fps compares very favourably with the factory quoted values of 350 fps for the Premier, 416 fps for the Senior, and 420 fps for the Hurricane. I think one can safely assume that the factory quoted values are going to be for highly optimised conditions.The Mercury was noticeably more comfortable to double-cock than the single-stroke any of these, but of course the two strokes took slightly longer to carry out.

Finally a combination of spings (A) and (B) was used, and now cocking required an uncomfortable increase in effort for the second stroke. Surprisingly at the single-stroke level, the muzzle velocity was practically the same as with the the weaker spring combination (A) plus (C), and a velocity of only about 220 fps was recorded. However, at the second stroke level, the velocity did increase markedly, up to 440 fps (about 3.5 foot pounds). However, in my opinion, the gain in muzzle velocity was not worth the increased effort of cocking, and I found the (A) plus (C) combination infinitely preferable.
So I decided to used the (A) plus (C) spring combination for accuracy testing.

The following picture shows the target from a string of 10 shots, with the gun charged at the two-stroke level, shot from 6 yards and using a bench clamp. As you can see, all 10 shots fell in an area about the size of a 20 p coin, and in fact the spread horizontally was much less than that. The vertical spread being larger than the horizontal spead was probably due to a very small upward creep of the pistol in its clamp after each cocking, as it the pattern did seem to proceed upwards as the number of shots increased. I was very pleased to see that there were no flyers. At the top of the picture are examples of the same type of pellet, fired at the same power level, point blank at a steel plate.

When a second accuracy test was caried out under identical conditions, but using the single-stroke power setting, a similar tight grouping of 10 shots was obtained, again with no outliers.


So having put several hundred pellets through the gun at various power levels, what are my conclusions?

On the downside, my gun is heavy at 1.7 kilos, which compares with 1.5 kilos for the BSA Scorpion and 1.9 kilos for the Milbro Cougar. However, the original Mercury prototypes would have been much lighter, as they were constructed largely of aluminium alloy, whereas mine was mainly steel. Another downside that was not obvious at first, and was a peculiarity of the double cocking principle, was that once the gun had been cocked to the higher power level it was not possible to de-cock it completely, and it could only be de-cocked to the lower power level. Any attempt to de-cock a second time resulted in the start of a second cocking stroke. Once the lower power level had been reached, the only way of making the gun safe was to fire it. This caught me out once, when I had forgotten I had cocked the gun to full power and after de-cocking I was under the impression the gun was safe. So I was carrying around a cocked and loaded gun without being aware of it. A surprise firing when I pulled the trigger (with no disastrous consequences I am glad to say) made sure that it never happened again. I can see that this aspect of the gun’s design would not be something that would recommend it to the market!

The gun has some definite positives, most significantly the ease and comfortableness of cocking compared to pistols of similar power. Another aspect that I liked was its versality when using it for general plinking. For most purposes the lower power setting was adequate, and cocking was effortless. However, when fancying a few longer range shots, or wanting to do real damage to a can, the instant availability of the higher power setting was a nice thing to have.

All in all, it was a difficult project (mainly due to problems of my own making) and took longer than expected, but I am glad that I undertook it. Not only did it help while away the hours of lockdown, but I now have a gun that looks good in my collection, one that is a working copy of a model that I would never be able to own, and it has also helped answer a few technical questions about an obsolete design. Equally important, I have a gun that is very enjoyable to shoot, and I am sure it will get quite a bit of use.