• Welcome to the DeeperBlue.com Forums, the largest online community dedicated to Freediving, Scuba Diving and Spearfishing. To gain full access to the DeeperBlue.com Forums you must register for a free account. As a registered member you will be able to:

    • Join over 44,280+ fellow diving enthusiasts from around the world on this forum
    • Participate in and browse from over 516,210+ posts.
    • Communicate privately with other divers from around the world.
    • Post your own photos or view from 7,441+ user submitted images.
    • All this and much more...

    You can gain access to all this absolutely free when you register for an account, so sign up today!

GeckoSub Mirage Evo - And Adventures in 3D Printing Speargun Parts

Thread Status: Hello , There was no answer in this thread for more than 60 days.
It can take a long time to get an up-to-date response or contact with relevant users.

Diving Gecko

shooter & shooter
Jun 24, 2008
1,698
461
188
Intro
I am setting out on yet another Mirage project. In short, I'll be using the Mares Cyrano 1.3 handle instead of an original Mirage handle and 3D printing some other parts. But first of all, credit where credit is due. This idea is not mine but Popgun Pete's as explained here. He planted the seed and it has grown well in the fertile soil that is my Mirage obsession;-)

But why even bother, you might ask? Well, I find that the 40-something-year-old Mirage handles are not the easiest to seal and keep in good working order. Also, while it is sacrilegious to some, I have been sawing off the sloped-back grip on two of my original Mirages as I never fully got used to them and finally, the lack of an active line release was beginning to bother me, too. I guess we can say that I am in love with the technology of the Mirage - but not its first execution beautiful as it may be on the surface.
To make a Mirage more to my liking I have been considering fabricating a handle from scratch, but with my current skills, it's honestly too much to take on. But trying to make a more modern off-the-shelf handle work as a replacement seemed like a somewhat more doable project.

Now, this is not a straight drop-in of a handle nor an easy conversion. Besides the active line release and more regular angle of grip, the Evo handle is a candidate pretty much only because the shooting barrel sits high in that gun - as it does in the Mirage. But to make it work, you would still need to make a new power regulator bulkhead, nose cone and a pumping barrel. The reason I am giving this a go now is because I have gotten into 3D printing and feel I can possibly make this parts successfully myself.

We will hit some speed bumps on the way - I can guarantee that and will document my thoughts, failures and solutions but for now, let start with some pics.

Some handles!
At the top, an original Mirage handle (V1 even).
Middle row: two heavily modified Mirage handles (the one on the right with an active line release).
And on the bottom, a brand spanking new Mares Cyrano 1.3 handle
YOiRpIt.jpg


Next up, let's say hello to my little friend Anycubic I3 Mega - it's because of him/her(?) that I am even venturing down this untrodden path:
DF5p9I0.jpg


The printer itself has been modified lately. I wont go into too much detail but in case there are some 3D printing dudes/dudettes in the audience, I have recently converted it from a Bowden to a direct extruder:
7fUYgr7.jpg

(The conversion is to better control retractions as I can't have blobs and underextrusion mess with my o-ring seals and also, I might need to print some flexible parts at some point).
 
Last edited:
Printer Calibration and Bulkhead Dummies
The Anycubic I3 Mega, at the time of my purchase, was - and probably still is - widely considered as one of two very good bang for the buck deals. If you have the money, and less time to tinker, the Prusa I3 MK3 is often mentioned as the best way to get going with minimal fuss.

Still, since I wont be printing the Millennium Falcon, Groots nor a skull pencil holder but rather structural parts which needs to be hold dimensions fairly well, some calibration was in order:
h1PYtJR.jpg


These are just some of the many tests I have been running lately. Thing is, while some of the calibration has to do with the mechanics of the printer, the filaments themselves also behave differently.

But once it was up and running fairly well and I had one of the filaments dialled in, I set about "reverse engineering" the bulkhead. Just figuring out where the locating pegs and the bore for the power regulator were took a while:
MWxo6LW.jpg


But I think on V8 of the dummy bulkhead, I am spot on:
PhRbt4Q.jpg


P.S.
I am actually incredibly tempted to print the Falcon. And Groot... and heck, pencil skull holders are cool, too;-)
 
Last edited:
Early Nose Cones
Early on, actually before I made the changes to the printer seen in the first post above, I printed a few versions of nose cones.
Some screenshots from the CAD:

FXbvoUE.jpg


And in real life:
Zcisu4I.jpg

LitDMxE.jpg


I still like the idea of a fiber optic sight at the front, so will incorporate that in the nose cone, as you can see on the top of the cones. Also, for a shooting line holder, I plan on inserting some stainless steel. There's a small step on the lower, right hand side of the muzzle to add a bit more material where the shooting like holder will be:
gTQ7lrJ.jpg

nQnNU7j.jpg

CEFGt7E.jpg


But I wasn't 100% satisfied with the printing of some of the really thin walls and I also need to deal with a compatibility issue of these parts so I took a break from this project to upgrade the printer itself and ponder how to overcome the issue.
 
Last edited:
High But Not High Enough - The First Real Obstacle!
I mentioned we would run into speed bumps and the first one is a pretty fundamental compatibility issue. While the Cyrano handle is a "high barrel handle" its barrel actually doesn't sit as high as it does in a Mirage.
In the Cyrano the barrel sits about 1.45mm lower than it does in the Mirage. That may not sound like much, but it has a knock-on effect which leaves very little room for the pumping barrel under it. And worse yet, it leaves almost no space to have a proper wall thickness between the two barrels and at the lower part of the nose cone.

Here's how the issue looks in the CAD:
zGCVRrP.jpg

LKTJrkJ.jpg


Having this little material is critical - potential a deal breaker - as the walls need to be of sufficient thickness to allow the o-rings to seal. Also, the walls see a bit of abuse when inserting the nose cone.

Now, the figures shown below aren't really set in stone - as they depend on the design of the o-ring seat in pumping barrel and the amount of compression on the o-rings and so on. E.g. the original o-ring bore design isn't really great but to make a better one, you actually have to make the bore for the pumping barrel even larger - resulting in even less wall thickness. But I have been fiddling around with the design for a while, so can share where I am at now:

I have about 0.6mm of wall thickness between the two barrels:
hXmgnmQ.jpg


And a smidge more (0.75mm) under the pumping barrel - and that's already with an o-ring bore/seat design that I am not super happy with:
99JDRAt.jpg


And yes, the walls are thin in the original Mirage nose cones, too but those cones were injection molded and I believe the plastic is thus inherently more homogenous than what 3D printing can produce. I don't feel confident I can print such thin walls and have them be strong enough for repeated assembly and disassembly and to be 100% impermeable.
So, what to do...?
 
Last edited:
  • Like
Reactions: landshark sa
High But Not High Enough - A Hack? A Solution?
The solution I have in mind, in all its simplicity is a bit of hack, really. I intend to just shift the barrels up a bit at the nose. This will give me space to increase the thickness of those thin walls. Yes, the barrels wont be parallel with the reservoir any longer. For the pumping barrel that doesn't matter at all, but of course, it's an issue for the shooting barrel. But how much of an issue is it really...?
Let's say I raise the front of the shooting barrel by 2mm and let's, for simplicity's sake, say the shooting barrel is about 100cm long. That means, if I am not mistaken, that if I am aiming 100% dead straight along the reservoir I will be shooting 1cm high at 5m from the tip of the gun, right...? I can certainly live with that, haha.
Also, I don't aim that precisely in the first place and one could argue that the offset counters shaft drop making aiming more instinctive;-). And it is probably well within the built-in accuracy of the gun and its setup in the first place.

One thing I can think of is the need to have just a tiny bit of slop in the bulkhead so that it doesn't push the barrel down at that point and incur a bit of a banana shape. But again, with the tolerances between the bulkhead and the reservoir and the bore for the barrel I think this will take care of itself.

Besides that, am I right in assuming this offset would be a non-issue? I could use your help on this one. Anything I haven't taken into account?
 
Last edited:
Mares used what they had expertise in to make the original "Mirage" and we know they sort of stuffed it up when they got the breathing wrong in the original version. Whoever designed the rear handle never actually considered what the rest of the gun required and it seems they designed the exterior of the gun first, then crammed the workings inside that "envelope". The "Mirage" should have had a longer travel gate for the 3 position power selector, but they wanted the gun to look more or less like their other guns and that included the size of the metal selector "shift pattern" plate that forms the gate. Modern guns don't use a metal plate gate, instead they use a metal stop in a plastic gate directly incorporated in the molding. A new "Mirage" can make do with a two position gate as we have discussed previously.

Although it looked nice, many divers did not like that raked handle, especially when they had to load the gun in other than calm conditions.

You may be able to use a thinner wall pumping barrel by using stainless steel tubing. There has to be room for the pumping barrel's rubber sleeve valve to open, or at least fit in the gun with the main barrel sitting just above it. As the main barrel holds the gun together the pumping barrel only has to trapped at either end, so it does not need to carry any screw threads as the main barrel does. The "Cyrano 1.3" main barrel already has breathing ports in the right locations, so it is the obvious main barrel to use. The original pumping barrel has a reduced OD at the nose end, but there is no reason for the barrel tubing to be as large an OD as it was in aluminium. Interestingly Mares originally used the same size (exterior) "O" ring on the nose end of the pumping barrel as they did for the bulkhead boss (interior) that plugs the rear end of the pumping barrel!! If you need to provide a shoulder at the front end of the pumping barrel then a short plastic bush fixed with a grub screw or two could be used to do the same job as the machined shoulder in the original. You can always flatten or create a curved indentation in the top of this bush where it will press against the main barrel OD.
 
Yeah, I have thought about a stainless steel barrel and plastic bushing held in place by a C-clip - I did something similar on another gun - but if the bushing doesn't actually protrude a bit into the bore of the nose cone, it probably won't really help to guide the o-ring into place. And for it to be able to protrude in a little, that bore would have to be enlarged and then we are back at the problem with the thin walls.

I feel like I have a pretty good idea by now about how to do o-ring seat design, what works and what doesn't, and as you point out, you do indeed need a proper shoulder to support the o-ring on insertion whether it's a bushing or a bigger OD for the pumping barrel.
The problem I have is with the original design - that shoulder is just too little of a step up and it's very difficult for me to insert any the original barrels into the nose cones without the o-ring getting pinched.
The front o-ring bore is 11mm and the shoulder behind it 12mm, if I recall correctly. That's just not much support at all. If Mares originally used the same o-ring as at the back of the pumping barrel, maybe that's because they realized they really needed to overstretch the o-ring at the front to have a chance of it staying in place when assembling the gun. Not kosher sealing design, haha.

A while back I had a Mirage nose cone that just wouldn't seal. As a last resort, I enlarged the bore for the pumping barrel (from 12mm to 13mm) and redesigned the corresponding front of the barrel and it has worked wonders. Instead of the shoulder just being 1mm larger than the o-ring seat, it is now 2mm larger and this did the trick completely. This is why I feel I really need to have a bigger bore in this DIY 3D printed nose cone than in the original. That, coupled with the fact that the shooting barrel sits lower in the new handle, is why I am running out of space.

This is actually also why I am not sure I benefit much from moving to, say, a 10x12mm stainless barrel. I can probably still only make the o-ring seat 11mm in diameter, so would definitely need a bushing behind it - and then as mentioned, we are back at the root of the problem.

I'll juggle the numbers and drawings a bit more to see if I have missed something:). Otherwise, I still think my best way out of this pickle is to to lift the barrels in the nose. It's unorthodox for sure, but the real world ramifications seem negligible. What do you think about that slightly curve ball solution?
 
Last edited:
Just looking through my stock of photos I see that the Cressi Saetta has a high mounted barrel, but I am only judging it by this photo. If the "Cyrano 1.3" handle proves too difficult then it is another option. I don't know anything about its main barrel breathing ports and have not seen much written about the gun at all. It looked to be a gun prone to bending up at each end "like a banana" due to what seems a very high barrel position.
gunsaetta3__84921_zoom.jpg
gunsaetta3__84921_zoom.jpg
 

Attachments

  • Cressi saetta handle.jpg
    Cressi saetta handle.jpg
    15.6 KB · Views: 339
Yeah, I downloaded and tried to do some measurements on a tech drawing of that handle around the time I first saw your thread. Can't recall why I opted against it. Maybe something as "shallow" as the old school line release or it's humpback looks;).


KX5rz3Q.jpg



Did some quick measurements in Photoshop. The resolution, though not too shabby, is not that great either so hard to get a clear reading. Also, for some reason the barrel reads as 19mm OD...
But I think the distance from the top of the barrel to the underside of the reservoir is about 4-4.5mm. If so, that's a really high handle and smack in Mirage territory. For comparison that same measurement on the Mirage is about 4.25mm and 5.7mm on the Mares Cyrano.
 
Last edited:
Using stainless steel for the pumping barrel means that you could make the nose end very thin in wall section, much thinner than it could be done in alloy. That is what I was thinking of as I found stainless steel barrels can be very thin in wall section, such as they are in a "Taimen" barrel. You would then only need a small diameter "O" ring and that may provide that little extra bit of thickness in the muzzle wall.
'
 
  • Like
Reactions: Diving Gecko
For some reason I doubled up on the barrel photo when I had intended to click on the Cressi Saetta muzzle, but here it is anyway.
Cressi saetta muzzle.jpg
 
Using stainless steel for the pumping barrel means that you could make the nose end very thin in wall section, much thinner than it could be done in alloy. That is what I was thinking of as I found stainless steel barrels can be very thin in wall section, such as they are in a "Taimen" barrel. You would then only need a small diameter "O" ring and that may provide that little extra bit of thickness in the muzzle wall.
'
How thin do would you think I could go?
Just speculating but perhaps I can get away with 0.25mm wall thickness... As you say, there are no threads and no real loads on that barrel.
I might ask Dima about it, if I can track him down - he's been AWOL in Belize forever. He might have a suggestion though he doesn't like to go too thin in his stainless barrels but that's on account of the threads and shock loading, I think. Oh, come to think about it, his sliders are incredibly thin in places. Hmmm... I'll start looking around to see what I can find.
 
Last edited:
I think that you could go very thin, the problem is to machine the material away evenly as the stock tubing has to run true in the lathe. One way to ensure this is to run a centering tool in the mouth of the tubing using the tailstock on the lathe, while using some lubrication on the centering tool (oil or grease). If it does not run true then you may get breakthrough at one section of the very thin tube. In order to reduce weight the "Taimen" barrel is very thin as it is machined away virtually for its full length. From memory, as I cannot find my notes on it, the 10 mm ID "Taimen" barrel is 11 mm OD for most of its length having been machined away everywhere except the ends which carry screw threads.

https://forums.deeperblue.com/threa...-project-polishing-inner-barrel.103994/page-2
 
You might consider putting an extension at the lower end of your muzzle that projects rearwards into the gun. That will give you a greater length to engage the pumping barrel into as the seal here is to keep the gun from leaking at this position, especially as the high mounted inner barrel guns tends to "banana". That is why Mares stuffed two very much overstretched "O" rings in there originally, using them as a sort of "rubber packing" rather than a dynamic "O" ring seal. There is no particular reason why the muzzle molding has to have a flat back, it is just that up until now they always have been, as far as I know that is for most of the "Sten" derivatives.
Gecko muzzle with lower ext.jpg
 
This is what the "Mirage" power selector gate should have looked like. The selector control shaft would have been shorter as it no longer has to reach further forwards, but at full power the brass control piston or plug will have pulled right back into the tunnel provided in the molding. This would have opened the airflow up, although there would still be that "dogleg" in the airflow channel.
 

Attachments

  • Mirage selector gate long.jpg
    Mirage selector gate long.jpg
    67.4 KB · Views: 381
  • Mirage regulator detail 2A.jpg
    Mirage regulator detail 2A.jpg
    96.7 KB · Views: 393
This is what the "Mirage" power selector gate should have looked like. The selector control shaft would have been shorter as it no longer has to reach further forwards, but at full power the brass control piston or plug will have pulled right back into the tunnel provided in the molding. This would have opened the airflow up, although there would still be that "dogleg" in the airflow channel.

Ah, another of those V1 vs V2 differences. In the V2 handle the brass bushing can't pull back in a recess at all. It stops where it would on a Sten. But then it's bulkhead is longer, too.
 
Connector Bushing for Pumping Barrel - V1 (Non-functional Draft)
On the original Mirage power regulator bulkhead, there's a small peg at the front that the pumping barrel slides onto.
YIQOpQJ.jpg


I could quite easily design that feature into my new bulkhead but printing it would be another matter actually printing it as there are some limitations to how a 3D printer can actually print. First of all, you can't print in free air and if you want to print without a lot of "supports" you need to orient the part so there's a flat facing the build plate.
In reg. to a bulkhead, that flat would be the front face of the bulkhead and if there's a protruding peg there, it complicates things a bit.

So, for now, I think I will just drop the peg and make a bushing that inserts into both the bulkhead and the pumping barrel. It can be done fairly simple, actually. I am planning on turning it in brass on my mini lathe.

And it helps there's plenty of space in that part of the bulkhead since the bushing can be made smaller in OD than the barrel. If inserting the barrel itself into the bulkhead, you run into exactly the same issues as at the nose cone where you are running of side wall thickness fast. It might be doable but doing it with a bushing seems a safer way for now.

Here's a first sketch of what I will be doing:
5OnfTda.jpg


I need to work on the bottom of the one-way valve a bit as in this first version, that o-ring will get squished pretty badly very fast. I need to use a smaller o-ring and add a recess which the o-ring sits in. That way, when the ball presses onto the o-ring the edges of the recess will stop the ball from traveling too deep into the o-ring. Too much travel and you reduce the life of the o-ring, too little and you wont get a seal.
 
Last edited:
Those valve "O" rings function as valve seats, so they don't really function as dynamic seals. Provided they do not split they can flatten out a bit and still work OK. An "O" ring moving on a shaft, or sliding down inside a bore is where they function as dynamic seals, but in the valve seat application they are static seals. A rubber washer could do the same job in those applications, but "O" rings are more convenient to use. The seal on the nose of the pumping barrel should be a static application, but when the gun tends to banana under increasing internal pressure the nose cone can move away at the bottom and that may cause a leak. Possibly you could bond the pumping barrel nose into the nose cone, or muzzle, and make it essentially one combined component. Early Nemrod pneumatic guns from the Silver series had few "O" rings, using rubber packings instead, but later versions switched to "O" rings throughout without any changes to the gun components. I only mention this as that pumping barrel nose seal does not have to be an ""O" ring.
 
Yeah, I downloaded and tried to do some measurements on a tech drawing of that handle around the time I first saw your thread. Can't recall why I opted against it. Maybe something as "shallow" as the old school line release or it's humpback looks;).


KX5rz3Q.jpg



Did some quick measurements in Photoshop. The resolution, though not too shabby, is not that great either so hard to get a clear reading. Also, for some reason the barrel reads as 19mm OD...
But I think the distance from the top of the barrel to the underside of the reservoir is about 4-4.5mm. If so, that's a really high handle and smack in Mirage territory. For comparison that same measurement on the Mirage is about 4.25mm and 5.7mm on the Mares Cyrano.
I just did a "quick and dirty" diagram to see how the "Saetta"
Mares Mirage Evo X.jpg
handle would fit.
 
Those valve "O" rings function as valve seats, so they don't really function as dynamic seals. Provided they do not split they can flatten out a bit and still work OK. An "O" ring moving on a shaft, or sliding down inside a bore is where they function as dynamic seals, but in the valve seat application they are static seals. A rubber washer could do the same job in those applications, but "O" rings are more convenient to use. The seal on the nose of the pumping barrel should be a static application, but when the gun tends to banana under increasing internal pressure the nose cone can move away at the bottom and that may cause a leak. Possibly you could bond the pumping barrel nose into the nose cone, or muzzle, and make it essentially one combined component. Early Nemrod pneumatic guns from the Silver series had few "O" rings, using rubber packings instead, but later versions switched to "O" rings throughout without any changes to the gun components. I only mention this as that pumping barrel nose seal does not have to be an ""O" ring.

True, but there's no need to squish even static seals more than necessary:). Supposedly, even static seals have ideal ranges of compression. From what I can read, you get fast diminishing returns when you start going over 30-40% or so for static seals. I do think a flat rubber gasket could do the job pretty well in a check valve like that, actually.

Good thoughts on perhaps seeing if there's a way permanently bond one of the ends of the pumping barrel.

I should have said, that my method right now, is to make it proof of concept without trying to overthink things too much, which I sometimes do. So, I will try to choose the handiest solutions and then perhaps in later versions, go for some more advanced ones.
 
Last edited:
DeeperBlue.com - The Worlds Largest Community Dedicated To Freediving, Scuba Diving and Spearfishing

ABOUT US

ISSN 1469-865X | Copyright © 1996 - 2024 deeperblue.net limited.

DeeperBlue.com is the World's Largest Community dedicated to Freediving, Scuba Diving, Ocean Advocacy and Diving Travel.

We've been dedicated to bringing you the freshest news, features and discussions from around the underwater world since 1996.

ADVERT