Page 1 of 3

Wide-range universal launcher

Posted: Tue Jul 24, 2018 7:37 am
by cprobertson1
Good day ladies and gents!

I thought I'd put a post together regarding my new launcher which is designed to fit rockets with 5-40mm nozzles, without changing any parts out.

While my previous launcher used a gardena nozzle (fairly common) - this particular system instead opts for utilising a conical filling nozzle (the "launch plate") which fits inside the throat of the rocket's nozzle, and creates a seal using an O-ring: with a launch-clamp preventing the rocket from breaking the seal as it is fueled - launch is accomplished simply by releasing the pressure on the clamp.
Capture2.PNG
Capture2.PNG (28.58 KiB) Viewed 119 times
The attachment Capture2.PNG is no longer available
As you can see from these CAD images, the seal between the nozzle and the launch plate can be accomplished for pretty much any size smaller than the launch plate diameter - though it is yet to be seen if the water stream passing over the O-ring for a few milliseconds during launch is going to be enough to blow it out of the seat! After that short uncertainty period, the jet of water should miss the O-ring entirely.

If it transpires that the O-ring is consistently blown out of the nozzle during launch, I have an alternative sealing method using a separate O-ring carrier ring, which fits between the nozzle and the cone and is deliberately left behind during launch (which may actually serve to reduce the weight of the nozzle as well - but at the same time creates a second path for water to pass around the seal under pressure - which may lower the possible range of pressures that can be used.

I have already tested this on a small-scale using a piece of wood in lieu of a hold-down clamp and results are promising. The test-nozzle has a 5mm diameter and is mounted on a hose which is connected to a regular water rocket (instead of launching, it exhausts through the test-nozzle) and it managed to not blow the 7mm x 1.5mm O-ring out when the launch plate was disengaged.

The launcher itself is fairly simple - consisting of a compressor fitting (on the end of a 30m hose) connecting to either a water reservoir (if the rocket is to be "fuelled" on the pad) or directly to the launcher.

The launcher has a T-junction - with the centre-connection going to the launch plate and the other connection going to a springloaded ballcock valve which serves as a way to depressurise the rocket quickly in the event of an emergency. The exhaust is vented via a 3D printed howell-bunger valve which directs a spray of water directly upwards rather than sidewards, with intent on reducing erosion to the surroundings/launcher/rocket. It is pretty much just a very inefficient nozzle :P)

The water reservoir consists of any number of water bottles - during pressurisation, water is forced out of the reservoir and into the rocket - after which the air simply passes through the reservoir and into the rocket. This obviously leaves the problem that there is now a compressed air vessel on the pad - as such, a number of small vessels are preferable to a single large vessel (to limit riskin the event of a catastrophic failure: a single vessel can dissipate it's contents quickly while a smaller vessel will be forced to vent through the other vessels, slowing the rate that the energy can be released.

I am not happy with the safety involved in such a reservoir system, so I'm going to add that at a later date though once I work out a safe way to do it - it will probably be simpler just to add a tube to the rocket that allows the ullage to vent until the reservoir is empty, at which point we close off the reservoir and compress/charge the rocket as normal - I'll figure that one out once I've got the launcher itself completed - since the reservoir-based fuelling method is more of a luxury than anything else!



The other difficulty with the launcher is the design of the hold-down clamps: which must have the following characteristics;
A) Fit varying nozzle-flange sizes (this launch plate supports up to 50mm: corresponding to a 40mm nozzle diameter - but the body of the rocket can become wider after that if needed)

B) Impede the motion of the rocket as little as possible (the rocket should slide out the clamp rather than having to force its way out)

C) Deform only minimally while providing enough downward force to counteract the pressure at the nozzle (200 PSI applied to a circle 40mm wide gives us around 1700 newtons - though the actual area might be greater as it is a cone which will reduce the equivalent force - but not by much (if anything))

D) Fail safe in the event that the nozzle seal is broken; the rocket should be incapable of launching.

E) Survive multiple launches (obviously useful!)



A) should be fairly simple - there are numerous ways to accomplish this: < shaped brackets are an easy option though there are others too! The lateral movement of the clamp required to allow an unimpeded launch is slightly larger than the depth of the flange: which will typically be 3-5mm (depending on engineering requirements under "C")

B) is accomplished by using a chamfered flange which can be seen along the top/outer rim of the nozzle. This fits into a correspondingly chamfered edge on the clamp - as the rocket moves up, it simply slides out of the clamp - the motion of the clamp can be assisted using springs if needs must.

C) This is probably the hardest part - particularly if I up the pressure even more than 200 PSI (though don't intend to do that anytime soon!). Using 200 PSI as a ballpark figure (worst-case scenario) and assuming a nozzle diameter of 40mm gives us around 1730 newtons - the equivalent of requiring around 173kg sitting on top of it to prevent a launch, if my calculations are correct (please forgive me if I've made a dumb error somewhere!) As such, I will probably be building the launch clamp assembly out of welded steel - which poses a whole other set of problems.

D) Interlocks can be used to prevent clamp movement: clamp arms can be designed so that both must be released before they will move. Clamps can be locked in place allowing redundancy in the locking mechanism.

E) Materials choice, waterproofing, etc.



I peronally think that "C" might be a show-stopper - what do you folks reckon (bearing in mind that 200 PSI is an absolute maximum - a deliberate overestimate: it's more likely to run at 100PSI - maybe occasionally going to 150PSI for smaller rockets (which won't have the force problem owing to a smaller nozzle diameter)

Re: Wide-range universal launcher

Posted: Tue Jul 24, 2018 7:39 am
by cprobertson1
Sorry - one of the attachments vanished for some reason...
Capture1.PNG
Capture1.PNG (23.54 KiB) Viewed 118 times
Capture2.PNG
Capture2.PNG (28.58 KiB) Viewed 118 times
Here are two sizes of nozzle fitting on the same launch plate: note the vertical displacement of the flange used to secure it - the launch clamp will need to be adjustable to compensate for this change (though we can deal with that by deliberately maintaining the same height of nozzle-assembly instead of changing the height of the clamps).

Clamp height can be adjusted with threaded bar incidentally!

Re: Wide-range universal launcher

Posted: Wed Jul 25, 2018 12:39 am
by anachronist
Making launchers is one of the things I like least about this hobby. A one-size-fits-all launcher would be useful.
It seems possible that one could also have a launch tube with that. It would provide an extra velocity boost before the water reaction mass starts being consumed... and it would also prevent your o-ring from blowing out because negligible water would come out until the rocket clears the tube.

Re: Wide-range universal launcher

Posted: Wed Jul 25, 2018 6:07 am
by cprobertson1
anachronist wrote: Wed Jul 25, 2018 12:39 am Making launchers is one of the things I like least about this hobby. A one-size-fits-all launcher would be useful.
It seems possible that one could also have a launch tube with that. It would provide an extra velocity boost before the water reaction mass starts being consumed... and it would also prevent your o-ring from blowing out because negligible water would come out until the rocket clears the tube.
I tested it last night up to 30mm - and at 50 PSI it HARD to hold that down. I accomplished it using a lever and fulcrum to push the rocket down - but even then it started to crack (but then again, it was just wood - so; I might need to upgrade to welded steel as I mentioned before... which ruins any chance of it being an easy build!)

Interestingly, the o-ring didn't blow out (at least at these pressures) - it would seem the seal is broken first around the inner circumference of the o-ring - which pushes it into it's seat, rather than the water attempting to go around the seat. So I may have accidentally solved my own problem! We shall see if my hypothesis holds at higher pressures though ;)

I'm currently looking at using paracord (or even rope) to hold down a collar, which is itself held together by pins (pull out the pins to release) - but even with 225lbs paracord, I don't think that will stand up to the force of larger nozzle sizes.

A launch tube is an interesting idea - but I then hit the problem that I'd need to attach a different size of tube for different nozzle sizes :P

At THAT point, I'd consider using a gardena style connector (or better yet, an airline quick release) and just attach different launch plates for each rocket. You could even make it a screw fitting if you really wanted.

---

Now the next question regarding launch tubes is the problem of the large nozzle diameters - am I right in thinking the force the launcher will have to hold down will be exactly the same for a 40mm nozzle as a 40mm launch tube jammed all up inside said nozzle :P?

From my preliminary design notes, it seems the single largest problem is holding the rocket down as we pressurise it (again, bear in mind that I can simply resort to not using high pressures! It's not like I *have* to use 200PSI! I'm just overbuilding it to that to ensure that it can definitely take less - belt and braces stuff :P

Of course, perhaps the most obvious answer to all this is just to reduce the nozzle diameter!

Re: Wide-range universal launcher

Posted: Wed Jul 25, 2018 12:20 pm
by anachronist
Now the next question regarding launch tubes is the problem of the large nozzle diameters - am I right in thinking the force the launcher will have to hold down will be exactly the same for a 40mm nozzle as a 40mm launch tube jammed all up inside said nozzle :P?
Yes, but the difference is where you can put the seal. With the seal on the base like you have it, you not only have to overcome the water pressure, but also exert additional force to squeeze the o-ring. If the o-ring is on the launch tube, then it's already squeezed by the nozzle, and the only downward force you need is to keep the rocket from lifting.

Re: Wide-range universal launcher

Posted: Thu Jul 26, 2018 3:19 am
by cprobertson1
anachronist wrote: Wed Jul 25, 2018 12:20 pm
Now the next question regarding launch tubes is the problem of the large nozzle diameters - am I right in thinking the force the launcher will have to hold down will be exactly the same for a 40mm nozzle as a 40mm launch tube jammed all up inside said nozzle :P?
Yes, but the difference is where you can put the seal. With the seal on the base like you have it, you not only have to overcome the water pressure, but also exert additional force to squeeze the o-ring. If the o-ring is on the launch tube, then it's already squeezed by the nozzle, and the only downward force you need is to keep the rocket from lifting.
Careful! The internal water pressure deforms it against the seat - it's just the same as a regular O-ring cylinder groove only it's at an angle - just the vertical force of the pressure that's a problem :)

Actually, the "upper" surface of the o-ring groove is missing: but the rest of the groove is still there. You can test that this works by taking a gardena nozzle and cutting the tip off, right above the O-ring (so the U-shaped groove becomes an L-shaped seat) - it will still seal BUT is more susceptible to debris - which isn't a terrible problem in our case - as its easy to pop out and clean

If we were to rotate the seat of the o-ring 45 degrees - so it had a flat surface above, pushing onto an angled surface (creating a "triangular crush seal") then we would need to provide compression to create the seal - but in our case, because the o-ring seat is perpendicular to the sealing surface, the pressure does the compression for us (though, as I mentioned, it should probably have a front-wall, the only reason it doesn't have a front wall is because it would be an extra step in printing them (as I'd need to trim the scaffolding off and then smoothe it all).

Debris might actually be a problem for this style of launcher - although I'd like to think you'd have the sense to give your nozzle a quick wipe if it got dirt on it, I think it may be more susceptible to ingress owing to the sheer size of the exposed surface. I did add a lip to the base-plate to try and angle the spray upwards during a launch, but that will only reduce the amount of dirt getting kicked up - not prevent it :P

I ran another couple of tests with it last night using a 20mm diameter nozzle and 100 PSI. Because the flange isn't terribly big, there isn't a whole lot to actually grab it by - so the stress at the contact surface is going to be pretty high. I'm currently using a wooden plank with a hole cut in it - but you can see where it's beginning to crush the wood around the rim (in the case of the lever I can just push the plank down harder as it deforms - but that won't be an option for the final design!


Say, see those zip-tie launchers, what sort of pressure can they take? Ah, smaller orifice... at 200 PSI they are only experiencing ~130N - it's more than an order of magnitude lower than my launcher will experience.

That said, 200 PSI is a lot of force at the larger diameter nozzles - at lower pressures and nozzle sizes it's much more manageable!

I hope my multi-nozzle launcher isn't just a pipe dream... get it!? :P

First things first, I need to rebuild my gardena launcher - just now it's a plastic box with a gardena connector sticking out the top.

My intention is to keep the gardena connector - but make the base-place detachable (it will be bolted down onto the top surface - but if you remove the bolts it should still be usable as a gardena launcher :D)

I'll sketch out my plans for the "base" launcher (that is, without the clamping mechanism added) later today, along with the various clamping mechanisms I've devised. My current favourite is two rods mounted on a ZZ-shaped armature - so upward force on the flange translates to tighter clamping pressure: the benefit of this is that by using threaded rods, the height of the upper surface (and thus the diameter of the clamping surface) can be adjusted - AND you can increase one more than the other to provide increased force (as a trapezoid acts as a lever and fulcrum if the vertices are articulated).

It's mechanically complex, but not terrifically so :P

The sketches will explain things better than words can, hopefully!

I'm going to dry to build the prototype out of wood, and then pressure test it to failure (I have 30m of airline so I can do it from a safe distance away - though I might see if my work will let me use their pressure testing chamber, though that would mean I couldn't actually test it by launching bits of pipe rockets ;))

Re: Wide-range universal launcher

Posted: Mon Jul 30, 2018 6:34 am
by cprobertson1
"Ooops..."

That is the sound a mad scientist makes when the top cap pops off their pipe-rocket at 370 PSI with a sound like a gunshot, replete with a rolling echo that chills your very soul... I am so glad I done that in outdoor bay at work.

I have however decided on a clamping method that uses a minimum of welded parts. It does, however, uses threaded steel bar that is welded to a piece of square bar. You could of course, drill a hole through some threaded tube and drop the threaded rod down inside it, and then add a nut on the opposite side - I'm just using welding as I have a bunch of welders who can put a good fillet weld on it :P

A you can see from the attachment, one of the threaded bars takes the majority of the pre-launch stresses - but in doing so, acts as the fulcrum for the clamp - which itself acts as a lever. As the rocket pushes up, it is held down by being unable to compress the threaded bar at the working side of the lever. If you need more force applied to the clamp, just raise the lever a little at that side.

To launch, just pull out the threaded bar that stops the lever moving. I've indicated "open slots" on both the upper and bottom surfaces of the support bar - I'm going to figure out a better way of doing that (and might even add a deliberate pivot point at the bottom side) - at the moment it will work but the wear on the lever will be significant (as it will jump down the threads, which will file away the wood with every launch. One of course doesn't need to use a threaded bar at this side - one could just as easily use a bit of wood! I just chose to use threaded bar as that's what I have lying around, and it'd allow me to adjust the amount of force applied to the lever.

Also, note that sketch isn't to scale - the threaded bar will only be about 100mm tall - no need to make it the length of the rocket if it's only holding down the flange 50mm up from the nozzle!

This was actually the hold-down clamp style used in the Juno and Titan rockets - only they didn't need to adjust the height, so they used rigid steel instead of threaded bar - but the principle is the same - it acts as a lever, and you pull out the support at the working side and voila - your rocket is now launching. Those clamps had to hold down the rocket at full thrust and release as soon as the command was received. You can also make it incredibly strong by doubling up on the fulcrum supports - and can exert great mechanical advantage just by lengthening the lever.

Re: Wide-range universal launcher

Posted: Mon Jul 30, 2018 1:53 pm
by anachronist
Even if you seal at the bottom, you can still use a launch tube. Even if the tube leaks, it will still provide an extra velocity boost once the rocket clears it. You could even have some minimum-sized tube (1/2" PVC) for standard bottles, and have sleeves to slip over it for larger diameter bottles. Your seal would still be at the base. The idea is to maximize the piston effect without consuming reaction mass. The presence or absence of a launch tube makes a significant difference.

Re: Wide-range universal launcher

Posted: Tue Jul 31, 2018 6:05 am
by cprobertson1
Indeed! I think I'm going to leave it out of initial designs for the conic-baseplate launcher for the sake of simplicity.

The big problem is that it makes the launcher less universal - that is - supposing I had a 10, 20 and 40mm nozzle - the max launch tube I can use for all three is 10mm. I'm currently looking at using a launch tube that clips into a gardena nozzle (in lieu of the the conic baseplate).

The alternative is to make the base plate exchangeable - which, owing to my plan to have a gardena attachment (one of the brass connectors) is actually really easy to do!

I think that will have to be the ultimate plan, unless I use a discarding-sabot style launcher - which makes everything more complicated!

So - standardised launch tubes: make them in a number of common, easy-to-acquire sizes - and just make the rocket nozzles the size of the nearest rod. They can be attached to the hold-down launcher with the gardena connector - and voila - just change the launch-plate around.

For everything else we can just use the conic sealing-base-plate that I designed above.

You could of course make a number of launch-rods with conic fairings that fit over the sealing cone and up the rocket's tailpipe - but at that point you are creating an extra seal - an superfluous failure point, especially when there's a perfectly good gardena connector right underneath the base plate!



So: the current plan is to get the launch-clamp system tested built and tested: and then upgraded to the launch-tube plates instead of the conic sealing plate.

So: the complete launch kit will have a base station with a gardena connector (which can be used for gardena nozzle rockets) - into which will fit either the universal conic base plate, or one of several launch tubes (will be specific to a particular nozzle diameter).

Sounds like a plan. Either way, looks like I'll have at least some of the launch clamp assembly built by the end of the weekend :D

Re: Wide-range universal launcher

Posted: Tue Jul 31, 2018 2:51 pm
by anachronist
cprobertson1 wrote: Tue Jul 31, 2018 6:05 am The big problem is that it makes the launcher less universal - that is - supposing I had a 10, 20 and 40mm nozzle - the max launch tube I can use for all three is 10mm. I'm currently looking at using a launch tube that clips into a gardena nozzle (in lieu of the the conic baseplate).
Well, the launch tube doesn't have to seal. It can leak. The point is that it provides a pressure surface for pushing the rocket without expending reaction mass. You can always fit a larger sleeve around the smaller launch tube to accommodate larger nozzles.

But as you said, an exchangeable base plate might be a better solution.

Re: Wide-range universal launcher

Posted: Thu Aug 02, 2018 4:44 am
by cprobertson1
anachronist wrote: Tue Jul 31, 2018 2:51 pm
cprobertson1 wrote: Tue Jul 31, 2018 6:05 am The big problem is that it makes the launcher less universal - that is - supposing I had a 10, 20 and 40mm nozzle - the max launch tube I can use for all three is 10mm. I'm currently looking at using a launch tube that clips into a gardena nozzle (in lieu of the the conic baseplate).
Well, the launch tube doesn't have to seal. It can leak. The point is that it provides a pressure surface for pushing the rocket without expending reaction mass. You can always fit a larger sleeve around the smaller launch tube to accommodate larger nozzles.

But as you said, an exchangeable base plate might be a better solution.
I was thinking more along the lines that the smallest nozzle I would be using is probably in the region of 5mm diameter - which would make the launch tube inefficient for the majority of my other nozzles :P

I like this exchangeable base plate idea though - they look remarkably easy to seal at the bottom (again, provided you have enough downforce from the launch clamps).

Still haven't found any plate/square tube in the skip to build the launch clamp with though - I guess I'll just be building the gardena side of things at the weekend with provisions for adding the launch clamps later. I do have a scuba swimming test next tuesday, so my time might be limited, depending on how much time I devote to training (I should probably train - the weather's crap just now anyway xD)

Still, it shouldn't take too long to get the foundation of the launcher made: here's hoping!

Re: Wide-range universal launcher

Posted: Thu Aug 09, 2018 6:10 am
by cprobertson1
I'm behind schedule! The weather was great at the weekend so I took a spontaneous camping trip... that's my excuse anyway ;)

That's me got the launch clamp support rods welded (see attachment) - just need to saw the lever into shape and we should be good to go clamp-wise! I think I'm going to armour-plate the back of the clamp so the wood doesn't wear down on the sliding contact surface (at least not to the same extent as bare wood)

Point of interest - but that weld is actually the strongest part of those supports - if you applied increasing force to the threaded bar, the threads, then the threaded bar, and then the smooth bar will all fail before the weld itself does! How cool is that!



The launcher is going to be mostly modular - the current plan (as already discussed) is to use a basic gardena launcher as the base (the "box launcher" I call it... because it is ultimately a wooden box with a hose and some control cables coming out the side, and the gardena connector and some energy dissipators on the side (trying to reduce the force the water jet makes on the ground)

The Box-launcher can be used on it's own to launch gardena-nozzle rockets - or you can add an adaptor plate and launch clamps onto that so it can handle larger-nozzle rockets (optionally with launch tubes): the clamps hold the rocket down against the plate and create a seal using one of a variety of methods (pretty much always an o-ring, though I do also have plans for a rubber gland that can withstand higher pressures).

That brings us round to issue #41, launch tube fixing.

I have several designs of launch-tube - but in order to rigidly affix the launch tube to the launcher, the launch-plate ends up having a lot of depth to it - which causes me mechanical problems (it creates a lever with the tall base plate acting as a fulcrum) - and material issues (taller base plate needs to be wider - more material)

But it has got me wondering - how much lateral force is there on the launch tube? When I picture this in my head - the rocket thrust is parallel to the launch tube - so the tube needs to be rigid enough to take the force of the rocket accelerating to the end of the tube (which will be a function of the mass of the rocket and it's acceleration)

Does that mean I can get away with only having to add vertical support to the launch tube and not waste material installing it in a deep recess?

You can picture this as pushing a pencil into some sand. If you push the pencil in a little bit, you can knock it over easily. The further you push it in, the harder it is to knock over. In the case of a water-rocket's launch tube - does it even matter it it can be knocked over (bearing in mind that the launcher can be designed with guide rails)?

Re: Wide-range universal launcher

Posted: Thu Aug 09, 2018 9:33 am
by anachronist
I doubt there's any lateral force unless you're launching the rocket at an angle. In theory, to function as a launch tube (that is, to give the rocket an extra velocity boost before the reaction mass starts being consumed), it doesn't need to be attached at all. As long as it's kept from shooting out the back of the rocket by your launcher, it will serve that function. I would think that the only reason you'd want to secure it to the launcher would be to seal it (assuming your rocket will be pressurized through the tube).

Re: Wide-range universal launcher

Posted: Thu Aug 09, 2018 10:26 am
by cprobertson1
anachronist wrote: Thu Aug 09, 2018 9:33 am I doubt there's any lateral force unless you're launching the rocket at an angle. In theory, to function as a launch tube (that is, to give the rocket an extra velocity boost before the reaction mass starts being consumed), it doesn't need to be attached at all. As long as it's kept from shooting out the back of the rocket by your launcher, it will serve that function. I would think that the only reason you'd want to secure it to the launcher would be to seal it (assuming your rocket will be pressurized through the tube).
Excellent - that's what I thought, but wanted to double-check!

In most (but not all) of the designs I've came up with so far, the force of the nozzle against the base (aided by an face-sealing o-ring) does most of the sealing - though I've sealed the base of the launch tube in all cases - I reckon that if it weren't sealed, water would be forced through the tube and out the unsealed areas - causing you to "waste" some of the gains you made by adding a launch tube in the first place.

Worst case scenario (in which the water can flow unrestricted through your tube during launch) you would lose the piston effect across the inner diameter of your tube: it'd only be pushing against the actual material of the tube (plus any resistance to the flow inside the tube): sealing it prevents that and allows you to push against the entire diameter of the tube.

Time to bring out the silicone gun ;)

Re: Wide-range universal launcher

Posted: Thu Aug 09, 2018 10:28 am
by cprobertson1
**Addendum: one could also add a one-way valve to the end of the launch tube and not have to worry about sealing it too well in this case - it'd prevent backflow during launch when the seal at the bottom is broken. Food for thought if I'm trying to keep things very simple (I can just silicone a valve in, or use a backflow preventer from the plumbing store, or DIY something if I'm feeling adventurous xD)