I think I made the point poorly before. The bolts make it worse, but even in the simply restrained case (no bolts, bonded to its surroundings), you are looking at stresses in the neighborhood of yield for the sabic plastic with a 100 psi load... This means that the 100 psi load is really an acceptable proof pressure, not an acceptable operating pressure (think 1/2 to 1/3 of that). This is all great work to make cannons more available, but it would be for naught with only one accident. Consideration for some analysis or testing must be made. A copy of Roark's Formulas for Stress and Strain will allow simple geometry to be checked (and should be available at any decent library that has a technical section, see local universities) the other option is to test the parts to failure. since we are unlikely to be able to test fatigue to failure (and the fatigue properties of plastics are a bit more dubious than that of metals, and harder to find) that means that there needs to be larger margins between the operating pressure, the proof pressure and the burst pressure than I might design in with a proper analysis tool. I can not emphasize enough the good work going into this thread. pardon my boat anchor of "check it before we grenade it" but I want to see this successful.
Absolutely - I've worked in engineering firms for near 30 years: not as an engineer, but CAD/Tech support, but one cannot fail to pick up something along the way. I have access to a whole bunch of nerdy engineers at work, and access to some rather nice software. I'll see what I can fish up as I develop this concept further. I've been wracking my brain to no end trying to figure out a way to isolate the magazine somehow all along. Come to think of it, a half dozen larger holes in an aluminum disk - say 3/8" holes, and that covering the lexan...just something to look inside and see the slides is about all one needs I think, to visually locate debree or bearings. What do people use for the ammo fill tube? Some sort of screwed in plug? That seems another "weak" part of the cover.
Greg, I understand your concern, but... isn't that a little extreme? In all the years I've been building cannons, I've only seen one magazine lid fail. And that was due to balls shooting out the ammunition ramps striking an excessively thin piece of acrylic, not due to pressure. Those rounds are a concern worth discussing. I've seen a lot of cannons covered in dents on the inside due to rounds getting shot out the ammunition ramps, and I regularly see people applying extra internal armor to their cannons to protect the magazine walls where those ramps are aimed. The basic operation of a Big Gun cannon naturally forces rounds in the ramp back up and out, to prevent machine-gunning. Those rounds can move with quite a bit of force, enough to break smaller, less-sturdy plastics. So if someone wants to use weak, brittle 1/16" acrylic for their lid, of course they're going to have problems. 1/8" lexan, on the other hand, is the standard for magazine lids in the WWCC. I have used 1/4" lexan before, when the local shop didn't have any 1/8" in stock. The thin blast shield lexan I'm using in the laser cut cannons is indeed very thin and experimental, but it is holding up fine with no signs of wear at all.
It is not extreme unless you can provide data that either the boundary conditions are wrong (and can be reduced such that the analysis says you are ok) or are ok with the risk of lawyers. My BC's are a swag at what really happens, and could be pessemistic. If they are, they can be modified to prove things are ok. That being said, we really do have to be careful because one enterprising lawyer who looks at something that was absolutely not designed per practices used in industry can destroy us all with only one accident. We deal with pressure vessels here at work, granted much higher pressures, but still pressure vessels. It is simply a matter of how much risk of liability you are willing to tolerate for the hobby as a whole with the understanding that one injury of a non-participant (or disgruntled participant) could cause us all a whole lot of problems. As the cannon designs are created with the intent to enable more people to make them, we have to keep that in mind. I'm happy to use the codes I have access to to do analysis on the designs to help improve them and prove they are safe, primarily because easy big gun cannon access is needed, and I want to see the hobby grow without any unnecessary risks. I welcome challenges to the boundary conditions, and better information to validate analysis. The point of designing properly is that you never ever get near a failure on a part that could cause liability/injury. The only way to test to that is to test to failure and verify that the failure is sufficiently far from operating conditions to reduce the risk to acceptable levels. The failure testing also shows whether the failure will cause a risk of injury or not. What we do at work is we proof at between 1.5 to 2x the design condition, verify no yield, and then expect burst to be above 1.5x to 2x proof pressure, depending on the design. We test to those conditions precisely because we want to reduce our risks...
Cool discussion - please keep it going with your insights and knowledge. It will most certainly keep people interested in maintaining a safe operating environment. I've been doing tons of studying and reading over the years, and know just about every clubs rules and construction guidelines by heart. (I was even toying with the 1/96 scale idea, but the HMS Nelson would have needed a VERY large pickup truck or a trailer...I doubt that the wife would have appreciated that.) Anywho, back to the cannons. I was (re)reading the plans that are floating out there that one can buy (for the various cannons), and I noted that they advocate PVC, Acrylic and very much smaller material thicknesses. For instance, the walls of the lathed magazine on one cannon is only 0.10" thick when finished, and the cover is held down only with 2 screws. We all know PVC will fail catastropically (and will shatter into sharp pieces) and is not the best material to use IMO for any portions of these cannons. Hence my idea of going with HDPE, 0.25" thick walls for the magazine, 1/8" thick lexan and more screws (I added 4 more to bring the number to 8 - but that may be overkill) Here is a picture of what the wafers would look like. Thomas
There is a key point missing in this discussion: the magazine is not, I repeat NOT a pressure vessel. The gas from firing vents out the loading tube and openings for the barrels. Unlike a fast gun cannon, air is pressurized and stored in the accumulator part of the cannon base, not the upper magazine. When the cannon is fired, gas will leak out of the breach and travel back up the feed ramps and vent to atmosphere. edit: The unsealed, open breech is one of the big losses of the Big Gun "Indiana" Cannon. Many people have tried to build a mechanism to seal the breach and eliminate the losses at the open breech, but they are generally much more complex to manufacture and maintain, and are more likely to malfunction or fail. On the other hand, it means that a cannon base can be built and have different magazines attached very easily. I've seen the same design cannon base used for fixed (non-rotating) triple torpedo cannons, fixed six barrel (2x triple) cannons, rotating three barrel and rotating four barrel cannons. I've also seen good magazines salvaged from bad cannon bases and given new life in a new ship.
The magazine does indeed see a pressure pulse. (magnitude in question right now). That makes it a pressure vessel in my book. it does not see steady, that is true, however with the pressures being equal, the transient situation is far WORSE from a life perspective.
To what extent is back-pressure in the magazine necessary for propelling the ball bearings down the barrels? To take an extreme example, what if the lid of the magazine was wire mesh -- what percentage of muzzle velocity would be lost?
Fluid Dynamics...fun. One should be able to adopt a pressure gauge to fit into the feeder barrel (after loading the magazine) and take a few shots, and watch the pressure gauge "jump".
You may have a hard time actually seeing the peak pressure, as the dynamic response of the gauge is not necessarily fast enough to catch the peak. The thing to keep in mind (and I would say I have done a poor job of typing my thoughts on this) is that thoellering has done a good job addressing what I long considered the highest risk item in big gun cannons, the pvc accumulator, in a manner that takes into account both normal operation and abnormal operation. next up is the question... what component fails next that actually poses a risk of injury, how, and what ways to address it. In my mind, that is either the joint between magazine and the rest of the cannon, the barrel retention, or the top of the magazine. Of those three, I think the magazine top is the higher risk (though it could be argued either way) and it must be considered for both worst case normal operation and abnormal operation. So what are the ways to address the risk, particularly with poor understanding of the dynamics of the cannon, and prove it has been addressed? A. over-engineer to a point where any worst case will not fail the part B. Engineer in a failure point that fails first (such as a burst disk or a deformation where gas vents well before stress becomes too high) but fails safe and how do we prove that it is good. Testing at operating parameters won't tell us enough, unless we want to run a statistically significant sample through orders of magnitude more shots than they will ever see in operation. Testing the tails multiplied by a factor, e.g. proofing, can help prove it, particularly in the fails safe design case. so that being said, what are the worst case normal conditions? I would say worst case normal would be last round being fired, as the pressure pulse would not be attenuated driving rounds back into the magazine, along with any impact damage that may come from the second to last round case hitting the lid. worst case abnormal would be obstructed barrels....
> thoellering has done a good job addressing what I long considered the highest risk item in big gun cannons, the pvc accumulator By using a standard commercial valve which allows a standard commercial tank for the accumulator? The more off-the-shelf bits the better, I agree.
Great job on the Cannon design. Looking at the 1st drawing,is that MRM-6 manifold from Clippard? (red) Somone who has experience with the cannons can add more information on how much air flow is needed and if this is going to work The information on Clippards list the MRM-6 as 1/8 NPT inlet with 6 -10/32 outlets. 5.9scfm at 50 psig. Phil
Yes - I'm not sure if that is at each outlet - I need only 2 of the openings for the H-39. I finally put all the pieces together (roughly) to see what it would (might) look like. Here is the album with more images: http://photobucket.com/H-39_Cannon_Design_TH I'm using 1/4" steel tubes. For the before mentioned cross sectional area: http://www.hobbylinc.com/htm/k+s/k+s7115.htm Steel Tubing: 1/4" OD, 0.028 Thickness Inside Area Clear Area: 0.0296 With 2 Barrels = 0.0592 With 3 Barrels = 0.0888 With 4 Barrels = 0.1184 Typical clippard 15/32 opening = 0.1103 So that should be OK so far also. By the way: here is a free 3D drafting program from Autodesk. http://www.123dapp.com/create Here is also the option of creating a 3D model (like I did above) and have it printed...Resin, ABS, Brass....heck, even Titanium. But that be spendy. In the album above, there is a quote for the magazine I created. $30 something for having the thing printed in resin...it has 4 out of 5 wrenches for strength. Might be something to look into. 3D printing is quite the thing now-adays... Thomas
I appreiate all the work youve done on this and even sharing the information! I do have a couple questions Are the barrels pressed in ? and is there plans in mind for barrel depression? or just the standard hose and clamp set up ? Phil
Good question: I had the following in mind. As with all things, as one works on one part, another part may need revising. Given the nature and complexity in trying to simplify things, I'm trying my best to incorporate safe handling of materials, pre-manufactured components, and keep the cost down as well. The convenient component is the 3D CAD work of course. Anything and everything can be form fitted, tested and I'm already looking at downloading a copy of Autocad Mechanical, which allows a user to perform stress analysis on parts and pieces. Its pretty cool stuff. Back to the depression. I had in mind to consider something of a range set point for the guns. I figure that each 1 degree in depression from X will change the range of the guns by about 5 feet with the barrel of my H-39. There should be a way to lock that concept in somehow. Say have the guns, when fully elevated be at a certain degree below zero, so that the range is set to about 20 feet, and then have 4 set points on the controller that will depress it from Long (20 feet), to Medium (15 Feet), Short (10 Feet) Point Blank (5 Feet). Since I learned that beyond 30 feet these bearings are somewhat erratic, that makes sense. That ships "roll" is also a given, but thinking that 2 ships sailing at parrallel course at about 2 ship lengths appart should be at about Short Range (10 Feet). Using relative ship lengths for a measuring guide should help with long distance range finding. With that in mind, I wanted to try to set up the guns so that the vertical riser and the elbow is locked in (a solid long piece) and bent to meet that 20 feet angle when the gun is at set point Long Range, and that the barrel depresses after the bend (with whatever method works best in the hobby - tubing seems to be the favorite thing to do?). I think it would require less force from the servo, so perhaps a micro servo might work. [I dont know much about servos (yet) so that is all just wishfull thinking on my part.] In part of all this thinking, is the thought of having a bearing firing straight up out of the cannon, and if one of those vertical riser tube connections would be broken or loose or missing...at least with the complete vertical riser and then the bend, that bearing would shoot someplace other than where your head might be, as I think that scenario could well play out when working on a ship. Also, it will allow supports for the barrels to be placed there on the vertical, and somehow tied into how the turret attaches to it. Once again, just brainstorming ideas that have not been put to paper or CAD yet. Other thoughts I've had consider several smaller volume chambers on a manifold system, or smaller acumulators so that the whole gun assembly will not be so tall, saving more space by using thinner wafers, or using aluminum in certain locations etc. For instance, none of the hole sizes are locked in yet, they are just rough and a bit oversized. All the Clippard Parts are to scale, and so are the wafers. The slides seem right for the 0.177" BB's. What drill bit sizes to use, the manufacturing process, the templates and all these things are in the rough draft right now. As I work on one thing, I find that I have to tweak another, slowly tightening the process. Thanks a ton for the great feedback from everyone! Its amazingly gratifying to finally put my thoughts onto paper and have experienced persons provide such awesome feedback. By the way, I contacted a company called "Airtronics" and had them send me some quotes: From: Airtronics [mailto:airtronics@airtronicsinc.net] Sent: Wednesday, July 18, 2012 9:54 AM To: Hoellering, Thomas Subject: RE: Quote Hi, The prices are below. The lead time is about 1-2 weeks. The only data I have is what the catalog tells us, 5.9 SCFM @ 50 psi, that should be flow from inlet to one outlet. At 100 psi I would expect around 15 SCFM. Thanks Denise From: Hoellering, Thomas [mailto:my-work-email] Sent: Tuesday, July 17, 2012 10:12 AM To: 'airtronics@airtronicsinc.net' Subject: Quote To whom it may concern: I’d like to receive a price quote (including shipping) for the following items: Clippard: 1 x AVT-24-4 $16.82 1 x MJV-3 $12.36 1 x MPA-5 $7.69 1 x MRM-6 $8.69 1 x MAR-1-2 $13.65 1 x 1/8” to 1/8” NPT connector (to connect the MJV-3 to the AVT-24-4 – please verify for me that this can be done) H0202 $2.54 Thank you, Thomas
Airtronics is good people, I've dealt with them before. If you want to keep your costs down, I've often seen clippard components in varying quantities and types on fleabay. Some used, some brand new. Fleabay isnt useful for consistent pricing, but it is useful for picking up cheaper one-offs for trying things out.
Good thinking - I will keep that in mind Fleabay...nice. So I'm working on my stern keel, and decided to go with stainless steel sleeves for my home-made stainless steel stuffing tubes. Tore out the brass tube I was using before (somehow it got a kink in it..). Anywho, I set a 1-1 scale print of the sideview of the gun I'm developing here at its proper location. Without having any of the parts, I can make adjustments to my layout and modify ribs. The center prop will cause major issues, but I might have an idea of how to litterally build the cannons around it. I need to print a front view of the assembly and see how that looks. Then play around with the CAD file a bit...I have the ribs. Easy enough to 3D them and see what I need to do to make things fit in the virtual model.
Micro servos won't work for the depression. The repeated shock of the cannon firing can strip the gears out of a standard servo, so it'll demolish a microservo.
Ahh - so many things I do not know. Thanks for sharing. I did a bit of rework on the design, in an effort to save space. I have a center prop to think about and had a cool idea last night while pondering. I often see the barrel sticking through the magazine, and supported by the mass of the gun below the magazine. So I offset the open area of the magazine that holds the rounds and made that smaller, so about 2-1/4" diameter "hole" offset on a 3" circle. And the barrels are now sticking in the "rim" and through the lexan. And I shaved another 1/8" off another waver, the one that feeds the air from the Rotating Manifold to the ... well, Manifold. Thats 3/16" thick now. Made a whole lot of difference. I'm thinking of a way to incorporate the flexible hose clamps into the support bracket for the barrels...clamp over the hose and barrel and hold them together. Not sure what or how yet. And perhaps have an attachement point for a clevis on the forward one to depress the barrels by up to 4 deggrees from horizontal. Thats the blue dashed barrel shown below. The cyan lines below is the prop shaft for the center prop. Now I gota find a place for motor. Here is a plan view of the wafers...some are larger to acomodate some sort of belt drive, sproket drive, or wheel drive, and or to allow more material where the screw holes or the barrel holes are a bit close to the outside edge. Kill multiple stones with one bird sort of thing...read from left to right, down and left to right: Lexan Cover (1/8" Lexan) - Magazine Rim (1/4" HDPE) - Magazine Slides (1/4" HDPE) - Manifold Top (1/4" HDPE) - Manifold (3/16" HDPE) - Base Plate (1/8" Aluminum or HDPE) Cheers, Thomas
