Speargun Build Simulator

[Jesse_Spiller]

Hey Guys,

I've been working on a Speargun Simulator for the last few months. I tried a few programs, but nothing was really shareable before I committed to learning Python/Django/CSS/HTML/JavaScript and renting my own server environment. This is the result:

Technica - Speargun Simulator (link)

These are the instructions on how to use it (link).

It works by equating the momentum and energies in these equations (link) that I derived as part of my thesis. I've tested 27 rubber samples from many manufacturers in various diameters to derive forces, energies and friction losses. Coronavirus has prevented pool testing so far, but the model should incorporate friction and hydrodynamic drag by the end of the year.

I'm looking to incorporate a graphical results display soon (something like the graph below), as well as a whole bunch of other features.

It's really good at quantifying the differences between two guns. For example, you can input two identical guns and change one parameter (eg shaft diameter) and see how it affects the performance (an example of this is linked here).

It's interesting to see how:
- The ratio of shaft weight to stock weight affects recoil,
- Conventional, roller and inverted guns compare to each other,
- Efficiency changes with different power systems and stock weights,
- The effect of various band diameters,
- Adding kicker bands adds or detracts from performance?
etc etc

I'd be keen to hear:
- Did you find any errors/manage to break it?
- How you think it could be improved?
- If there's any features you want added?
- Did you find any interesting trends or relationships?

JS.

 

[Auspicious Spearo]

Hi JS, Nice plan and could be a really nice tool for spearos keen on the theory! Straight off the bat though, there is no mention of band stretch. What is it assumed to be? Could this be an input by user?
I will spend a bit more time on it in the coming days.

 

[Jesse_Spiller]

Hi JS, Nice plan and could be a really nice tool for spearos keen on the theory! Straight off the bat though, there is no mention of band stretch. What is it assumed to be? Could this be an input by user?
I will spend a bit more time on it in the coming days.

Hi ND,

The band stretch is based of data from stretching the bands to 350%, leaving them stretched for 30 mins, and recording the energy they kept stored. The graph below is the average of 8 different 16mm rubber brands. The blue region represents the energy lost, about 30%.

I didn't test past 350% or put %band stretch as an option for a few reasons:
- Stretching past 350% caused permanent deformation of the rubber, which made checking results impossible.
- The force started increasing sharply post about 320% (this is actually really bad for efficiency).
- It increases the complexity of the simulator (I had this as an option previously).

Thanks for your observation, I will put the 350% disclaimer on the website.

 

[Kodama]

In that case it might be better to use 320% for the calculation some in reality we rarely stretch bands over 320-340% when using small id bands. With regular bands often less. So perhaps using 320% would give more real life results.

 

[Jesse_Spiller]

The code has been setup to take streatch % as an input, but it adds 3-6 more options that might confuse people. It's sitting at default 350% for now.
If you've got suggestions that would make the simulator easier to understand, or perhapps more visual, I'll look at including band stretch as an input.

 

[Andrew the fish]

intersting calculator. I like the recoil energy and efficiency, interesting to know. But perhaps, the recoil accelleration will be more useful number, not just the energy, because this is what moves the gun, and makes us miss the fish. For that, the Calculator needs to have weight of the gun factored in. Obviously, we also need shaft velocity at distance, like at the muzzle, 1m, 2m etc. Or shaft momentum rather, since shaft length, diemeter and hydrodynamic efficiency will combine to produce the most important factor - penetration at distance. Actually momentum is more important than kinetic energy or velocity for example, because this is better translated into the pentration in real world. Easy to calculate too, if we know the "ballistic coefficient", if we can call it that.

Also perhaps much needed factor is total force. This is the weight trigger is holding, and what bends the barrel . Without the force limit built into the calculator, this is all dream land.

 

[Jesse_Spiller]

Hi Andrew, thanks for the feedback.

Obviously, we also need shaft velocity at distance, like at the muzzle, 1m, 2m etc

I agree wholeheartedly; penetration at distance is THE critical factor. To determine this, I need to determine how much energy is lost due to friction underwater (both the gun and the shaft) then do some penetration testing. Unfortunately, Corona has prevented this so far. There is 30 kgs of ballistic jelly currently sitting in my garage . This should be complete by the end of the year.

perhaps, the recoil accelleration will be more useful number, not just the energy, because this is what moves the gun, and makes us miss the fish.

The force is what moves the gun; you can calculate the acceleration using A=Force/Mass, but it wouldn't be very useful as the acceleration lasts for varying periods of time, depending on many inputs. The only objective measure of recoil is 'free recoil', as all other measures need to consider how the firearm is secured, which is well beyond scope. Free recoil 'energy' is the simplest number to interpret as it doesn't change. You can also have free recoil 'velocity' (which factors stock weight). A higher velocity makes the same recoil energy feel sharper, whereas a lower velocity makes the same energy feel dull - but these are very much open to human interpretation.

Or shaft momentum rather, since shaft length, diameter and hydrodynamic efficiency will combine to produce the most important factor - penetration at distance. Actually momentum is more important than kinetic energy or velocity for example, because this is better translated into the pentration in real world. Easy to calculate too, if we know the "ballistic coefficient", if we can call it that.

The ballistic coefficient can only be experimentally determined - that is part of the next major project. The coefficient will give shaft speed at distances.
Momentum, velocity and kinetic energy are all related, but kinetic energy is the most important factor for penetration depth of low-speed non-deforming projectiles, along with shaft diameter and fish cavity strength (see attached):

Xp = penetration depth, Yc = fish cavity strength.

Also perhaps much needed factor is total force. This is the weight trigger is holding, and what bends the barrel . Without the force limit built into the calculator, this is all dream land.

Thanks for your really good idea of displaying the total trigger force. Trigger mechanism force will be included in results within the next few days.
It might also be worth calculating roller and/or pulley forces...

Thanks Andrew, you've given me a lot to think about. The simulator will be better for your input!

 

[DanielPi]

via pimion​

Hi JS,
An impressive calculator and research topic in general. I'm currently in my third year of the mechanical engineering program at PSU and I've been immensely unlucky with my research on the Design and Optimisation of Spearguns. I want to change my research topic (though I've been working on it for 3 weeks already)...
May I include a reference to your thesis publication (if any) or use the link to the calculator itself? Thanks.

 

[Jesse_Spiller]

Hi Daniel,

How have you been unlucky? Has COVID-19 caused you issues?

What exactly is your research topic? Have you read Worthington's paper (link)?

If you want to change topic, perhaps you could focus on an aspect of speargun design?
- Spear hydrodynamics and accuracy factors is huge and unexplored.
- What causes shaft whip/buckling (Newton's column theory?) and can you model it?
- Nobody has done good enough experimentation with natural latex (me included) to model it - this could be great.
- What about experimenting and designing something better than a flopper head, or a breakaway? Neither designs are drag efficient.
- Can you make shafts from tungsten? They would be thinner for the same sitffness.
- What about a better trigger mechanism? Apart from the ErmesSub double roller, they are all the same design forward or reverse mech.

You can use whatever of mine you want but it won't be published for about another 9 weeks.

JS

 

[popgun pete]

You want to look at the Russian forums, the Russians have been working on speargun calculations for years, but the emphasis would be on pneumatic and hydropneumatic guns due to cold water spearfishing where bands don’t work so well. When I was communicating with Sergiy Kravchenko during the "Black Sea" gun development he indicated what the performance figures should be using these calculations. However the predicted results did not completely accord with the experimental results which meant that despite all the factors being considered they had left something out, or some of the input parameters were wrong.

 

[popgun pete]

Just remembered this one. https://forums.deeperblue.com/threads/ultimate-speargun-properties-calculator.110849/#post-973964

 

[Kodama]

- What about experimenting and designing something better than a flopper head, or a breakaway? Neither designs are drag efficient.
- Can you make shafts from tungsten? They would be thinner for the same sitffness.

I have been using the Sigalsub Ghost shaft with great succes. I have replaced the rubber band to hold the flopper in place with a sliding shaft cone and can say that this shaft is extremely accurate. Obviously the accuracy is not determined by the shaft but I dare to contribute part of it to the hydrodynamics of the spear. The sharkfins are minimal and the flopper disappears completely into the shaft.

Tungsten was considered by Spearq8 in the past but if I remember correctly it was ruled out because of production costs. however it is likely that you could increase performance by using it as a spear material.

 

[Jesse_Spiller]

I have been using the Sigalsub Ghost shaft with great succes. I have replaced the rubber band to hold the flopper in place with a sliding shaft cone and can say that this shaft is extremely accurate. Obviously the accuracy is not determined by the shaft but I dare to contribute part of it to the hydrodynamics of the spear. The sharkfins are minimal and the flopper disappears completely into the shaft.

I haden't heard of this shaft - thanks very much for mentioning it!
It makes shortshafting possible...

 

[Jesse_Spiller]

Thanks for pointing out some interesting references Pete,

Just remembered this one. https://forums.deeperblue.com/threads/ultimate-speargun-properties-calculator.110849/#post-973964

That is the Excel spreadsheel Speartool (link); the product of Jacob Worthington's 2017 thesis: An Analysis of Speargun Performance (link).
Jacob kindly provided me with a copy of Speartool and his raw rubber data; it is a good foundation on which to build.
Key differences:
Speartool only does conventional and roller guns, not inverted.
Speartool doesn't factor recoil or band movement.
Jacob didn't let his rubbers sit when testing, so noticed little hysteresis. I left mine loaded for 30 mins, during which they lost about 30% energy.
Jacob has incorporated barrel flex, which I have omitted.

You want to look at the Russian forums, the Russians have been working on speargun calculations for years, but the emphasis would be on pneumatic and hydropneumatic guns due to cold water spearfishing where bands don’t work so well. When I was communicating with Sergiy Kravchenko during the "Black Sea" gun development he indicated what the performance figures should be using these calculations. However the predicted results did not completely accord with the experimental results which meant that despite all the factors being considered they had left something out, or some of the input parameters were wrong.

Could you link the specific forums? Threads? Are any in english?

Thanks, JS.

 

[Kodama]

Regarding loss of power after time you may find this test interesting.

 

[Jesse_Spiller]

Regarding loss of power after time you may find this test interesting.

The guy had a good idea, but the test wasn't useful at all. All shots penetrated and reached the limit of the shooting line; he needed to either increase the taret thickness or density to make any substantial comparisons.
The only thing he was able to observe was his opinion that the recoil felt similiar, which a human can't really quantify very well.
I'd like to see it with a better target, and recorded penetration.

 

[popgun pete]

http://apox.ru/forums/forum/65-podvodnaya-ohota/ is one of the main forums in Russia. On-line translators do a good job of translating into English. Vlanik put up one of his calculators here on Deeper Blue which is an Excel spreadsheet, a search should find it and it is most likely in the pistonless gun thread. The file name is vlanik.xls. Reference to it here https://forums.deeperblue.com/threads/vlanikgun.90450/page-3

 

[Jesse_Spiller]

Hey guys,

There's been some more work done on the simulator. Thanks for all the feedback, especially Kodama - I have shown my appreciation by implementing some of the most popular requests.

There are three new features:
- You can now enter bandstretch (from 300-380%)
- It factors drag (guesses shaft deceleration over 8m)
- It estimates penetration at ranges up to 8 metres.

Apart from rubber experimentation, no testing has yet been completed (thanks COVID-19). The model is only theoretical at the moment and it's limitations are discussed more in the attached document (Limitations.pdf) if you're interested. It needs more testing to account for friction, define a drag coefficient and determine target cavity strength.

There were questions regarding how the rubber energy worked, see attached Rubber_energy.pdf for a more detailed explanation.
The experiment results are summarised by these graphs for 14mm, 16mm, 18mm & 19mm bands.

The black dots are the experiment data points; you can see the spread across 14 & 16mm band diameters was quite large. This was due to many factors such as inner bore diameter and whether the rubber was dipped or extruded. The lines are the average for each diameter - this is what's used in the simulator.

There were some questions regarding efficiency and recoil - how are these determined?
Efficiency is how much is the energy in the bands actually goes to into the shaft. Different guns waste more energy on recoil and in the bands themselves, exactly how this is worked out is attached as Equations_of_Motion.pdf

Drag and penetration are just guesses at the moment. The framework for them to be semi-accurate is there but experimentation is needed to determine two variables that can't be done theoretically (drag coefficient and cavity strength). In the meantime, I'm using a drag coefficient of 0.82 (same as a long, thin cylinder - Wikipedia) and a cavity strength of 20 Mpa (slightly weaker than human skin). Some background on this is attached as drag_penetration.pdf.

There was also some confusion about combining different guns types in the simulator. It can be definately be done, for example 1 roller band and 1 conventional band is the same as a roller with a kicker.

Kodama came up with the idea to canvas spearos for specifications of their guns (specifications aren't on many manufacturers websites). He made a form that we could distribute and build a library of gun data. Ideally, you could select your gun and the simulator fills itself out...
This is definately feature worth investigating - we're talking about exactly what data is needed at the moment; don't want to make it too hard, else less people will contribute.

The next feature addition will be graphs - the data table is ugly and hard to read.

Please keep the feedback coming. I want to know if you break it or get an error - you guys are the Alpha testers.

Regards,

JS.

 

[Jesse_Spiller]

Re-linking for convenience:

Simulator: http://www.technicaspearguns.com/simulator.html
Instructions: http://www.technicaspearguns.com/instructions.html

You can copy and paste the results hyperlink into a reply on this forum; others can click it and your inputs/results will appear.

JS.

 

[johnakerblom]

Re-linking for convenience:

Simulator: http://www.technicaspearguns.com/simulator.html
Instructions: http://www.technicaspearguns.com/instructions.html

You can copy and paste the results hyperlink into a reply on this forum; others can click it and your inputs/results will appear.

JS.

Thanks for all the in-depth research. Very cool and informative.

I've built traditional guns, roller guns, and an inverted roller gun. Lately, for my new blue-water build, which will be a gun approximately 58" in total length with a 50"span from the muzzle to the wishbone of the spear. This is the longest the gun can be where I can reach pre-tensioned roller bands to and load from my hip.

My problem, like I think a lot of other gun builders experimenting with new designs, is that it's really tough to make apples to apples comparisons between band types. Generally, with a 58" long, 11/32 diameter spear, I know that four 16 mm (5/8") bands stretched to 350% about maxes out the power where adding another band won't accelerate the spear any faster because the band retraction speed is the limiting factor. With traditional band guns, it's easy to add and remove bands until you find the sweet spot where you max out the power for whatever purpose you are seeking without overdoing it and needlessly increasing recoil and loading effort for no gain.

But I don't know exactly where that point is with rollers or inverted rollers. And based on my limited understanding of physics and the guns I've built, I have some serious doubts whether inverted rollers with four or less bands have enough power to take advantage of the pullies (i.e. I don't think most of them are powerful enough to accelerate a thick spear beyond the band retraction speed, which is the only real advantage inverted rollers offer - besides narrower muzzles, recoil, and other ancillary concerns).

To that end, I built a rig using a winch, a crane scale, some lumber, and some hooks to take static measurements in 2" increments of each type of band from 0" to 50". While my methodology is not perfect, I think it is good enough for comparative purposes. I am wondering if you, or anyone else on this site, could take a shot at calculating the total force/energy created by each of the bands based on the measurements in my chart. Presumably there is a way to simplify the data to make it easier (i.e. assuming a simple linear plot vs. the more complex one implicated by the raw data).

No idea if this is a lot of work or something you could do in five minutes. Thought I'd ask on the off chance you found this interesting.

I figure it might be helpful to other builders since this size gun is about the biggest you can build and still be able to load with the butt pad on your hip (at least for people 6' tall or shorter with a similar reach to me. Unlike traditional band guns where you can build a six foot gun and grab the slack bands, with a roller you have to be able to reach all the way to the muzzle.

Anyways, thanks for contributing the the pool of knowledge out there. A lot of speargun building is still trial and error and its great to see people with expertise wade in with ideas and information.

John Akerblom