Spring Piston Simulator Project

[MartyMcFly]

Total Recall Recoil!

Time sure flies when you're working, said no body, ever! Finally, I've stolen sometime away from real life to provide an update post. Recently, I've been thinking about recoil and how to implement it. Why recoil? Well, I think it would be cool to know how recoil depends on the rifle's characteristics.

We can say confidently that recoil will largely be dependent on the mass of the rifle, the mass of the piston, the power of the spring, and the mass of the pellet. The larger the rifle's mass, the lower the recoil. On the opposite end, the higher the power or more massive the piston, the bigger the recoil. Furthermore, old man Newton told us a long time ago that every action has an equal and opposite reaction, and something about conservation of mass and energy. In other words, I've used more of Newtons laws to workout recoil.

That said, the key things I added to the simulator was tracking of momentum, specifically, tracking of pellet, piston, and rifle momentum. Keeping account of these quantities gets us recoil velocity and recoil energy.  I think the main take away is that the rifle's recoil velocity is mostly determined by the piston's velocity. Intuitively it makes sense, the piston assembly is the only thing with significant mass and velocity that is moving around in the rifle. Therefore, controlling the weight and velocity of the piston gives us a way to control recoil, IF we keep the rifle's mass constant. We could of course raise the mass of the rifle to reduce recoil, but who wants to carry around a Howitzer...

Anyways, getting back on track! What's the point of simulating recoil? Well, I hope that we can experiment with different piston designs (i.e. mass, stroke length, etc.) and be able to explain what kind of recoil impulse the rifle will have.

Attached is the new output with recoil graphs - hopefully they make sense!

-Marty

[JPSAXNC]

I wonder about the idea that an action has an opposite and equal reaction. Because action requires energy so some energy must be lost in the action, leaving less energy for the reaction? Also the piston with more mass creating more recoil I agree with if the piston is moving fast, but what if the piston with more mass is moving slower?

[MartyMcFly]

I wonder about the idea that an action has an opposite and equal reaction. Because action requires energy so some energy must be lost in the action, leaving less energy for the reaction? Also the piston with more mass creating more recoil I agree with if the piston is moving fast, but what if the piston with more mass is moving slower?

Thanks for the question. You are correct, some energy is lost. The following is where most of the losses come from:

1. When initially compressing the spring, we lose some energy into heat and vibration of the spring.
2. When decompressing the spring some of the energy is lost again via the spring vibrating and twisting.
3. The piston seal creates friction, which turns some of the forward motion into heat.
4. Next the compressed air heats up and gives up some of its energy into the metal of the gun before it is transferred to the back of the pellet via pressure and air velocity.
5. When the air hits the pellet skirt it deforms under pressure, which absorbs some energy in the form of heat again.
6. Next the pellet acting against the lands of the barrel causes friction and is absorbed into the metal as heat.
7. Lastly, some of the pellet's energy is imparted into the column of air in the barrel, which is expelled when the pellet leaves the barrel.

So even though the pellet provides an equal and opposite reaction to the airgun, some of this energy never makes it to the pellet because it is dissipated into those energy sinks. That said, I think it's worth clearing up a few things about recoil. Felt recoil has a lot to do with mass and acceleration. Two guns of the same power might be giving off the same energy via recoil, but we will typically experience worse recoil from the lighter gun. This is due to the fact that the heavier gun is accelerated into the shoulder at a slower velocity, which means that the energy transfer is taking place over a longer period of time and feels less jarring.

Lastly, slowing down the piston (not too much) will result in less felt recoil because of the time component mentioned above. But when we do that, it impacts accuracy because the shot cycle gets longer. We also allow heat more time to dissipate into the metal (rather than keeping pressure high), which wastes some energy again.

-Marty

[WhatUPSbox?]

Recoil is interesting.
If using the artillery hold approach, then the range of unconstrained motion may be relevant (for example I got ~8mm in the past). This should be influenced by the mass ratio and piston stroke length.
If using a tight grip, then the force-time curve is of interest. In this case differences between spring and gas ram should show up.

Great work.

[MartyMcFly]

Recoil is interesting.
If using the artillery hold approach, then the range of unconstrained motion may be relevant (for example I got ~8mm in the past). This should be influenced by the mass ratio and piston stroke length.
If using a tight grip, then the force-time curve is of interest. In this case differences between spring and gas ram should show up.

Great work.

I haven't thought that far ahead yet. The approach right now is that the rifle is suspended in air and there are no other forces acting against it.

-Marty

[MartyMcFly]

Failure is not an option! Perhaps we can make it a feature?

This weekend has not been productive. As noted previously, the simulation treats the compression chamber, the TP, and the area behind the pellet as one volume. So today, I tried to tackle the issue of splitting up the compression volume into three parts, with mass flow occurring from one volume to the next.

The results were not pretty. First, the code wouldn't stop running, it went into an infinite loop. I know that the problems are coming from the mass flow calculations. As the volumes are changing throughout the stroke the sim is running into cases where the derivatives are either exploding in size or we're getting divide by zero problems.

I've set some floors on how small the compression chamber can get to avoid those divide by zero errors, this allowed the code to run and terminate. But now the simulation terminates too early and the output is garbage. I think that the second problem is that I'm not handling the reversal in upstream pressure (when piston bounces) correctly. Unfortunately, figuring out what is going wrong is starting to look daunting as the amount of new code added is significant and it is all linked together.

I was getting a bit too excited about moving to the next step, only to be reminded how feeble my efforts have been. I will continue to work on this problem, but it might take a few steps back before I can make the next step forward.

PS. I found that GTAs forum software censors the word s.u.c.k. It's kind of funny, because with the reverse pressure its appropriate for this conversation. 

-Marty
 

[WhatUPSbox?]

PS. I found that GTAs forum software censors the word s.u.c.k. It's kind of funny, because with the reverse pressure its appropriate for this conversation. 

-Marty

Funny, my wife is on a couple of Facebook gardening groups. She tells me that if they use the word Hoe, it gets banned. 

[subscriber]

I wonder about the idea that an action has an opposite and equal reaction. Because action requires energy so some energy must be lost in the action, leaving less energy for the reaction? Also the piston with more mass creating more recoil I agree with if the piston is moving fast, but what if the piston with more mass is moving slower?

The spring and piston have the same momentum as the rifle.   The rifle is heavy and moves much slower than the lighter piston.  But kinetic energy = 1/2 x mass x V x V    .   So the piston has much more energy than the rifle because it has proportionately more velocity.  If this principle did not exist, firing a centerfire rifle or shotgun would impart so much energy to you that you would not be keen to repeat the action.

The pellet has so little momentum by comparison you can almost ignore it.  In any event, the pellet gets some of the energy from the spring.  Usually closer to a 1/4 than 1/2 of the input energy due to inefficiency in converting spring energy to air pressure to pellet velocity.