20 replies to this topic

[Doom]

Here's the tl;dr version:

• Front-weight darts to make them stable.
• Make darts consistently. Keep the shape consistent. Keep the mass consistent (and measure it to check).
• Make darts well.
• Use systems that remove human variation from the gun, like piloted valve systems (i.e. "backpressure tanks").
• Increase the dart mass.
• Increase the muzzle velocity.
• Use a stiff barrel.
• Use a sight.
• Use the barrel length that maximizes performance.

These changes are very likely to improve accuracy and precision. The remainder of this thread is details about why these changes will probably improve accuracy and precision and some requests for testing to see what works best and what doesn't work.

-----

In light of some recent discussion of rifling as a potential cure for inaccuracy and imprecision of Nerf darts, I decided that we should discuss how to actually improve accuracy and precision. It's a shame that so much effort has been focused into rifling, which probably can't work. Let's focus on what stands a good chance of working.

(I have already explained why rifling most Nerf darts probably can't have any significant good effect and how tests so far have only shown that rifling doesn't seem to improve precision or range.)

Wikipedia has a good page about the difference between accuracy and precision. Basically, accuracy is how close the darts are to their intended target and precision is the spread around where the darts hit.

What is most central to accuracy and precision is repeatability. We want the trajectories to be as consistent as possible between shots. Consequently, we'll be looking to eliminate sources of variation.

Dart stability seems to be necessary for high accuracy and high precision. Stability here refers to resistance against overturning or perhaps even fishtailing. If a dart overturns of fishtails significantly, it is subject to aerodynamic forces that are inconsistent from shot to shot, so the accuracy and precision are affected greatly. The range is generally also very significantly affected.

We know that front-weighted darts have their center of gravity far in front of the center of pressure (where the net aerodynamic forces act), and this leads to stable darts. The darts rotate about their center of gravity, so when the center of gravity is basically at the tip of the dart, the dart basically can only move the length of the dart into the air stream. This applies a force that will rotate the dart back until it is parallel to the flow. This is why front-weighted darts are stable. If they started to become unstable, they'd correct themselves.

There are many other sources of variation that are worth looking in to. We want to ensure that darts are consistent. A few things to keep consistent: dart masses, weight distributions, and drag characteristics. The weight distribution can be kept consistent by using the same materials for all darts and carefully positioning and sizing all of the dart materials. The drag characteristics basically are shape of the dart; we don't want some darts to be bent, have different tips, have hot glue sticking off one end for some darts, etc. Here's the short version: making better and more consistent darts will help accuracy and precision.

Another source of variation is muzzle velocity. If your gun shoots strong on one shot and weak on the next, then obviously where you aimed for the first shot is probably not where your dart will land for the second. Muzzle velocity can be made more consistent by identifying sources of variation in the gun-dart system. Dart mass must be kept consistent here again. If a valve is used, it's better to use a piloted valve as they open basically instantaneously and the time to open is relatively independent of how quickly the user pulls the trigger. If a spring gun is being engineered, try to design the system to make the catch release at the same rate regardless of how quickly the user pulls the trigger. I could go on; you get the idea.

The actual dart mass is also important in addition to its consistency. Heavier darts are more resistant to wind and other transient forces that will push your darts off course. This, unfortunately, conflicts with some people's safety guidelines, but perhaps those guidelines can change in the future.

The actual muzzle velocity is also important for similar reasons. Higher muzzle velocities can lead to lower transit time (the time from leaving the barrel until hitting the target) for the dart, which should reduce the total impulse applied to the dart from transient forces like wind. And as we know, lower impulses mean lower changes in momentum, which means the dart will be more on target. Of course, the muzzle velocities could increase so much that drag causes the transit time to increase. I think this is an area where we need to do some testing before figuring out what works well. Increasing muzzle velocity could potentially counteract some of the limitations in accuracy and precision seen with dart mass restrictions.

The Spud Wiki's page on spudgun accuracy notes that barrel vibrations can cause inaccuracies. I recall this being mentioned on some real gun ballistics websites too. I'm not too sure how significant this effect is for Nerf, but it might be more pronounced for more flexible barrel materials like PETG. A few tests to see if the barrel stiffness (the product of the modulus of elasticity and the second moment of area) has an effect on accuracy seem to be justified. I have a table listing the stiffnesses of some common barrel materials.

A number of other things on the gun can be changed to improve accuracy and precision. Sights are an obvious addition. Simple iron sights are plenty adequate for most Nerf guns, though, I am intrigued by the possibility of using ladder sights with angled shots, but that's for another post. The sight radius, the distance between the front and back sights, can also be increased to allow you to be more precise when aiming. Longer sight radii make aiming errors more noticeable.

These are the main things I'm considering when looking to improve accuracy and precision. Does anyone have any other ideas or thoughts?

I think we'll have to do a lot of testing to figure out some other things. I'm curious about the relationship between precision, dart mass, and muzzle velocity. I had planned to do some tests to determine this relationship last summer but I never got around to it.

Edited by Doom, 27 April 2011 - 09:09 AM.

[taerKitty]

I'm more of a try-it-and-see, so I can only offer empirical data, but I think ballistic coefficients (i.e. head shape) may be a factor. I suspect a more aerodynamic dart head will be more accurate, and certainly longer ranged.

[Doom]

I can think of one reason why more aerodynamic darts might be more accuracy or precise. More aerodynamic darts can maintain their speed better than other darts, so their transit time is reduced, which should reduce the total perturbative impulse (as previously mentioned) from wind, etc., that will push them off course.

I'd be interested in whatever tests you do. We all should do more tests and share the results.

Extra air shot out of the barrel when the dart leaves the barrel is called muzzle blast. This could have a negative effect on accuracy and precision because it's difficult to predict. Porting is one way to avoid this problem. Another way is to use barrels that are just long enough for the dart to leave without excess gas escaping. Thankfully, this coincides with the barrel length that maximizes performance (for low-speed guns, at least), so using the optimal barrel length can improve accuracy and precision.

Barrel length brings to mind another item in the myth category along with rifling for Nerf. Some people seem to think that longer barrels somehow make things more accurate or precise. This is probably true to some degree, that is, barrels that are shorter than the dart probably aren't very accurate. Some people seem to think that constraining the darts to move in a straight line helps somehow. But this is based on a poor understanding of Newtonian mechanics. When the dart leaves, it's a ballistic projectile. Where it was in the past is irrelevant; the conditions when it leaves are all that matters.

Once the barrel is long enough, then its length does not influence accuracy or precision unless increases in length change the sight radius, vibration characteristics (i.e. natural frequency) of the barrel, muzzle blast, etc. And it's obvious that longer is not always better here.

[taerKitty]

I'd be interested in whatever tests you do. We all should do more tests and share the results.

As I said, I can only provide empirical data.

Extra air shot out of the barrel when the dart leaves the barrel is called muzzle blast. This could have a negative effect on accuracy and precision because it's difficult to predict. Porting is one way to avoid this problem. Another way is to use barrels that are just long enough for the dart to leave without excess gas escaping. Thankfully, this coincides with the barrel length that maximizes performance (for low-speed guns, at least), so using the optimal barrel length can improve accuracy and precision.

I seem to recall a thread (by you, perhaps?) to determine optimal barrel lengths for airguns.

Also, what about 'telescoping barrels', such as 6" of CPVC at the breech end, hammered into 12" of PVC? I would think that extra air would be able to escape around the side of the barrel in a more controlled fashion, and evenly around the dart so it doesn't destabilize it.

Edited by taerKitty, 27 April 2011 - 12:53 PM.

[k9turrent]

Also, what about 'telescoping barrels', such as 6" of CPVC at the breech end, hammered into 12" of PVC? I would think that extra air would be able to escape around the side of the barrel in a more controlled fashion, and evenly around the dart so it doesn't destabilize it.

Well, I have two barrels that come to mind:

1. I got titan with a 10" petg barrel in a 24" pvc sheath. It feels like that the darts are firing more consistent, and I don't recall any fish tailing out of the barrel; I have seen seen it happen coming out of BBBB regularly.

2. On the same titan, I have an 3-shot inline shotgun barrel, 5" petg in 8" pvc, so the grouping appears to be tighter than with a 5" pvc sheath.

Edited by k9turrent, 27 April 2011 - 01:50 PM.

[shmmee]

When I do test the rifled barrels vs smooth bore, I think I'll throw in a ported barrel as well - though I don't know what the "optimum" barrel length is for my PAS. (Using 10" of cpvc. Dart fit is certainly snug, but not "twist fit".) If I got lucky and nailed the perfect barrel length solely on guess work, porting won't display any effect. Maybe I'll have to test two additional short barrels (5" or so) ported, and unported to see if there is any difference. If porting can be proven to make a difference in a mid power blaster like my stock PAS, it should have an even greater effect on the high end ones like snaps.

Ooh! Porting a barrel may be an easy, effective way check your barrel length! Coat the ports very lightly with dish soap or other bubble blowing solution, and fire a dart. No bubbles, or bubble solution inside the tip barrel = too long of a barrel, you're drawing a vacuum before the dart leaves the barrel. If you get bubbles, that means there is excess pressure that could of been used to accelerate the dart. If it just sputters violently, it's too short. That's purely theoretical of course. I don't know if ports on a barrel can even blow bubbles. Perhaps a similar test can be done with a weak tape (painters tape), or glitter/chalk line dust. The chalk line dust may work better with only a single small port drilled.

Edited by shmmee, 27 April 2011 - 02:06 PM.

[durka durka]

Ooh! Porting a barrel may be an easy, effective way check your barrel length! Coat the ports very lightly with dish soap or other bubble blowing solution, and fire a dart. No bubbles, or bubble solution inside the tip barrel = too long of a barrel, you're drawing a vacuum before the dart leaves the barrel. If you get bubbles, that means there is excess pressure that could of been used to accelerate the dart. If it just sputters violently, it's too short. That's purely theoretical of course. I don't know if ports on a barrel can even blow bubbles. Perhaps a similar test can be done with a weak tape (painters tape), or glitter/chalk line dust. The chalk line dust may work better with only a single small port drilled.

I like that idea a lot. In fact, it would be easy to have an entire set of ported barrels specifically for testing optimum barrel length.

[taerKitty]

I like that idea a lot. In fact, it would be easy to have an entire set of ported barrels specifically for testing optimum barrel length.

Only problem with drilling ports in a barrel is deburring the inside.

[shmmee]

Only problem with drilling ports in a barrel is deburring the inside.

I'm no gunsmith, but ports should be placed close to the tip of the barrel. I'd guess towards the last 1/2" to 1/4". Should be easy to clean out with a small knife, or hobby file. You are right though, there is definitely a right way and a wrong way to deburr drilled ports holes. My first instinct would be to grab a .5" drill bit. But if the drill bit expands the barrel ID enough for air to pass around the dart - you've defeated the purpose of the ports.

[atomatron]

For consideration as you didn't really touch on this:



The barrel wobbles in the coupler, this probably isn't a problem for most but it does happen.

[Daniel Beaver]

For consideration as you didn't really touch on this:

http://youtu.be/dITQvgxZS7k

The barrel wobbles in the coupler, this probably isn't a problem for most but it does happen.

I was just going to mention this. I looked at a frame-by-frame of that video awhile back, and it doesn't seem that the barrel bends very much; the wobbles is in the joint between the barrel and the coupler. And that's PETG, so I would bet that PVC sheathed barrels are fairly stiff.

Edited by Daniel Beaver, 29 April 2011 - 03:42 PM.

[Boot]

After reading this thread I realized that in all my builds I've focussed almost entirely on ROF, Range, and minimizing blaster size. I figured it would be nice to try to build the most accurate blaster I could. The blaster itself is still being worked on (and hopefully will be done soon), but I noticed a few interesting things along the way.

I actually had the good fortune to be allowed to mess with my schools high speed camera, and so of course proceeded to do some nerf tests . Although rifling probably will have some effect on a minor level, after watching a dart being fired in slow motion it becomes apparent that the effect wouldn't be great. A whole range of other issues have greater effects than rifling. When you have an overly short barrel (even by a matter of a few cm) as the dart exits the barrel excess gasses will continue to accelerate the end of the dart resulting in fishtailing or tumbling. To see this in real time it is easier to use a longer dart. This is because fishtailing is easier to identify. This effect is generally not as profound with short stephans (because there is little distance for the back of the dart to be accelerated. A solution to this would be over estimating barrel length, or potentially drilling holes at the the front of the barrel to make a short 'faux barrel' that will keep the whole dart aligned when exiting the barrel.

When the barrel is too long the only thing I could notice conclusively was a (generally consistent) drop in range that did not lower consistency. Of course the solution to both these problems would be to find the optimal barrel length. For my blaster that was done by wasting about 1m worth of barrel material... (if it works, the porting method would be an incredibly useful method, since a ported barrel could be used to find optimal length on a number of blasters).

A VERY general rule I came up with that seemed to work was that the optimal barrel length will (depending on the barrel material and tank) be about 3/5 of the volume of air at standard pressure pumped into the tank. That is to say, if your pump and barrel are equal in diameter, and to fill your tank you pump 5 5" strokes your barrel should be 15" long. Of course this is completely dependent on the tank and barrel material, and will change drastically between different setups, however it offers a reasonable estimate to start from when looking for optimal barrel length.

Doom summarizes the situation well. A stiff barrel (like a CPVC Breech) of perfect length, powered by a quick release QEV (like a BS tank) will result in one of the most accurate blasters out there. Another thing to consider is that springers do remove human variation from the firing system, however they have a tendency to 'jiggle' when fired. If improving accuracy on a springer a stock would probably be necessary.

EDIT: Oops...I just realized I basically repeated Doom's second post, except he did it much more succinctly and effectively... Well, at least the whole thing about the high speed camera supports Doom's knowledge...

Edited by Boot, 30 April 2011 - 08:14 PM.

[shmmee]

Firstly, Boot - you rock! Is there any chance of getting some of that footage posted, or possibly even re-shooting and comparing non ported and ported barrels? That kind of testing can really only be done effectively with a high speed camera.

Secondly, the initial (and objective) testing I've done with rifling (comparing a gentle 18:1 twist, a tight 9:1 twist and a smooth bore have been so obviously against rifling that I've had a tough time forcing my self to set up the "precision shooting through paper test" I promised everyone and gather hard data. The darts weighted with a #6 washer swam like a fishy when shot out of the rifled barrel. I shot a heavier silicone head dart, and that looked like it flew perfectly straight, but it almost torqued the head off after the second shot. It's possible I'm not using a heavy enough dart. I think I'm going to have to wait till my 3 rd generation silicone head darts are done before doing more testing. They'll have more surface area to bond with an may be able to handle the stresses better. They'll also be heavier than the #6 washer. I also added a ls spring to my pas, which allows me to add another 2 inches to the barrel (confirmed by comparative testing 10" to 12") So the fishtailing with the 10" rifled barrels may be caused by the increase in power and not the rifling. I could add ports to the tip, but I only want to change one variable at a time while testing.

[Doom]

I've been busy, so there's a lot to reply to.

1) I seem to recall a thread (by you, perhaps?) to determine optimal barrel lengths for airguns.

2) Also, what about 'telescoping barrels', such as 6" of CPVC at the breech end, hammered into 12" of PVC?

1) Yes, I posted a thread about mathematically determining optimal barrel lengths. I've been tinkering with different analytical models in my (nearly nonexistent) free time, so I'll probably post an update about all this eventually.

2) This sounds like a good idea to test. It'd hard to predict how well it might work.

Ooh! Porting a barrel may be an easy, effective way check your barrel length!

This is a pretty creative idea. Good thinking. This sounds like it should work, but I personally would find it to be a hassle. If you want something cheaper than a chronograph to do measure optimal barrel length, I would suggest a ballistic pendulum.

For consideration as you didn't really touch on this:

http://youtu.be/dITQvgxZS7k

The barrel wobbles in the coupler, this probably isn't a problem for most but it does happen.

That's a very interesting video. Thanks for posting it. I see more than wobble in the coupler too. We should test springer vs. air gun accuracy and precision (for about the same muzzle velocity, dart mass, etc.) to see if these effects are significant. We should also test couplered barrels vs. firmly fixed barrels.

It's also worth noting that it seems that only the first vibration mode is excited. This simplifies modeling of the barrel's vibrations, if we want to do that to get a rough idea of what we have to gain by using different barrel materials or configurations.

1) When you have an overly short barrel (even by a matter of a few cm) as the dart exits the barrel excess gasses will continue to accelerate the end of the dart resulting in fishtailing or tumbling. To see this in real time it is easier to use a longer dart. This is because fishtailing is easier to identify. This effect is generally not as profound with short stephans (because there is little distance for the back of the dart to be accelerated.

2) Oops...I just realized I basically repeated Doom's second post, except he did it much more succinctly and effectively... Well, at least the whole thing about the high speed camera supports Doom's knowledge...

1) It's nice to have experimental verification of this. The bit about short darts is interesting. I'd be interested in seeing the footage like shmmee, if that is possible.

2) What you've posted is definitely appreciated. Explaining what I said in a different way is good. Also, I don't know very much. I just think about what little I do know. A little knowledge goes a long way.

Edited by Doom, 05 May 2011 - 10:40 PM.

[Boot]

Is there any chance of getting some of that footage posted, or possibly even re-shooting and comparing non ported and ported barrels? That kind of testing can really only be done effectively with a high speed camera.

1) It's nice to have experimental verification of this. The bit about short darts is interesting. I'd be interested in seeing the footage like shmmee, if that is possible.

2) What you've posted is definitely appreciated. Explaining what I said in a different way is good. Also, I don't know very much. I just think about what little I do know. A little knowledge goes a long way.

Sorry for how late this reply is. If I ever get another chance I will post videos up. This time however I had to delete the files right after reviewing them because the raw files are gigantic since our cameras also shoot in HD (I believe in 720p). Next time I will see how much I can compress them to web worthy sizes. Before then however I found a few similar videos on Youtube (with the help of a proxy ).

The first is THIS, an awesome 1000fps video shot by Automatron.

In the first shot we kind of see the wobble effect Doom described, where the dart exits the barrel and the tail is further accelerated, causing wobbling and fishtailing. This would point to too an overly long barrel.

In the second shot the dart exits smoothly as would be ideal (This difference is probably due to discrepancy in the two splitfire tanks)

This fishtailing can also be seen in THIS video, posted by 0engineer (who I don't believe is a member on this site).

Edited by Boot, 06 May 2011 - 07:40 AM.

[Irish8]

An update on barrel ports. My Quadshot only has 8" barrels but a perfect plunger and turret seal as well as it's original and Longshot springs. As you can imagine the barrels are about 2" too short and my stephans were fishtailing like hell for about the first 10'. I drilled two holes (one on each side of the barrel) just smaller than 1/8" in diameter, a centimeter back from the ends of the barrels. The darts now exit the barrels straight and only fishtail when I get a goofy dart with a looser fit (not many do that, most of my darts are "virgin" because I just made a fresh batch). Also the accurracy at 30' shrunk from about 1' in diameter to roughly 4" a 33.3% 66.6% accuracy increase. I haven't been able to do a range test yet but last time I checked (about a year ago and before I had 100% seals on the plunger and turret) it was tapping the low 90's. Either way the accurracy was been drastically improved on my favorite primary, I also did this to my PAS and pipe-ring SNAP. On my end of things having barrel ports has been proven to greatly decrease dart tumble and fishtailing which theoretically SHOULD improve the range as well.

Edited by Irish8, 16 May 2011 - 01:59 AM.

[Doom]

Thanks for doing some tests Irish8.

Not to be a buzzkill, but how many trials did you have?

To be clear (as I was in previous rifling threads) just doing a few shots and seeing a difference in the average is not enough to show there is a difference. You need to use statistics to show the difference is statistically significant, that is, more than what you might expect from the variation seen normally.

Edited by Doom, 15 May 2011 - 08:11 AM.

[Eik]

Has anyone produced a rifled barrel?
Spudfiles has a how-to rifle pvc that I found interesting.

[Irish8]

Thanks for doing some tests Irish8.

Not to be a buzzkill, but how many trials did you have?

To be clear (as I was in previous rifling threads) just doing a few shots and seeing a difference in the average is not enough to show there is a difference. You need to use statistics to show the difference is statistically significant, that is, more than what you might expect from the variation seen normally.

I did substantial trials with the Quadshot because as I said before, it is my favorite blaster, and I was really afraid that if the ports didn't work I would have an even bigger sproblem on my hands. I fed 60 stefans through it before I drilled the ports to see where my accurracy diameter was. After drilling and deburring I have now fed over 100 stefans through it and my accurracy diameter shrunk (as stated in my previous post). Also if you were wondering I made sure to stand in the exact same place in each trial and I propped my Quadshot on my kitchen counter to assure consistancy. The only thing that I have not been able to verify yet is range but I'm quite sure that it will be getting more because the target I shoot at is a cardboard project board draped with a beach towel (to stop darts from getting stuck in it and to hide all the pockmarks) and before testing the ports (I had already made the seal improvements) and the stefans were bouncing back an average of 6'-7' and now that the barrels are ported they bounce back down the hall anywhere between 10'-12' which suggests an increase in power delivered to the dart since it is not being "blown" off target.

I repeated the same types of trials in my other 2 blasters and recieved comprable results out of each. I don't know what your standard is for how many trials should be run but doing this was enough to convince me. Next week once finals are over my friend and I are going to do range tests in his cul de sac. I'll post ranges once this has been finished.

I know that I don't have many posts here on Haven but that doesn't mean that I'm a noob. I'm not trying to pick a fight or anything Doom, I just wanted to clear that up since I haven't been very active on the boards and I know that it could be a concern.

[Doom]

Has anyone produced a rifled barrel?
Spudfiles has a how-to rifle pvc that I found interesting.

I linked to a thread here where someone rifled a barrel. Look at the first sentence past the divider of the first post.

The other thread I linked to has some tests too. The link "Rifling debunked" in my signature also has some tests.

I did substantial trials with the Quadshot because as I said before

It doesn't seem that you did any statistical analysis, so just testing alone is not enough.

If you recorded the data, I would suggest posting it so that we all can analyze it. As I've said, just observing a difference in the average (for example) is not enough to say there is a difference. You have to use statistical tools to analyze the data to see what's going on.

I know that I don't have many posts here on Haven but that doesn't mean that I'm a noob. I'm not trying to pick a fight or anything Doom, I just wanted to clear that up since I haven't been very active on the boards and I know that it could be a concern.

I didn't call you a noob, so please don't imply that I had or that I might think you are one. My posts might come across as combative, but I assure you that this is not intentional. The point is to have good data supporting a hypothesis.

Your data probably is good as the idea behind porting is based on sound science, however, let's be sure.

Edited by Doom, 15 May 2011 - 11:32 PM.

[Irish8]

It doesn't seem that you did any statistical analysis, so just testing alone is not enough.

If you recorded the data, I would suggest posting it so that we all can analyze it. As I've said, just observing a difference in the average (for example) is not enough to say there is a difference. You have to use statistical tools to analyze the data to see what's going on.

I did record my data. I just didn't want to put it up and possibly clutter the thread with extra stuff that some people might not want to look at, I was trying to paraphrase to be helpful. I just wrote down what the results of firing were before and after the ports to observe the difference and if you want to see them I would happily put the data up.

I didn't call you a noob, so please don't imply that I had or that I might think you are one. My posts might come across as combative, but I assure you that this is not intentional. The point is to have good data supporting a hypothesis.

Your data probably is good as the idea behind porting is based on sound science, however, let's be sure.

I'm sorry if it sounded like I was getting defensive Doom . I wasn't trying to imply that you were calling me a noob because I knew you weren't, but some people might think "What's this guy who's only posted 24 times doing?" I was just trying to clear things up, and I'm sure you were just trying to be positive that the tests were done thoroughly.

And I know what you mean by "let's be sure" I was nervous as hell when drilling the ports to the point where I had to take a 20 min break before trying again. The one thought going through my mind was "Please dear GOD don't let me mess up my blaster!!" That would probably be anyone's fear of trying something new like this out even if there was precedent.

Edited by Irish8, 16 May 2011 - 01:55 AM.