[cockerpunk]

Lord Odin, on Jan 16 2009, 10:33 PM, said:

How many balls of each type and total were used in this test? Just curious about this.

Just so I'm not making wrong assumptions, could you please explain what the Vmean, Hmean, and Vmean/Hmean are?

I'm trying to hold off on making conclusions until I know what I'm looking at.

3 balls each with the seams horizontal and vertical. with a double check of seams at 45ish degrees.

Vmean is the mean of the vertical seam orientation, Hmean is the same thing. V/H is simply one divided by the other.

[BOBBYTUCSON]

awsome test and results. this is just another reason why techpb rules!! but now , will we have results on high , mid, low quality rps paint? id like to prove to my teamates that dxs is better than rps!! lol , i hope hope to see how evil does compared to gold

[cockerpunk]

BOBBYTUCSON, on Jan 16 2009, 10:24 PM, said:

awsome test and results. this is just another reason why techpb rules!! but now , will we have results on high , mid, low quality rps paint? id like to prove to my teamates that dxs is better than rps!! lol , i hope hope to see how evil does compared to gold

no other brands, thats not the point.

[Lord Odin]

Let me see if I'm reading this right.

It seems to me that deformation, shell strength, and shell thickness are pretty similarly related with minor exceptions. The thicker the shell, the stronger they seem to be and more deformation they can withstand before failing.

All balls seem to be stronger with the seam horizontally than vertically, which would make sense to me. The seam should be the strongest point on the ball and if forces inside the ball are pushing on the seam, it should hold out longer than if it were placed vertically. The Gold looks like it has a strength that is closer to being uniform than the others because they are much stronger at the seam and would require more force to break them. I'm assuming that's what the Vmean/Hmean is looking for: strength uniformity.

Was the distributed load test to see what would happen to the ball if pressure was added from all sides?

[brycelarson]

Lord Odin, on Jan 17 2009, 08:32 AM, said:

The seam should be the strongest point on the ball and if forces inside the ball are pushing on the seam, it should hold out longer than if it were placed vertically.

the seam is actually the weakest point on the ball.

[Lord Odin]

brycelarson, on Jan 17 2009, 08:47 AM, said:

Lord Odin, on Jan 17 2009, 08:32 AM, said:

The seam should be the strongest point on the ball and if forces inside the ball are pushing on the seam, it should hold out longer than if it were placed vertically.

the seam is actually the weakest point on the ball.

From external forces but what about internal forces?

[brycelarson]

Lord Odin, on Jan 17 2009, 08:51 AM, said:

From external forces but what about internal forces?

since the internal forces aren't what cause barrel, loading or impact breaks - I don't know that it's an issue - and I don't see what you mean exactly.

In this test the balls always broke at the seam first - the break propagated to other places on the ball - but the seam was the weak point.

[Lord Odin]

brycelarson, on Jan 17 2009, 10:58 AM, said:

Lord Odin, on Jan 17 2009, 08:51 AM, said:

From external forces but what about internal forces?

since the internal forces aren't what cause barrel, loading or impact breaks - I don't know that it's an issue - and I don't see what you mean exactly.

In this test the balls always broke at the seam first - the break propagated to other places on the ball - but the seam was the weak point.

I think I screwed up in thinking with internal forces instead of external. I'll try to explain what I mean.

If you place a ball with the seam horizontally, the pressure points are at the poles. As the force increases at the poles, the equator of the ball (the seam) is where it will naturally want to expand. According to the data, it takes a lot of pressure in this position before the ball breaks. So forces are acting on the seam from inside the ball.

Now when the ball is rotated with the seam vertically, the pressure points lie on 2 points of the seam. When its compressed, it still wants to expand at the equator but that line only intersects the seam at 2 points. So you have external and internal forces acting on the seam. The balls don't look like they hold up as well in this position and require less pressure before failure.

I agree that the seam is the weak point, now that I think about it. It just depends on where the ball is contacted and with how much force before it fails.

[cockerpunk]

does anyone have a background in marital science? if i start tossing terms out like site defect, void space and force multiplier, will people be confused?

[Poe]

cockerpunk, on Jan 17 2009, 12:22 PM, said:

does anyone have a background in marital science?
...

If only it was a science... maybe my wife wouldn't yell so much.

[cockerpunk]

Poe, on Jan 17 2009, 02:15 PM, said:

cockerpunk, on Jan 17 2009, 12:22 PM, said:

does anyone have a background in marital science?
...

If only it was a science... maybe my wife wouldn't yell so much.

oh lol, firefox didn't pick that one up!

thats certainly true too, relationships are hardly a science.

does anyone have that kind of background? im looking up Wikipedia entries and im not finding a whole heck of a lot.

[Merc4Hire]

cockerpunk, on Jan 16 2009, 07:25 PM, said:

BOBBYTUCSON, on Jan 16 2009, 10:24 PM, said:

awsome test and results. this is just another reason why techpb rules!! but now , will we have results on high , mid, low quality rps paint? id like to prove to my teamates that dxs is better than rps!! lol , i hope hope to see how evil does compared to gold

no other brands, thats not the point.

Why not make it part of the point?

not because I want to be smug about my favorite brand or whatever, but I thought the purposes of punkworks were to:

1) determine what factors work better, in a qantifiable way.

2) get a large enough sample of available factors for the info to be useful.

3) apply it by choosing to buy or make the stuff that works

4) discourage false perceptions and encourage manufacturers to make stuff that works.

{5) be secretly smug about knowing the real truth} {officially not a factor}


This is supposed to be applied science, knowing how much compression is necessary to break paint and how much is nifty, but the bottom line of all of this is people want to know what paint/ barrel combo is going to work best for them. Why shouldn't we know if the compression factors are different for other brands?

Wouldn't a more useful test be to determine which classes and brands of paint perform best? To me paint performance would be defined by 1) shoots straight consistently 2) doesn't break in my pod, hopper, breech, or barrel. 3) does break on opponents. 4) is less effected by weather, or predictably works well in known weather 5) is conistent case to case on the previous 4 points 6) cost per previous factors.

This test seems to address parts of #2 and #3 and (maybe#4 depending on what other data ou are holding), but wouldn't it be more applicable test be to fire a bunch of different brands of paint at a target that simulates a player, and record breaks on target versus breaks in gun, at specific temperatures and at a fixed range and fixed velocity mean? Then periodically repeat the test as new products become available, and to test for continued quality.

Don't get me wrong though, this is a very cool test. I just have a hard time seeing how we can use the information you gathered to make our stuff work better. How can I use this to help me make better paint purchasing decisions?

[cockerpunk]

Merc4Hire, on Jan 17 2009, 03:53 PM, said:

cockerpunk, on Jan 16 2009, 07:25 PM, said:

BOBBYTUCSON, on Jan 16 2009, 10:24 PM, said:

awsome test and results. this is just another reason why techpb rules!! but now , will we have results on high , mid, low quality rps paint? id like to prove to my teamates that dxs is better than rps!! lol , i hope hope to see how evil does compared to gold

no other brands, thats not the point.

Why not make it part of the point?

not because I want to be smug about my favorite brand or whatever, but I thought the purposes of punkworks were to:

1) determine what factors work better, in a qantifiable way.

2) get a large enough sample of available factors for the info to be useful.

3) apply it by choosing to buy or make the stuff that works

4) discourage false perceptions and encourage manufacturers to make stuff that works.

{5) be secretly smug about knowing the real truth} {officially not a factor}


This is supposed to be applied science, knowing how much compression is necessary to break paint and how much is nifty, but the bottom line of all of this is people want to know what paint/ barrel combo is going to work best for them. Why shouldn't we know if the compression factors are different for other brands?

Wouldn't a more useful test be to determine which classes and brands of paint perform best? To me paint performance would be defined by 1) shoots straight consistently 2) doesn't break in my pod, hopper, breech, or barrel. 3) does break on opponents. 4) is less effected by weather, or predictably works well in known weather 5) is conistent case to case on the previous 4 points 6) cost per previous factors.

This test seems to address parts of #2 and #3 and (maybe#4 depending on what other data ou are holding), but wouldn't it be more applicable test be to fire a bunch of different brands of paint at a target that simulates a player, and record breaks on target versus breaks in gun, at specific temperatures and at a fixed range and fixed velocity mean? Then periodically repeat the test as new products become available, and to test for continued quality.

Don't get me wrong though, this is a very cool test. I just have a hard time seeing how we can use the information you gathered to make our stuff work better. How can I use this to help me make better paint purchasing decisions?

well first and foremost, we are not product reviewers. only when a product claims to be different or advantageous in some way that we have evidence that it is not, is when we become product testers. i genuinely despise reviews, we are not the consumer reports of paintball.

but point per point -

1. we can do this without testing brand of paint, we need only look at the quality of the paint.
2. we can do this without testing brand of paint, we need only look at the quality of the paint.
3. we can do this without testing brand of paint, we need only look at the quality of the paint.
4. we can do this without testing brand of paint, we need only look at the quality of the paint.
5. that secretly is the point of all this - good call.

me and bryce have talked alot about this issue, and to maintain objectivity and analysis on the science, and not the products themselves, we wont want to do brand vs brand testing. you are welcome to do that, but we will refrain from doing that.

unless of course the brand changes an important factor, like the method by which a gun works. we can't test a spool vs a poppet with dye guns, we need to change brands.

the bottom line assumption is that in terms of paint is that we tested a wide range of paintballs, grades, qualities and types, that is enough for us to make generalizations about all paintballs, unless the producers of paintballs start marketing something new. to my knowledge no paint company markets anything other then higher quality, cheaper price, thicker fill and such.

also, to prevent any type of product testing, we tend to tell our sponsors that we will only test there goods, we will never test sponsored goods vs un-sponsored goods. so, in theory, we could do a paint test, but all the paint companies tested we would buy, or they would all donate to us.

so, there are alot of reasons why we wont be doing brand comparison tests.

EDIT what i am try to say is that brand in and of itself is not a factor. only when one brand is fundamentally different then the next is it a factor, and then its the fundamental difference that is the factor, not the label on the box. if one brand made paintballs not with encapsulation machines, or with a non-gellitine shell, or something of that nature, then we would defiantly take a look.

This post has been edited by cockerpunk: 17 January 2009 - 05:07 PM

[Merc4Hire]

thanks for the response. kind of my point though is that it is very seldom that you can do a direct quality tier comparison between brands. You can't buy brand x midrange paint, with a shell strength of aprox 77% stronger than gold and buy rand y midrange paint, with a shell strength of aprox 77% stronger than gold, and expect them to behave similarly. There are so many other factors that affect performance. Turns out brand y is sloppy and the diameters vary by a lot more than brand x....

I get why you hate reviews, but a direct comparison test is really useful. There isn't a lot of room for opinion when you say that the numbers show that running the same.685 cp barrel at 280, it took brand x's $40 a case an average of 2 cases to get a barrel break, and brand y's tended to break about every 3 hoppers worth. You don't have to say that you like procaps better than rps, you can say that with the same setup, one broke more paint, one bounced more, and one had a better standard deviation on the target. you did it with barrel bore ratio's. Or your tests might show that in price tier they really were about the same on the target even though a bunch of factors were different. To me that means a lot more than rating just one factor of the paintball, and is easier to test. Without testing every factor about the quality of the paint, you assume that shell strength is the only thing that determines quality of paint, and you know better.

Before you can say that a certain factor determines quality, such as shell strength, you have to compare several paintballs that are about the same in that factor and see what they do over the chrono and on the target. Otherwise you are just picking something easy to measure and generalizing. ie from our sample, people who are 6' tall can lift more than people who are 5'7" look at the numbers, decide for your self, 6' people are better.Conclusion: date a tall person.

I also understand your obligations to sponsors and your care to avoid letting those make skewed tests. I respect that a lot. Your barrel test is great. however the length comparisons only seem to directly apply to barrels that arrange their porting and control bore lengths similarly to CP. Things like starting the porting at the end verses the middle would seem like a ,fundamental difference'. Your tests are excellent, but they don't cover everything yet. I don't expect you to get every single little thing eliminated. It just isn't definitive yet.

I really hope I haven't offended you, I think punkworks is one of the coolest things to happen to paintball in since about 2003, and wish it had been around since the eighties.

[cockerpunk]

Merc4Hire, on Jan 17 2009, 07:52 PM, said:

thanks for the response. kind of my point though is that it is very seldom that you can do a direct quality tier comparison between brands. You can't buy brand x midrange paint, with a shell strength of aprox 77% stronger than gold and buy rand y midrange paint, with a shell strength of aprox 77% stronger than gold, and expect them to behave similarly. There are so many other factors that affect performance. Turns out brand y is sloppy and the diameters vary by a lot more than brand x....

I get why you hate reviews, but a direct comparison test is really useful. There isn't a lot of room for opinion when you say that the numbers show that running the same.685 cp barrel at 280, it took brand x's $40 a case an average of 2 cases to get a barrel break, and brand y's tended to break about every 3 hoppers worth. You don't have to say that you like procaps better than rps, you can say that with the same setup, one broke more paint, one bounced more, and one had a better standard deviation on the target. you did it with barrel bore ratio's. Or your tests might show that in price tier they really were about the same on the target even though a bunch of factors were different. To me that means a lot more than rating just one factor of the paintball, and is easier to test. Without testing every factor about the quality of the paint, you assume that shell strength is the only thing that determines quality of paint, and you know better.

Before you can say that a certain factor determines quality, such as shell strength, you have to compare several paintballs that are about the same in that factor and see what they do over the chrono and on the target. Otherwise you are just picking something easy to measure and generalizing. ie from our sample, people who are 6' tall can lift more than people who are 5'7" look at the numbers, decide for your self, 6' people are better.Conclusion: date a tall person.

I also understand your obligations to sponsors and your care to avoid letting those make skewed tests. I respect that a lot. Your barrel test is great. however the length comparisons only seem to directly apply to barrels that arrange their porting and control bore lengths similarly to CP. Things like starting the porting at the end verses the middle would seem like a ,fundamental difference'. Your tests are excellent, but they don't cover everything yet. I don't expect you to get every single little thing eliminated. It just isn't definitive yet.

I really hope I haven't offended you, I think punkworks is one of the coolest things to happen to paintball in since about 2003, and wish it had been around since the eighties.

rigtho, we do general tests. and we make general recommendations based on those test. we can learn alot from these general tests, and if we have a standard type of test and a standard of equipment used to measure that test, we can repeat certain types of tests and compare test resealts that might be separated by a large amount of time. such as the "standards" we have used for accuracy and ball brittleness (rig has yet to be reveled).

obviously finding the exact optimum would require a 6 or 8 dimensional matrix of testing, with all manner of extreme precision measuring. it requires that you spend time with your setup to get the ideal. we can't test every setup with every paint with every loader and barrel and bolt and air tank and dwell setting and ....

often times the ideal is impossible to achieve (the fundamental problem that faces engineers across the world), so using a series of general ideas and relationships made off our test will give the best indication of what to use.

we dont use hard and fast rules here for a reason, we say things like "underboreing by about .003 or .004" or "we did not see a statistical difference" and "depending on the porting pattern, 12 to 14 inches one piece underbore is probably your most efficiency barrel"

and above all else we look at the scale of the outputs we get, and we try to understand that in a real world way. often we can see a resualt, but it just isn't large enough to be useful and such.

[Lord Odin]

I wouldn't rule out product comparisons completely. They do have their place, especially in paintball testing. The testing that is being done is exactly what you guys are intending it for: to find the underlying causes and effects to explain behaviors generally. However, those results aren't always applicable. For example, lets hypothetically say that shell thickness is found to be the main contributing factor to accuracy. It explains it but people aren't able to apply it because of a lack of information. Paintball manufacturers don't publish their specs of their products and people have no way to tell which brand is better or worse. That can apply to anything from barrels, to bolts, to tanks, etc. By comparing products, people can see where they line up with each other and what their specs are.

Like you said CP, right now, you're simply after finding what the causes are and paving the way with creating testing standards. Once you know what does what, those tests can be repeated so that product comparisons can be done later on.

[cockerpunk]

anyway, back to the test at hand.

i think my leading questions would be these -

1. does seam orientation matter? what mode of failure are we looking at, and how does the seam effect that failure?

2. on page two, how does the method of loading change the load capability?

3. general thoughs in this one - note how much deflection a paintball can take under normal (ie non-impact) loading. how does this related to bore size choices?

after we explore those different questions and any others that arise in that discussion, well post some more data.

This post has been edited by cockerpunk: 17 January 2009 - 11:16 PM

[Troy]

Since I'm one of the first to respond to CP, I'll grab some of the low hanging fruit... the orientation of the load on a paintball is very critical, if any of the paintballs are loaded vertically, they are FAR more likely to fail, then if they are loaded horizontally. With deflection numbers of at least .09 its no small wonder why we didn't see any statistical difference with a .003 underbore, which represents only about 3% of the amount it can be deflected until it brakes.

Troy's notes:
1) This test is very relevant to bolt strikes and underboring, but unfortunately not on how paintballs perform in the stack. Because a paintball is loaded in the stack with three points of pressure it's still unclear how or if the seam effects paintballs in the stack. I think we have reason to believe that this is a very critical point where many failures occur, so it would be fairly important to test it. If I were you, I would use a clear cylinder the width of a typical feedneck and press a paintball between two ball bearings with different seam orientations in 3D.
2) There certainly weren't many repetitions in your test, I would have liked to see a lot more.
3) I would be interested how much resilience is added to a paintball if it is in a bore. I'd like to see a paintball being pressed in between the two bolts in the middle of a freak bore, and see how much more pressure needs to be applied to break it.

[Merc4Hire]

Ok troy that was pretty astute, and I like all 3 of your points. I stand corrected on the practical application

[Leftystrikesback]

cockerpunk, on Jan 17 2009, 09:22 AM, said:

does anyone have a background in material science? if i start tossing terms out like site defect, void space and force multiplier, will people be confused?

I'm focusing on the subject for my masters. I'd certainly like to hear your thoughts, though I'm not sure how you're going to toss out those terms since the gelatin shell of a paintball is not crystaline.

Do you have any photos of the cross sections of the paintballs so that we can look at what the seam looks like? From paintballs that I've cut in half, the seam is actually thicker than the rest of the ball. This is interesting, because you would think that would mean that it is the strongest part of the ball, which Bryce mentioned is contradictory to this test, since the seams were the main sites of failure.

But the reason the seam is there in the first place is because it is where the two halves of a paintball are 'stuck' together while they are warm and soft. So it makes sense that the seam is thicker, and also makes sense that it would fail there, since there could be imperfect bonding between both halves, leading to a weak spot along the seam.

Also, do you think we should be assuming the paint inside pressurizes turning the ball into a pressure vessel? I have no idea if that is a valid assumption but thinking of things this way seems like it explains the vertical seam failure.

Hoop stress would be acting to force the two halves of the paintball apart when the seam is in the Vertical position. In the horizontal seam situation, hoop stress goes along the seam (in the strong direction) and radial stress would be acting on opening the seam. And if the squished paintball can be approximated as a cylinder pressure vessel (and a cow as a sphere ) then hoop stress is much greater than radial stress, and we get our result that the paint ball fails sooner when the seam is positioned vertically.