[Egomaniacal]

Molybdenum, on 29 June 2011 - 08:52 PM, said:

It's far too consistent to be shooter action.

I don't know how you can conclusively say that it's not shooter action, golf swings are pretty consistent too.

cockerpunk, on 30 June 2011 - 08:43 AM, said:

our tank had about 3700 psi.

Really? That's the only thing you got from that entire post...

Egomaniacal, on 30 June 2011 - 01:08 AM, said:

Why no vertical measurement? I thought part of the idea of holding it was so you could get a complete measurement without restricting yourself to one dof... but you've limited it anyway by only measuring one direction.

This post has been edited by Egomaniacal: 30 June 2011 - 09:17 AM

[cockerpunk]

Egomaniacal, on 30 June 2011 - 09:03 AM, said:

Molybdenum, on 29 June 2011 - 08:52 PM, said:

It's far too consistent to be shooter action.

I don't know how you can conclusively say that it's not shooter action, golf swings are pretty consistent too.

Egomaniacal, on 30 June 2011 - 01:08 AM, said:

Why no vertical measurement? I thought part of the idea of holding it was so you could get a complete measurement without restricting yourself to one dof... but you've limited it anyway by only measuring one direction.

we only have the equipment for one DOF, sorry, we don't get paid or supported, we have what we have.

im just tryign to think of how the forward kick can be part of the human response, but i think the "lightly held" data we also took should provide the answer your looking for. i bascially just held the gun in one hand (loosly), and shot it. and human response should be pretty minimal.

Egomaniacal, on 30 June 2011 - 09:03 AM, said:

Molybdenum, on 29 June 2011 - 08:52 PM, said:

It's far too consistent to be shooter action.

I don't know how you can conclusively say that it's not shooter action, golf swings are pretty consistent too.

cockerpunk, on 30 June 2011 - 08:43 AM, said:

our tank had about 3700 psi.

Really? That's the only thing you got from that entire post...

Egomaniacal, on 30 June 2011 - 01:08 AM, said:

Why no vertical measurement? I thought part of the idea of holding it was so you could get a complete measurement without restricting yourself to one dof... but you've limited it anyway by only measuring one direction.

i told you how much air was in the tank, and we wieghed the setups ... what more do you want?

[brycelarson]

Egomaniacal, on 30 June 2011 - 09:03 AM, said:

I don't know how you can conclusively say that it's not shooter action, golf swings are pretty consistent too.

these peaks are within like 5 msec of each other. There's no way that a human flinch presing the gun forward would be that consistent.

I know we would be talking about some sort of trained, autonomic response - but human reaction time is considered good / normal in the 200 msec range. That's longer than this entire sample. Human muscles can contract at a maximum rate of about 55 msec.

So, yes, I feel confident that graph isn't showing Gordo's shoulder pusing forward.

[cockerpunk]

brycelarson, on 30 June 2011 - 09:32 AM, said:

Egomaniacal, on 30 June 2011 - 09:03 AM, said:

I don't know how you can conclusively say that it's not shooter action, golf swings are pretty consistent too.

these peaks are within like 5 msec of each other. There's no way that a human flinch presing the gun forward would be that consistent.

I know we would be talking about some sort of trained, autonomic response - but human reaction time is considered good / normal in the 200 msec range. That's longer than this entire sample. Human muscles can contract at a maximum rate of about 55 msec.

So, yes, I feel confident that graph isn't showing Gordo's shoulder pusing forward.

you are failing to discount that i am indeed, the batman. just sayin'

[Egomaniacal]

cockerpunk, on 30 June 2011 - 09:19 AM, said:

we only have the equipment for one DOF, sorry, we don't get paid or supported, we have what we have.

im just tryign to think of how the forward kick can be part of the human response, but i think the "lightly held" data we also took should provide the answer your looking for. i bascially just held the gun in one hand (loosly), and shot it. and human response should be pretty minimal.

I'll be interested in seeing that. I realize that human reaction is ~200ms, but if a person applies a simple forward push that could easily take place over a few tens of ms during the firing cycle to compensate that could explain the net forward kick of the Ion. The human brain is aware of the decision to make an upcoming action long before that action is made. Think about it like adding a flat/smooth close to linear curve to your data. Nothing I know about Physics can explain why you guys are measuring a net forward displacement if you were measuring the marker alone. I suppose the only explanation I can come up with is a restoring force of the marker against cockerpunk's chest.

And christ, I feel like I'm pulling teeth here. I'm not asking you to buy more gear. Would it be so difficult to take the sensor you have and rotate it 90* for a few shots? I mean I know it won't be measuring the same shot simultaneously with the horizontal measurement, but as you've pointed out numerous times the action of the marker is pretty consistent.


I appreciate what you guys do and know you're doing it for free (to us), I just figured you'd want suggestions to make the most out of the equipment you've got. I thought that last one was pretty obvious, sorry I wasn't more clear about it.

cockerpunk, on 30 June 2011 - 09:19 AM, said:

i told you how much air was in the tank, and we wieghed the setups ... what more do you want?

I made two suggestions for dealing with noise, asked (along with Jack) how consistently you can align the sensor - to which your response was "it's good" , and I was curious if you could weigh an empty vs full tank for posterity's sake. Mass and its distribution are kind of important when considering forces and acceleration.

This post has been edited by Egomaniacal: 30 June 2011 - 10:34 AM

[brycelarson]

Egomaniacal, on 30 June 2011 - 09:50 AM, said:

I'll be interested in seeing that. I realize that human reaction is ~200ms, but if a person applies a simple forward push that could easily take place over several hundred ms during the firing cycle to compensate that could explain the net forward kick of the Ion. Think about it like adding a flat/smooth close to linear curve to your data. Nothing I know about Physics can explain why you guys are measuring a net forward displacement if you were measuring the marker alone. I suppose the only explanation I can come up with is a restoring force of the marker against cockerpunk's chest.

And christ, I feel like I'm pulling teeth here. I'm not asking you to buy more gear. Would it be so difficult to take the sensor you have and rotate it 90* for a few shots? I mean I know it won't be measuring the same shot simultaneously with the horizontal measurement, but as you've pointed out numerous times the action of the marker is pretty consistent.


I appreciate what you guys do and know you're doing it for free (to us), I just figured you'd want suggestions to make the most out of the equipment you've got. I thought that last one was pretty obvious, sorry I wasn't more clear about it.

cockerpunk, on 30 June 2011 - 09:19 AM, said:

i told you how much air was in the tank, and we wieghed the setups ... what more do you want?

I made two suggestions for dealing with noise, asked (along with Jack) how consistently you can align the sensor - to which your response was "it's good" , and I was curious if you could weigh an empty vs full tank for posterity's sake. Mass and its distribution are kind of important when considering forces and acceleration.

I don't know where the forward force sum comes from either. On an unbalanced spoolie it's possible that the forward speed of the bolt is so much faster than the retun force that there is a total movement of the gun in that direction.

as to sensor orientation - hell yeah. We plan to stick that little bastard all over the gun in all kinds of orientations. As we often say - the first test is a test of the test. These are the results we can use to tune the test, see it's strengths and weakensses and get a better understanding of what we are capable of measuring.

As to orientation - I'm not quite as confident as Gordon. I'm thinking that we can do it better. I'm not sure what sort of scale change we will see. I think that a 0 degree, 22 degree, 45 degree setup would be interesting to compare. That'll help us understand better how much orientation matching matters.

I can certainly toss a tank on the scale and see empty v full. In this case we chose to weigh the entire rig instead of worrying about individual components too much. Obviously looking at the Protege with std rig v with just a 13/3k. That test proved that we do need to be quite careful with mass. My thought on it at this point is that we have to let the gun matter - even though it's the smallest individual mass. I think for the next test I would like to set a standard mass for tank, hopper and paint. We can toss them on the scale then adjust up and down with total paint count starting at a pod.

[cockerpunk]

i think your assumption that is a net forward response is not correct. while the peak forces might be higher in the forward direction, they are very short and tall = not much area. the sustained backwards push while slower, probably has more area = net negative position. i am going to keep wrestling with the intigration in an attempt to dervive position from this data, but so far i ahve not been able to compute anything i trust.

yeah, i don't see a reason why we can't mount the sensor to pick up verticals. we were not even sure the human rig was repeatable in this test so we didn't even bother. i don't see a reason why we couldn't.

wieghing a tank is easy, we can do that.

[hill]

Any plans to do the testing in a non human rig? To see the difference, if any.


I would also be interested in seeing the test run with the marker off, to see if pulling the trigger produces any readings.

[brycelarson]

hill, on 30 June 2011 - 10:49 AM, said:

Any plans to do the testing in a non human rig? To see the difference, if any.


I would also be interested in seeing the test run with the marker off, to see if pulling the trigger produces any readings.

yeah, we'll do some sort of clamped test at some point.

And yes, we plan to do a non gased and turned off test. I figure we can use a bunch of various setups to figure out what parts of the cycle are what. for example, a sniper degassed would show us hammer and valve alone.

[hill]

brycelarson, on 30 June 2011 - 10:59 AM, said:

hill, on 30 June 2011 - 10:49 AM, said:

Any plans to do the testing in a non human rig? To see the difference, if any.


I would also be interested in seeing the test run with the marker off, to see if pulling the trigger produces any readings.

yeah, we'll do some sort of clamped test at some point.

And yes, we plan to do a non gased and turned off test. I figure we can use a bunch of various setups to figure out what parts of the cycle are what. for example, a sniper degassed would show us hammer and valve alone.

Very cool stuff. Wish I was better at the computer gadgetry.

[Egomaniacal]

brycelarson, on 30 June 2011 - 10:32 AM, said:

I don't know where the forward force sum comes from either. On an unbalanced spoolie it's possible that the forward speed of the bolt is so much faster than the retun force that there is a total movement of the gun in that direction.

as to sensor orientation - hell yeah. We plan to stick that little bastard all over the gun in all kinds of orientations. As we often say - the first test is a test of the test. These are the results we can use to tune the test, see it's strengths and weakensses and get a better understanding of what we are capable of measuring.

As to orientation - I'm not quite as confident as Gordon. I'm thinking that we can do it better. I'm not sure what sort of scale change we will see. I think that a 0 degree, 22 degree, 45 degree setup would be interesting to compare. That'll help us understand better how much orientation matching matters.

I can certainly toss a tank on the scale and see empty v full. In this case we chose to weigh the entire rig instead of worrying about individual components too much. Obviously looking at the Protege with std rig v with just a 13/3k. That test proved that we do need to be quite careful with mass. My thought on it at this point is that we have to let the gun matter - even though it's the smallest individual mass. I think for the next test I would like to set a standard mass for tank, hopper and paint. We can toss them on the scale then adjust up and down with total paint count starting at a pod.

Thanks.

cockerpunk, on 30 June 2011 - 10:35 AM, said:

i think your assumption that is a net forward response is not correct. while the peak forces might be higher in the forward direction, they are very short and tall = not much area. the sustained backwards push while slower, probably has more area = net negative position. i am going to keep wrestling with the intigration in an attempt to dervive position from this data, but so far i ahve not been able to compute anything i trust.

yeah, i don't see a reason why we can't mount the sensor to pick up verticals. we were not even sure the human rig was repeatable in this test so we didn't even bother. i don't see a reason why we couldn't.

wieghing a tank is easy, we can do that.

cockerpunk, on 29 June 2011 - 04:12 PM, said:

...

I think you'll need to average your shots together to get rid of that noise. But that still looks like it'll be positive to me.

[brycelarson]

Gordon - are you sure that up is forward?

[cockerpunk]

brycelarson, on 30 June 2011 - 11:33 AM, said:

Gordon - are you sure that up is forward?

idk, you have the sensor ... go check? maybe im wrong.

[brycelarson]

where did you order them? I'm at work but we can pro just look them up.

This post has been edited by brycelarson: 30 June 2011 - 12:15 PM

[cockerpunk]

well you have to look how we mounted them on the gun, in the corner is a little sign convention.

EDIT - this is not the exact one (ours is +/- 5g) but its the same board, note the sign convention and how we have it mounted (you can also see it in the video) - http://www.sparkfun.com/products/849

Y positive is backwards

i guess i was wrong, probably want to edit that thanks!

This post has been edited by cockerpunk: 30 June 2011 - 01:56 PM

[paintballjla]

cockerpunk, on 30 June 2011 - 08:43 AM, said:

Jack Wood, on 30 June 2011 - 04:59 AM, said:

Are you happy with the consistency of the orientation of the sensor relative to the axis of the gun/barrel? And how do you feel that the tape does the job of securing the sensor to the gun? When we looked at this a couple of years back we made a rig where the sensor bolted to a clamp, and the clamp bolted to the barrel. There was very little potential for damping to creep into the results.

yes, i am happy with the alighnment and consistency. we could next time we test do the same gun again and see if the same thing comes out to prove it.

we didn't want to add any weight, nuts, bolts, and fittings all add mass to the system. we want something as unentrusive as possible.

I would like to see the same gun tested again at a different time, because my fear with using hands it that a persons grip my change from day to day and begin of testing to end, as there hand gets tired ect. I think that i would be cool to make some thing that secures the sensor the same way every time. I think that the gain in weight can be minimized depending on design but the gain in repeatability will out weigh the small additional weight. Besides the weight is not going to be the same for each shot anyways unless you are counting each paintball and not just having a pod of paint and "full tank" of air for each, if you are so worried about weight then do you count each paintball in the hopper and then add another each shot, and do you refill the tank up to the same pressure after each shot? The reason i am asking is because i dough you go to this extent to make sure the weight is the same, so what would a little bit more mass be. I may be wrong but first i dont think that a well made mount rig would change the data to much, This could be tested by comparing one test with tape and one with a more repeatable rig in terms of gun to gun. And besides if the additional weight added to the gun from an attachment rig would change all the data form every gun tested the same and could possible be nullified by seeing how much if any it changes the results by doing a bolt/clamp on mount compared to tape and go from there to decide what you want to do.

And i my be reading the wrong but it appears that you have take data in only the foward back ward direction would it be possible to get more data form the test. i.e. more accelerometers placed in other places to measure other directions. and on other parts of the gun to see just how much the placement of the accelerometer changes the data.

And on the note of borrowing guns i think that it will be important to know exactly what you are getting data form, because yes it is easy to tell what physical changes have been made but with so may guns you have a whole list of things that you can change in the board that could affect how much the guns kick like dwell ect. I sujest that you request that the guns broad be reset to factory setting for a data set so a fair comparison can be made between guns.

And video high speed or not or possibly audio would be cool. Because the data is awesome but it would be even cooler to have something to compare it to like. I know that it would be difficult to see like at what point in the data acquisition that the ball left the barrel etc.

This post has been edited by paintballjla: 30 June 2011 - 02:14 PM

[cockerpunk]

paintballjla, on 30 June 2011 - 02:05 PM, said:

cockerpunk, on 30 June 2011 - 08:43 AM, said:

Jack Wood, on 30 June 2011 - 04:59 AM, said:

Are you happy with the consistency of the orientation of the sensor relative to the axis of the gun/barrel? And how do you feel that the tape does the job of securing the sensor to the gun? When we looked at this a couple of years back we made a rig where the sensor bolted to a clamp, and the clamp bolted to the barrel. There was very little potential for damping to creep into the results.

yes, i am happy with the alighnment and consistency. we could next time we test do the same gun again and see if the same thing comes out to prove it.

we didn't want to add any weight, nuts, bolts, and fittings all add mass to the system. we want something as unentrusive as possible.

I would like to see the same gun tested again at a different time, because my fear with using hands it that a persons grip my change from day to day and begin of testing to end, as there hand gets tired ect. I think that i would be cool to make some thing that secures the sensor the same way every time. I think that the gain in weight can be minimized depending on design but the gain in repeatability will out weigh the small additional weight. Besides the weight is not going to be the same for each shot anyways unless you are counting each paintball and not just having a pod of paint and "full tank" of air for each, if you are so worried about weight then do you count each paintball in the hopper and then add another each shot, and do you refill the tank up to the same pressure after each shot? The reason i am asking is because i dough you go to this extent to make sure the weight is the same, so what would a little bit more mass be. I may be wrong but first i dont think that a well made mount rig would change the data to much, This could be tested by comparing one test with tape and one with a more repeatable rig in terms of gun to gun. And besides if the additional weight added to the gun from an attachment rig would change all the data form every gun tested the same and could possible be nullified by seeing how much if any it changes the results by doing a bolt/clamp on mount compared to tape and go from there to decide what you want to do.

And i my be reading the wrong but it appears that you have take data in only the foward back ward direction would it be possible to get more data form the test. i.e. more accelerometers placed in other places to measure other directions. and on other parts of the gun to see just how much the placement of the accelerometer changes the data.

And on the note of borrowing guns i think that it will be important to know exactly what you are getting data form, because yes it is easy to tell what physical changes have been made but with so may guns you have a whole list of things that you can change in the board that could affect how much the guns kick like dwell ect. I sujest that you request that the guns broad be reset to factory setting for a data set so a fair comparison can be made between guns.

take these one at a time -

sure, we can easily repeat these tests for hands consistency and for sensor mounting, but like the debate about hands consistency, im just not seeing that as as large a factor as everyone else is. so it might be couple of degrees off, thats less then 5% error - i'd be more then happy with a test rig with less then 5% error. you can't feel 5% less kick, so there isn't much point in measuring it. that being said, its an easy test to do, so i dont see a reason why we couldn't.

secondly, we only have one input, so we can only simultainously measure one dimention of kick.

third, we would have to varify of course that the gun is operating properly and under normal conditions of course. if we can, we can post the settings used (if we know them). but like any test, i'm not gonna take one sample as representative of every gun like that. im far more interested in the shape and profile then i am the values anyway (others might be different).

[brycelarson]

cockerpunk, on 30 June 2011 - 01:38 PM, said:

Y positive is backwards

i guess i was wrong, probably want to edit that thanks!

guess that changes things a bit. Makes sense though - the first major spike would be the rammer. It's up - meaning it's pushing the rig back. makes sense.

paintballjla, on 30 June 2011 - 02:05 PM, said:

I would like to see the same gun tested again at a different time, because my fear with using hands it that a persons grip my change from day to day and begin of testing to end, as there hand gets tired ect.

the fact that there was incredibly minimal change shot to shot when holding the gun indicates that the hand holding system is very repeatable. As to whether there is an overall, low frequency shift in data from shooter reaction isn't clear.

paintballjla, on 30 June 2011 - 02:05 PM, said:

Besides the weight is not going to be the same for each shot anyways unless you are counting each paintball and not just having a pod of paint and "full tank" of air for each, if you are so worried about weight then do you count each paintball in the hopper and then add another each shot, and do you refill the tank up to the same pressure after each shot?

we did add paintballs to bring the count back up after chrono and after every couple shots. The rigs with tank, hopper, paint and gun all weighed upwards of 8 pounds. Each ball shot weighs something near .007 pounds. Each shot off the tank uses up about 1/1500th of the air in the tank. So 5 shots would change the rig weight by about .4%. Assuming that mass has a liner relationship to recoil (which we'll examine) that's pretty much nothing. even 50 shots in is pretty minor.

paintballjla, on 30 June 2011 - 02:05 PM, said:

And i my be reading the wrong but it appears that you have take data in only the foward back ward direction would it be possible to get more data form the test. i.e. more accelerometers placed in other places to measure other directions. and on other parts of the gun to see just how much the placement of the accelerometer changes the data.

yes, we plan a ton of tests.

As to future testing - we've got several dozen ideas. Everyone please feel free to keep suggesting new ones.

[Egomaniacal]

That makes a lot more sense. The first spike would be the hammer and paintball being pushed forward (rig pushed back), downward spike is hitting the valve (pushing forward), second upward spike is the paintball actually accelerating (pushing the rig back), the small downward motion is the bolt being accelerated back (rig pushed forward) - is it attenuated by continued gas expansion behind the paintball? Then the final upward spike is the bolt hitting the boltstop (pushing rig back).

Looks like the human element is less than I thought it'd be. Looks nice guys, I'm looking forward to more results!

[paintballjla]

cockerpunk, on 30 June 2011 - 02:14 PM, said:

secondly, we only have one input, so we can only simultainously measure one dimention of kick.

So you could take a reading from a different spot/directional configuration just on at the same time? I think that this would be cool just to see other directions, and how much it changes if mounted different places. And how hard would the problem of only one input be so solve is that one input into the digital oscilloscope or collection software?

And by no means am i saying that you test rig is not accurate i would just like to see some numbers that tell me that. I agree with you that it will probably only have marginal error i would just like to see the same gun run at a different time to ensure that you can get the same data as the first test. I think form what you have described as the testing procedures to be this would be relatively painless and it would make me a believer as probably may others that it dose what you say it dose.

I'm interested to see what the affects of may things are. To name a few changes in dwell, and then i would like to see if the barrel has any thing to do with kick, like dose bore or length change it.

This post has been edited by paintballjla: 30 June 2011 - 10:16 PM