280 fps everytime, and ideal accuracy

**athomas** · 10-03-2002, 10:04 PM

How about a barrel that auto adjusts to the size of the ball in the breach. The barrel can still have rifling to help reduce wobble in the event that there is odd shaped balls. And then apply the acceleration algorithm.

Now, once its calibrated, you could adjust the final velocity by inputting a preset number.

I wonder how fast the processor would have to calculate the values. Would it actually be of any benefit over good quality control in preset regulated valves that are used today? I think this thread will make a good topic of conversation.

**Redkey** · 10-04-2002, 11:53 AM

I've thought about this too...

I figured the gun would be set up to adjust the velocity *after* every shot. The microchip would make adjustments to the solenoid timing so that the following shot would have a slight correction applied to it. The larger the variation from the goal the stronger the correction.

I have a cocker setup to use a ram to do the firing... I just need to build or buy a SSR to protect the digital IO lines of my DAQ card. I'm cheating on the solenoids though... I'll be using a bank of 24 volt SMC VQ1101Ns plumbed with 1/4 line. Too big to comfortably fit on a gun... but easy to work with.

Another way to do it would be to have some way to electronically control the firing air pressure. Not sure how this would be accomplished though.

I suppose you could also use some kind of iris to control the size of the air flow paths between the valve and the bolt.

**sniper1rfa** · 10-04-2002, 01:33 PM

the BEST way would to have a pressure transducer hooked up to your board, and calculate the dwell based on the pressure in the chamber. unfortunately, you could only do somewhat inaccurate math with the current chips.
you could probably do it to within the fps though.

**athomas** · 10-07-2002, 08:36 AM

The mag could be made to work like this because the velocity of the ball is directly related to the pressure in the front chamber. Now all we need is an electronically controlled pressure regulator instead of the spring regulator that we use now. I don't know if there would be any noticable benefits, because the mag regulator is darned good the way it is. The real benefit would be that you would only need to calibrate over the chrono for a certain barrel and paint and then it would be all pressure dependent after that. You could increase the psi by a 1 or 2 and gently increase your fps instead of guessing with an allen wrench. The setting would be the same every time because of the electronic pressure regulation. Consistency is the name of the game for any gun.

**314159** · 10-07-2002, 05:26 PM

athomas, the setup i think of would not work for the mag. because the setup that i have in mind would calebrate itself for the current shot. not off of the previous shot.

the multiple velocity sensors down the barrel would mesure the deviation from the acceleration path of a shot going at 280 fps and adjust the dwell time to compensate for it.

**athomas** · 10-07-2002, 07:03 PM

I wasn't talking about calibration off the previous shot. My first method indicated the same method as yours with some added variables.

My second though using just the mag, electronically adjusts the pressure instead of having a mechanical pressure regulator. The velocity of any gun is based on pressure within the dump chamber. If every other variable is fairly constant and the pressure is regulated exactly the same every time, then the veleocity will be the same every time. The biggest shot to shot inconsistency in todays paintball guns is in the pressure regulation, but that is something we can change. I was just throwing out the idea of a more precise electronically controlled pressure reguation for the front valve chamber. If this is at least consistent, then we have more control over our shots.

Another inconsistence is the variances between cheap paintball. We don't have any control over that one, even if we measure the acceleration profile of the prevous shot and adjust for the next one. The next ball might be totally different so the previous information would be useless anyway.

Just some added thoughts and ideas to look at.

**AGD** · 10-07-2002, 11:51 PM

Nice idea and theoretically possible. The problem is that a lot of the variation in FPS is due to the paintballs inconsistencies. When we shoot nylon balls here in the shop the FPS variation is quite good and when you go back to regular paint you can see the difference.

With a compensation system that corrected on the next shot you would just be chasing noise. You can't correct on the same shot because the air chamber pressure is too far upstream to change the ball half way down the barrel. The only possibility would be opening ports on the side of the barrel to instantaneously vent pressure. considering the ball spends 3-4 ms in the barrel you would have to be quick about it.

AGD

**athomas** · 10-08-2002, 06:43 AM

I guess I should change the "biggest shot to shot inconsistency in todays paintball guns is in the pressure regulation" to read "the biggest controllable inconsistency in todays paintball guns is in the pressure regulation". At least by controlling the regulation, we remove another variable.

To control velocity by opening vents in the barrel would work if they could react fast enough, but the inefficiencies would be very large. You would always have to have a little more air than is needed for every shot just in case it is needed. That excess air would have to be vented every time. Any air not used to push the ball is wasted air and contributes to the inefficiencies of the gun.

**athomas** · 10-08-2002, 07:23 AM

As indicated previously by 314159, I think the adjustment of the dwell time in a different type of gun setup would work. That way, real time compensation for paintball inconsistencies would be in effect. The problem would be for the equipment/computer to know the shot profile of the barrel being used. I suspect a number of shots would have to be taken and the computer would have to calculate an average shot profile. Maybe an ongoing average could be used such as the profile for the last 100 shots. That way, any changing conditions such as moisture or temperature would gradually be taken into consideration. Any single inconsistency due to a bad paintball would not have a large detrimental effect on the overall consistency of the gun. In fact the gun could compensate for the bad ball because it could adjust the dwell time to give an acceleration profile similar to any other ball.

**314159** · 10-08-2002, 05:44 PM

time for some numbers

i am changing 280 fps everytime to 275 fps everytime, (to get as close as comfertable to the 3 shots under 280 fps as i feel good about.)

the paintball is traveling at 3300 inches per second at 275fps(275ft/1sec x 12in/1ft)

i change my idea to use single beam break beam chronos, i will need a pair of break beam sensors per chrono (the next chrono down the barrel can use one of the beams from the previous sensor). this will get a more accurate measurement (it does not assume that the paint is .68 in as the above method does)

if these sensors are spaced every inch, the front of the ball would break the consecutive beams just slightly under 1/3300th of a second.

to get +/- .1fps accuracy, i would need to take 3,300,000 readings per second. i can get a scenix microcontroller that can do 50 million instructions per second, so that leaves me 15 instructions per timing sample, which should be enough to do a sumation routine. the current thought is to have one scenix microcontroller dedicated per pair of beams, and to have a main microcontroller controlling the logic. after a microcontroller that was dedicated to a break beam got a reading, it would stop it's sumation routine and dump the result to the parent microcontroller in a serial fashion.

currently i think that it is very posible to calebrate the shot while you are shooting it. you can calculate the acceleration of the paintball by the velocities obtained by the break beam chronograps down the barrel. you can compare the current acceleration of the paintball curently being fired, to the acceleration of a paintball being fired from that gun at 275.0 fps depending on how much the acceleration is over or under this benchmark figure, the dwell time will be adjusted acordingly. all this processing will be done before the paintball hit 275 fps, or near 275 fps this will let the paintball hit 275 fps on the dot without any venting out of the barrel. all this is to be computed by the main microncontroller.

**314159** · 10-08-2002, 05:48 PM

instead of a nice array of scenix microcontrollers, i would only need 2, when one is done monotoring it's gates, it could move up 2 gates, almost playing leap frog.

i could also have sensors every half inch. and have 2 setups as described above, each watching a gap of 1 inch, offset by .5 inch.

note: this idea is not just turning off the airflow when the paintball hits 275 fps.

this makes me wonder if the accuracy of a paintball is based on it's velocity consistcy more than it's true roundness of the shell.

**Pand0ra** · 10-10-2002, 11:54 AM

Another thing which could work is to change the barrel diameter on the fly.

The solution is the magnetostriction. If the barrel is made in the correct material, it could be possible to change his diameter by using some magnetic coils.
The only thing I wonder is if it's possible to do a variation of diameter in the range of 0,5mm. And if the barrel wouldn't be too brittle.

The measurement of the speed of the paintball is just a detail in my opinion. It would also be nice to find a way to do the measure from behind the ball. Maybe ultrasonic waves could work, or a nice optical system.

@++

**Pand0ra** · 10-10-2002, 12:02 PM

Another idea came to my mind too.

You could also use multiple releases of gaz behind the ball, over the whole length of the barrel, to correct errors in the velocity.
This is a bit like the old V3 made by the germans in 1945, but less dangerous :).

@++

**Redkey** · 10-10-2002, 05:59 PM

the problem with optical sensors...

What happens when you break a ball and get zilch for feedback data?

Would a gun that could correct it's velocity be tourny legal? What would stop someone from having their gun setup to reduce the velocity after a certain length of time or after a button was pushed?

280 fps everytime, and ideal accuracy

280 fps everytime, and ideal accuracy

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment