So I was Thinking about a New Gun Design.....

[om3n]

For everyone that says "go back to basics and make it cheap" my response is that I don't do that.
AGD

?!
Why is Tom talking about this in terms of HIM making this gun??

Please mean what I think it means....

[nerobro]

For everyone that says "go back to basics and make it cheap" my response is that I don't do that. China, Tippmann and everyone else does that just fine. There is no possible way to make a marker in this country that is any where near good value for the money.

But a midrange, or high end gun? I'm not even entirely sure if that's possable. The financial realities of production are something you're a lot better at than I am.

Carburetors are much simpler and cheaper than fuel injection, how did we ever get cars with computers? Perhaps the age of paintball innovation is over with, are we now in the twilight years?

This is a sad state. :-/ Nearly everything went to two designs by the time I stopped really following paintball. Spool valves, and linked hammer guns.

I do think the electronic regulator concept can work. And it would be a uniquely AGD thing. I think it could be done at a competitive price. Though in this case, the hardware is the easy part, (I think... ) the software is going to be where all the magic happens. With a gun that more or less has it's operating method defined by software, a "lets get it to market" software could be released, and as new features came up, the gun could be brought up to the new standards either by flashing, or new chips.

getting technical again
Anyone know what sort of electrical requirements injectors have?

[cockerpunk]

no doubt the rise in electrical control systems in both cars and paintball guns can be paralleled. we used to live in a world where engineers knew there was a dynamic range of conditions and then made some assumptions and picked an average value that would work in the vast variety of systems. now we live in a world where you can stick a sensor on it, measure it, optimize it, then change it to get an ideal all the time (or damn close).

but i wonder why the closed loop systems that are so common in modern vehicles and other appliances (washing machines, blenders ... god you name it) have not taken off in paintball. my initial thoughts are that no one in the industry is any good at programing. it would be a pity if that were true. a closed loop system could be made idiot proof very easily. combining a few sets of sensors (eyes or pizeo-electrics) and a barrel tip chrono, you would never have to adjust your gun for any reason, the gun would be self-setting up. much like how car today optimize there timing and such for performance, or fuel economy.

idk what the next revolution in gun design will be. my thoughts are a gun and loader that are integrated to truly stop all possible barrel breaks, but we will see if that is the case. there are a few reasons why i think that, but i want to know what tom thinks. the industry is certainly positioned for the next breakthrough, nothing new or revolutionary has come out since the Mini, which wasn't that big of deal.

idk Tom, what do you think is the next revolution?

[nerobro]

*feedback systems*

Sensors are expensive. And running blind works really well for applications like this. As much as cars have sensors out the wazoo, at full throttle, say.. half the sensors are ignored. Or the car runs off of a reading that was taken at startup, not something that's constantly updated.

An all singing all dancing gun was tried in the past. It didn't get far. For a lot of reasons... I don't think the market could support a gun like the cyber9000. Features that would allow for self velocity regulation would require ugly add ons, They also couldn't be trusted to work consistantly. A gun with a built in chrono, is a gun with just one more part to have fail. Guns need to be sexy to sell.

With an injector, and either a nice input tranducer (like the angel airs had..) or feeding it with a "decent" n2 system, we could precisely meter out air. Precisely enough that most of the velocity variation would come from things that we can't really control.

Because it's worth mentioning, current motorcycle EFI runs blind, or nearly so. That's not far from what we'd be doing.

I hate to say it, but I agree that the quality of programming in paintball guns usually leaves something to be desired. So does the engineering (or complete lack thereof...)

Still I think we're to the point where acutally "doing" something would be the next step. Either call up siemens or bosch, and see what they have to say about injectors, or go scrounge one and throw it on a barrel, see if we can get sufficent velocity.

[cockerpunk]

Sensors are expensive. And running blind works really well for applications like this. As much as cars have sensors out the wazoo, at full throttle, say.. half the sensors are ignored. Or the car runs off of a reading that was taken at startup, not something that's constantly updated.

An all singing all dancing gun was tried in the past. It didn't get far. For a lot of reasons... I don't think the market could support a gun like the cyber9000. Features that would allow for self velocity regulation would require ugly add ons, They also couldn't be trusted to work consistantly. A gun with a built in chrono, is a gun with just one more part to have fail. Guns need to be sexy to sell.

With an injector, and either a nice input tranducer (like the angel airs had..) or feeding it with a "decent" n2 system, we could precisely meter out air. Precisely enough that most of the velocity variation would come from things that we can't really control.

Because it's worth mentioning, current motorcycle EFI runs blind, or nearly so. That's not far from what we'd be doing.

I hate to say it, but I agree that the quality of programming in paintball guns usually leaves something to be desired. So does the engineering (or complete lack thereof...)

Still I think we're to the point where acutally "doing" something would be the next step. Either call up siemens or bosch, and see what they have to say about injectors, or go scrounge one and throw it on a barrel, see if we can get sufficent velocity.

sensors used to be expensive.

they are not anymore.

the accelerometer that makes the wii system (littereally without the wii wouldn't work) is less then a 2 dollar part. you can buy them on any electronics site for cheap cheap cheap. pizeo-electriccs are even cheaper, and eyes are also VERY cheap. if you didn't know, most encoders use break beam eyes like in out guns, and encoders are SUPER cheap. stepper motors on the output side are easy to get a hold of today too, up to relatively large torques even. we arn't talking about CnC hyper accurate sensors here.

that is a key reason why control systems have taken over our daily lives so much, because they are cheap and effective. its the reason why electronics have taken off in paintball so much too. you used to have to spend hours with an expert to fine tune your trigger, now your trigger control system comes from china for 5 bucks (PCBs are super cheap), and is better then any expert can tune a mech.

while clumsyness might be an issue, that does not seem like a reason to not attempt a fully electronically controlled gun. as i said before, the payoffs would be awesome. i for one, hate having to setup and tune guns, you would not have to do any of that. not to mention reliability in this game is a huge marketing position.



the only reason i can think of that anyone wouldn't attempt to have more electronic control on there guns is that the current people writing the code are not talented enough to control those types of things. or they do not realize the goal in and of itself.

[luke]

For everyone that says "go back to basics and make it cheap" my response is that I don't do that. China, Tippmann and everyone else does that just fine. There is no possible way to make a marker in this country that is any where near good value for the money.

AGD

[luke]

Perhaps the age of paintball innovation is over with, are we now in the twilight years?

AGD

[snoopay700]

sensors.

I agree, although i do love tinkering and stuff it would be nice to just be able to go out to the chrono range, shoot off a few shots, shoot a few over the chrono to be sure your chrono is accurate, and then go play. The only thing you would ever have to do is change modes, and that can be done fairly easily.

[cockerpunk]

I agree, although i do love tinkering and stuff it would be nice to just be able to go out to the chrono range, shoot off a few shots, shoot a few over the chrono to be sure your chrono is accurate, and then go play. The only thing you would ever have to do is change modes, and that can be done fairly easily.

i keep a couple of mech cockers on hand whenever i get the urge.

sometimes though, you just want your damn gun to work so you can freakin play the game.

[nerobro]

sensors used to be expensive.

they are not anymore. *snip*

that is a key reason why control systems have taken over our daily lives so much, because they are cheap and effective.*snip*
while clumsyness might be an issue, that does not seem like a reason to not attempt a fully electronically controlled gun. as i said before, the payoffs would be awesome. i for one, hate having to setup and tune guns, you would not have to do any of that. not to mention reliability in this game is a huge marketing position.

the only reason i can think of that anyone wouldn't attempt to have more electronic control on there guns is that the current people writing the code are not talented enough to control those types of things. or they do not realize the goal in and of itself.

Your point stands up to an extent. The IS board was $10 in parts, including the PCB, at 20 count quantities. Lots of sensors are cheap now. Lots of sensors were always cheap. Eyes are cheap (photoresistors/transistors/whatever) are less than a nickel each, leds are a few pennies so long as you don't want them to be "laser". (that term makes me ill...) Now, mounting those sensors, and putting them in useful places, is a lot more difficult.

Also a lot of the "sensors" used today aren't serious sensors. More often than they'd like to admit they're just resistors placed so the microcontroller can sense the current being used. I think that hits upon why things have gotten cheap. Processing power is now at nickel or quarter cost, instead of dollars. You can take advantage of cheap sensors due to cheap microcontrollers. This is a leap that the paintball industry has already made.

A collar that had two gates in it to measure paintball velocity would cost very little electrically. But you'd need to have parts machined, or plastic molded. Those things aren't cheap to start doing . And if you're really smart you'd have some fancy way of protecting hte wires going from said collar back to the gun. Maybe it could be sold as an accessory? I don't see it needing to be intergal to the design.

I think the goal of production for a "completely self compensating" gun would not be hard to achieve. We're talking cheap parts, obviously. However trying to sell said gun... that would be a lot harder.

As for deeper electronic control of the guns... most of the designs out there are quite brute force. They are consistent because the inputs to them are large, and robust. like 7-10ms on times for solenoids.

With the injector design, we could be playing with outputs that are in the khz range, rather than hz range. :-) I want to see a data sheet for the peizo injectors.

A gun with the injector would be quite consistent. It would only need once a day velocity setup, and that would be the end of it.

When developing a product, one needs to keep their mind open to futher expansion, but not get caught up in feature creep.

[cockerpunk]

thats exactly what i mean. the issue isn't that sensors are expensive, its that writing the code to control them is far more complex then we seen in most paintball guns today.

thats the only reason i can think of why they arn't more prevent.

[oldironmudder]

Found this video that shows an injector working.

http://www.youtube.com/watch?v=u-ZC1FaNDh0

http://www.etas.com/data/RealTimes_2006/rt_2006_01_28_en.pdf

For this reason, the actuator of a
piezo injector consists of 350 stacked
layers of piezo crystals whose combined
expansion at 150 volts adds
up to roughly 40 μm. This is sufficient
to effect reliable injector valve
switching


Compared to conventional electromagnetic
injectors, the piezo injector
is four to five times faster.

Because the fuel injector's opening
or closing interval has shrunk to a
mere 100 μs, up to seven injections
per cycle are feasible at an injection
pressure of 1800 bar (26,107 psi)
(planning for 2000 bar is underway).
Even very small fuel volumes of only
1 mm3 per lift are easily accomplished
with piezo injectors.

Im still looking for better specs.



Oh ya, found this picture of a 6.4L Powerstroke injector.

[Psi]

Perhaps the age of paintball innovation is over with, are we now in the twilight years? AGD

I seriously doubt you believe that Tom. I believe that innovation in respect to rate of fire has gone as far as it needs to, but some people just can't help themselves and will push it further. Lets hope they are as wise as you and find other markets for it.

Where innovations need to continue is in accuracy and range. But look who I'm talking too :) Which brings me to the alternating vortices in your first post. How can they not be random? I understand they are alternating. But with a trip boundary layer caused by the seam and the inability to control the orientation of the seam, it would seem that the vortices would always be random.

Have you seen something with all your super duper gadgets? I would go look with my own super duper gadgets but Fairbanks spent all my super duper gadget money rolling BMWs into little balls of aluminum. I come seeking knowledge all masterful and all knowing Tom.

[vf-xx]

Im still looking for better specs.

Oh ya, found this picture of a 6.4L Powerstroke injector.

Hrm. That's going to make for a long gun....

[Dend78]

this is a walk in the park fit for an AK-47, M-15 ...

what the heck is an M-15