PoeticInjustice
more magical than the magicalest
teh cut3 on3's bitch
My name is Inigo Montoya. You killed my father. Prepare to die.
Posts: 300
|
Post by PoeticInjustice on Mar 3, 2005 23:31:45 GMT -5
just thought I'd throw it out there to see who out there actually knows what they're doing for the inevitable problems I'll run into designing and building the controller at some point in the future
|
|
DarkNeopagan
damn i love posting here!
1st Sargeant
"One case, One kill" Broadsword Motto
Posts: 185
|
Post by DarkNeopagan on Mar 4, 2005 12:38:39 GMT -5
Just let me know what it is, if i can't answer it, someone i work with will be able to.
|
|
PoeticInjustice
more magical than the magicalest
teh cut3 on3's bitch
My name is Inigo Montoya. You killed my father. Prepare to die.
Posts: 300
|
Post by PoeticInjustice on Mar 12, 2005 1:39:02 GMT -5
this is what I have of the schematic so far has fire selection control, the breakbeam circuits, power supply, safety and trigger. the safety is in hardware, if it's not on, trigger switch doesn't get connected to the microcontroller. what it doesn't have yet is solenoid drivers (likely a power transistor or MOSFET), flyback diodes to kill back EMF from the solenoid valves, decoupling caps, debouncing on inputs (might not need that, not sure?? or I can possibly do it in software), and it doesn't show where connectors would be going from the board to all the stuff, just shows wire instead (same circuit theoretically) oh, and which input/output going to each pin on the microcontroller can be changed, it's pretty unimportant really, as I don't need to do anything fancy with the more advanced I/O functions. So the order will all change according to what would fit on a board easier. I have to figure out what my solenoid valves are going to be and what their power ratings are before I can size appropriate transistors and flyback diodes to drive them. After I finish figuring this schematic out, I'm gonna look into either having a few PCB's done by some company, or try to do them with copper clad and etch 'em myself, hoopefully without killing myself from the fumes in the process. As far as the valves go, I've calculated that I'll need between a .1 and .2 Cv rating on my ram control valves, and ideally they'd be able to handle 250 PSI. I haven't figured out the necessary flow rating for the firing valve itself. Work in progress I guess. I should be able to figure it out via the number of shots available in a given sized tank, and work it out from there as far as figuring out a valve that can flow about 200x the amount per shot, per second (instead of 100, to account for partial duty cycle, etc etc) It really looks like the valves are going to be the most difficult part to figure out, and also likely the limiting factor in rate of fire. meh. But, most 12v and even 24v valves tend to be very low wattage, so, I may be able to get a more powerful coil to do the job faster, and just have higher power consumption. current electro guns get 10's of thousands of shots on a 9v battery, I'm looking for *maybe* 5000, on 1 or 2 ~1500 MAH 7.2v nicads, so I have a *lot* more power available to work with, and that should buy me some serious flexibilit in selection. I may actually figure out some hard specs and have em sent out to a few valve manufactures to see if I'm even sane as far as that goes. This is really getting to be a run on paragraph here. thoughts?
|
|
Tharnex
yah....i spend a lot of time here
Sargeant
Posts: 122
|
Post by Tharnex on Mar 13, 2005 23:20:54 GMT -5
hmmm, as far as shots per tank and in that department, i think i might be able to help you or at least point you int he right direction. CO2 is very inconsistant when it comes to constant pressure and volume due to the fact that its part liquid when it leaves the tank. most markers have expansion chambers and/or regulators to increase the volume for the cO2 to expand thus giving you less of a chance that liquid cO2 will enter you gun and damage your internals. This is why most players convert to HPA this will elimate all chances of your marker becoming a solid block of ice. if you are going to fire fast, then i HIGHLY recommend the use of HPA and for that beast i take it that the efficency will not be the greatest so i would get a large tank, a 4500 117ci woudl be good, because it coudl hold a lot of shots or find one in your price range. I saw a vid of this gun attached to a canister of CO2 and lets just say that it fires until it could not fire anymore .
|
|
PoeticInjustice
more magical than the magicalest
teh cut3 on3's bitch
My name is Inigo Montoya. You killed my father. Prepare to die.
Posts: 300
|
Post by PoeticInjustice on Mar 13, 2005 23:58:30 GMT -5
yea, CO2 is highly unpredictable due to it's liquid nature at tank pressure, and massive evaporative cooling at high levels of use. I have no doubt that this thing would freeze solid on max rate of fire full auto in a short time. but it would still run it in short periods without any problems most likely.
As far as air consumption per shot and valve sizing based on that, that link I posted a while back (which now seems to be dead?) said I can expect 400-500 shots on a 3000PSI 44ci tank.
44in^3=.0255 ft^3 *3015 PSI=76.8 standard cubic feet of air, over 400 shots=.192 SCF per shot, =19.2 SCFM at 100 shots per second.
so the firing valve needs to flow about twice as much as the ram control valves, based on those rough calcs.
problem with figuring Cv here is that I'm essentially exhausting to atmosphere, so I guess that my drop in pressure would essentially be the same as my input pressure? not sure here. delta-p is used in calculating Cv. bleh, at any rate, assuming full pressure drop at 40 SCFM (roughly twice the flow above) it gives a Cv of about .3. assuming only a 25 PSI delta-p, I need a Cv of .5-.6, which I don't think I'm going to find at the required speeds. :*(. I'll have to figure this one out somehow, it's beyond my knowledge. oh well.
so I guess I need to find 2 valves with .1 to .2 Cv, and one with about .3 to .6 Cv that can operate better than 200hz (5 ms cycle time, not gonna happen, but the close to that, the better!)
that was long winded.
and yes, this'll need compressed air of some sort for sure. otherwise you'd be holding a portable ice maker.
<edit> FUCK. that's not 19.2 standard cubic feet per minute, it's 19.2 standard cubic feet per SECOND, so doubled for duty cycle that's about 240 SCFM, requiring close to a 3 Cv valve. argh. this is gonna require some more careful engineering here. shit. shit. shit.
I'm really hoping the figures on that page that I'm basing consumption off of were way the hell off. </edit>
bah, after checking again, I do need to find a better guide to air consumption per shot, all the different tank/pressure/shot combinations they list come out to largely varying consumptions, some nearly half the .192 SCF/shot listed above. bah.
no more brainpower left to figure it out tonight methinks
though I did think of a new mechanical design improvement. news at 11!
|
|
PoeticInjustice
more magical than the magicalest
teh cut3 on3's bitch
My name is Inigo Montoya. You killed my father. Prepare to die.
Posts: 300
|
Post by PoeticInjustice on Mar 16, 2005 23:28:45 GMT -5
I just set up a link so just about all my available resources on the project are up on my site 69.244.139.89/project/ and are free for the browsing
|
|
Tharnex
yah....i spend a lot of time here
Sargeant
Posts: 122
|
Post by Tharnex on Mar 17, 2005 1:18:16 GMT -5
maybe you should think to limit the cycles of the marker, due to the feed rate available you to....normal shake and shoot (9bps max), agitating hopper (10-15bps), force feed (17-32bps), i don't se you getting any higher than those and as you know the higher you go the more expensive it is. i found a qloader set on ebay for like $50, came with everything and 5-100 round q-pods. Take a look into it woudl be the cheapest around. now with the reduced cycles on the gun, air will last longer and then a possible port to CO2 can be made. Note, 10bps is still pretty nasty, and you have to think of the hopper size, a 200 round hopper will only last 20sec tops, and that is a lot of paint too. soooo quick bursts would be best for a lot of reasons: conserve air, 'heating' up gun, paint saving, and you still retain the desired effect of scaring the enemy. Also the marker will not fire if there is not a ball in the breech so, that will limit your rate of fire too ( in a good way) you don't want to make a flame(paint)thrower now do we .
|
|
PoeticInjustice
more magical than the magicalest
teh cut3 on3's bitch
My name is Inigo Montoya. You killed my father. Prepare to die.
Posts: 300
|
Post by PoeticInjustice on Mar 17, 2005 2:13:27 GMT -5
the current idea is to have an adjustable max rate of fire when in full auto mode, and for burst fire mode, fire as soon as the electric eye shows the gun is ready
yes, q-loaders are the best thing out there, but ammo capacity sucks at 100 per loader
I've been thinking of some belt-fed designs for a focred feed loader capable of delivering high rate of feed, and I've looked at patents for about a dozen or so motorized and/or spring powered loaders.
the perfect loader would be something you could strap on your back, and run a tube to the gun, and that's what I'm trying for. if I don't have the time to get that to work, then I buy a q-loader set.
|
|
Tharnex
yah....i spend a lot of time here
Sargeant
Posts: 122
|
Post by Tharnex on Mar 17, 2005 14:13:16 GMT -5
if you want to do something lke that then you might want to look into something called a warpfeed. it will send the balls through a tube and stuff liek you want and its keeps a constant ball pressure. if you want something on your back, then that means a really long tube, thus more paint required to take up space in the hose. there shoudl be a paintball 'tank' on the ground by the gun(or on the gun itself) and you can pour the paint into that and then it will feed to the gun thus shorter hose, less volume of useless paint ect. www.airgun.com/warp_feed.shtml
|
|
PoeticInjustice
more magical than the magicalest
teh cut3 on3's bitch
My name is Inigo Montoya. You killed my father. Prepare to die.
Posts: 300
|
Post by PoeticInjustice on Mar 17, 2005 15:59:29 GMT -5
yup, warp feed is cool. I've looked at the patents on it. stock warp feed is too slow though
|
|
Tharnex
yah....i spend a lot of time here
Sargeant
Posts: 122
|
Post by Tharnex on Mar 17, 2005 20:24:54 GMT -5
you can bump it it 22, pretty good for what we are aiming at. think about it, 10bps, @ 300, no one can dodge that. so my question to you is what are you trying to achieve in doing this, what do you want out of this 'uber' gun.
|
|
PoeticInjustice
more magical than the magicalest
teh cut3 on3's bitch
My name is Inigo Montoya. You killed my father. Prepare to die.
Posts: 300
|
Post by PoeticInjustice on Mar 21, 2005 0:37:28 GMT -5
Well, shoot for the best is what I'm going for. I want an insanely high rate of fire, so in ( x number of shots )burst mode, it'll rip em out and put em all damn near on the same target. On another note, i came to yet another realization that I was trying to do something I didn't need to.. I assumed that the bolt would have to be able to cover a fair bit of gap on either end of the feed port to seal, but in the forward position it only needs to seal in front. duh. this means I can get away with a single sensor for bolt position sensing, because the bolt can now be shorter than it's stroke. so it'll clear the sensor if the bolt is in either position fully. then it just has to remember internally which way its going. this also means less reciprocating mass because of a smaller bolt, which is definately good! that other mechanical revelation I made mention of earlier was that I can move the pneumatic ram that actuates the bolt to the side of the gun. easier to draw than to describe. doing that instead of having the ram behind the bolt like in the original drawing does several things for me 1. significantly shortens the overall length of the gun 2. solves the problem of having to find a way to keep the bolt from rotating and thus blocking whatever porting is in it to actually fire the paintball 3. makes it to where less force is applied on the bolt forward stroke than on the reverse stroke. this probably makes it less likely to break balls. the bolt forward stroke is now the retract stroke of the actuator which has less force because the bore is effectively reduced by diameter of the rod. It also means that the bolt will pull back faster after a shot, which is good as well. there might have been something else that design does as well, but I don't remember it. <yawn>
|
|
PoeticInjustice
more magical than the magicalest
teh cut3 on3's bitch
My name is Inigo Montoya. You killed my father. Prepare to die.
Posts: 300
|
Post by PoeticInjustice on Mar 21, 2005 12:54:34 GMT -5
|
|
PoeticInjustice
more magical than the magicalest
teh cut3 on3's bitch
My name is Inigo Montoya. You killed my father. Prepare to die.
Posts: 300
|
Post by PoeticInjustice on Mar 21, 2005 20:17:29 GMT -5
hah! freebies! I was bored, and thinking about the feed mechanism, so I went around looking for any dead hardware with belts in it. found a long since replaced printer
got 2 stepper morors and a 28vdc reversible motor for driving a belt that's about a foot long loop, so 2 ft total. also have drive circuits for all of the motors, if I can figure them out...
four optical interruptors, and one reflective optical sensor
very, very cool, this'll give me some goodies to play with
course, it wasn't entirely free. my hands are now more or less covered in black ink..bleh. oh well, lotsa goodies! anybody got dead printers? <grin>
|
|
PoeticInjustice
more magical than the magicalest
teh cut3 on3's bitch
My name is Inigo Montoya. You killed my father. Prepare to die.
Posts: 300
|
Post by PoeticInjustice on Mar 22, 2005 11:58:17 GMT -5
this is what I have so far for the general flow of the code, I think it's getting fairly close coders out there, any thoughts? necessary values 2 bits for bolt position and direction 2 bits for shots remaining in burst mode value indicating trigger was released at some point in the last cycle or before initialize pins for I/O as needed. need 6 inputs, 2 possibly analog. need 3 outputs, all digital. need additional outputs if using pick and hold for faster solenoid actuation. fire bolt forward solenoid and wait some delay, and/or wait for sensor input to ensure gun is in forward position perform tests? enter main loop main: check to see if gun is in the middle of firing 'what portion of firing are we in? gun has following 'states' in order 'ready, fireing, bolt back travel, waiting for ball, bolt forward travel, back to ready
check if trigger pressed, if not, set value indicating trigger has been released (necessary so we don't go full auto all the time!!)
(possibly need a delay, make sure trigger definately released for more than 1 or 2 ms) if trigger released, and released not set,
if status ready if semi check for trigger pressed and released, if both, call fire return if burst check for trigger pressed and released, if both, set remaining shots to max check for remaining shots,if >0 call fire decrement remaining shots return if full check for trigger pressed, call fire return goto main
if status firing check timer for sufficient delay to fire ram return, if so fire ram return valve, set status to bolt back travel check timer for firing valve dwell, if reached, shut off valve goto main if status bolt back travel check timer for firing valve dwell, if reached, shut off valve 'in both states so varying dwell times can't cause problems check timer for return valve dwell, if reached, shut valve off. check optical interruptor for bolt being in end positon. if bolt is in end position, *has changed state! remember, single
sensor!* change mode to waiting '(note, we are going to need to keep track of this input over time. I think an ideal situation is, as soon as gun is fired, monitor for
a change in signal to verify that bolt has moved forward to block sensor. after we're sure it's been blocked for some period of time (1
ms or 2 should be sufficient?) then we can set a bit to tell us that the minute it becomes clear we know for sure we can shut off the
valve) goto main
if status waiting check ball presence optical sensor to see if ball is in place, if it is, change to bolt forward travel mode and fire bolt forward
solenoid goto main
if status bolt forward travel check timer for forward valve dwell, if reached, shut off valve check bolt position sensor, if bolt forward, shut off valve change status to ready goto main
goto main
fire: reset timer1 open fire valve set status firing set trigger released false return
|
|