Wicked Air Sportz: Turbo Rev

[Vegeta]

Lets talk about this controll of 'Torque'. Adjusting the torque of the motor using amperage/voltage variations? No... won't work. Thin about it. The rev motor spins at one speed, but that speed could be controlled by varying amperage/voltage. BUT. Speed and torque are two VERY diffrent speed. I don't care if a motor is spinning a wheel at a high RPM, if it doesn't have force (torque) behind it, it is not goign to propell the car. TheRev board it like a servo. it is a motor and has gears to change the high speed and low torque of hte electric motor down to a lower speed yet higher torque that is needed. The WAS turbo rev board CANNOT effectively change torque by just increasing the load going to motor. They wopuld haev to have to change gear ratios in the rev motor to control things. Notice I said effectively.

And also, I donnot thing they can be changing amperage going to the motor from the battery using that chip. Not on that board. Someone who knows a bit more on how it operates might know but I don't think it would be possible.

[314159]

http://www.rcdesign.ru/eng/electronics/esc/

this link is of a motor speed controller using the pic12c509 (identical to the pic12c508 except for a little more ram and rom).

[Vegeta]

Im referring to Torque as hte force in hte rotation of the motor. More torque means more power to move heavier things. Less torque would mean that the revvy impeller would not be strong enough to jostle any paint around, especially on a full hopper, no matter what speed.

The question is- can that ESC controll torque? Lowering hte speed of the motor will not nessecarily increase torque... will it?

[314159]

in an electric motor, lowering the speed will not increase the torque.

as far as controlling the speed of a motor by sending it a pulsed waveform. the one in the link above operates at 3khz, this either turns the motor completley on, or completley off.

if the motor was on 100% duty cycle. as an example it would have the same torque as when the motor is on at a 50% duty cycle (when the motor is operational) and 0 torque when the motor is inoperational. if you take the speed that the motor is fed the on/off signal down to an extreme, let's say 4 hertz. it becomes easier to see that the paddle has the same torque behind it when it is operational when it is being rapidly being turned on/off as when it is operational 100%.

so torque is not an issue.

[nerobro]

well if you're changing the speed the motor is turning you're also changing the torque it's producing. the lower the rpm an electric motor turns the higher it's torque, peaking at 0. Also by turning the motor on and off you're effectively lowering the torque by having the motor off a portion of the time.

[314159]

you could easily average it and get and average torque.... but when the motor is on, there is the same voltage/current/power going through the motor for that instant that it is on (operating at a reduced duty cycle), to when it is operating at a 100% duty cycle. so the motor would output the same torque at that instant.

by varrying the duty cycle of the motor, you can controll the speed, having it produce the same torque (while the motor is moving).

if you had something that would require 15 units of torque to move, and the motor produced 20 units of torque when it was on, and it operated at a 50% duty cycle. the mean average would decieve you. because it would be able to be moved.

[Vegeta]

...Have wefound out if the WAS Turbo rev board really does all this or is that just more false advertising?

I am still cracking up everytime I read that tidbit about hte 'Artificial Intellegence software' that this thing is supposed to have.

[nerobro]

I think we can safely say that their claims are incorrect. Controling a motor at 4mhz is bad for the motor. there is no way for it to properly determing the rate of fire youre' shooting, at least with any kind of reliabilty.

[314159]

4mhz = the clock speed (speed of internal oscilator, can run up to this speed)

1mhz = the instruction speed (1/4 of clock speed)

you can switch on/off the motor at whatever frequency you want within reason



almost institaneous rate of fire = 1 / (the difference of the time between the tailing edge of 2 balls)

the tailing edge of 2 balls = the moment in which the eye goes from blocked, to open

[nerobro]

1mhz of instructions are very quickly eaten up if you're trying to find BPS, and send the proper control impulses to the motor ;-)

[manike]

I thought you guys might be interested in some quotes for the last time I clashed with WAS over the turbo rev. The following quotes come from the thread http://www.pbnation.com/showthread.p...threadid=16097

Originally posted by WickedAirSportz
The artificial intelligence software uses an adaptive algorithm. This means that the feed rate will increase or decrease based on the time frame between balls.

Your example shows that there will be delays caused by the bolt opening/closing. What most people can not comprehend is that the TurboRev actually syncronizes the feeding to match the delay. This keeps the flow of balls steady, and gravity working.

This made me laugh my cotton socks off. There is no way a system which looks at where a ball is at the top of the feed tube can synchronise the dropping of a ball with when the breech is open or shut. Especially since it is using a rotating paddle to randomly knock balls towards the feed hole. To think it could 'synchronise' even hitting a ball at the right time to feed it, is laughable.

He tried to explain a form of SHM that would make balls at the bottom of a stack fall faster due to impacts at the top of the stack, timed to co-incide with when the breech was open... To quote a good friend of mine, who will remain nameless unless he wishes otherwise, the chances of that happening are like a Baseball player being blindfolded and expected to hit a home run from a fast ball pitched at him from anywhere out of the crowd...

You will notice that WAS once again tried to talk about something he knew nothing about (The Warpfeed), and tried to discredit my maths (he was wrong again) and you can see my reply to which he never responded again... I wonder why?

He also stated that the algorythm required at least 16 and no more than 32 shots (I think those numbers are right but it's from my memory) to be fired for it to work out how to feed properly... so what happens if the string you fire is shorter than 16 or longer than 32? Apparently it fixes what it does at one state after 32 shots fired... what happens if you don't fire consistently over those 16-32 shots (and lets face it unless you are in FA you won't...)

You might also enjoy this thread on the subject...

http://www.pbnation.com/showthread.p...threadid=21281

wow this was nearly a year ago that I wrote all that but it's familiar huh?

I came to the conclusion back then that this guy was full of hype and a BS artist!

manike

[Pand0ra]

The torque is very simple to adjust when you deal with permanent magnet DC motors.

You've simply to limit the current in the motor.
The lower the current, the lower the torque will be.
The maximum speed of the motor is determined by the voltage

If you compare two motors of the same type and size, but with different windings, you'll see the one with the lower internal resistance will have the highest torque.
The drawback is in general a loss in speed.

Other kinds of DC motors use a winding for the stator.
In this case the current in the stator determines the torque.


Back to the board now...

First of all I don't see a real need of torque control, except to avoid the destruction of components or gears in case of the rotor gets stuck.

I don't see how they could implement it here.
To implement a torque control, you need some kind of feedback.
The PIC needs an amplifier to drive the motor (in this case a transistor, probably a MosFet).
Hence there is no feedback at this level.

You can always put a shunt in series with the motor to measure the current. The problem is you need at least an amplifier to increase the amplitude of the feedback (the value of the shunt is very low, the tension measured also).
There's no A/D converter either in the PIC.
This could be solved with the usage of a VCO, to create a frequency proportional to the tension. Not implemented on the board from what I see (and it's quite hard to implement on the PIC, as the only counter is already used by the PWM probably).

I come to the conclusion there is no torque control on this board (maybe it's a consequence of the pwm control, but it's definitly not a real torque control).


Now back to the brain...

It's not possible to measure the real number of balls which leaves the loader with the gate implemented in the loader.
The reason is simple: you never now when there's no gap between two paintballs in the feed.
How could a normal barrier detects a movement of the paintball if the gate is always cut? It's impossible.

The measure is inherently false. As a consequence, whatever the processing done in the PIC, the result is incorrect.
I don't see any reasons to do heavy processing on this kind of datas...


The story would be totally different if there was an intellifeed to the hopper (is there one?).
In this case the pic could increase the speed of the motor till there's no more big gaps between the paintballs.
This would result in a decrease of the power consumption, and increase the life of the battery.

Just my 0,02 Euros

@++

[billmi]

Originally posted by Miscue
Well, looking at the TurboRev board... I don't believe it has the hardware to adjust power/speed... just turns on and off. I'm not an EEG dude, so not sure. Anyone know?

I haven't thrown it on a waveform recorder, but Jim Drew says he's pulse code modulating the output to the servo. This makes absolute sense as the paddles appear (to me, subjectively) to spin slower than with the X-Board. I haven't noticed a change in paddle spin rates as Drew claims (hence scoping the output would be pretty revealing).

Re: the previous comment of not having a 12v reg, PCMing the output will give a lower average voltage, making the vreg unnecessary, if the PIC is controlling a transistor that can handle the full voltage (18v).

PCM is an ideal method for adjusting motor speed as opposed to changing voltage, because you get more torque at lower speeds.

See you on the fields,
-Bill Mills

[billmi]

Originally posted by Butterfingers


The increase in feed rate could ALONE be attributed to the "WAS modified" Vortex Impeller that was in the revy.

No Vortex impeller, WAS modified or otherwise was used in the test.

http://www.warpig.com/paintball/tech...up/index.shtml

See you on the field,
-Bill Mills

[billmi]

Originally posted by manike
To quote a good friend of mine, who will remain nameless unless he wishes otherwise, the chances of that happening are like a Baseball player being blindfolded and expected to hit a home run from a fast ball pitched at him from anywhere out of the crowd...
[/B]

You forgot the rest - that has to be done while the stadium is moving. :-)

See you on the field,
-Bill Mills

[Koosh]

You guys seen the interview Jim Drew did with http://www.paintballreality.com/ ? I would reccomend reading it, if you haven't, here it is cut and pasted for your pleasure!

PBREALITY: Can you explain to us a little but about the

TurboRev and how it works?

JIM DREW: My definition of how it works is always disputed. There are some that

feel this technology is not possible. The results we have stated, as well as independant sources such as WARPIG, clearly show that the technology works. So those that don't have an open mind to technology, please skip this description.



The TurboRev uses a computer chip (more powerful than what we used to land our astronauts on the moon). This chip is programmed with routines that 'learn' patterns in the feeding and alter the motor speed and torque to achieve the best feed rate, and all in real time. By deliberately attempting to synchronize the "smacking" of the balls above

the stack in the feed tube with the cycling of the bolt, you can accelerate the balls into the breach. It is a similar concept to the desktop conversation pieces that have metal balls

suspended from strings. You pull back the first ball and release it, upon contact with the line of balls, the last ball rockets away from the rest, with the middle balls not disturbed. Its a simple idea, but the implementation is quite a feat. It took about 3 months of daily work to get the software to work properly. 3 months to write 500 bytes of code.



There is more, but I think this is the only part that pertains to this discussion...

[Jeb_Hoge]

You know what I'm waiting for, though? BLUETOOTH. "Real wireless intellifeed using Bluetooth technology!"

[pete-is-god17]

im all for you taking the this to the source and stickin it to em

rock on all u smart dudes, rock on

[WickedAirSportz]

I'll tell you what, since the days of agitated hoppers are numbered, and Wicked Air Sportz will be releasing our own force feed hopper in the future, I would be happy to openly discuss the technology that makes the TurboRev the fastest feeding device for the Revolution.

As Bill Mills pointed out, he conducted his tests with a 1st generation TurboRev with the stock 4 blade impeller. After Bill's tests, I did change the software a little bit to help in those situations where there are fewer balls in the hopper to start with.

Now, if people would like an informative discussion about this technology, I am open to it. The technology is quite possible, and it works - I don't think Bill got a big check to lie on the Warpig results.

Some of the posts here are dead on the money about how it works, and others are so far off in left field. It seems that most of the people posting are in their youth, and are probably not very experienced with assembly language programming, or even the basic tricks you can do with the PIC micros. If you think you need multiplication or division instructions, you are thinking too high level. There are no such requirements in the TurboRev software, although lookup table pointers are certainly built in realtime (even with the limited RAM).

The concept of there never being a "gap" between two balls is wrong. Out of months worth of testing, I have only seen a few times where two balls were actually touching together (no space between) right as they passed the dead center portion of the infrared beam. Remember, the beam is modulated at a particular pattern to eliminate reflections from the balls false triggering the receiver, and to eliminate problems with ambient light.

Basically, what I would like from this discussion, at the end of course, is a public apology. When we are done, the information that I am going to give would allow others to produce a competing product. As I stated, we are looking toward the future with our new hopper (which will likely be as doubted as everything that I have ever worked on in the past). You have to ask yourself one question, if nothing else - why does the TurboRev feed faster than any other aggitated hopper in existance?

Let me know if we can have a rational convesation here.


Perhaps when this discussion is closed, we can move on to the Equalizer doubts.

[Butterfingers]

We are not doubting the claim that your product works. Your products usually do. Not once did we say your products DIDNT work.

However, we find it misleading that you must hype up products to make it look better than it really is.

Most of your products are doubted because you are making outrageous claims. Yes they work but not the way you describe it.

State of the art microcontroller?

Common its a $4 microcontroller. You're making it sound like this thing has the cray supercomputer in it.

I say if you dont wan't people do doubt your products don't make bogus claims, all we want to see is performance.

Look at the halo... similarily according to your standard the halo uses "space age electronics more powerful than the computer used to put the astronauts on the moon" also.

If the guys at halo made that claim we would be cynical too.

The bottom line the consumer dosent like being fed a line of crap for a product they can judge the performance of by themselves. It turns alot of us off.

[cphilip]

Yes the smoke is a bit thick. We are slaying hype here not your product. Stick to the truth and sell the snake oil to the kiddies.

[WickedAirSportz]

State of the art microcontroller?

Well, perhaps you guys should be taking your complaints up with Microchip, because they clearly list in their advertising that their processors ARE the state of the art in technology.

If your arguments are about how things are worded, well then I am sorry - that will probably not change. It's accurate, and usually taken from some other source (like the Microchip website and documentation in this case).

Buzz words are used for advertising purposes, that is how products are sold. You will find such words in our advertisements and statements, but you will not find anything that is inaccurate.

[WickedAirSportz]

We are not doubting the claim that your product works. Your products usually do. Not once did we say your products DIDNT work.

Ah, here is another point that is a bit personally irratating.

There are those that are doubting that this technology is not possible (even though the results certainly prove otherwise). Just because YOU can not create the technology, does NOT mean it can not exist. I think this is probably the biggest issue in the "are UFO's real" debate. Technology beyond our scope of thinking is certainly not possible - we are just not thinking hard enough.

[Xen]

Nothing you say is inacurate eh? Well gravity itself can't feed faster than 13 bps and you site claims that the turborev can feed 16 bps. Since when does an impeller become an instrument to force feed (i.e exert constant pressure on the balls) the paintballs into the breech? Seems like you trying to misslead people.

[Miscue]

Originally posted by WickedAirSportz
I'll tell you what, since the days of agitated hoppers are numbered, and Wicked Air Sportz will be releasing our own force feed hopper in the future, I would be happy to openly discuss the technology that makes the TurboRev the fastest feeding device for the Revolution.

As Bill Mills pointed out, he conducted his tests with a 1st generation TurboRev with the stock 4 blade impeller. After Bill's tests, I did change the software a little bit to help in those situations where there are fewer balls in the hopper to start with.

Now, if people would like an informative discussion about this technology, I am open to it. The technology is quite possible, and it works - I don't think Bill got a big check to lie on the Warpig results.

Some of the posts here are dead on the money about how it works, and others are so far off in left field. It seems that most of the people posting are in their youth, and are probably not very experienced with assembly language programming, or even the basic tricks you can do with the PIC micros. If you think you need multiplication or division instructions, you are thinking too high level. There are no such requirements in the TurboRev software, although lookup table pointers are certainly built in realtime (even with the limited RAM).

The concept of there never being a "gap" between two balls is wrong. Out of months worth of testing, I have only seen a few times where two balls were actually touching together (no space between) right as they passed the dead center portion of the infrared beam. Remember, the beam is modulated at a particular pattern to eliminate reflections from the balls false triggering the receiver, and to eliminate problems with ambient light.

Basically, what I would like from this discussion, at the end of course, is a public apology. When we are done, the information that I am going to give would allow others to produce a competing product. As I stated, we are looking toward the future with our new hopper (which will likely be as doubted as everything that I have ever worked on in the past). You have to ask yourself one question, if nothing else - why does the TurboRev feed faster than any other aggitated hopper in existance?

Let me know if we can have a rational convesation here.


Perhaps when this discussion is closed, we can move on to the Equalizer doubts.

Rational Conversation? I suppose we should mimic you with conjectures and statements we have not attempted to back up in a logical fashion.

Once again, you've made comments w/o any supporting evidence/data. Your authority on the matter is self-proclaimed. You have not been able to answer the simple questions : WHY? and HOW? other than that it's because you say so.

So, some things are in left field. WHICH claims are incorrect? Support with evidence.

I do not see the necessity of an apology because the information presented: A - Has no ill-intent. B - Invites and does not inhibit counter-argument to test the validity of the claims. C - Has not been disproven by you or anyone else, and thus stands as the best flow of reasoning to base our understanding of the matter.

"It seems that most of the people posting are in their youth, and are probably not very experienced with assembly language programming, or even the basic tricks you can do with the PIC micros."

This is your response to my post I assume. Reason is timeless and undiscriminating. A logical proceeding is no less sound or less valid if born from the mind of a five year old or a sixty year old. Don't tell us what our ages are, explain why the logic is wrong.


<marquee><B>PROVE IT......PROVE IT......PROVE IT......</B></marquee>

[WickedAirSportz]

Since when does an impeller become an instrument to force feed (i.e exert constant pressure on the balls) the paintballs into the breech?

There is not constant pressure on the balls.

I will be back tomorrow to check the responses. If people would like to know how this works, and have an open mind, with an open discussion, that would be great.

Once the discussion is complete, there will be nothing left to "prove" to anyone, and the nay-sayers will probably feel a little silly.

[Miscue]

Show us the code, so we don't have to bother with getting a chip programmer to get it ourselves. Provide commenting on the code that shows where the chip does all the things you say it does.

[cphilip]

Well Jim heres the thing...no one here is saying your product is not better than say a VL revy. No matter how it works. We just hate to see you stoop to what we feel are gross exagertions to sell it. Thats the point here. Maybe you feel compelled to do that but around here that doesn't sit real well. We perfer honesty and technical data. Facts. If that doesn't sell then well we would rather not condone its use to do so. See?

It's nothing personal at all. It's just the general attitude we cultivate here on AO

[WickedAirSportz]

Show us the code, so we don't have to bother with getting a chip programmer to get it ourselves. Provide commenting on the code that shows where the chip does all the things you say it does.

Not a chance... I am going to do something even better. I am going to hold your hand, and walk you through the technology, make you answer the questions, and let you build one yourself.

BTW, I was not foolish enough to not lock the PIC chip.

[Butterfingers]

WAS,

The test results don't prove that the "AI technology" is possible. It just proves your loader feeds faster.

Sorry a change in the loaders logic does not consititute "AI"

Nor... is a $4 pic a state of the art microcontroller.

Perhaps we have diffrent definitions of "AI"

To me the optical eye picking up changes in feed rate and comparing it to a pre made chart does not constitute AI.

Or is it even possible given the limited capabilities of the microchip?