PDA

View Full Version : The "90psi" graph...



Kweasi
07-12-2002, 10:34 AM
...I was giving that graph a good stare and
noticed that it shows the barrel as being in
a vacuum for a good portion of the sample time.
WTF!?
.
.
.

http://www.automags.org/forums/attachment.php?s=&postid=299290

^Right along the bottom here; notice how the
scale shows the pressure as being less than
zero? How can this be?

-Kweasi

nerobro
07-12-2002, 11:26 AM
It would appear that the gauge wasn't "zeroed"

314159
07-12-2002, 11:55 AM
maby it was the "vacume assisted feeding" that was listed on the boxes of the first angels ;)

aut911
07-12-2002, 12:17 PM
i can just see it now, the vacum asisted angel.

comes complete with super tight vacumefeed and charger.


ittl even vacume up that nasty coke spill!

nicad
07-12-2002, 01:30 PM
that is pretty funny..
but did you know that the PGI Mayhems have a vaccume assist feed? its in the bottom of the breach directly below the ball feed hole. actually pretty cool- too bad they drink air like a mule. :)

314159
07-12-2002, 01:38 PM
to bad that stock mayhems max out at about 9 balls/sec =[

Kweasi
07-13-2002, 12:51 AM
So I take it you guys agree that the graph is inaccurate?

That leads me to be suspicous of that 90psi peak pressure figure.

Can anyone clarify?

I greatly appricate the REAL DATA that this forum provides, but I'm a brutal skeptic and can't accept what it suggests without some explanation.

-Kweasi

nerobro
07-13-2002, 03:05 AM
The data is fine. the scale is off. the machines used to do this kind of data collectoin either require re-zeroing frequently, or you need to use some logic and place the graph in the right place.

AGD
07-13-2002, 09:50 AM
The sensor is dynamic in order to be able to respond quickly. With our setup the baseline of the dynamic sensor moves around so you do need to shift the graph up or down to zero to get the right numbers.

AGD

Kweasi
07-14-2002, 05:04 PM
Thank you for your helpful replies. I figured it was something long those lines.

Does that 90psi figure include the scale compensation?

Also, I played around with that 90psi figure and by applying P=F/A I found that at 90psi the ball is exposed to over 30lbs of force (see below). That seems funny to me, I'd think that it wouldn't take nearly that much force to get a ball to start moving, seems to be that subjecting a poor little paintball to 30lbs would bust it in short order. Anyone have any thoughts on this?

-Kweasi

The Math:
P=F/A
Since the back half of the paintball is the area under influence of the pressure, and since it's dome shaped, if we draw vectors perpendicular to the surface of the ball then all the vertical components will cancell, and we're just left with the horizontal vectors. For that reason we can consider the area of the ball under influence to be circular. Thus:
A= (pi*D^2)/4 where D= 0.68
Thus A= 0.363in^2

so:
F= PA
F= (90)(0.363)
F= 32.7lbs

Redkey
07-15-2002, 01:33 AM
Sure you can get the ball moving with much less pressure/force. However, you'll have to apply that lower force for a much longer period of time. Because the ball has to accelerate quickly you have to hit it pretty hard to get it up to speed before it leaves the barrel.

The theory behind "low pressure" guns is that they hit the ball with less pressure for a longer period of time.

Redkey
07-15-2002, 01:54 AM
AGD

The pressure transducer is just like a load cell that uses a strain gauge bridge to "sense" the load. There are two ways to "zero" the transducer.

You should be able adjust the zero point and gain of your signal conditioning unit to have the output voltage at zero pressure always be the same. This is only possible if you transducer does not have the signal amp build into it... if so, you are stuck with the second option...

The other way to do it is to have labview perform a software zero before data is collected. Basicially this just shifts the entire dataset so that your initial voltage levels correspond to zero pressure.

What is the max capacity or your transducer? If you are trying to measure 100 psi using a 1,000 psi transducer that could also explain some of the low pressure drift you are seeing.

bjjb99
07-15-2002, 02:02 PM
Kweasi,

The even distribution of pressure against the rear surface of the paintball is what keeps it from breaking. A similar effect can be seen if you take a plain old grocery store egg and squeeze it in your hand (take any rings off first); it is surprising how hard you have to squeeze before you make a mess. Smooth, convex structures are quite resistant to compressive forces of this type.

BJJB

SLICEnDICE
07-15-2002, 11:25 PM
I was under the impression from my reasurch that most pressure transducers come from the factory calibrated already.

bjjb99
07-16-2002, 07:49 AM
Yes, most sensors are calibrated at the factory. Some pressure sensors are pre-calibrated for linearity (good dV/dP), leaving the zero offset to be corrected at the time of collection. All sensors are subject to drift with external pressure and temperature. It is good laboratory practice to calibrate your sensitive instrumentation immediately prior to, and immediately following, any data collection. You then know how the instruments were drifting at the time you ran your test, and can correct for that drift during data processing.

BJJB

SLICEnDICE
07-16-2002, 01:52 PM
I see clearly now and understand.

Kweasi
07-20-2002, 06:58 PM
I'm still not clear here....

It's my understanding that the "90psi" and hense the 32lbs force figures are the "threshold" before the ball starts to move. That is to say the ball remains stationary untill the pressure behind it reaches the threshold and overcomes the forces holding the ball still.

But this does not compute- if that were the case than how can a bolt supplying only 8lbs force (ie Matrix) chamber a ball?

-Kweasi

Doc Nickel
07-20-2002, 07:14 PM
Kwayze- keep in mind that, in the millisecond-times we're looking at here, it's also a matter of "that's how much pressure managed to reach the breech before the ball's friction and inertia were overcome and it began to move."

In other words, the chamber- the breech and space behind the ball- was filling faster than the ball could 'react'.

Liken it with a golf ball being hit: The ball actually compresses a little and deforms, THEN starts to accellerate off the tee.

I'm not saying the paintball deforms, I'm saying that the force builds up before the ball begins to move- and the faster the force builds, the higher the apparent pressure before that movement.

So it's not so much that it takes X pressure to get the ball going, its a matter of that's how fast that particular marker, at that pressure, with that valve dwell time and that particular valve and bolt, can pressurize the breech, and to what level, before the ball's intertia is overcome and begins to accellerate.

I'd be interested to see those same transducers in Nova, which is supposed to operate normally at 90 psi- I'd wager the breech pressure before the ball moves is far lower, but the "peak" as depicted above is spread out over a far longer time.

Doc.

sniper1rfa
07-20-2002, 08:48 PM
im with doc, also, there is the sticktion (is that actually a word?) before the ball unseats itself.

Kweasi
07-22-2002, 10:35 AM
AH HA! That clears it up a good deal. Much thanks.

But...
If different valve set-ups do in fact cause different relationships between peak-pressure and ball movement, does that suggest that "low-pressure" markers *really are* gentler on paint?

-Kweasi

sniper1rfa
07-22-2002, 11:00 AM
technically, yes, but you wont notice it, as there is no way in hell that 90 psi will break a ball (or come remotely close) with only a little friction and its own inertia opposing its movement.

athomas
07-22-2002, 09:02 PM
The pressure in the chamber is about 350psi. When the bolt moves forward it releases the pressure behind the ball. When the pressure behind the ball is greater than the friction and stationary inertia forces on the ball, the ball begins to move. The rate of acceleration of the ball is directly proportional to the pressure exerted behind it.

Edit: This explaination is for a mag valve.

So, as air/gas builds up pressure behind the ball, it starts moving. Initially, the rate of pressure increase is greater than the rate of speed increase. Then as the speed increases due to acceleration, the area behind the ball increases faster than the chamber can pressurize it. Even though there is 350psi in the chamber, the pressure never gets above 60psi behind the ball. This is where the pressure graph shows a decline in overall pressure behind the ball. At some reduced pressure, the forces behind the ball will not equal the frictional forces in front and along the sides of the ball. The ball will start slowing down.

This happens very fast. To be able to measure it with a degree of accuracy takes precise instruments. The data is collected electronically and is subject to calibration/offset, but relationship to high and low is always constant. You just have to know where zero is and calibrate the data accordingly.

Kweasi
07-22-2002, 10:36 PM
Offsetting that graph to "zero" it, effectively moves the whole thing up 5 units. That would mean the actual "theshold pressure" is 95psi, would it not?

-Kweasi

athomas
07-23-2002, 05:55 AM
I agree. The corrected data may have to be moved up 5psi. That would mean the actual psi behind the ball is 95psi.

Kweasi
07-23-2002, 01:08 PM
(Musta read my previous post before I edited it and corrected my error- just clarifying).

-Kweasi