[cockerpunk]

data - http://spreadsheets.google.com/pub?key=pym...HdSUAE5ioZ_zbNA

testing rig video - http://www.youtube.c...h?v=wzjRf22Ma-0

test video - http://www.youtube.c...h?v=4HnKpM6KN-c

thanks to mike video - http://www.youtube.c...h?v=KXPbUZUvvLI

discuss!

This post has been edited by cockerpunk: 28 December 2008 - 12:22 PM

[Lord Odin]

Very interesting data. Couple of questions.

1. Do you have specs listed anywhere for the rifled barrels?
2. Do the one piece barrels have the same bore size at the breech as they do at the muzzle? In other words, do they taper along their length and if so, are they the same?
3. The two piece test is rather interesting. There doesn't appear to be a trend amongst the different length barrel fronts. Was there porting on the fronts?
4. What was being tested with the Dye barrels that wasn't covered in the 12" various bores? Is the barrel different?

People have generally recommended 12-14" for a barrel length and its really cool to actually see some numbers to back that up. Looks like they were pretty close.

The 12" front/various back is really interesting. It seems that the paint to barrel match (see top note for paint size) in that instance showed the best efficiency. That's actually opposite of previous bore tests. Any ideas why that may be? Are the freak and CCM kits you guys used in the other tests different from the CP kit?

[brycelarson]

Lord Odin, on Dec 28 2008, 12:23 PM, said:

Very interesting data. Couple of questions.

1. Do you have specs listed anywhere for the rifled barrels?
2. Do the one piece barrels have the same bore size at the breech as they do at the muzzle? In other words, do they taper along their length and if so, are they the same?
3. The two piece test is rather interesting. There doesn't appear to be a trend amongst the different length barrel fronts. Was there porting on the fronts?
4. What was being tested with the Dye barrels that wasn't covered in the 12" various bores? Is the barrel different?

People have generally recommended 12-14" for a barrel length and its really cool to actually see some numbers to back that up. Looks like they were pretty close.

The 12" front/various back is really interesting. It seems that the paint to barrel match (see top note for paint size) in that instance showed the best efficiency. That's actually opposite of previous bore tests. Any ideas why that may be? Are the freak and CCM kits you guys used in the other tests different from the CP kit?

1. Nope, based on the mean velocity we assume they're huge. We tried to measure - but with the rifling it's really hard. Gordon got... I think .689 - but that would most likely be the "peaks" of the rifling.
2. yes - same diameter throughout.
3. yes, check CP's website for a picture - I don't have any. The two piece barrels have two extra rows of porting - otherwise all CP barrels were very similar in porting.
4. Just to throw something else at very similar bore into the test.

As to your last point - we think it's because of how small paint tends to be. Since CP makes a .682 at the smallest we don't feel that we were able to actually underbore all that much.

[Promaster928]

This is great stuff guys. I can't wait for the accuracy part.

Brycelarson, You mention wanting to underbore more then .682. It seems from your results that the .685 was best size bore to the paint you were shooting which was around .684.

So that would mean the best results came from the barrel being just a little overbored? I'm just trying to better understand all this.

[1NOnly]

Excellent job! Can't wait for the accuracy portion of the test.

[Compulsion]

What's with the wicked spiking on those first shots?

In many cases the first shot is 20 or more FPS higher than subsequent shots. Do you have creep in the gun?

This test might benefit from two or three clearing shots before beginning to take measurements. At the very least, remove those outliers from the calculations of Mean and SD.

This post has been edited by Compulsion: 28 December 2008 - 04:17 PM

[Lord Odin]

Compulsion, on Dec 28 2008, 03:14 PM, said:

What's with the wicked spiking on those first shots?

In many cases the first shot is 20 or more FPS higher than subsequent shots. Do you have creep in the gun?

This test might benefit from two or three clearing shots before beginning to take measurements. At the very least, remove those outliers from the calculations of Mean and SD.

That's the beauty of standard deviation. Outliers won't affect the SD as much as it would say mean or range. Picking and choosing which shots to include or exclude isn't a good idea. That can get you into all sorts of problems. Its better to just include all data and analyze it from there. Besides, the bell curve will always have outliers towards the extrema.

That being said, a few clearing shots might be a good precautionary measure. Especially if there are asymptomatic problems with a gun.

[brycelarson]

Sorry, forgot to note - we knew going in that gordon's emag shoots hot on the first shot after sitting for a while. We wanted a sample size of 20 - so sampled 20 then removed high and low - leaving 20 data points per barrel.

Odin - I'm sure that's why you got different means - the first shot would certainly push the data sets up a few fps.

[Compulsion]

Lord Odin, on Dec 28 2008, 04:56 PM, said:

Picking and choosing which shots to include or exclude isn't a good idea.

You're right, of course. It's unscientific to massage your data.

Lord Odin, on Dec 28 2008, 04:56 PM, said:

That being said, a few clearing shots might be a good precautionary measure. Especially if there are asymptomatic problems with a gun.

At least try a few of the tests with a different gun. See if that initial spike persists.

Who knows, the spike is caused by something outside our current model. Maybe there really is something special about the first shot.

The most exciting phrase to hear in science, the one that heralds new discoveries, is not "Eureka!" but rather, 'hmm... that's funny...'
--Isaac Asimov

[brycelarson]

the first shot being hot is standard on gordon's emag - not on any other gun we've tested with. I don't know why - I assume that the valve reg has a little leak allowing the dump chamber to slightly creep up as the gun sits without venting the chamber. one shot and it's back on line - which is why we chose to take 22 samples and dump the high and low.

[cockerpunk]

dropping the top and bottom value in a set is common practice to remove the outliers.

it works perfectly in this case, we recorded 22 values with each, drop the fastest and slowest, and still have a data set of 20 values.

i am also a bit disappointed to see that we couldn't get a very good underbore going. an odd thing to point out is that in our last test the "matched" did the worst, or worse then over boring anyway, and this test shows the opposite.

me and bryce have a theory as to why, we think.

[Lord Odin]

brycelarson, on Dec 28 2008, 04:16 PM, said:

Sorry, forgot to note - we knew going in that gordon's emag shoots hot on the first shot after sitting for a while. We wanted a sample size of 20 - so sampled 20 then removed high and low - leaving 20 data points per barrel.

Odin - I'm sure that's why you got different means - the first shot would certainly push the data sets up a few fps.

Ahh, that would explain the difference in numbers I was seeing. Thanks for clearing that up.

[TechPB-Mike]

awesome job fellas, holy shit CP hooked some brothers up!!!

Travis you're the MAN!!!

[Troy]

brycelarson, on Dec 28 2008, 04:28 PM, said:

the first shot being hot is standard on gordon's emag - not on any other gun we've tested with. I don't know why - I assume that the valve reg has a little leak allowing the dump chamber to slightly creep up as the gun sits without venting the chamber. one shot and it's back on line - which is why we chose to take 22 samples and dump the high and low.

I guess if you really wanted to, you could do an outlier Q test, but dropping one high and one low is an acceptable amount of data massaging.

[chewiestmonkey]

what would be your guys theory as to why the matched paint to bore worked better this test?

This post has been edited by chewiestmonkey: 28 December 2008 - 08:25 PM

[AndreS]

very nice work!!!

[WihGlah]

Sorry - also posted on PBN - and I see Complusion also noticed that the e-mag shoots hot on the first shot

HOWEVER - from a statistical standpoint there is another anomaly:

from the second shot to the 22nd, there is a trend for the velocity to creep up about 20fps.

(for example, the 0.685 one piece 10", shots 2 - 13 are all below the mean, shots 14 - 22 are all above.)

also - 21 shots in a row with an upward trend??

That's not random - and it should be. Something else is going on.

This post has been edited by WihGlah: 29 December 2008 - 03:50 PM

[brycelarson]

automag shoot-up.

[Iram]

Yeah, an automag may not have been the best choice of guns to test with.

How far back (in inches) is the porting on the CP barrels? Do all the barrels have the same amount of porting?

[brycelarson]

Iram, on Dec 29 2008, 03:50 PM, said:

Yeah, an automag may not have been the best choice of guns to test with.

How far back (in inches) is the porting on the CP barrels? Do all the barrels have the same amount of porting?

our rate of fire on the tests was very consistent - so the effects (if there were any) would have been the same on all barrels - therefore as a comparative test - it's still valid.

WAIT A SECOND! in order to drop the high and low I sorted by lowest to highest - this isn't the shooting order.

don't have a barrel in front of me - but about 3" of porting. The porting pattern is identical - except the 2 piece fronts have two extra rows of ports (see pics)

EDIT

I measured:

10" one piece - 10 rows of porting over 3.5"
12", 14", 16" one piece - 12 rows over 4.25"

10" two piece tip - 13 rows over 3.75"
12", 14", 16", 21" - 16 rows over 4.5 inches.

Attached File(s)

•  two_piece.JPG (9.41K)
  Number of downloads: 17
•  one_piece.JPG (5.78K)
  Number of downloads: 14

This post has been edited by brycelarson: 29 December 2008 - 05:13 PM