simple diy ballistic chronograph
-
farcticox1
- Sergeant
- Posts: 1080
- Joined: Sun Mar 31, 2013 12:37 pm
- Has thanked: 69 times
- Been thanked: 96 times
so you shoot your phone and see how far it penetrates
-
farcticox1
- Sergeant
- Posts: 1080
- Joined: Sun Mar 31, 2013 12:37 pm
- Has thanked: 69 times
- Been thanked: 96 times
Well, it appears to work quite well
and it's free, I'm going to give this a try
and it's free, I'm going to give this a try
-
farcticox1
- Sergeant
- Posts: 1080
- Joined: Sun Mar 31, 2013 12:37 pm
- Has thanked: 69 times
- Been thanked: 96 times
Finally getting round to giving this a try, I'm going with the emitter/receiver setup. I have downloaded Softchrono but not installed it yet, it's going on an old laptop I have. Anybody successfully using this have any tips? I want to keep the led pairs as close together as possable and still get good readings. 
-
jimmy101
- Sergeant Major
- Posts: 3199
- Joined: Wed Mar 28, 2007 9:48 am
- Location: Greenwood, Indiana
- Has thanked: 5 times
- Been thanked: 17 times
- Contact:
Here is a bit simpler design;
from http://www.inpharmix.com/jps/Jims_chrono.html
It is just a cutup piece of the same pipe used for the barrel. Obviously this is much easier with plastic than metal.
You really don't need emitters with this design (daylight is MUCH brighter than any emitter) and the chrono isn't a permanent part of the gun. This design is still overkill, the detectors are in the little tubes on the bottom, but it would probably work fine if those tubes were omitted and the detectors were just glued into the holes of the main part.
As far as the detector spacing; you can calculate how far apart they need to be based on the expected velocity range, how accurate you want the measurement to be and how reproducible you think the gun's muzzle velocity is.
To start with, it is silly to try to make a chrono accurate to +/-0.1 fps if the shot to shot variability is 50 fps. Similarly, you probably don't need a chrono that can measure Mach 5.
This is a typical calculus problem on measurement, accuracy and error propagation. But my calculus is too rusty so we’ll just manually crank through the numbers.
Assume your gun(s) muzzle velocities range from 100 fps to 800 fps. 100 fps is pretty slow and 800 fps is pretty good for a homemade device. If you are using the soundcard on your laptop (or phone) as the data logging device it probably has a maximum sample rate of either 40 or 44 KHz. Using the 44 KHz (44000 samples/second) number, that means you can get a data point once every 1/44000 seconds, which is 23 microseconds (23x10^-6). Let’s say we want to go up to 800 FPS plus or minus 10 FPS. If your gates are 1 foot apart at 800 FPS it takes the round (12 inch)*(second/800 feet)*(feet/12 inch) =0.00125 seconds (1.25 milliseconds) to travel that distance. Since your data points are 23 microseconds apart the error in timing is 1.25/0.023 (converted the times to milliseconds), which is 1 part in 54 or about 2%. So the error is about 15 fps (2% of 800 fps) just due to the sample rate of the sound card. At 800 fps that 2% error is probably less than how much shot to shot variability you have so the accuracy is probably adequate.
At a muzzle velocity of 100 fps it takes 0.01 seconds to cross the gates and your error is 10/0.023 (times in milliseconds), which is 1 part in 435, or about 0.25% and 0.25 fps. Your device is much more accurate than it needs to be but that is fine if you also want to be able to do 800 fps with the same setup.
What if you want the device to be shorter? Let’s cut the gate separation down to just 3 inches. Your timing at 800 fps, across 3 inches, is (3 inch)*(second/800 feet)*(feet/12 inch) = 0.31 milliseconds (1/4 of the previous case with 1 foot gate separation). Your sound card accuracy is still 1/44000 seconds so the error is 0.31/0.023 which is one part in 13. At 800 fps the timing error is +/- 62 fps (800/13). That accuracy is probably getting to be more than the shot to shot variability so 3” gates at 800 fps is probably a little too short. At 100 FPS the error would be (3 inch)*(second/100 feet)*(feet/12 inch) = 2.5 milliseconds and the timing error 2.5/0.023 which is one part in 108 or a little less than 1%, less than 1 FPS and more than accurate enough.
So it looks like if you sample at 44KHz and for the velocity range of 100 to 800 fps that a 6” gate separation is probably adequate. If you velocity range is less than say 500 fps than a 6” separation would always be good enough.
If you sound card maximum sample rate is 20 KHz instead of 44 KHz then the gate separation would have to be about twice as long to get the same accuracy.
The other consideration of gate separation is how long the shell is and how you wire the two detectors. For a worst case scenario you might just say the gates must be farther apart than the length of the slug. That way you are sure only one detector is active at a time and you don’t need to worry about the signals mixing. In many cases though, the detector will still work properly if that isn’t the case. It just means that the signals for 11 01 00 10 11 (the pattern of on/off for the two detectors as the slug passes them) must all be different enough to be recorded by the sound card. Alternatively, if your sound card has a stereo input you could wire the two detectors to different audio channels and the detectors could be closer together than the length of the slug but the two signals would always be distinct since they are recorded on different audio channels.
from http://www.inpharmix.com/jps/Jims_chrono.html
It is just a cutup piece of the same pipe used for the barrel. Obviously this is much easier with plastic than metal.
You really don't need emitters with this design (daylight is MUCH brighter than any emitter) and the chrono isn't a permanent part of the gun. This design is still overkill, the detectors are in the little tubes on the bottom, but it would probably work fine if those tubes were omitted and the detectors were just glued into the holes of the main part.
As far as the detector spacing; you can calculate how far apart they need to be based on the expected velocity range, how accurate you want the measurement to be and how reproducible you think the gun's muzzle velocity is.
To start with, it is silly to try to make a chrono accurate to +/-0.1 fps if the shot to shot variability is 50 fps. Similarly, you probably don't need a chrono that can measure Mach 5.
This is a typical calculus problem on measurement, accuracy and error propagation. But my calculus is too rusty so we’ll just manually crank through the numbers.
Assume your gun(s) muzzle velocities range from 100 fps to 800 fps. 100 fps is pretty slow and 800 fps is pretty good for a homemade device. If you are using the soundcard on your laptop (or phone) as the data logging device it probably has a maximum sample rate of either 40 or 44 KHz. Using the 44 KHz (44000 samples/second) number, that means you can get a data point once every 1/44000 seconds, which is 23 microseconds (23x10^-6). Let’s say we want to go up to 800 FPS plus or minus 10 FPS. If your gates are 1 foot apart at 800 FPS it takes the round (12 inch)*(second/800 feet)*(feet/12 inch) =0.00125 seconds (1.25 milliseconds) to travel that distance. Since your data points are 23 microseconds apart the error in timing is 1.25/0.023 (converted the times to milliseconds), which is 1 part in 54 or about 2%. So the error is about 15 fps (2% of 800 fps) just due to the sample rate of the sound card. At 800 fps that 2% error is probably less than how much shot to shot variability you have so the accuracy is probably adequate.
At a muzzle velocity of 100 fps it takes 0.01 seconds to cross the gates and your error is 10/0.023 (times in milliseconds), which is 1 part in 435, or about 0.25% and 0.25 fps. Your device is much more accurate than it needs to be but that is fine if you also want to be able to do 800 fps with the same setup.
What if you want the device to be shorter? Let’s cut the gate separation down to just 3 inches. Your timing at 800 fps, across 3 inches, is (3 inch)*(second/800 feet)*(feet/12 inch) = 0.31 milliseconds (1/4 of the previous case with 1 foot gate separation). Your sound card accuracy is still 1/44000 seconds so the error is 0.31/0.023 which is one part in 13. At 800 fps the timing error is +/- 62 fps (800/13). That accuracy is probably getting to be more than the shot to shot variability so 3” gates at 800 fps is probably a little too short. At 100 FPS the error would be (3 inch)*(second/100 feet)*(feet/12 inch) = 2.5 milliseconds and the timing error 2.5/0.023 which is one part in 108 or a little less than 1%, less than 1 FPS and more than accurate enough.
So it looks like if you sample at 44KHz and for the velocity range of 100 to 800 fps that a 6” gate separation is probably adequate. If you velocity range is less than say 500 fps than a 6” separation would always be good enough.
If you sound card maximum sample rate is 20 KHz instead of 44 KHz then the gate separation would have to be about twice as long to get the same accuracy.
The other consideration of gate separation is how long the shell is and how you wire the two detectors. For a worst case scenario you might just say the gates must be farther apart than the length of the slug. That way you are sure only one detector is active at a time and you don’t need to worry about the signals mixing. In many cases though, the detector will still work properly if that isn’t the case. It just means that the signals for 11 01 00 10 11 (the pattern of on/off for the two detectors as the slug passes them) must all be different enough to be recorded by the sound card. Alternatively, if your sound card has a stereo input you could wire the two detectors to different audio channels and the detectors could be closer together than the length of the slug but the two signals would always be distinct since they are recorded on different audio channels.
-
jimmy101
- Sergeant Major
- Posts: 3199
- Joined: Wed Mar 28, 2007 9:48 am
- Location: Greenwood, Indiana
- Has thanked: 5 times
- Been thanked: 17 times
- Contact:
I left out one more source of error and uncertainty; the accuracy of measuring the distance between the detectors. An advantage of a longer detector is that the uncertainty in that distance has a smaller affect compared to a shorter detector.
-
farcticox1
- Sergeant
- Posts: 1080
- Joined: Sun Mar 31, 2013 12:37 pm
- Has thanked: 69 times
- Been thanked: 96 times
Wow, that's great. I was sort of thinking the same thing myself with the spacing, I have made a jig with 3" & 6" spacings to start, I will try the emitters first as I will be shooting it in my workshop which doesn't have great lighting.
And a junction box to make connecting easier, the 1.5V battery will go in here as well. I am wiring them in parallel with 1k resistors, is that still the best option
I am expecting 300fps upwards, maybe over 400fps, I have several to test so I really don't know what I'm going to see yet
- BFG.jpg (66.38 KiB) Viewed 5480 times
- Bolt.jpg (45.82 KiB) Viewed 5480 times
- Full auto.jpg (54.1 KiB) Viewed 5480 times
- Pistol.jpg (64.8 KiB) Viewed 5480 times
- Semi auto.jpg (47.4 KiB) Viewed 5480 times
- Single shot.jpg (68.61 KiB) Viewed 5480 times
-
jimmy101
- Sergeant Major
- Posts: 3199
- Joined: Wed Mar 28, 2007 9:48 am
- Location: Greenwood, Indiana
- Has thanked: 5 times
- Been thanked: 17 times
- Contact:
That looks like a good setup. You might need to adjust things a bit but your setup will probably work. I went with the open design to avoid having the chrono jig act like an extension of the barrel. The 1K resistor should be about right and overall the chrono setup is pretty insensitive to details like the resistor.
Using emitters is a good idea if you are working under lights. The sun is constant but lights, both incandescent and florescent, fluctuate in intensity at 120 Hz (twice your mains frequency). Indeed that can be used as a rough test of your setup, you should be able to use a sound recorder or a software based oscilloscope on the PC to record your shop light's 120 Hz signal. You might download http://www.zelscope.com/. That will work with any audio range AC signal (the sound card can't measure DC) and a mic or a phototransistor is a good test system to try.
Using emitters is a good idea if you are working under lights. The sun is constant but lights, both incandescent and florescent, fluctuate in intensity at 120 Hz (twice your mains frequency). Indeed that can be used as a rough test of your setup, you should be able to use a sound recorder or a software based oscilloscope on the PC to record your shop light's 120 Hz signal. You might download http://www.zelscope.com/. That will work with any audio range AC signal (the sound card can't measure DC) and a mic or a phototransistor is a good test system to try.
-
farcticox1
- Sergeant
- Posts: 1080
- Joined: Sun Mar 31, 2013 12:37 pm
- Has thanked: 69 times
- Been thanked: 96 times
Got the 5mm Infrared Emitter/Receiver pairs, which is typically which ?
is this correct?
- s-l400.jpg (6.54 KiB) Viewed 5317 times
- main-qimg-c6263bd614a7f5af3611fe8a7ea2a83f-c.jpg (7.61 KiB) Viewed 5317 times