Optics questions for building a chronograph for shooting

[Etesla]

Howdy folks,

I'm trying to make a chronograph that measures the speed of a projectile like this one:

https://www.amazon.com/Caldwell-720001-Ballistic-Precision-Chronograph/dp/B00HTN5DTE

I want mine to work with anything that shoots something with a velocity between maybe 200 and 4000 feet per second with 1 fps precision. I want the physical 'window' I have to shoot my projectile through in order to get a reading to be maybe 5"x5" at the smallest.

Overview of current strategy:
Right now, I plan on using a photodiode, probably in 'photoconductive' mode, along with a transimpedance amplifier as my main sensor. The sensor would detect changes in light caused by the projectile.
There will of course be two of these sensor things a fixed distance apart. I'll call each of the two sensor things a 'sensor plane'. When a projectile crosses the first sensor plane you start a timer, and when it crosses the second sensor plane you stop the timer. You measure the time between when each sensor plane is triggered by the projectile, divide the fixed known distance by that time, and get your speed. Easy.

I understand how to design and implement the electronics part of the project, I've done photodiodes reacting to fast changes in light before in other systems. However, I'm unsure of how to do the 'light' related stuff in terms of placement of the sensor, how many sensors to use per 'sensor plane', and where or if to include a light source. Remember, I want a reliable 5x5" ish target area through which I shoot to get the thing to work. I also want to guarantee that the sensor only senses a projectile starting the moment the projectile crosses the 'sensor plane'. In other words I want the 'sensor plane' to actually be a plane, and not an upwards facing cone or something weird and unpredictable like that.

Specific Questions:
As far as I can tell there are two general optical strategies. Either you detect a decrease in light due to the projectile casting a shadow over the sensor, or you detect an increase in light due to the projectile flashing light back at the sensor. I see pros and con's with each approach, and I need someone with more experience to tell me which approach is more normal / easy to implement / what the commercial units do.

Other questions:
For each 'sensor plane':

If I go with detecting a shadow:
Do I just use the suns light or do I make my own LED array?
Do I bury my sensor in a hole/slit in the body, or do I put it out in the open? How do I make sure the thing is detecting a projectile passing the plane, and not a projectile that has 'almost' reached the ideal sensor plane?
Do I need multiple sensors facing multiple LED's, or just one sensor?
Does it work more like a break beam sensor, or like a high speed high gain light intensity sensor?


If I go with detecting a glint:
Do I need an LED shining in the same parallel direction as the sensor?
What LED / photodiode wavelength can I use so that sunlight doesn't interfere with this strategy too much?
Does this really work? It seems like the signal from a 'glint' is probably pretty weak and could be washed out by the sun?
Do I need to have my thing in a dark enclosure for this strategy to work?
Should my sensor and possible LED source point towards the ground or towards the sky?

Thanks in advance! I'm really looking for someone who owns a chronograph and who can tell me how it was constructed in detail regarding the placement and quantity of optical components, or someone who has built one of these successfully before and has good evidence that it was accurate. Quick sketches would be great if you have an idea. Thanks in advance!

[Marco]

Just brainstorming ... how about using two wide angle infrared LEDs in the borrom corners of the window, with narrowband IR photodiodes next to them, a strip of retroreflector tape in between and a mirror strip on top?

PS. to see how existing systems do it, just read the patents ... unless you want to sell it, then don't.

[pwlps]

I didn't try to study all your experimental constraints but if it can give you an inpiration here is a simple technique I used in a project requiring a fast tracking of a light source motion.  There were two photodiodes side by side with a small screen between them (the diodes were actually mounted on a a rotor of a stepper motor), the diodes were connected back to back with a TIA detecting the differential current. The angular resolution of this setup was of course many times better than anything that could be done with a single diode detection. Maybe in a similar way you might use an array of such photodiode pairs to detect the target position.

[cncjerry]

I built a chrono and it had an overhead light bridge in an octagonal ring.  It worked ok using IR diodes and detectors for arrows and i could occasionally get it to trigger with a pellet rifle but never with my  300 Win Mag.  There was a circuit someplace that i followed.  If I remember right, you tapped off the A/C pulse and used an arduino to calculate the speed between pulses.  By the time i was done with the construction of the three rings (averaged them and helped with reducing missed projectiles) I had a fair amount of money ($70, compared to commercial products) and time in it, just wasn't consistent enough and would only get my .300 now and then.  I found one that didn't cost all that much more and works perfectly with a PC interface, etc.  Then my son shot the arm, still don't know how he did it, everyone laughed if course.  I repaired it and it still works fine.  Amazingly low cost.  If you want the make, send me a PM and I'll dig it out.

Jerry

[cncjerry]

One other thing i remembered, look up coil chrono, for some reason that comes to mind.  That used a coil of magnet wire for the detectors.  That worked pretty well but the loop of wire wasn't that large.

[mag_therm]

Looks like a bundle of problems, as you mention, doing it optically out-doors.
I was a bit involved with measurement of military projectiles acoustically to give velocity and coordinates.
That was using the trailing cone of sound wave front of the supersonic projectile.
Accurate enough for the requirements it also worked for sub sonic projectiles from tank, machine gun etc.

At that time there was consideration of detecting the triboelectric charge of the projectile passing over capacitive sensors,
 but I don't know if that went anywhere.

[Wimberleytech]

I know you did not ask for this, but I will post it.  This is the schematic I reverse engineered of a Pact Model 1 chronograph.

[tautech]

IMO your 1fps accuracy is optimistic.
Had a Shooting Chrony for some years and at one time we placed 2 behind one another and they agreed within ~5fps but no better.



While they may not be seen as a metrology grade instrument they are fine for the bulk of ballistic requirements when cartridge ES fps variability is rarely in single digits due to a whole # of variables in manufacture or even the best handloads that in most cases far surpass any manufactured load.
The Chrony uses photo diodes or transistors with ambient light diffused via the translucent sky screens although they can be used without the sky screens on white sky overcast days. Indoor illumination accessories were available for these and they mounted where the translucent screens would be.

Before the days of semiconductors velocity measurements were done by projectiles breaking fine wires set apart at a precise distance.

[cncjerry]

I bought a Competition Electronics version of the one posted with PC interface for $99USD.  It was relatively accurate and repeatable.  I chrono'd every shot for years and it worked great.  It had a little piece of code that would insert directly into and Excel sheet while open or maybe I wrote it, can't remember.  My son hit the right forward arm and it folded over the bullet and then hit the one in the back and folded that one over too.  They were stuck together.  Pretty amazing.

I would carry a piece of paper with me and during the range breaks I would have the paper clipped to the back arms with a laser bore site installed in the weapon and aimed at the target.  I would then walk it out a little ways and align the laser dot on the center of the paper.

By the way, just remembered, there is a chrono now that fits on the end of a rifle.  I dont know how accurate they are as the projectile is only measured over a couple inches if that.  they might be just for air rifles or such.

Depending one what you are trying to measure, there's another way using audio.  You record from a central spot the sound of the weapon firing and the projectile hitting the target (or the hitting the microphone, ha!).  You know the distance from muzzle to target and the time from the recording.  This has its limits, of course.  You get more accuracy by using the speed of sound as well.

jerry

[rstofer]

Looks like a bundle of problems, as you mention, doing it optically out-doors.

Even for the big guys.  Given a sensor at the bottom of the trapazoidal shaped window, there is often a sky screen over the top to smooth out the background light.  Then there is time of day - more specifically angle of the sun.  Then there is the time of year - altitude of the sun.  Sometimes it all works out.

The best solution is to put the sensor/window inside a large box with an internal light source.  Exclude extraneous light.

I use this chrono for pistol:
https://www.midwayusa.com/product/101506602?pid=773378

And this for rifle:
https://magnetospeed.com/pages/how-it-works

The advantage of the magnetospeed is that I don't tie up the entire range for an extended 'cease fire' while I try to get the chrono aligned with the bench in 3 dimensions.  No fair firing into the ground, the screen needs to be high enough to shoot into the backstop.

I also have a couple of Ohler chronos.  Their latest incantation has 3 sensors:
https://oehler-research.com/chronograph/

If I were going to build a chrono, I would just buy the sensors:
https://www.midwayusa.com/product/1015063648?pid=183896

And the skyscreen:
https://www.midwayusa.com/product/1015066255?pid=878318

Or just buy the whole system:
https://www.midwayusa.com/product/101506602?pid=773378

Oh, wait!  I did buy the whole system...

I don't prefer chronos that have the electronics downrange.

[rstofer]

Over 2 foot sensor spacing, the difference in time for 3000 fps and 3001 fps is 222 ns.  I don't know how you plan to do it inside an Arduino but it should be pretty easy.  200+ ns is a lifetime in electronics terms.

[tautech]

Sky screens.
When I was given a Chrony it had already had a sky screen leg/wire shot out however our traditional #8 gauge fencing wire is a perfect substitute  however the translucent plastic screens were missing.
With a little previous knowledge of watching guys use these things at my range I deduced correctly that all that was needed was a white background IF there was already sufficient ambient light.

The substitution solution was the white plastic road edge markers for the photocells to have as their background and they have worked sufficiently well over the years to not need anything better. However if just appropriate length sections of these some 90mm wide plastic markers are used they can act as a sail should there be any wind so I ran them over the saw bench and split them in half.
YMMV
 

[Wimberleytech]

I love posts from fellow gun owners!!  Cool.

[Etesla]

Alright, it sounds like the usual method is to detect the projectiles shadow, and that the suns light is the only light source I should use (no LED's). I will use this method because it seems fairly simple.

I'm now curious about two things.
1: How does the diffuser work / how can I replicate it? Is it translucent? is it opaque? Could I just put a white piece of paper cut out in the same shape in the same place, or is there some special stuff going on that I don't understand with the diffuser's shape/ material?

2: The sensor that lives under the diffusers. Is there physically only one photodiode/ other sensing element, or are there a couple of them side by side? How far below the surface does this sensor / array of sensors sit? Is there a screen that has an optically important function between the sensor and the air? Is the cavity the sensor sits in 1" deep, 2" deep, 0" deep? Is the shape of the cavity the sensor lives in special, or is it just a hole, or a slit, or what?

I would love to hear from cncjerry, rstofer, and tautech about these specific design features.

[tautech]

For the Chrony the sensors are in the small cutout front and rear of the foldout box.



The skyscreens are ~400mm above them. Opaque is probably a better description of them than translucent however as I've pointed out just a white background with sufficient ambient light is sufficient for reliable projectile detection or even just a white sky cloudy day without any skyscreen needed.

[cncjerry]

I don't know, but for $99 I just can't see you making one that works better than the CE or others.  But heck, i've spent 10x that on projects so I had what I wanted.

The diffuser is easy to make, heck you can make it out of strips of milk jugs.  But though the CE works that way, I'm not sure if you wouldn't be better off with I/R diodes on the top screen.  After that, if I remember, you use a capacitor to tap off the pulse, amplify it, square it and use it to interrupt the processor.  The one I built had three vanes so not only did you get the initial velocity, you also knew that other parameter that escapes me that has to do with how fast the projectile is slowing down as it is measured through the 2nd trap.

The first test in building it is to get the pulses to show up reliably on your scope.  You have to make sure the sensors aren't saturated.  I didn't do it at the time, but if you put sensor strings right next to each other you could probably factor out the noise, don't know.

One of the guys mentioned about setup time.  Dont underestimate the importance of fast setup at the range especially a public range.  That/s why I used the borescope and paper.  I also just checked my borescope and glad I did as the batteries were starting to leak, so thanks!

[Etesla]

For the Chrony the sensors are in the small cutout front and rear of the foldout box.

Is the sensor literally in open space with a roughly 1"x.25" slit in the enclosure above it? If so, how far down into those slits are the sensors in inches? Also can you get a good view of how many physical photodiodes / phototransistors are used in each sensor?

I'm doing this project for fun to fill some time during the next two weeks. I want it to be useable, and if It's not I want to understand why.

I'll probably get some parts tomorrow to start doing testing on my versions of the sensor subsystem, but I want to learn as much as I can from the pro's before I really get into it.

[tautech]

For the Chrony the sensors are in the small cutout front and rear of the foldout box.

Is the sensor literally in open space with a roughly 1"x.25" slit in the enclosure above it? If so, how far down into those slits are the sensors in inches? Also can you get a good view of how many physical photodiodes / phototransistors are used in each sensor?

Tricky to get good pics but you should get some idea from those attached.
There only appears to be a single sensor in each unit however it's behind a lense that's recessed ~1/2" into the molded housing. Top/outside of lense is gently curved and what can be seen of the actual sensor through the lense looks like it's a clear diode.
If I get some more time later I can unscrew the sensor housing and pop it open for some better detail pics of it all.

[Etesla]

Those are some very helpful pictures. Thank you so much for taking the time to do that! It's people like you that make this forum great

[tautech]

More Chrony info
Sensor = BPY6243
Support IC = TL064CN

In attached pic note there are 2 lenses and the bottom of the concave dish in the one closest to sensor is ~2mm from the bulb of the sensor. Top face of that lense is flat.
Top lense is convex on the external (top) side and flat on the bottom face. Spacing between them is a shade under 17mm. Both are of some gin clear acrylic.

[StillTrying]

Is the sensor literally in open space with a roughly 1"x.25" slit in the enclosure above it? If so, how far down into those slits are the sensors in inches? Also can you get a good view of how many physical photodiodes / phototransistors are used in each sensor?

I've always assumed that there's a lens on the photo diode(s) that turns the beam into a narrow strip, similar to the lens that turns a laser dot into a thin line. But I've never seen one of the sensors.