Projectile chronograph

[umka]

Hello everyone

I am looking at building a projectile chronograph, actually to measure arrow speeds for archery, and was wondering if anyone else had done it or similar.

My biggest concern is the sensing circuit I don’t know what to use, any ideas Please help as I don’t know how to accurately do this for such a small item that is so inaccurate.

For the control I guess I can use the CCP example provided by CCS with the 2 interrupts being triggered on a PIC16F877a to give the time period over a know distance.

[Guest]

I think your only chance is Doppler Radar. There are low cost products on the market to do this. Otherwise you will need a rather large array of infrared sensors and at least one emitter focused with some rather strange optics. Perhaps if you can shoot the arrow through a space of 2 feet or so the sensor array would not be too big. But the circuitry to sense all these receivers seems like it would cost more than a doppler approach. Mattel makes a very low cost Doppler Radar toy that can be hacked. Perhaps start with this?

HTH - Steve H.

[SherpaDoug]

I have always wondered how bullet chronographs work. They seem to have two triangular hoops spaced apart that the bullet goes through. I don't know how they sense and have never seen one firsthand. If the bullet was magnetic it might be an inductive sensor, but not for lead....
_________________
The search for better is endless. Instead simply find very good and get the job done.

[dan king]

I too have wondered how they work. I borrow one periodically for our paintball events and am always impressed how repeatable and reliable it is.

Anyhow, the sensor looks like a cmos array and uses outside light to illuminate an opaque shield that is mounted above the sensor. The arrangement makes an upside down triangular shape with the backlight making the top.

Other than that, I'm not sure since I can't take it apart since it's borrowed. Costs approx. $100 US.

Dan

[newguy]

Ever watch Mythbusters on the Discovery channel? They often use a bullet chronograph for all their gun/arrow/projectile myths. I've never taken one apart, but based on what I've seen I'd say it's optical in nature. The two triangular-ish "gates" have a semicircular white top section that is obviously used to reflect light from their apex (if that's the right word). I think you'd need a linear optical array to detect the passing of a bullet or arrow, something like this one: http://www.melexis.com/Sensor_ICs_Infrared_and_Optical/Optical_sensing/MLX90255BA_387.aspx

The Sharp IR distance sensors use a similar linear optical sensor arrangement to sense distance, but I don't think you'd be able to hack one to extract the sensor without completely destroying it.

[SherpaDoug]

That MLX90255BA chip is not going to time a bullet. Its max clock speed is 800k pixels/second with 132 pixels per line for a net time resolution of 165us/sample. A rifle bullet at 5000 ft/sec would not spend more than 1 or 2 samples between detectors. For reasonable resolution you need a sensor at least 100x faster.

Could it be a single photo transistor with a slit aperture to view the whole target above it? It would require good level detection to see a 22 cal. bullet flash by, but it might be last enough.
_________________
The search for better is endless. Instead simply find very good and get the job done.

[Ttelmah]

There are lots of different bullet detection systems used on chronographs. Funnily enough one of the 'best', is the simple sheet of paper, with aluminium foil on each side. You simply detect when a contact occurs between the foil faces. this works better for a lot of ammunition styles, than the more common 'modern' system, which looks for rapid changes in the IR reflection. Basically the units have an IR transmitter, and receiver 'side by side'. The transmitter has optics to give a fan shape beam, that is narrow in the other direction. The receiver has a narrow band optical filter in front, making it relatively insensitive to other light sources. The output from the detector, is fed to a differentiator circuit with a high gain amplifier. It gives massive output swings for fast changes in the reflection, but nothing for slow changes. You detect on the rising edge of the output from this. The only 'caveat', is you have to be far enough from the muzzle, to not detect the smoke as a false trigger.

Best Wishes

[newguy]

The slit idea was what I was getting at. Position the array to scan or view just the white target of one of the gates. You only need one "hit" to determine that a bullet is in view. The second gate works the same way. Since the distance between the gates is known, and the times that a bullet traversed the first and second gates is measured, the speed of the bullet can be determined. You'd need a slit/linear detector to find the bullet instead of a standard beam because the bullet/arrow can be anywhere; you can't count on it to break a narrow beam only.

Slit aperture with a phototransistor might not work. A bullet would block such a small amount of light that the total received photocurrent would exhibit a very shallow dip. That small dip would be extremely difficult to detect.

[SherpaDoug]

I think Ttelmah has answered how they work. The key is the differentiator (analog) and keeping system noise low.

The linear array chip takes too long from the time the photons hit the chip to when the signal is clocked out the pin. You can't read any given cell in the array at a high enough rate. It might work for arrows, but not bullets.
_________________
The search for better is endless. Instead simply find very good and get the job done.

[umka]

Well I'm getting a lot more interest in this topic than I thought I would so thanks.

I too think Ttelmah is right, it is an optical sensing circuit of sorts.

So does the IR fan reflect off the skirt of the triangle above the emitter and sensor? It may be best to just buy the sensors designed for a chronograph already made with the optics as they are not that expensive.

[Ttelmah]

The top screens, usually have two purposes. They act as 'sky screens', to help block unwanted light, and provide a baseline signal. This raises the level of IR being detected by the sensors, so that the current through them can be adjusted to a level nicely in the centre of the available output swing range (this is done either by a processor, or an integrator and feedback). then there is normally a 'dip' in the signal as the bullet passes (though in some cases with nickel jacketed rounds you instead get a momentary increase, which causes problems on some systems - careful design needed...). The key is the fast response in the differentiator.
I built a number of different systems doing this some years ago, some dealing with projectiles much faster than any you will get from a 'propellant' based system...

Best Wishes

[umka]

I am ony rally wanting it for archery to test arrows so absolute max is 400fps but normally about 200-320 fps range is expected

[SherpaDoug]

If I were to build a simple one for my own use I might use paper screens connected to microphone elements. The tin foil clad paper Ttelmah mentioned sounds good too.
_________________
The search for better is endless. Instead simply find very good and get the job done.

[Ttelmah]

I'd not be at all surprised, if for an arrow, you couldn't 'get away' with simply a photo detector, with a slit mask spaced a little above it, and a differentiator. It'll tend to give a number of false triggers (when birds go over, or when clouds pass etc.), but would be cheap, and very easy to make/try.

Best Wishes

[libor]