4
\$\begingroup\$

Me and my friends are trying to build a laserharp driven by arduino as a learning exercise. Currently we are using a phototransistor to interrupt the laser light and play music.

However the activation area of the phototransistor seems to be very small and requires it to be placed quite precisely in the path of the beam.

I am unable to look for something like a large activation area phototransistors... if they exist?
Or is it possible to use some kind of lensing arrangment (which should be cheap) to focus light from a wider area to the transistor.

As another possibility, I would also like to look into solar cells. Since there area is quite adequate they are perfect for my application. However I don't know if their response time would be fast enough (close to phototransistor) and secondly from where do I get a single solar cell?
Since the laser is just a point on cell, using a solar panel is quite inadequate for my purpose.


More details: Sorry for not having them before.

I have one 8mW green laser (power is not precise, I hacked it out of a home projector.. but is getting my job done)

We incident the laser on a mirror, that is being rotated by a stepper motor. Currently 4 steps clockwise, and then 4 anticlockwise with a delay of around 4 ms on each step. This produces 4 beams of laser from a single beam.

As a pick I am currently using a phototransistor with Vcc. on the collector, and an emitter grounded through the resistance with a floating base.

When the phototransistor is used to block one of those lasers, if the laser happens to fall on the activation area of the phototransistor then it gets biased and I can read that. This combined with the step number of the stepper motor lets me determine which beam is being interrupted(realize, there is only 1 beam at a time)

The problem is that since the activation area of the phototransistor is quite small, its inconvenient to get it in line of laser. This creates a non-interactive user experience. I need to remedy this.

\$\endgroup\$
  • 1
    \$\begingroup\$ I'll be honest I don't understand the problem - maybe I'm being stupid and maybe a little diagram would help explain the problem to me? I've seen large activation photodiodes - maybe these'll do? Why are you unable to look for something like this? \$\endgroup\$ – Andy aka Apr 25 '13 at 20:49
  • \$\begingroup\$ I have added more details, not sure what kind of diagram do you need. Circuit diagram? And no I have not been able to find any cut photodiodes, can you direct me to a link? My transistor looks like this: essentialscrap.com/toycar/electronics-phototransistor.jpg \$\endgroup\$ – Dhruv Kapur Apr 26 '13 at 2:04
  • \$\begingroup\$ So you are holding the phototransistor by hand, like a guitar pic? And waving it where the laser beams are? \$\endgroup\$ – Bobbi Bennett Apr 26 '13 at 2:35
  • \$\begingroup\$ Yep... exactly. Currently its in a small prototyping bread board hooked to the main breadboard via really long wires. \$\endgroup\$ – Dhruv Kapur Apr 26 '13 at 8:36
  • \$\begingroup\$ All this is starting to sound like a barcode scanner with a voltage to frequency converter on the scan voltage and a voltage controled gain on an amplifier fed from the scan generated frequency and gain controlled by the detector signal. Kind of like air-piano with real sound. A smart phone and camera might have an APP to mimic that (or someone will rush to make one now). \$\endgroup\$ – KalleMP Oct 31 '15 at 11:22
2
\$\begingroup\$

For a somewhat larger photodetector, use a 10 mm, through-hole, diffused package (not water-clear), green LED as your sensor.

Something like this one:

Diffused 10mm green LED

The diffusion will provide better sensing when the laser is off-axis to the LED, compared to a water-clear one.

The reason to use a green LED is, LEDs when used as photodetectors are most sensitive to their predominant emission wavelenth, or actually slightly shorter wavelengths.

See this link for LED as a light sensor, which touches upon the above.

\$\endgroup\$
0
\$\begingroup\$

As you say, the laser is a small point, so will only shine on a small fraction of a solar cell at a time. It seems to me you have to get your mechanical issues fixed anyway so that the beams are where the player expects them to be. When you get this right, a detector the size of T1-3/4 LED should be big enough. You wouldn't want the laser misaligned by more than the radius of one of these anyway. If you can tolerate a little more slop, add a diffuser over the sensor.

\$\endgroup\$
  • \$\begingroup\$ The mechanical issues are pretty much fixed. Laser's are at a point where the player expects them to be. And can you please give me a link to a detector you think might work for me? And yes, I'll try to add a diffuser today, but first will have to search for some ideal diffusers. Last time I used a translucent polythene but it didn't work well. \$\endgroup\$ – Dhruv Kapur Apr 26 '13 at 2:12
  • \$\begingroup\$ @Dhruv: If the mechanical issues are fixed, then you should be able to hit the targets you have. In any case, a thin piece of paper could be a good enough diffuser. \$\endgroup\$ – Olin Lathrop Apr 26 '13 at 12:15
0
\$\begingroup\$

It seems to me that the laser beam is at one of four angles - lets say, for simplicity: -

0º, 10º, 20º and 30º. Let's also say it has a beam divergence of 1º.

Now the trouble with this is that there are gaps in between where nothing happens so I would make the stepper motor fill these gaps with many more sub-divisions. So from -4º to +4º around each note position, the note would be pure. In the 2º areas between notes you can mute the note.

I don't think anyone should believe that playing a musical instrument is ever going to be easy but if you think of a guitar, you can hold the note with a finger behind one of the frets and, providing you are either directly behind the fret or nearly all the way back to the fret that would dictate the next semitone lower, it works!!

Same with a piano - you don't have to hit the key exactly dead-centre - you can be a little to the left or the right.

If you develop the laser positioning in this way you'll have a bunch of angles that correspond to exactly the same note and maybe you can develop vibrato effects within the single note by "wobbling" the photo-transistor (on the hand) to create this.

Have you ever looked up on google what a theramin is - this uses an EM wave to detect hand positions and produce notes with both a decent frequency range and decent amplitude rangle. One hand controls frequency (pitch) whilst the other controls amplitude.

In short, more laser positions are needed with a little more sophistication on interpreting the positions. I'd like to hear a demo when you have it working, seriously!

\$\endgroup\$
0
\$\begingroup\$

I have used fragments of a broken solar cell, about 1cm square, for a laser pickup. It was good up to mid-range audio, though I had no reason to test it exactly. Surplus solar cells are quite popular for making homebrew panels, so they should not be too hard to find.

You would have to combine the solar cell with a transistor to duplicate what your photo-transistor was doing, of course.

However! .. I just got the 8mW laser comment. That is too much to be flashing around off of hand held reflective bits of silicon, so skip the solar cell fragment. Don't wear rings when you play, either!

On the other hand, if you just de-focus your laser a bit, so that it spreads out as wide as the uncertainty of your pickup placement, that would work as well. With a laser diode, that would involve just pushing (or unscrewing) the collimating lens in (or out) a bit.

Finally, you could put a diffuser in front of the phototransistor, like this: enter image description here

I do not know the QE of your detector, lets guess it is 0.44 A/W * (Q=200) = 88 A/W, and the aperture of the transistor, say 3mm. With the card 100mm away, and 8mW scattered into 4 pi steradians, that gives (9e-4/4pi)*8mW = 0.6uW or 4.5uAmp signal.

Turns out the phototransistor shown is probably this one. BPW14N At 1 mA per mW/cm^2, to get the same 4.5uA current, you need 4.5uW/cm^2. That is 8mW scattered into a sphere of radius 12cm. So my guess was pretty close.

That will give you 0.45V with a 100K resistor. Close enough to consider setting it up and trying. I think 100mm is too far away, try 20mm or so.

\$\endgroup\$
  • \$\begingroup\$ Are there any creditable papers/research work that test solar cells for such applications? or can you tell me more about how you did it? Plus I can't de-focus from the source, although at sensor area it might make more sense. Please see the added details of how it works. And finally where can I buy such cells online? Thanks :) \$\endgroup\$ – Dhruv Kapur Apr 26 '13 at 2:07
  • \$\begingroup\$ Sorry, the performance wasn't recorded. It was using a .5 mW HeNe laser with a bit of the beam reflected back into it from the bit of solar cell, which was also connected straight into a microphone jack. It produced really nifty environmentally responsive heterodyne sounds. Since you do not have solar cells in your kit, don't use 'em! 8mW is a pretty strong laser, but hacking it to get a wider beam as I said by moving it's lens a bit seems your best bet. Next best is to put a diffusor -in front of- the detector. Say, a ping pong ball slipped over the transistor. \$\endgroup\$ – Bobbi Bennett Apr 26 '13 at 2:51
  • \$\begingroup\$ type in 'broken solar cells' in google. \$\endgroup\$ – Bobbi Bennett Apr 26 '13 at 3:02
  • \$\begingroup\$ I guess I'll go with defusing the sensor, which one would be the most ideal diffuser. And how much losses can I expect... like currently, the laser is the only thing that biases the transistor. I am afraid that if I diffuse the beam poorly then it might not have enough power left and I might need to DC bias using the 3rd terminal and want to avoid that. \$\endgroup\$ – Dhruv Kapur Apr 26 '13 at 8:39
  • 1
    \$\begingroup\$ @DhruvKapur : A bit of scotch tape makes a very nice forward scatter diffuser. \$\endgroup\$ – Bobbi Bennett Apr 26 '13 at 16:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.