Nuts & Volts - April 2005

Project

delay compensation. In that case, all your measurements will be off by a fixed amount. You can subtract this amount manually to find the speed of light, but it’s not very elegant.)

Let’s step through the circuit. First, you must use a Schottky 74S86 (not the 7486 or the 74LS86). The 74S86 has a symmetrical, 10-ns, low-to-high and high-to-low delay. Of course, the other families have different propagation delays. Remember, we want to resolve one ns. The first thing we do is buffer the reference signal. The XOR with one input connected to ground does this. Then

Parts List

F o r

E l e c t r o n i c s

NUTS & VOLTS E v e r y t h i n g

Transmitter

R1 R2 R3 Rx C1 C2, C3 D1 D2 Dx Y1 SW1 U1

10M 1/4 W
1.5K 1/4 W
1K 1/4 W
Part of laser pointer, see text
0.1 µF 25 V
33 pF 25 V
Red LED
1N4001
Key chain laser pointer (see below)
4 MHz crystal
SPST switch
74HC04 hex inverter

Note: There are many sources for these inexpensive lasers. I’m quite sure I got mine from All Electronics a few years ago. They currently have them listed as LP-506 on their website (www. allelectronics.com) for $4.75. Hosfelt Electronics has a similar item, 75-374, for $5.95 at www.hosfelt.com Officemax item 11134783 appears to be similar at $10.99.

Receiver

1K 1/4 W

5-20 pF variable cap

0.1 µF 25V

74S86 Schottky Exclusive OR Do not substitute.

GP1FA551RZ Sharp fiber-optic receiver IC ( 13.2 MHz speed)

Available from Digi-Key, 425-1109-5-ND $2.38 each ( www.digikey.com)

R1, R2

C1, C2

C3

U1

MOD1

Optical Bench

4” x 4” x 1/8” clear plastic (see text)

Credit card/wallet Fresnel lens magnifier 2” x 4

(approximately)

Edmund Scientifics ( www.scientificsonline.com) part 3038456 @ $1.95 each looks identical to the one I used. Officemax part 11035435 ($9.99) might be useable, but it has a handle. A web search brings up many listings.

Misc. 12” x 12” wood base, small wood strips to mount parts, screws, etc.

Beam-splitter Lens

we delay the signal with R1 and C1. This provides about five to 20 ns of delay. Then, the signal is inverted by connecting an input to a logic high. Next, we delay this by five to 20 ns with R2 and C2. Finally, the signal is buffered again with the last section of the XOR chip. The delay is 30 ns for the three XOR sections and 10 to 40 ns for the delay circuits for a total delay of 40 to 70 ns.

There are some subtle points to note. Two resistor-capacitor (RC) delay circuits are used because the XOR output has asymmetrical drive. It sinks much more current than it can source. This means that the capacitors will discharge faster than they charge. If only a single RC delay was used, the rising edge would be delayed more than the falling edge. By using two RC delays — one on an inverted signal and the other on a non-inverted signal — we can control the delay of each edge and can make the delays symmetrical. (Note, if your receiver has a significantly different delay, you may have to adjust the values of the RC circuits. For more delay, increase the resistor and/or the capacitor values, and for less delay, reduce them.)

The other point is a very useful property of the XOR gate. A logic low at one input means the output follows the other input. A logic high at one input will invert the signal at the other input. This means that you can control the inversion (or phase) of a signal with a logical value without changing the propagation delay. If you do a lot of digital design, you will find this little trick valuable.

Unfortunately, there is a problem with the receiver. The actual IC is a small three-pin device buried in a large plastic housing, which is normally mated with a fiber optic cable. It’s very difficult to align the laser light to shine deep into the small hole used for the cable. While that is possible, it’s much easier to remove the IC from the housing. I know because I tried it both ways. Look closely at Photo 2. At the front is a large, black component. That’s the optical receiver. If you look to its immediate left, you will see a small, clear, plastic object. That’s the actual receiver IC that has been removed from the housing. A better picture of the receiver and housing is shown at the bottom right in Photo 1.

I was successful in removing the receiver IC with two different methods. The first, shown in Photo 1, was to very carefully saw the sides of the housing until I reached the cavity that held the actual IC. However, once I did that and was able to examine the inside of the housing, I noticed that two small retention barbs held the IC in place. These barbs are accessible where the IC leads exit the cavity. The next time, I used an X-acto razor knife to cut these barbs away. With a little inward pressure on the face of the IC through the hole (using a toothpick) and gentle pull on the leads, the IC was removed.