People have always been fascinated by the fury of the heavens. Electronics prodigy Richard Hodgkinson created a lightning distance timer so he would no longer have to manually calculate the approach or retreat of a thunderstorm. He recycled a 70 KHz crystal from an old device for the oscillator. "Let there be light," he declared as two HP 45MGC670 surface mount LEDs were attached to allow him to see measurements in the dark. And his project wouldn't be complete without two 1.5V AA cells which are the heart of his timer. Now his creation is alive, all he needs is a wicked thunderstorm.
Lightning distance timer
Imagine you're right in the middle of a thunderstorm; the lights have gone out, and there's nothing better to do than time the number of seconds between a lightning strike and the thunderclap in an effort to see whether the storm is approaching or receding.
By the time you've divided your answer and found out the distance, the chances are that you'll have missed the next and probably far more spectacular lightning strike.
Why do you have to perform the division? Why not modify a stopwatch to do this as well as measure the time delay? - This is the exact theory behind this device. Fortunately, it is surprisingly easy to modify a stopwatch to perform the necessary division to convert the time into distance - all that is necessary is to change the crystal by a suitable fraction - and the stopwatch now automatically reads in suitable units.
The harder part is finding a suitable crystal. Just about all electronic quartz clocks today use a 32.768Khz crystal as the timebase oscillator, dividing this frequency down in order to achieve all the necessary frequency clock signals to allow the passage of time to be quantified.
With a little maths, we can find that, taking the speed of sound to be 760 mph at sea level, that a suitable crystal to produce a reading where seconds now equals miles should be around 6.9177Khz, or 11.1418Khz for a reading in kilometres (See the bottom of the circuit diagram for the calculations used)..
Regrettably, finding crystals with such low frequencies proved impossible through all the normal electronics retailers, and thus a little lateral thinking needed to be employed to find a suitable crystal.In the case of this specific project, a 70Khz crystal (Marked ELSIE A 70.00 KCS 1318 D), encased in a glass envelope with a B7G valve base, was found in an old box of parts that probably originated from an electronic organ of some description.
The 70Khz crystal lies extremely close to the 6.9177Khz required for readout in miles, except that it differs by a factor of ten.
As a result, miles are read out on the x10 seconds figure, with tenths of miles on the corresponding seconds digit of the stopwatch display. Within the limits of convenient use, the readout of the stopwatch is now limited to displaying a maximum of 6 miles due to rollover onto the minutes digit - However, 6 miles equates to a time delay of around 28.5 seconds; a delay that I personally have never experienced between lightning strike and the arrival of thunder. The calculated error between the required frequency (x10) and the crystal frequency used is also in the order of 20 yards per mile - Hardly significant.
The original 32.768Khz crystal could have been made to be selectable with the new 70Khz crystal in order to preserve the original function of the stopwatch, but in this design case space proved to be so well utilised that no space was left to fit such a switch. Any other versions of this project could quite easily use a variety of different crystal frequencies for the modification, providing that they relate in some way to one of the two required frequencies.
A further issue lies with the design of the stopwatch and its new role. Clearly, anyone who wishes to build such a device will probably make use of the cheapest possible stopwatch that they can obtain. In this lies the problem that the majority of stopwatches don't feature any kind of a backlight - And with the lights gone, or at least turned off, reading the display could prove difficult.
In order to achieve a kind of backlight for the stopwatch used, without modifying or altering the LCD display or the associated elastomer connection strips, two HP 45MGC670 surface mount LEDs (Mounted on stand-off wires) were arranged on a extremely small piece of stripboard to shine directly into the side of the LCD glass. With a little adjustment, it proved possible to create a backlight (Perhaps it should be deemed a side-light?) that provides sufficient illumination to make the characters just about visible on the display.
One comment about the contacts I found in the stopwatch: They're not solderable. It took quite some time and effort to file away just enough of the plastic LCD support to allow wires to be (carefully) soldered directly to the stopwatch PCB. On that note, the front of the stopwatch was actually cut out and mounted behind the front of the new case for practical ease.
Now that LEDs feature in the design, another problem arises. LEDs draw far more current than the stopwatch unit and they require a greater forward voltage that the 1.5v button cell can provide. As a result, this necessitates the use of conventional cells, namely two 1.5V AA cells, the use of AA over AAA cells being determined by the lack of any spare twin AAA cell holders at the time of construction.
The LEDs also require some means to turn them on and off, and a substantially larger case than the original stopwatch case to house both the larger cells and the oversized crystal.
Following on, the next logical question was "Why don't I make it so that the whole thing is turned on and off via the same switch?" This requires a dual pole switch in order to control the 1.5V feed to the stopwatch (Derived from one 1.5v cell in the 2-cell series chain) as well as the 3V feed to the LED's. Another issue is that most stopwatches default to the time mode rather than the stopwatch mode when power is first applied.
With insufficient space to mount a mode button on the new case, a small circuit consisting of little more than two resistors, a electrolytic capacitor, and a few germanium protection diodes (Just to make sure that any reverse voltages stored on the capacitor at power-off can't even get close to +/- 0.6v to damage the stopwatch) provides a positive 1.5v pulse at power on of around 200ms length, causing the stopwatch to automatically flip into stopwatch mode. Some stopwatches may require shorter or longer pulses, so choosing the values really is a matter of guesswork.
Since a DPDT switch was used as the power switch, even more protection was offered to the stopwatch unit in the form of a discharging path via a 10R resistor the instant the unit is switched off.
Due to a new case being used in place of the original stopwatch case, the original spring terminals pressing onto a piezo sounder mounted on the rear half of the old case could no longer be used, resulting in the use of a KPEG153 piezo transducer located just to the side of the on/off switch.
An interesting observation was that the stopwatch appears to drive the piezo in a similar way to a H-bridge amplifier, deforming the element in both directions in order to achieve maximum volume possible (Despite this, the volume of the acknowledgement beeps from the stopwatch are still relatively quiet, with either the original piezo sounder or the new KPEG153 transducer).
N.B: Note the use of a neck cord on the case - The case is rather smooth and the 70Khz crystal used is both fragile and (unfortunately) irreplaceable.
And despite all this work, there still hasn't been a thunderstorm to test it with! So much for the dry weather!
|Quantity||RS Part #||Part description|
|1||512-3580||2xAA Battery holder|
|1||512-3552||2xAAA Battery holder (alternative, doesn't require battery snap)|
|1||489-021||Battery connector (only required for 2xAA battery holder)|
|1||278-682||24hr std shockproof digital stopwatch (suggestion only)|
|2||247-6667||SMT chip resistor|
|1||435-6767||Green LED chip|
Cable tie eyelet for neck cord
B7G Valve base for crystal
2 OA91 axial diodes
Please note the 70Khz crystal and crystal valve are not supplied by RS Components and will have to be sourced elsewhere.