As I was saying before we ran out of time and space, there are still a few more circuits I'm adding to my car (Electronic Design, July 6, p. 113 and August 3, p. 98).
Directional-light flasher: I was going to build a new directional-light flasher circuit, for my OLD 1968 Beetle, because the old flasher circuit had worn out and gotten flaky. But since I was replacing the whole car, I naturally figured that I wouldn't have to build that circuit.
Ooops! I noticed on my new (1970) VW, that when I hit the directional signals, the indicator light on the dashboard would turn on instantly. But the actual blinker signals on the outside of the car would not turn on for 1 second—in the second half of the cycle.
WELL, I didn't like that one bit. I may have been driving around for a lot of miles in cars that did that, but I don't have to put up with it.
I like to hit my blinkers early, so other drivers can tell where I want to go—for example, to change lanes. But I don't like the idea that when I hit the switch, nobody sees my blinkers until after a 1-second delay. Do YOU KNOW if your blinkers actually start blinking so that OTHER DRIVERS can see them right away? I have gone on record saying that I hate cars whose windshield wipers do not begin to move until 0.4 seconds after you turn them on. I hate that delay.
The circuit di agram shown in Figure 1 will start flashing the actual directional lights, right away. No relays, no contacts to wear out.
The diode D2 is a little catch circuit, so the first BLINK is not longer than normal. (I hate when the first blink is very slow like that.) Q1 is normally turned ON while waiting for a load to be connected. As soon as a load current flows through R1 (which is 4 ft of 20-ga. wire), A1 starts to pull the 1-µF capacitor down. Then A1 and A2 form a conventional triangle-wave oscillator. A3 drives the LED indicator and makes a tick sound in your radio speaker—but only if BOTH bulbs are working. If one bulb goes out, the LED indicator will not light. (Trim that pot.) A1 and A3 should be a FET type, such as an LF411 or 1/2 LF412, so its common-mode voltage includes the + rail.
This circuit uses a milliampere or so from a +23-V supply, which is generated by the little rattler in Figure 2. You can use almost ANY op amp here. You could use a CMOS gate, and get slightly larger output, but I would not trust CMOS circuits to run on +14 V, in case of transients. It's easy and safe to get adequate output from an op amp (23 V from 14-V in) that would survive 36 V.
Horn modulator: What is a horn modulator? I got a nice letter from a guy who had read my column on Reflex Response, where I tried to warn a dog to get out of the high-speed lane (Electronic Design, Dec. 5, 1991, p. 125). He explained that if you really want your horn to work well, and get the warning across to a dog (or a person), CHOP IT at 4 Hz. I asked, "Won't it work better at 2 Hz?" He said, "No, 4 Hz." So, I built this horn chopper that will allow me to chop the horn at 4 or 2 Hz (Fig. 3).
Of course, I will make it easy to turn this off, and keep a "dc" horn path. In this circuit, I did not put in a catch diode, so the first BEEP is longer than the rest. It seems to work. I like it.
Have you ever watched the strobe lights on an airplane at night, and wondered if there wasn't a better way to track it? The first time I saw this was 24 years ago, and I was impressed: the strobe light would blink twice, with a spacing of about 0.3 seconds, and then repeat after about 2 seconds. I immediately realized, even in the first 4 seconds I ever saw this, that these blink pairs were EASIER to track than the ordinary, dumb, evenly spaced blink, blink, blink. I really like that scheme for the strobe lights on planes. But, I don't think I will build this into my horn....
Back in July, I observed that you might want to turn your radio on and off with a FET switch instead of the crude pnp Darlington I showed there. That would be easy to do by connecting up another transistor, similar to the ones shown in Figure 1, to turn the radio on and off. The FET's ON resistance can be as low as you'd like—using a tiny bit of current from the +23-V supply.
Did you hear that Sidney Darlington passed away recently? In his obituaries, they listed a lot of things he invented besides just the compound transistor connection that everybody knows. My old colleague from Philbrick-Nexus, Al Pearlman, still claims that he invented the "Darlington" first, but Darlington published first.
All for now. / Comments invited! RAP / Robert A. Pease / Engineer rap@webteam.nsc.com—or:
Address: Mail Stop D2597A National Semiconductor P.O. Box 58090 Santa Clara, CA 95052-8090
Reprints
