Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Tailight sequencer not using PIC

Status
Not open for further replies.
Attached is a modified circuit that will drop out of the Sequential Mode if the brakes are pumped but then stay on. Now if the brakes are pumped, both lights will sequence until about a second after the pumping stops and the brakes steadily applied. C9, D2, and R8 were added so that the Pulse Detector responds only to pulse edges, not DC levels. Previously a continuous "high" on the input would cause the Pulse Detector output to stay high. Also had to modify the values of R1 and C1 to accommodate this difference.

I also changed Q1 to a MOSFET (M1) since I realized that the 5V reverse base-emitter voltage rating was being exceeded (that's not modeled in LTspice). M1 can be just about any N-MOSFET.
Edit: An NPN transistor, such as a 2N2222(A) could also be used in place of MOSFET M1 (with a ≈50k resistor in series with the base).

There were a few other changes to correct another race condition I noticed.

Tailight Sequencer 2.GIF
View attachment Taillight Sequencer 2.asc
 
Last edited:
This is a good addition to the circuit.

I see that you have added R7-C8 to delay the pulse detect into the reset input of the first counter. I was wondering about that. Now as the first pulse comes into the clock of U3 the reset will be sure to be low so the counter will start. This was an ambiguous state before without the delay.

I think there may be a problem with the input circuitry. Either the voltage from the turn signal switch is there or it is an open circuit. It does not go to ground. There is no way for C4 to discharge except through the input impedance of U2.

Still trying to make sense of the D2 U4 combination. Is the polarity wrong on C9?

That Federal regulation must be for newer cars. On the older non-computer cars, the turn signal flashes slower when a bulb is burned out, as it is a thermal device. Less current through the flasher longer to heat up.

This is an old car built before the “We know what’s good for you” intrusive do gooder standards. Small real chrome bumpers, no computer controls big brother OnStar monitoring system, etc. I can’t see where adding DOT approved lights from another vehicle is going to be a legal problem, especially as there will be three times as much light emitted.

We did add disc brakes and three point seat belt shoulder harnesses. Also electronic ignition, sway bars, and radial tires. The car handles better than when it was new, puts out less smog and stops better. If you count all the materials used and processed, the energy needed to transport the material, run the factory, etc. it is ecologically sounder to fix up an old car than make a new one.
 
The input is not open, it has the impedance of the front turn signal bulbs to ground, which is quite low when the bulbs are off. That should be more than sufficient to bring the input low.

Yes, I had noticed that the polarity was wrong for C9 but forgot to change the schematic. :eek:

D2 has nothing to do with U4. D2 clamps the capacitor output spike to +12V when the capacitor input signal goes high.
 
Last edited:
Oops, forgot about the front turn signal.

Thanks for the answer about D2.

I'll try to breadboard the circuit soon to see how it works. So far the last half sequencer works OK. Think I'll change R3 to a pot for varying flash rate.l
 
I was getting discouraged at having to learn software to make a circuit board so investigated the offerings on the 'Net. There are light sequence modules available commercially but none of them do what this circuit does. If you want the lights to NOT sequence upon the first press of the brake pedal, you have to run a wire to the turn signal indicator lamp on the dashboard. That is not an option considering the age of the vehicle and fragility of the wiring under the dash.

I was able to get the drive to the tail light bulbs figured out. I'm using a 2N2222 off the output IC to drive a power Mosfet that turns on the bulb. Current draw for each bulb is a bit over two amps, which the Mosfet handles easily without a heat sink. Using a 10 k resistor in the base circuit of the transistor so the chip output isn't loaded down. Another 10K pull-up resistor on the collector of the 2N2222 that keeps the gate of the Mosfet at Vcc until a signal comes in. The part is an FQPF22P10, 100 volts 13 amps and only about a dollar each from Mouser.

BTW, DipTrace turned out to be not so hard to learn. I really like the Autorouter function and the ease at changing components around on the board. Going to try a simpler battery discharger circuit first and send it in to get a board made for practice.
 
Why not just drive the MOSFET directly from the output IC?

If you need an inverted signal just replace the CD4071 OR gate at the output of my circuit with a CD4001 NOR gate.
 
I'm not to familiar with Mosfts, used to old school bipolar transistors. Apparently they don't work on current, but voltage. The 4001 solution sounds OK. There is not much "gate" current required?
 
I'm not to familiar with Mosfts, used to old school bipolar transistors. Apparently they don't work on current, but voltage. The 4001 solution sounds OK. There is not much "gate" current required?
A MOSFET gate looks like a capacitor so there is essentially zero gate current. The 4001 will readily drive them.
 
That Federal regulation must be for newer cars. On the older non-computer cars, the turn signal flashes slower when a bulb is burned out, as it is a thermal device. Less current through the flasher longer to heat up.

And you can bet there is a federal regulation addressing that as well. Any modifications should address all pertinent issues.
 
Last edited:
I'm not to familiar with Mosfts, used to old school bipolar transistors. Apparently they don't work on current, but voltage. The 4001 solution sounds OK. There is not much "gate" current required?
Also note that U10 must remain a CD4071 OR gate.
 
Changing the outputs to NOR gates leaves U10 as the only OR gate left. If U10 still has to be an OR gate couldn’t it be replaced with two diodes and a 10k resistor for a so-called “Mickey Mouse Logic” setup? The cathodes of the diodes would be tied together at the reset input to U5 with a 10k resistor to ground. This way U6 U7 U8 could be replaced with NOR gates and one wouldn’t need an extra chip for U10. Hate to have a whole IC used just for the OR function on the input of U5. Or am I over thinking this?

BTW, I think that a flash rate of 60 to 100 per minute will satisfy Federal and State regulations. That would be for the front turn signals. However, that just seems too fast for sequential flashers. Has anybody measured the cycle time of the sequential flashers on, for instance a new Mustang? I think I’ll change R3 to a variable resistor for fine tuning of the sequence rate.
 
Changing the outputs to NOR gates leaves U10 as the only OR gate left. If U10 still has to be an OR gate couldn’t it be replaced with two diodes and a 10k resistor for a so-called “Mickey Mouse Logic” setup? The cathodes of the diodes would be tied together at the reset input to U5 with a 10k resistor to ground. This way U6 U7 U8 could be replaced with NOR gates and one wouldn’t need an extra chip for U10. Hate to have a whole IC used just for the OR function on the input of U5. Or am I over thinking this?
.......................
Remember, you can share the chip for U10 between the two circuits. I would rather add one IC (and it's likely cheaper), than adding four diodes and two resistors that also has inferior performance (which may or may not matter in this circuit).
 
Do you want the taillight sequence found in older sedans? If I recall, those were cascading taillights that turned on in sequence like a VU meter.
If you want a single lamp chaser, like those found on arcade game faceplates, you will need more logic control. I have an arcade faceplate that uses a Z80 CPU & PIO to flash border strings of LEDs.
Have you checked auto wiring schematics or contacted the automakers to see how they designed these taillight chasers? Those are done with flashers in series with the taillights.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top