taillight converter - the other way

[rotarymike]

I'm trying to build a circuit that will take the output of a standard 4-pole trailer connector and control a trailer with separate brake and turn signals (IE, a backhalf of a car). I'd prefer to not use a micro since I would like to power the circuit off the trailer connection.

Standard 4-wire connector is thus:
L
R
Running Lights
Ground

I've got that I can use an AND gate to get this output:

L R | L1 R1 B
=========
0 0 | 0 0 0
0 1 | 0 1 0
1 0 | 1 0 0
1 1 | 1 1 1

Where L1 and R1 are the left and right turnsignals on the trailer and B is the brake. I also think I can use an XOR in parallel with the AND, the XOR output controlling a switching transistor in the L to L1 line and R to R1 line. That is, when the XOR output is high, whichever turn signal is on will power the trailer light; when it is low (either both L & R = 0 or L & R = 1) neither turn signal will light. That gives me this truth table:

L R | L1 R1 B
=========
0 0 | 0 0 0
0 1 | 0 1 0
1 0 | 1 0 0
1 1 | 0 0 1

which is closer. My problem is thus: brakes with one turn signal on. In the circuit I have described above, when the turn signal is interrupted the AND will switch off - leaving me with flashing brakes as well as a turn signal... which because it's an interruption and not a signal will make the XOR work backwards and allow the wrong turn signal to flash.

Any ideas, short of a microprocessor?

Mike Harrington
mike (a) rotarymike DOT com

[Bigglez]

My problem is thus: brakes with one turn signal on.

(1) Brake is only ON if both L and R come on together.
(2) Brake remains ON if either L or R is solid.
(3) Brake is always OFF if both L and R are off.

What you need is a memory that knows both L and
R were on together, even if one of them goes off first
(because its a turn signal).

A flip-flop or latch will do this and doesn't have to be a
uC or even a logic IC. A pair of PMOSFets would be
easiest.

To SET the latch, AND L and R. To RESET the latch NAND
the L and R. The latch will ignore the conditions of either
the brakes on and one turn signal or one turn signal and
no brakes. The latch will follow the 4-way flashers, where
both L and R go on and off together, flashing both L and R
and the Brake lights.

Power for the brake lights can come from two power diodes
that OR the existing L and R signals.

[PICmeUp]

The other thing you could do, depending if the 4-wire source supplies enough power, is as follows:

Connect R1 to R through a NC (normally closed) relay controlled by L.

Connect L1 to L through a NC relay controlled by R.

Connect B to L, R or running lights (if those are always on) through two NO (normally open) relays in series, one controlled by L and one by R.

[Bigglez]

The other thing you could do, depending if the 4-wire source supplies enough power, is as follows

That will work, you have implemented the logic from a
previous post using relays.

The downside is that mechanical relays will eventually
fail, and are prone to dirt and vibration form the
automotive environment.

Would be interesting to hear back form the OP on how
they solved the original problem.

[jollyrgr]

This is one of those circuits that will drive you nuts until you figure out how to make it work. The first time I saw this circuit in use was when I started adding third brake lights to old cars. On cars with amber turn signals the brake light was separate and did not blink during a turn. You would hook the third brake light right to the brake circuit.

For the cars with a red blinking turn signal (blinking the brake light) a module came with three wires that connected to the third brake light. One wire went to the left light, one went to the right, the other was the output. I could not figure out how this thing worked EVERY TIME no matter how fast the blinker was on the lights. The module was made to be in "dirty" conditions (next to the wiring harness under the car) so it was potted in epoxy and you could not see the circuit.

I asked an electrical engineer as well as PhD electronics instructors how such a thing could work. All took the immediate answer of AND gate and chided me for not knowing. Then I'd point out that if it was an AND gate one input would be a clock signal (blinking turn signal) and the third light would blink. Then they'd say "capacitor on input" which I pointed out could not be because it would work no matter how slow it blinked. (I'd turn my key off while the turn signal was blinking and the blinking light would go out and the third light stayed on.) When I'd point that out to them they'd tell me they'd think about it but none ever found the answer.

It took some time but then I figured out how to make this work. Hopefully I remembered it well enough to draw the schematic (below) correctly.

Here in brief is what happens. Assume the driver is making a left turn and signals as such. The left turn signal is activated and is an OFF blink. The driver applies the brake and the RIGHT brake light is on. This provides a GATE signal at SCR2. D2 would conduct but since SCR1 is not conducting, no current flows. When the left light comes on SCR1 now has a gate signal and both SCR1 and SCR2 are triggered and because of the current from both D1 and D2 the SCRs can conduct. When the left turn signal goes dark current is still being supplied from D2 so both SCRs still conduct powering the load. If the driver stops braking the maximum time the load will be active is until the left turn signal goes dark. When this happens the SCRs stop conducting. Even if the signal continues to blink there is no current or trigger at D2 so the load is not activated. The diodes are needed to prevent the signal from the opposite light from crossing over causing BOTH lights to be on with the turn signal.

There is ONE problem I foresee in using this as a brake controller. If you activate the hazard blinker (sometimes called a four way) it will pulse the brakes. If you activate your hazards while driving (say in a "slow" condition) it would be a problem.

[Bigglez]

It took some time but then I figured out how to make this work. Hopefully I remembered it well enough to draw the schematic (below) correctly.

Is this from reverse-engineering a commercial third brake
light controller? Or, is this your solution? Tested or not?

There is ONE problem I foresee in using this as a brake controller. If you activate the hazard blinker (sometimes called a four way) it will pulse the brakes. If you activate your hazards while driving (say in a "slow" condition) it would be a problem.

Brake lights have priority over four-way hazard flashers.

On vehicles with combination Brake and Turn signals (i.e.
only one light circuit on each side of the vehicle, even if it
has multiple bulbs inside the fixture), the four-way flasher
is connected across the brake pedal switch.

A vehicle with four-way flashers active will still show both
rear lights as solid when the brake pedal is pressed. Useful
if the four-way flasher is left on by mistake. Or, the vehicle
is driven with them on - while not legal at least the traffic
behind knows when the brakes went on!

[Bigglez]

This is one of those circuits that will drive you nuts until you figure out how to make it work.

Here's a tale of a circuit fault that did drive me nuts!

My 1979 Ford Mustang had red turn and brake ligts
at the rear. One day (around 50k miles of use) the
left turn signal would also flash the instrument panel
lights. The right turn signal operated normally.

There was no mods or add-ons to the vehicle and
it was in good running condition. The vehicle did not
have a third brake light (pre-dates the law) or a
trailer hook up.

The fault persisted inspite of my efforts to check
by inspecting the flasher relay, the fuses and fuse box,
and the turn signal "stalk" switch assembly, plus as
much wiring as I could find under the dash.

Days later I found the problem, and had a good laugh!

I'll post the answer after some of you have a go
at solving it.

[jollyrgr]

Is this from reverse-engineering a commercial third brake light controller? Or, is this your solution? Tested or not?

It was impossible to reverse engineer as the circuit was a solid mass potted in epoxy. The above circuit is what I came up with. No, I have never bench tested this. Simulation software was non existent (at least for me) when I came up with this in the late 80s. If I had some SCRs available I'd bench test it.

Someone else came up with this idea:
http://www.uoguelph.ca/~antoon/circ/3brake.htm

Their circuit would require that the logic module have a ground connection. I know the modules I installed did not have one. (See links below for more on this.) Possibly if the caps and resistors were eliminated at the gates of this circuit the ground for the logic would not be required. It has another advantage over mine. My circuit has three PN voltage drops to the load where the one at the "uoguelph" link above would only have one.

Commercially available products abound. See this:
http://www.carparts.com/LOGIC-CIRCUIT-F ... _10618.car

or this:
http://www.buyautotruckaccessories.com/ ... ic-module/

or this one with instuctions showing it is a three wire device:
https://www.custompickup.com/shop/index ... revstart=0

Another option would be to get a third brake light trailer hitch. See this:
http://www.wheelskin.com/cr-019r.html

WARNING: Some lighted hitches only light up with the brake or one of the turn signal! This hitch cover appears to have more than two wires. Many are LED and would not have current available to appy a trailer brake.

As to my comment on four way/hazard. Yes the brakes have priority. But with the add on third brake lights they flashed with the hazards on. The same would be true with my circuit or the linked circuit above. Thus it would pulse the trailer brakes while driving.

[jollyrgr]

Days later I found the problem, and had a good laugh!

I'll post the answer after some of you have a go
at solving it.

If it is the same problem I've had then it is a broken filament in one of the dual filament light bulbs. The 1157, 2057 and equivalent have a filament for the tail light (on when the headlights are on) and the brake/turn signal filament. One of the filaments failed and shorted to the other on the left side. When you turned left power was back fed from the one filament into the "tail light" circuit which also fed your dash lights.

I've had a few goofy things like this happen because of shorted filaments. Sometimes the problem would be intermittent as the filament kept bouncing around and would make normal contact with its original connection (self repairing light bulb; almost). Other times it would be shorted to the other filament. I've had these types of bulbs blow fuses. But as soon as enough current was there to blow the 20A fuse, the fault would be gone. (Usually this happens to me when the temperatures are 10F or lower and it is too cold to work on the car.) Turn signal in one direction, no problem. Turn signal in the other direction would blow the fuse once a month or so.

[rotarymike]

thanks for the inputs guys. I think I'll be able to breadboard the SCR solution when I'm finished with classes in late November - now till then is pretty much booked solidly by school. OTOH, I'll have almost two months to go crazy waiting for grades, so a little grease therapy will be in order. One of the major priorities on my auto white-board is to finish the half-a-car trailer so I can paint it and my car at the same time.

Mike

[Bigglez]

thanks for the inputs guys. I think I'll be able to breadboard the SCR solution when I'm finished with classes in late November...

I think that was a concept (and not a working schematic).
I'm troubled by the lack of current limiting feeding the
SCR gates. Be careful!

I'm looking forward to hearing of your progress towards the
end of the year. We'll also have a new resident for 1600
Pennsylvania Ave by then!

[rotarymike]

I've got some solid-state relays left over from another project (reengineering the failure-prone Jeep headlight switch) and I could use those for current handling. Although... I was considering converting the taillights in the trailer to be LED. Not the cheapo LED bulbs, but a round circuit board with a bunch of high-intensity LEDs for each.

[Bigglez]

I've got some solid-state relays left over from another project.

Interesting approach. SSRs are usually associated with
AC power circuits, and as such are scaled to 120 or 240VAC.
Operating them on 12V DC may be a problem. Also, an SCR
or Triac (the core in an SSR) is a latching device, and
load current must be reduced to zero (or reversed) to shut
them off. In AC power control this happens every 1/2
cycle of the AC.

Although... I was considering converting the taillights in the trailer to be LED. Not the cheapo LED bulbs, but a round circuit board with a bunch of high-intensity LEDs for each.

We recently had a thread about LED tail lights, and there are
vendors springing up with either OEM LEDs in bulk or Lamp
replacements that use LED elements. You can probably keep
you existing trailer wiring and lamp sockets and plug in LEDs.

[psycho]

If it is the same problem I've had then it is a broken filament in one of the dual filament light bulbs. The 1157, 2057 and equivalent have a filament for the tail light (on when the headlights are on) and the brake/turn signal filament. One of the filaments failed and shorted to the other on the left side. When you turned left power was back fed from the one filament into the "tail light" circuit which also fed your dash lights.

Brilliant! Though I have never ran into this problem, I would not have thought of this. It makes perfect sense!

Kevin

[Bigglez]

If it is the same problem I've had then it is a broken filament in one of the dual filament light bulbs. The 1157, 2057 and equivalent have a filament for the tail light (on when the headlights are on) and the brake/turn signal filament. One of the filaments failed and shorted to the other on the left side. When you turned left power was back fed from the one filament into the "tail light" circuit which also fed your dash lights.

Bingo! You nailed it. I had most of my friends and
co-workers stumped for a while. Even kept the
"magic bulb" on my desk for a while. The fat filament
had failed and fused to the thin filament support...