Sequential LED's (like the New Mustang's Turn Signals)

[Falcon69]

Drawing

I'm confused though, on the schematic here http://homemadecircuitsandschematics.blogspot.com/2011/12/how-to-make-car-led-chasing-tail-light.html

It shows pin 15 (MR) and Pin 5 (Q6) connected on the 4017.  I'm confused.  Q6 is one of the 10 outputs. 

Is it drawn wrong?

[Falcon69]

oh!  Wait, I think I understand.

In that schematic, they are using only 6 outputs, technically, Q6 is number 7, so, when it triggers the Q6, it goes to MR which resets the clock and starts back at Q0, therefore, no need to use gates on the other outputs.

Since I will only be using 3 outputs (1 for each LED assembly, left/right), I need to connect Q3 (pin 7) to MR (pin 15)

Is this correct in thinking?

[Falcon69]

oh

I just realized that if I take out that flasher module from the vehicles circuit board, the front lights won't flash right.  I'll need to trace the wires and run one all the way to the back, where my circuit will connect to the rear taillights.  bummer. more work.

[jlmoon]

no what the sequencer (4017) is being reset on count #6.. 0 - 5 then reset the pulse train at #6
so in effect if either on L or R.. you get 6 counts 0-5 and then a reset to start over
you could do the same after 0,1,2.. but then it might look funny to your eyes, there are so many ways to sequence a set of LEDs and depending on how you wire them is how they will appear.  Can play tricks on your eyes. 

The website diagram will work as well , they are using diode enable or logic instead of gates.. I prefer gates.. , something about the flexibility in them..

[tjb1]

I'm not sure what lights you have but its likely GND, PARK, BRAKE/TURN.



I've done a lot of wiring with those and the vehicles I have worked with use the same wire for the brake/turn and just flash or hold it steady for brakes.

[Falcon69]

tjb1

Yes, they are pretty much the same, just oval instead of round.  They just hook up to the battery for 12 volts.

So parking would basically just be the dimmed lights (white wire)

[jlmoon]

oh

I just realized that if I take out that flasher module from the vehicles circuit board, the front lights won't flash right.  I'll need to trace the wires and run one all the way to the back, where my circuit will connect to the rear taillights.  bummer. more work.

Yes, I was wondering when you were going to address that part of the project.  My solution was to OR (summing mode) the three outputs together and drive yet another gate of another mosfet for the front turn signal so when any one of the three turn signals are in (1) mode.. the front turn signal would be on as well and during the rest time somewhere between Output #3 and reset back to #0 the front lamp will be off. 
 
To do this project is not as easy as just hooking up some logic and relays.  Back in the early 60's if my memory serves me correctly they used to use a drive device to sequence those lights on the old Ford Thunderbirds.. I had one, but it has been so long ago now.. forgot if it was a stack of flashers or a drive motor that did that.  Maybe someone else on here will chime in on that one

[Falcon69]

I think what I am going to do is just tear the dash apart and find the switch wires of the turn signal and tap into those and run wires all the way back.  Place the flasher module back in for the front lights.

[jlmoon]

what type of vehicle is it.. year and model?
that still won't resolve the logic needed to sequence the turn signals, mute the braking signal and deal with the front lamp flash mode.. it's gonna be a    but you can do it!

[Falcon69]

it's a 2002 Dodge full size 4x4

Here's the schematic so far.  I figured no need for resistors after the LED's because I'm sure the assemblies have them built in already.

[jlmoon]

no resistors needed.. yes, they build those in on the boards

[jlmoon]

it's a 2002 Dodge full size 4x4

Here's the schematic so far.  I figured no need for resistors after the LED's because I'm sure the assemblies have them built in already.

what are those semi's.. Q1.... driving the LEDs.. and what are you going to do with the anodes on your diode trios on each side?

[Falcon69]

those are NPN transistors to power the LED's instead of powering them from the 4017. Should I replace those with mosFETs?  Each of those LED's shown are each of the 3 LED assemblies on each side.

[Falcon69]

oh, right, D6 and D9 should be connected for the brake light

[jlmoon]

actually on the Right LED group your anodes don't go anywhere
you will also need to use mosfets unless you're prepared to do some tricks with forward bias voltage drops on those transistors
need those Vds to be low so you don't loose any brightness due to voltage drop and you can use small heat sinks, and the calculations begin!

I wouldn't advise NPN.. , you loose .6 - .7 V across the B-E junction. , the mosfets drop like almost nill.. take a look at some of the mosfets.. IRFZ44.. or take your pick
at 10 Volts  Vgs.. the R_drain to source (Rds) is like in the milliOhms .. so very little power will be dissipated across that active junction.  Something to consider. 

[Falcon69]

Okay, i will try and find some surface mount mosfets that will work.

I ultimately wanted to design a circuit that i could just hook up the wires already at the rear, but, with the flasher module up front, not sure how to do that.  At this point, for me, i think this circuit will work, as long as I can tap in the turn signal shifter and brake light and parking light controls and run the wires back.

I may need to add a resistor as a dummy load for the flasher module so the front lights don't flash erratically.

I was recently in an accident, and my right turn signal on my car flashes super fast, and I'm guessing that's because the light on the rear no longer works, hence, the idea of a dummy load.

I will grab my nephew and head to Radio Shack and pick up some test parts.  I hope they have these chips in stock.  If all works well with this schematic, I will buy some surface mount components and design the PCB.  Since I have to make 10 boards, I might as well buy 10 sets of parts.  He can sell them to his buddies and get back his money and thus, basically, get a free circuit board for controlling his lights.

[jlmoon]

Okay, i will try and find some surface mount mosfets that will work.

I ultimately wanted to design a circuit that i could just hook up the wires already at the rear, but, with the flasher module up front, not sure how to do that.  At this point, for me, i think this circuit will work, as long as I can tap in the turn signal shifter and brake light and parking light controls and run the wires back.

I may need to add a resistor as a dummy load for the flasher module so the front lights don't flash erratically.

I was recently in an accident, and my right turn signal on my car flashes super fast, and I'm guessing that's because the light on the rear no longer works, hence, the idea of a dummy load.

I will grab my nephew and head to Radio Shack and pick up some test parts.  I hope they have these chips in stock.  If all works well with this schematic, I will buy some surface mount components and design the PCB.  Since I have to make 10 boards, I might as well buy 10 sets of parts.  He can sell them to his buddies and get back his money and thus, basically, get a free circuit board for controlling his lights.

the right signal is not working either because your right turn signal wire is shorted to ground (makes the thermal breaker - flasher) open and close much more rapidly than normal or you were hit hard enough the filament fused inside the bulb and created a direct short due to impact force while it was hot or active with current.

I would breadboard that circuit and get it working with LED's on a bench before going out and cutting / modifying any wires in that truck.. trust me on this one, you will be better off and learn alot in the process.
Good luck!

[Falcon69]

yes, I plan to bread board it.  I use to install car alarms, so tracing the wires to find the correct ones won't be a problem, just a pain.

I am going to check at radio shack today, to see if they stock those parts, though, I doubt they do.  They seem not to stock too many parts these days, like they used to.  I might have to drive out to Wilsonville, OR and buy them at Fry's electronics.  Long drive though from Vancouver.

[jlmoon]

Regarding those mosfets and how to use them, here is a link to help you out .. Enjoy!
https://www.youtube.com/playlist?list=PLjzGSu1yGFjUgTVkhb4WslOVDNAZ3K4yR

[Falcon69]

Okay, so, I tried this circuit on the truck that I found on this website http://homemadecircuitsandschematics.blogspot.in/2013/03/sequential-bar-graph-turn-light.html .

The sequencing of the LED's work as intended, just like they do on the new Mustangs.

However, When the brake pedal was applied, they still flashed.  Also, when turn signal (or brake) is turned off, they keep flashing but dim until they are lit no more.  When I turn on my headlights, NOTHING happens.

So, I took that schematic, and I redrew it with modifications.  I am not sure it will work, but seems to when I run a simulator.

The LED assemblies I am using have 3 wires: Red (Brake/Turn), Black (Park/Tail), & White (Ground).

I will be using 3 LED Assemblies per side: Right and Left.

I only drew the left side tail lights for the schematic.  The other side of the Hex inverter and Quad Nor Gate will be used for the right side tail lights.

I may need to replace all the Transistors with mosFET's, but will try it with Transistors instead, because of cost. Seeing how there will be a direct battery hook up and batteries usually put out 14.5volts, I should be okay on brightness for the LED's even with the voltage drop of the transistors. In theory anyway.

Do you see anything that I can change with this circuit?

I appreciate your help, and so does my nephew, who I am doing this for.

Jason

[Falcon69]

So, I modified the circuit I found, with some help, and this works, except for one problem.

I'm hoping someone can help me out on this one.  I'm stumped.

C1 (470uF capacitor) is there because the turn signals and hazards are connected to the vehicles turn signal switch, which flashes with the relay.  I have tried to find a direct connection with a continuous flow of 12 volts from the turn signal switch, and can't find one.  If I take out the flasher module, then the front lights on the truck do not flash.

So, C1 keeps power supplied during the 'OFF' times of the flashing, so it will keep Q1 turned on.

The problem I have is that when the turn signal is turned off, the LED's still flash for a few seconds and dims to nothing while the Capacitor discharges, and then finally turning off Q1. This poses a problem. When the brakes or park lamp are activated, they will not turn on Q4 or Q5 until the capacitor totally discharges and turns off Q1, allowing Q4 and Q5 to turn on (depending if brake or park lamp is activated).

Is there a way to fix this?  Anyway to discharge the cap quickly so the brakes will activate the LED lights?  This could cause on accident if there is a delay in the brake light turning on. I can't place a transistor in there, because it will just turn off and on as the lights flash.  I've tried change the values of R4 to 10k, 1k, and 100k. Anything larger than about 30k won't even turn on Q1.  This changes nothing, only makes LED's slightly brighter if anything.  I've tried changing the values of C1 to 4.7uF, 0.1uF, and 0.01uF and that really screws up the flashing of the LED's and I can't adjust the flashing sequence via R10.  So I'm stumped here, can't figure out how to fix this problem.

Also, one other thing, is a resistor needed between the collectors of Q4 and Q5 so that when the brakes are applied while the park lamp is on (the dimmed lights when headlights are turned on), the park lamps will turn off?  I'm just thinking if electrons take the path of least resistance, when Q4 turns on, the LED's would ground out through Q4 and not even go through Q5.  Or does this not even matter, since the LED's are the same LED's for both operations (no separate LED's).  I guess I could just simply move R8 to the collector side of Q5 to solve that problem, if there was one.

FYI, the circuit does work as intended, accept for the problem explained above which I beleive is in part because of C1 capacitor.  Other then a couple relays to change the polarity of the input connections, the circuit works fine, so I'd rather not mess with too much if it is not needed to fix the capacitor problem.

I appreciate your help with the capacitor problem.  Thanks in advance.

Jason

[nuno]

I don't know about the capacitor (I guess you would have to go get a turn signal control signal instead of just looking at the flashy signal, it's the only way to know if the turn signal is commanded on or off), but I think you have quite some simplistic "power supply" (and no fuse). The electrical environment on a car is noisy and spiky. With the engine on you have some 14.5V, and there are 2 versions of the CD4017 (and of the 4000 series CMOS chips in general), one that takes up to 12V and another that takes up to 15V. But even using the one up to 15V it's a very small margin and there are the spikes from other stuff connected to the 12V system. I have nothing more specific than a googled example of input power conditioning:
http://electronics.stackexchange.com/questions/76588/automotive-power-filtering-protection-feedback-needed

[Falcon69]

Yes, I checked the data sheet.

The 555 accepts up to 16 volts, as well as the 4017

The final circuit that will be in the truck permanently will have a 2 amp fuse.  I checked online and each of the 6 lights draw 243mA each.  So, a 2 amp fuse should be good.  I will also be putting in a 7812 regulator as well.

I may have found a fix for the flashing of the LED's after the turn signal is turned off.  I am trying to make a Missing Pulse Detector circuit work for the input of the turn signals.  That circuit uses another 555 and turns the pulse signal of the flashers into a continuous 12 volts. 

Jason

[Falcon69]

I tried to use a missing pulse detector circuit.  I tried to use several different kinds.

I can't make them work.

I have a truck that I need to change the flashing of the turn signals/hazard lights into a solid continuous 12 volts.  Not a flashing or pulsing. I have located the wires in the steering column that are connected to the switch.  I was hoping they were a solid 12 volts, then switched at a relay, but they don't. The switch connection also flashes.  I need to change that into a solid 12 volts and will need to switch on and off quickly for braking.



Anyone have any ideas?  I'm stumped, been working on it for about 30 hours (2 days combined).

[rs20]

It's very scary how you've got all these power transistors doing the logic work as well. This approach seems inadvisable (see below), but there are several specific problems with your circuit that jump out at me:
 -- There's no resistor between Q2's base and pin 7 of the CD4017 (which can be ground). See the little arrow in a transistor? There really is a diode there. This means a huge current will rush through Q2.
 -- Again, there's no resistor on the base of Q3. This means capacitor C1 will get very quickly discharged by Q3. This is why the capacitor isn't succeeding at keeping that signal high for a decent while.

(There may be more issues, but I don't think addressing them is the answer.) Another reason doing your logic in the power side is bad is because so much voltage is getting burned off by all these transistors. Looking at D6, current flowing through it has to go through Q2, D2, and Q4 just to get out. That's over a volt lost right there, assuming the transistors are biased correctly. You may well eventually succeed at this approach, but it's more difficult and you don't gain much.

I'd advocate using logic gates or transistors* (with the CD4017 and brake/park signals etc as inputs) to decide when the LEDs should be on, and then just having a simple, single transistor per LED to do the high power work. Much more neatly compartmentalised, much much easier to debug. Pic attached.

   * Even thyristors???

PS/  One other thing -- just like the way that the voltage supply is normally positive and at the top, and ground is at the bottom (you got these things spot on), schematics are generally laid out (where possible) such that "data" or "control" flows from left to right -- so the way that the 555 timer drives the CD4017 to its left, which in turn drives the transistors to its left, is a bit confusing. Just a stylistic thing.