need diagram for LED Police lights

Wendy

Joined Mar 24, 2008
23,415
Actually it requires replacing a resistor, if you read post #38 I describe what does what.

The problem is I think your time measurements are way off. I thought the original YouTube video was to your specs, 3 flashes per second with 1 second between the flashes 3 seconds). The second was to the cop lights I have seen 8 flashes per second with 2.1 seconds rate (this works out to around .78 seconds gap per side).

There really isn't room for a pot, if I had laid it out that way it could have been done. I put a pot for the number of flashes because tolerance drift would have been a PITA.

Original
...... Your specs ................... My board
3 flashes/.5 seconds...........3 flashes/.5 seconds, each flash is 0.083 seconds.
.5 seconds gap..................1 second gap

Revised
...... Your specs ................... My board
3 flashes/.25 seconds ........3 flashes/.31 seconds, each flash is 0.0625 seconds.
.5 seconds gap ..................75 seconds gap

OK, I misread the specs a bit.

I'll drop the .75 seconds gap down to .5 seconds, but this is the last change. I've got a full day today, if you want to change the numbers a bit you have a little time. I'll do it tomorrow morning.
 
Last edited:

Wendy

Joined Mar 24, 2008
23,415


This should be the final version with the correct timing. Soon as I get it done I'll post a YouTube video link. I'll also post a revision minus the board layout that can have individual currents set to each LED. It will be slightly more complex but probably more reliable.
 

Attachments

Last edited:

tom66

Joined May 9, 2009
2,595
I have to say, I think this project would have been easier done using a 4017 or equivalent type decade counter; you'd need about 2 or 3. Basically it works like the traffic light circuits on the 'net, which have varying times and sequences in the red/yellow/green states. You'd still need a clock source, so it would have to contain a 555 timer or astable oscillator or another type of oscillator.

But the circuit is designed now, so I won't submit my idea beyond this paragraph.

I still like the 555/556 timer idea, it just doesn't seem that elegant combining so many timers.
 

Potato Pudding

Joined Jun 11, 2010
688
I did find it hard to see the red lights flashing in the video and would have liked seeing the working lights with the room lights off.

That is impressively well done and finely detailed. I almost agree with Tom that there are easier ways to do this but it is at least partly a matter of what you like to use.

I think I could do it with transistors easily enough because I like to work with transistors. Also like Tom I doubt I will go to so much effort and detail.
 

Wendy

Joined Mar 24, 2008
23,415
I'm not quite through with the drawings. The driver is a bit of a failed experiment, if I started from scratch I would do it slightly differently.

The red LEDs only look dim because the white and blue are high intensity and drown out the red color (which is also high intensity).

If you read through the thread I thought the timing would have been different. Regardless, it would have taken more than one 4017. Look at the sequence...

<flash><space><flash>space<flash><space><space><other side flash><space><flash>space<flash><space><space>

12 to 14 steps, and the associated circuitry would have used more parts (2 4017s and 555). However with the next schematic it will be 3 7555s, or 1 7555 and a 7556.

It may not look it, but I put a lot of thought into this design. A µc would have worked better, but since I don't do µc...
 

tom66

Joined May 9, 2009
2,595
I was thinking of some way to mux the clock input with turning the LEDs on and off.

So you would essentially have it with one 4017, where it would just control which LED group turned on; a current sinking transistor would switch the output on or off.

Only one 555 timer (oscillating at the frequency of the flashing lights) required, and one 4017...

I'm not sure how well it would work. I'll have to think about it.

A microcontroller makes all this obsolete, however I still like analog/digital mixed circuits, it seems too easy to just write a program for a PIC16F690 or similar chip, using a breadboard I could have a police light flasher up in 30 minutes maybe.
 

tom66

Joined May 9, 2009
2,595
OK, here is a quick schematic/simulation of my idea. It's difficult to explain. Done in Paul Falstad's excellent Java Circuit Simulator.

Since you said you scrapped your driver idea, you could consider integrating some of the ideas of this into your circuit.

What were the problems with the driver anyway?
 

Attachments

Wendy

Joined Mar 24, 2008
23,415
I like it. It has the same general parts count, but it would work with some modification. You are pushing this design to get 10ma, for 20ma you would need transistor drivers, one for each LED. Wookie would suggest a IC for the transistors, but I think that would end up taking more space.

The problem with my driver is it has no independent control of the currents through the LEDs. This design has that...



For the first design (Ver. 1.1C) you could do this.



If anyone wants to try either of these designs let me know, they need some polish.
 

Attachments

Last edited:

Potato Pudding

Joined Jun 11, 2010
688
I think I went too long on the Strobe Timer. What would be better is if I had inverted the Current strobe on signal to make it normally off.

And I am not sure I like it being a mode switched from low power slow pulses to 15Hz High Power.

Take the resistor away from between the base and collector of Q6 and the strobe just turns off and on to the High Power 15Hz.
 
Last edited:

Potato Pudding

Joined Jun 11, 2010
688
I also need to create some slave strings of transistors and LEDs so that the 12 quantity will fit.

While I was doing this I started to think about using the Flip flops as a chopper to create a Higher Voltage supply that could be stored in a capacitor. The Stored Over Voltage could be used by the RC car like a Nitrous Shot or Turbo. I am not sure what the tolerance for overvoltage would be but I would think that the 500ms you might get from a capacitor before it drops 70% of the way to the 7.2 supply would be fairly harmless as long as it were no more than 18 Volts.
 

Wendy

Joined Mar 24, 2008
23,415
So where are the 3 pulses per side? I don't see it.

The original circuit I created used a flyback circuit to increase the voltage to drive large strings of LEDs. It worked, maybe too well, but overall the circuit is super efficient overall.

The design I built works, I'm suggesting alternatives if the red LEDs are still too dim. Most of the changes are in the wiring harness, which is a piece of cake to do.
 

Wendy

Joined Mar 24, 2008
23,415
OK, but unless people are using your software the schematic is totally unreadable. The graph does nothing for most folks, OP and myself included.

If you are referring to the first schematic I don't see the pause between the flashes.

He is driving 4 red and 2 white on one side, and 4 blue and 2 white on the other. It is not easy to balance. That is where a lot of my final effort went.
 

Potato Pudding

Joined Jun 11, 2010
688
These are samples of the three different diode currents. and you are right that I missed the three pulse and three pulse. If I modified to add another pulse frequency and the component AND gates to control the LED's, I don't see why I couldn't do that too.

I had read that part of the requirements without it registering.

The component count is just going to get worse which normally makes it more difficult to set up. IC's can reduce components and increase reliability so much that given two options I would probably try to build an IC plan before a discrete system.

But for planning it out myself I find it simpler to think in terms of discrete components.
 
Top