Is there anyone that can lend a hand at creating a circuit that will replicate a Traffic Lane Control.
such as:
http://www.wanco.com/products/detail.php?prd_id=52&type_id=10

The final display will be part of a scale model of one of these and would use 25 - 3mm LEDs

I would like to be able to control which arrow function is active via remote.

Is there anyone who can help me on this...

Thanks

 

Hello,

Take a look at Bill's pages:

http://forum.allaboutcircuits.com/blog.php?b=378

Bertus

 

I was indeed just checking his pages out... Thanks...

Hoping perhaps he can assist me with this since I think this might be just up her alley from the looks of some of the circuits I'm seeing...

Thought perhaps I should add this image of the programmed effects.

 

You could drive each LED using its own MOSFET or bipolar transistor and use a diode matrix configured to turn on the proper LEDs for each pattern. For remote, you could use a demultiplexer so that you have only 4 conductors plus ground and +Vcc going from your remote control box to the display. All this is really easy. If you need a sample schematic, let me know and I'll get one posted up in a day or so.

 

I would be very interested to see what you have in mind...

 

I'd use a microcontroller and Charlieplexing.
http://en.wikipedia.org/wiki/Charlieplexing
You could do the whole thing using just six I/O pins for the LEDs.
Just about any 14- or 18-pin PIC could do the job.

 

If you were going to use a microcontroller you could use two of these. They make them in yellow but Rapid don't stock them. Shouldn't be too hard to find.

http://www.rapidonline.com/Electron.../LED-Displays/17.3mm-Dot-matrix-display/29387

 

I'm hoping to find a microcontroller solution to build this around.
Amount of discreet parts needed, actual foot print needed and the over all power of these things would be perfect for this situation...
The plan as it stands now for the actual display is to use these led's

http://www.kingbrightusa.com/product.asp?catalog_name=LED&product_id=WP4060ED

The Intention is to have the main circuit board in the blue shaded area, and the displays of course above...

Now mind you, This is a 1/32 model, the over all length from ball hitch to bumper is 4 inches, The display board will be 1.5" x 3.0" (4'x8' in real life)

 

if a microcontroller is used along with some LED charliplexing/multiplexing, then something like a PIC16F88 should work, you can use the internal oscillator function to minimize components on the board, plus if you can manage to prototype with surface mount components, you could do everything in SMD's (even the LED's, then apply a blob of hot glue on each LED opening to diffuse/spread out the light)and have the whole circuit as part of the LED display itself....

My .02

B. Morse

 

If you wanted to make the programming a bit easier you could use a 40 pin PIC like the 16F887. One pin for each LED plus spares. It's only 12mm square in quadpack format or the full size PDIP is 2 inches is long if you're worried about soldering the tiny one.

 

It can be treated like a 5 row by 14 column display, use 2 4017 decade counters to "scan" the columns and use 5 pic I/O's to use as the row drivers.... not much coding required, plus since the LED's will be multiplexed, the current consumption of the circuit should be minimal since only 3 LED's at the most would be on.....

B. Morse

 

Interesting exercise.

It seems to me that the rows above and below the center row are always mirrored and so I suspect a solution might require less output pins then one might expect.

Cheerful regards, Mike

 

It can be treated like a 5 row by 14 column display, use 2 4017 decade counters to "scan" the columns and use 5 pic I/O's to use as the row drivers.... not much coding required, plus since the LED's will be multiplexed, the current consumption of the circuit should be minimal since only 3 LED's at the most would be on.....

B. Morse

Did you perhaps mean a 5 x 7 matrix?

I don't see any advantages in your three chip solution. First of all when you reduce the duty cycle you normally increase "peak" current in order to realize the same "average" current and brightness. Reducing the duty cycle just for the sake of reducing "average" current and brightness doesn't seem like a very good design goal for this project.

If reduced size and "average" brightness are the primary goals for this demo' then I would recommend;

(1) Using a 14-pin PIC with 10 pins driving a 5x5 matrix at 20% duty cycle or 11 pins driving a 4x7 matrix at 25% duty cycle.

(2) Using 5 pins of an 8-pin PIC to drive a 4x5 Charlieplexed matrix with the LEDs in the rows above and below the center row connected in parallel (mirrored) and driven at twice the duty cycle of the center row to even out the brightness. The program would appear to be controlling 16 LEDs at 1/9th (11%) duty cycle.

The code required to drive a Charlieplexed matrix is very similar to the code required to drive a standard multiplexed matrix.

The LED patterns and time delay values can be stored in tables and can be selected one at a time using a single push button switch and controlled using a relatively simple sequencer routine.

Regards, Mike

 

It can be treated like a 5 row by 14 column display, use 2 4017 decade counters to "scan" the columns and use 5 pic I/O's to use as the row drivers.... not much coding required, plus since the LED's will be multiplexed, the current consumption of the circuit should be minimal since only 3 LED's at the most would be on.....

B. Morse

Um... Im not sure where only 3 LED's would be lit... if you refere back to the pic you can see most arrows are going to at minimum would have 5 Leds to create a Simple Arrow >

 

I believe he's referring to the number of LEDs that would be lighted at any one time in his suggested "multiplexed" display design. Up to three LEDs in a matrix column would be lighted at one time then you would light up to three LEDs in the next matrix column, and so on. The columns are lighted sequentially at a fast rate so that persistence of vision makes them appear to be lighted continuously.

Regards, Mike

 

Did you perhaps mean a 5 x 7 matrix?

I don't see any advantages in your three chip solution. First of all when you reduce the duty cycle you normally increase "peak" current in order to realize the same "average" current and brightness. Reducing the duty cycle just for the sake of reducing "average" current and brightness doesn't seem like a very good design goal for this project.

<snip>

No, I actually meant 5 x 13 (I first counted 14 columns for some reason ).
If you look at the display as a grid and count every vertical column there are 13 of them....


And I didn't say this circuit would give him an advantage over any other circuit..... was just throwing ideas his way.... with the possibilities of adding other "modes" of display in the future...

here is a version I made with an F84A driving an 8x8 LED Matrix, basically the same setup here ...http://www.youtube.com/watch?v=JBwQuF46CHU

So I will step out of the way ....

B. Morse

 

Ahhhh Yeah, Now I see what your saying with 3 on at a time... makes Sense..

I spent the better part of last night reading on those 2 controllers.. going to have to re read some parts this morning now since I fell sleep at the computer and now not sure I got all i read lol..

 

if a microcontroller is used along with some LED charliplexing/multiplexing, then something like a PIC16F88 should work, you can use the internal oscillator function to minimize components on the board, plus if you can manage to prototype with surface mount components, you could do everything in SMD's (even the LED's, then apply a blob of hot glue on each LED opening to diffuse/spread out the light)and have the whole circuit as part of the LED display itself....

My .02

B. Morse

Having everything on the display board it self would be awesome... then the amount of wire that has to go up the left and right supports would be down to a minimum... Potentially just power and data?

Other then have it remote controlled, which that would make it the easiest and provide less handling, a dip switch hidden in the box could be used to select the patterns I guess...

 

If the MCU were a surface-mount type on the rear of the display board and the control were a miniature switch on the same board, you would only need +5v and ground supplied to the circuit. You could make one of the sign board supports a copper or brass tube, with a small-gauge wire going through it.

If you wanted remote control, you would need at least two wires going through a conductive support tube, or you would need to do something like an IR remote control, which could get complex in a hurry.

 

I all ready have a PIC IR remote control that I have done for some lighting in a similar model...

http://www.rentron.com/remote_control/TINY-IR2.htm

Early thoughts were that I would end up running 32ga wire wrap up the support columns to get everything needed there...