Best practices using 555/4017 & many LEDs

Freddythunder

Without going back into the thread, I think you said 6 blues, but each segment is 10 no matter the color. Will this 2 547s be enough for 10 blues? I hope so because I ordered 50. Although that was only $5.

You are correct about using the other side of the first 555 astable to click on a third 555 and it would be more synchronized. I may approach that later but I think I'm going to start with a 556 for each 10 LEDs. I still have the whole other side to build later. It's getting to the point where I'm spending too much money on this one prop. I still have a whole movie to budget. And a car to buy!! oh my! I'm thinking of just doing the 8 segments of blue and red and have them blink randomly. I've seen police cars that are totally random and perfectly synchronized so I think I can still sell the effect if they are random.

Besides. I love the idea of only having ONE segment, circuit and all breaking on film day and having one or two backups built that I can swap out with a screwdriver and quick connect to power. Probably a few more bucks that way but I think it'll pay off in the long run.

Windadct

I scanned over the threads 4 pages - and know this is relatively late. But ideally for that quantity you should use a LED Driver - they actually flash the LED at high power faster then your eye can detect - the allows the LED to run a normal power (average) and the eye picks up the higher brightness.
Example - If you are looking at a cars LED taillights - and waive your hand in front of them you will see the strobe effect from this.

Freddythunder

Thanks windadct, but the end result is to have the rate slow enough that It doesn't look solid at all. I'm adding in potentiometers so I'll be able to slow that rate down actually. But I appreciate the input!

NascentOxygen

It's the level of current that is the important factor. If you increase that series resistor then you can use more parallel strings of two blues, although reducing their brightness in the process. At one stage you said they were too bright for the camera, so this is okay. (I tried to get you to determine the lowest satisfactory current for the blues, but you still haven't.) Are you using 5 parallel strings of two blues each with a 390Ω series resistor? You should increase that to 470Ω to avoid tempting fate.

For the circuit values I gave, this was how I summarized it: http://physicsforums.com/showpost.ph...9&postcount=40
So you'll use how many 556's? A dozen or more? That will be eye-catching, to say the least! Reminds me of the time late one Christmas Eve when I strolled down to the corner to have a sticky beak at a colourful Christmas tree decoration that seemed to be lighting up the street, only to discover it was the lights on a police car! (No, I hadn't been celebrating ....)

Freddythunder

Hi! I used that led calculator I mentioned a few times and then upped the resistance a touch. If they are too bright for the camera I can always adjust the camera or fix it in post. Sorry for not getting the lowest current like you asked. I should have but I need to get this project completed! It's driving me nuts! I made the array 4, 4, and 2 with resistors 47, 47, and 330 ohm respectively. Then I have a different array for reds which I tested the yellows in the same configuration; you saw that video. Here's the blues:

http://www.youtube.com/watch?v=jUB2m0NO9Uo

New issue. I made the actual circuit board and the timer is working, but the transistors are not. Is there a pretty good chance of getting bum transistors? I've Quadruple checked everything even looked at voltage readings and it just seems the second transistor is dead. Could I have killed it with the soldering iron? It was a terrible pain in the butt soldering on those pad per perf boards. I'm going to get a quote for printed boards and see if I can afford it.

I'm planning on using a 556 for each segment of lights. I'll have 8 to start in the overheads which I found a 4' by 10" clear fluorescent light fixture lens that will work absolutely perfectly and will probably spread out the light more too. As time progresses I'll have a few more chunks of LEDs on the pushbar and back window and I can use this same circuit over and over! I've added a 500k ohm trimmer on the first 555 so I can adjust the on off rate per segment.

And you were celebrating! Sticky beak is wine!
Blue lights are still going! Even my wife likes it!! She said she's proud of me :)

Freddythunder

Hey I really still appreciate the help. I don't think I would have made it this far without help! Btw

NascentOxygen

You put 4 blues in series? Using 12 volts? I'm surprised you got them to light, though it means you have almost no control over the current in them. I didn't want to have even 3 in series, because the constraints are so tight when using those small signal transistors for the switching. The blues have a higher voltage across them, and 47Ω gives next to no control over the current. Excessive current is likely to damage the LEDs.
No, not with the soldering iron. You most likely killed it with far too much current. Good job it was destroyed now, otherwise it would have happened at a more inconvenient time, e.g., when you are running it off the car's electrical system. Failure is inevitable when you overload that small transistor by making haphazard changes. If you hope to have reliable operation, you have to follow a calculated design.

Since you are driving the LEDs with the high current 555, we don't now need to be so frugal with drive current as when using a low output current 4017. I'll take another look at operating 10 blues using one pair of transistors.
I'm confused. You say the transistor isn't switching, yet you say the LEDs are still flashing?

I'll take a look at some changes to allow 10 blues.

NascentOxygen

To operate 10 blue LEDs, I suggest that you refer back to the schematic here and make the following changes:

change R₁ to 2.7kΩ
change R₂ to 47kΩ

and operate 3 strings of 3 blue LEDs in series with 220Ω ½W
in parallel with a single blue LED in series with 560Ω ½W

If the 560Ω gets noticably warm, replace it with a 560Ω 1W (or, equivalently, a 220Ω ½W in series with a 330Ω ½W).

Fingers crossed!

Freddythunder

I thought I did by following the calculator...

Sorry for the confusion; I have two identical circuits, one on the breadboard that was running the blue LEDs, and the one on the little circuit board that had never lit up anything. I ran a single LED to the 556 to check it's operation and it was working, but nothing was going through the transistors. So I don't think I burnt it out. I will try the new resistor values you suggest. Thanks.

NascentOxygen

Here's an idea if you get printed circuit boards made. Lay it out so each board supports 1x 556, 2x pairs of transistors, and 10x LEDs. When soldering on the blue LEDs, you have one pair of transistors driving 6 LEDs and the other pair driving 4 LEDs. When soldering on yellow or red LEDs, use a few jumper links to drive all 10 LEDs from one pair of transistors, leaving the space for the second pair of transistors empty.

NascentOxygen

Small signal transistors do get killed without much effort. It is easy to get their leads mixed up. Check that; otherwise, replace it and see whether the circuit starts working.

NascentOxygen

I'll add a note for when you run your circuit off the car's electrical system...

Refer back to the circuit schematic, you'll see capacitor C1. It's there as representative of an attempt to filter potentially destructive spikes that may damage your semiconductors. I suggest that you make C1 using 3 capacitors in parallel to make it closer to ideal: 470uF, 10uF, 0.1uF all in parallel, rated at least 25VW.

If I were building this for myself, I'd use a more thorough filter than this, but as you are working to a tight budget, try those and hope for the best.

Good luck with it.

Freddythunder

...And then 4 months later...

Hey Nascent, what does putting the caps in parallel do? I've never heard of that, just the difference between putting resistors in parallel vs series changing the overall resistance - does putting caps in parallel do anything besides offer a different capacitance? I have also seen some sort of regulators that put 4 diodes in a circle.. Do you have another idea to clean up the car power. As this project seems to be taking me years to accomplish, maybe I can build a power circuit next July or so....

In other news, I'm learning that soldering is not my strong suit and I get really frustrated doing it! I've posted asking where I could find some hobbiest to do it for me and see the cost to frustration ratio. Any ideas?

NascentOxygen

Welcome back. Again.

My suggestion for different capacitors in parallel is an attempt to produce behaviour closer to the ideal. While big capacitors (electrolytics) store a lot of charge, they are slow to react, and fast spikes can still get past them. Smaller capacitors are faster to react, so can better absorb fast spikes, but they can't absorb much energy. Placing a few different ones in parallel is a way to get a large capacitance which can still react fast.

I suggested that as a crude protection against transients on the vehicle power supply when the engine is running. If you power the display only while the engine is off, and is kept off, there should be no spikes in voltage.

There are regulator ICs available, or something could be constructed using a couple of transistors, but the circuit you built is probably sufficiently robust that the capacitors will do.

Sorry, I have no suggestions. But if you got those printed circuit boards constructed, the soldering should be as straightforward as possible. When soldering, the surfaces must first be cleaned of oxide, and the soldering iron constantly wiped on wet paper to wipe off the lead oxide.

They say practice makes perfect, though in the case of soldering, it should read "Lots of practice ....".

Good luck!