Running Light circuit - help needed

Antenociti · Post by **Antenociti** » Fri Dec 12, 2008 11:27 am

Theres a bit of a story to this so please bear with me:

About a year ago i produced a lit landing-pad for a wargamer which featured running lights that ran in to the center from each corner - 5 lights in a row so that they went 5-4-3-2-1-repeat.

I used a commerical kit ( link ) bought from maplins but changed the included LEDs to superbirghts - then ran fibre from each of the LEds to the relevant holes on the landing pad - i.e. from each LED there were 4 fibre strands with each strand going to one of the four rows of lights.

It worked fine and the customer lvoed it but the fibre tip didnt really illuminate that well even using 2mm fibre and having "made" a diffuser to go on top of each one: great with dim lights but easily missed in bright lighting or daylight.

I now have to make another landing pad witha similar arrangement: 5 lights running in toward the centre of the landing pad but this time only three rows of them.

The only way to get bright enough light (I believe) is to run a single LED to each hole in the (acrylic) landing pad or, in other words, to run a total of 3LEDs from each one of the 5 points on the Kit. (there are available 8 points for LEDs but am not using more than 5 on the landing pad)

I've tested and purchased some very bright green 1.8mm LEDs ( link )

These can be poked up in to the surface of the landing pad by drilling out a hole in the acryllic and they serve a double purpose as the landing pad itself is laser-etched with panel lines and so forth - so as well as getting the bright spot of each LED you also get a side-glow effect along the etched panel lines - it looks very funky.

The problem i have is that I haven't done and electronics work for over a year and for the life of me I cannot work out if my cunning plan will work using the LED-Running Light kit.

I dont even know if it will be capable of running 3 18mm LEDs from each of the "ports" on it and if it will what resistors I should use.

Now I'm kind of hoping that one of you electronic savvy folks will be able to help me out (using the above links to the items) {or these that go right to the pdfs link and link and link })

so can i change the reisistors to allow me to run 3 of those 1.8mm green LEDs from each of the "ports" on the Kit ..and all off a single 9v battery?

(and if not any better ideas that spring to mind would be greatly appreciated!)

en'til Zog · Post by **en'til Zog** » Fri Dec 12, 2008 6:34 pm

en’til Zog
First off, a 9 volt battery should be able to run all 3 LEDs in SERIES off of one port. Typically, a GREEN LED will need around 2 to 2.4 volts so 3 in SERIES will use up 6 to 7.2 volts, leaving around 3 down to 1.8 volts needing to be ‘dropped’.

Say, 3 volts to be safe, and at 0.020 amps ( 20 mA ) you’d need a 147 ohm resistor or slightly higher. Or use the actual voltage needed by the GREEN LEDs, figure the left over voltage, and divide that by 0.020.

HTH

Antenociti · Post by **Antenociti** » Sat Dec 13, 2008 6:45 am

certainly does en'til zog - i've got some 150k resistors so can start now - thanks a lot.

en'til Zog · Post by **en'til Zog** » Sat Dec 13, 2008 7:31 pm

Y're welcome!

Antenociti · Post by **Antenociti** » Mon Dec 15, 2008 1:08 pm

got some confusing infro from elsewhere saying i need 1500Ω resistors, not 150k or 150Ω .....

any thoughts?

Sparky · Post by **Sparky** » Mon Dec 15, 2008 3:21 pm

Just a quick check of the math should help:
I checked the link you sent there is a technical specs (thank the maker)
Forward Current the LED wants 20mA:
I = 0.020 (max 0.030)
V Forward Voltage of the LEDs = 3.3 Volts (max 3.8)
Voltage of supply = 9V

3 LEDs in series (voltage adds current is constant)
I = 0.020 Amps
V = 3.3 + 3.3 + 3.3
to find the resistor size:
Vsupply - VLEDs total / I = R
9 - 9.9 / 0.020 = - 45 ohms
Well you can see that the Power supply will not over drive the LEDs, no resistor is needed (some extra power is needed).

If you will be using a regulated power supply, you may put a 'safety' resistor in there, but its not needed as long as the power supply isn't coming in more than 9.9 volts.

Sparky · Post by **Sparky** » Mon Dec 15, 2008 3:29 pm

The confusing info may have been generated by slipping the current you put into the formula, IE if its a webpage based calculator, make sure you put the correct current in there. If it asked for millamps make sure you put 20, if it asked for amps make sure you put 0.020.

Also note that your Green LED (from Rapid there) is newer, and wants the 'standard' 3.3 volts at the 'standard' 20 mA. This is typical of the newer LEDs but not of older LEDs, where each color had its own voltage and possibly different current requirements. Its been made much simpler by manufacture's making sure each color in a series/model of LEDs uses at least the same current.

en'til Zog · Post by **en'til Zog** » Mon Dec 15, 2008 5:18 pm

You da man, Sparky!

Antenociti · Post by **Antenociti** » Mon Dec 15, 2008 5:23 pm

received this advice:

The green LEDs are 3.3V each. Then their series total voltage is 9.9V. They will not light when the circuit is powered by a 9V battery that has a diode in series with it.

Use a 12V battery, 3 green LEDs in series and no series resistor on each output of the IC. The current will be about 11mA in each output because the CD4015 limits the current.

Sparky · Post by **Sparky** » Tue Dec 16, 2008 12:02 pm

It's actually worse than that. (I hadn't noticed a diode in the original circuit, diodes eat 0.7 volts off hte power supply voltage).

I did a circuit with a Johnson counter, and found that 2 of outputs were noticeably dimmer than the others. I changed chip, LEDs and nothing helped. The conclusion is that some of the chip's outputs couldn't actually source as much current as the others. The effect was irritating, as we were making a ring of LEDs.

http://www.kc6sye.com/techmages_2_5_06_c.html

jwrjr · Post by **jwrjr** » Tue Dec 16, 2008 11:37 pm

Regular CMOS ICs don't source much current. If possible, try the same IC in the 74HC family. Those are usually good for 25ma per output.

Sparky · Post by **Sparky** » Wed Dec 17, 2008 2:31 pm

It would've been fine if all the outputs sourced the same current, but they didn't the number 1 spot and number 8 spot where half the LED light of the others.

dsl40083 · Post by **dsl40083** » Fri Dec 19, 2008 6:04 am

Any thought about if the kit mentioned in the first post will do as a chasing light for the fusion reactor for the Jupiter ? Running 4 white LEDs from each LED outtake.

What should the resistor for the LEDs be changede to?

Will it need more power than the 9V?

/Thx

kitty · Post by **kitty** » Sat Jan 03, 2009 11:28 am

If you want to light a model, you usualy close the model when you are done, i.e glue it all together.
Since your model is made of plastic and since you usualy have no access anymore when done you would want to reduce the heat emission and not use LEDs in series, since that would mean that when 1 led fails all the leds connected to that output fail.
So wisest is is to have each LED connected to the output with its own resistor and reducing the voltage.

That would mean you'd have to increase the output current of the output since every LED would need about 20mA and there are components that can do just that called transistors.
Specificaly darlingtons, which are 2 transistors in 1 connected in a specidic way.
A transistor is nothing more than a valve that controls the high current through it by applying a low current to an input line.
Connecting the base( the input line) of a transistor to the output of the timer will allow multiple LEDs, each with their own resistor to be connected to the transistors emitter.
That also gives you a chance to regulate the brigtness of those LEDs since switching the darlingtons cause a ramp up/down in the higher current going through them.
I.e if you make them blink fast enough they will appear to have a constant brightness and you can control that apparent brightness by switching the darlingtons off and on faster or slower (frequency).
You can get cheap darlington arrays that are 8 darlingtons in a 18 pin DIL housing (ULN2803A).
Connect the inputs of those to the outputs for the leds, the power line to your +5 to +12 V or whatever, the common to your - and you'll have 8 outputs on the array that can each drive 500 mA (that's like 25 LEDs with their own resistor)

If you use parallel resistor-LED pairs you lower the voltage of the circuit, so you won't have to turn as much power into heat with the resistors to reach the required voltage drop.
I usualy have a 5V powersource for models.
Since the darlingtons i use have a voltagedrop of 1.2V between emitter and collector (-) i calculate the resistors for a voltage drop of 0.5V (5V -1.2V - 3.3V ) if i use modern LEDs).

btw. a LED is basicaly a tiny hollow mirror with 2 electrodes at its focus in a tiny vacuum where the electrons jump from 1 electrode to the other, much like electric welding or a bolt of lighting.
So it's not advisable to look straight into the modern bright whites with narrow angles.

en'til Zog · Post by **en'til Zog** » Sat Jan 03, 2009 8:38 pm

Um, Kitty?

I've got some LED display devices that have been plugged in, with a wall mounted power supply - a "wall wart", that have been running continuously for a third of a century: well over 33 YEARS constantly. The often mentioned "Life" of an LED is something like 100,000 hours - to HALF BRILLIANCE, not to extinction or burn out. I have an LED clock I got in the early 1970's which has been in constant use since then, and all the LED segments are nice and bright, although some are a bit brighter than others.

Properly used LEDs are almost as durable as solid wire. Properly.

So burnout shouldn't be a problem, and LEDs in SERIES do save on power drain and heat build up, and do increase battery life.

LEDs are NOT little vacuum bubbles - the core of an LED is a semiconducting CRYSTAL which emits light when tickled by an electric current. Different crystal chemistries yield different emitted colors. Crystals emitting UV heavy "Blue" light are often used to stimulate phosphors which emit other colors: BLUE + YELLOW phosphor light = WHITE light, for instance.

jwrjr · Post by **jwrjr** » Sat Jan 03, 2009 11:26 pm

Actually, if you know what to look for you can get light bulbs rated at 10,000 hour life expectancy ... but you won't get them from Radio Shack (or the equivalent). Flashlight bulbs are, of course, right out.
As for transistors, I prefer power mosfets (specifically, the BS170). That one is rated at 500ma current (in a TO-92 package), is dirt cheap, and needs no resistor between the controller and the transistor.

kitty · Post by **kitty** » Sun Jan 04, 2009 11:28 am

en'til Zog wrote:Um, Kitty?

I've got some LED display devices that have been plugged in, with a wall mounted power supply - a "wall wart", that have been running continuously for a third of a century: well over 33 YEARS constantly. The often mentioned "Life" of an LED is something like 100,000 hours - to HALF BRILLIANCE, not to extinction or burn out. I have an LED clock I got in the early 1970's which has been in constant use since then, and all the LED segments are nice and bright, although some are a bit brighter than others.

Properly used LEDs are almost as durable as solid wire. Properly.

So burnout shouldn't be a problem, and LEDs in SERIES do save on power drain and heat build up, and do increase battery life.

LEDs are NOT little vacuum bubbles - the core of an LED is a semiconducting CRYSTAL which emits light when tickled by an electric current. Different crystal chemistries yield different emitted colors. Crystals emitting UV heavy "Blue" light are often used to stimulate phosphors which emit other colors: BLUE + YELLOW phosphor light = WHITE light, for instance.

First of all, wether leds are used on series or parralel, they use the same amount of power.
3 x a 3.3v led in series consume as much power as parallel.
(3 x3.3V) x 0.020A =0.198 W and (3 x 0.020A) x 3.3V = 0.198W.
the differerence in power consumption is dependent on the voltagedrop by the resistors.
Depending on the operating voltage and required current of the LEDs and the powersupply, parallel can be more efficient.

second: As to heat, think of this: when you have to drop 3 Volts in 1 resistor the heat is disipated from 1 source, when you have to drop 3V in 3 resistors that heat is dispated over 3 sources and thus the absolute temperature of those resistors will be lower.

Third: I didn't say that LEDs are tiny vacuum bubbles, i said they basicaly are tiny mirrors with 2 electrodes at its focus in a tiny vacuum where electrons jump from 1 electrode to the other.
The operative term is basicaly.
I still stand by that statement, although there's a lot more to it, but im not interested in giving a lecture about semi-conductors , crystaline matrices at a sub-molecular and sub-atomic level.

Fourth: I have had LEDs giving up on me within a year.
And for what reason? who knows because after replacing those the replacements have worked for years.
Noone can guaranty a 100.000 hours life expectancy, that is an average.
Noone can even guaranty a 1 year lifexpectancy.

Fifth: There also is that little matter of the resistors, they do break down occasionaly as well.
The more heat they produce, the more likely to burn out.
And when using LEDs in serie , when the resistor fails so do multiple LEDs.
So purely for redundency reasons it's wise to use LEDS paralell and have a powersupply that uses as low a voltage as possible.

Sixth: I almost forgot to mention that more and more microcontrollers are used for complicated lighting of models, and they usualy have an operating voltage of 2V - 5.5V.
A consequence of that is that you either have to use multiple external powersupplies or a voltageregulator inside the model, which means big time heat dissapation if you start with a 9 or 12V powersupply and thus wasting power.
If you design your lighting to work on a single 5V powersupply you can use that energy for light instead of producing unwanted heat and you only have 1 powerlead and a common going into the model without having to use voltage regulators inside.
It also means that you just cannot use LEDs in series (unless they are older types that run 1.8V etc.)

en'til Zog · Post by **en'til Zog** » Mon Jan 05, 2009 11:54 pm

I think what we're saying here is that there are several ways of handling a model lighting project. Some people are more comfortable with one approach, others use different techniques. So people who are starting out might try working in differing ways and see which they like best, and/or which works best for them.

Myself, I like discrete logic chips, some folks like PICs, and others....

Antenociti · Post by **Antenociti** » Sat Jan 10, 2009 9:32 am

well its up and running now - here it is on the still WIP landing pad fti:

thanks to everyone again for the help!

http://www.youtube.com/watch?v=zQNSQjie46g

jwrjr · Post by **jwrjr** » Sat Jan 10, 2009 11:21 pm

en'til Zog is correct. As one author said: "If you only see one solution to a problem, it means that you don't understand the problem."