Question for the E-Whiz'z

[Lemosteam]

On my layout I have been wiring my lighting.  The enginehouse has two constant lit LED outside lights and a third candle LED inside to make a flicker flame for the furnace. 

Whenever I connect the flicker LED to the circuit, the flashing of the LED makes the other two LED flicker as well.

I think I know why, and that does not matter.

What I want to know is there anything I can add inline to that LED to prevent the flicker to the other LED I want constant, or is it just simpler to use a separate power source for just that LED?

I have a switched battery solution planned, but wanted to explore any other options.

[peteski]

"E-Whiz"?  I like it!   
I need to know more details about the power supply and how all  those LEDs are connected to the power supply.  Just draw a diagram on a piece of paper, take a photo, then post it here.

[Lemosteam]

OK I have not connected the flicker LED to the layout circuit which is just a 3.5v wall wart, as I am afraid to do this because of what I think is happening in the next paragraph. 

The wall wart serves two sets of bus wires (one for the left side of the layout and one for the right.  The bus wires are just + and - speaker wire that I tap into for each building or light.  The smaller LED I use 500ohm resistors to connect their negative wire to the - bus.

The enginehouse has two LED in parallel, and when I also connected the flicker LED in parallel to them and to my handy dandy test battery rig below, all of the LED would flicker together.  I think it was because the circuit was having current fluctuation from the flicker led.

[peteski]

So when you use 3.5V power supply you do use a single (500 ohm) resistor in series with several parallel-connected LEDs? I think that is how you do it, but not 100% clear on that.


So, in this example you originally had a resistor in series with 2 LEDs in parallel.  Those LEDs consume constant amount of power (current), so the resistor is dropping constant amount of voltage.  No flicker.

When you add a 3rd flickering LED in parallel with those other 2 LEDs (and still using that single resistor), the flickering LEDs power consumption will fluctuate with the flickering.  That will change the amount of current (and voltage) across that resistor, causing the other LEDs to also flicker (since the flickerign LED sucks more and  less current as it flickers).  I hope that is clear enough of an explanation.

Solution is to separate the flickering LED, giving it its own resistor.

[Lemosteam]

No. Wwhen using the 3v tester as shown, none of them have a resistor.  I am only adding resistors to the single or multiple LED when they drop down through the layout.  The flickering LED has never been attached to the layout circuit, only the two entry lights have been connected to the layout. 

I only had this issue when bench testing the LED's in the enginehouse.

The two led are on the outside of the enginehouse at left.

[peteski]

Sorry John, I'm getting more confused.  As I explained,  the flickering LED needs its own power feed (with a resistor, if powered by 3.5V).

Now you are stating that there is no problem with the flickering LED when installed on the layout?  Did I mentioned that I'm confused?  Maybe I'm just dense.

Can you restate the problem (if there is one) and *EXACTLY* how you power all the LEDs in the engine house?  As I mentioned, a simple hand-drawn diagram woudl be ideal. I'm sure you're capable of that.   

[jagged ben]

Okay you'll correct me if I'm wrong, but I'm assuming you have the three LEDs in parallel with the power source, and no resistor because you're assuming the LEDs can actually withstand the 3.5V.

I'm a bit surprised that the power source apparently isn't robust enough to maintain a stable voltage, but I guess it isn't.  Quite possible that the 3.5V is regulated by components that are effectively acting like one resistor for the three LEDs in parallel.   Given this, I don't think that adding individual resistors to each LED will help.   Almost certainly the most straightforward way to deal with it would be to put the flickering LED on its own power source.

[peteski]

I cringe at John's LED powering scheme, but it is his layout.  We still need some more specific details because he did mention some 500 ohm resistor in his earlier post.   A simple schematic diagram of how everything is connected would be better than 1000 words.

John works with CAD for a living - I'm sure he is capable of scribbling a simple diagram.

[Lemosteam]

Here’s the schematic. I may have parallel and series mixed up? When all three are connected to each other and then to the battery tester, none are overly bright. It’s just a tester to make sure they light. I don’t leave them on for any length of time. None of my LEDs have burnt out using it.

What happens is as the flicker led flickers, the other 0201 LEDs mimic the flicker.

The layout mimics the battery with a positive and negative bus for the lighting carrying a constant 3.5v. All negative connections to the bus receive a 500ohm resistor. I just tap into the bus wires close to where the wires come through the bench work, just like track feeders.

My fear is that if I connect the flicker led to the bus wire it will make all of my lights flicker.

If there is no way to prevent that, I have already made a battery switch to power the flicker flame LED, but I have to lift the engine house whenever I want to turn it on, which is why I want to connect to the bus instead.

[peteski]

You have it correct John,  You can also mix serial and parallel connections within the same circuit, but we'll leave that alone for now.



Anyway, I now have a clear picture. 

You are testing these LEDs with a 3V lithium coin cell.  As Jagged en mentioned, the battery's internal resistance limit how much current it can supply, even when shorted out.  In your tester, the battery itself limits the current going through the LEDs

Imagine the batteries internal resistance  as a resistor connected in series with one of the batteries terminals (either one).  So you end up with that resistor connected in series with your parallel-connected LED cluster.  As I  mentioned in my earlier post, the flickering LEDs current varies with its flickering, changing the overall current draw of all those LEDs.  So the voltage across that internal resistor will also fluctuate with the flicker.  And because all 3 LEDs are connected in parallel being feed from that resistor, they will all exhibit a degree if flicker.

When you use your "big" power supply, it also has internal resistance, but it is so small that the changing current draw will not noticeably affect the voltage, so only the flickering LED will flicker.

[Lemosteam]

I am worried about the effect of the flicker led on the REST of the LEDs on that circuit.

[peteski]

I am worried about the effect of the flicker led on the REST of the LEDs on that circuit.

Are you planning on permanently running them from that small battery?  This will not damage any of the LEDs, but the effect will not be welcome visually.  Not much you can do, again due to the high resistance of a small battery (that translates to the battery unable provide steady current under varying load).  Why aren't you powering it from your large wall-wart supply used for rest of our layout?

If you were powering your LEDs more conventional way (using higher voltage power bus, and resistors for individual LEDs or clusters of LEDs, you wouldn't have any of these issues.

[Lemosteam]

I am only running the flicker on a battery in the enginhouse office and I will only turn it on manually by lifting the structure, which is on magnets.

I do not want that making all of the rest of the lights on the layout flicker, I unless there is some way to prevent the flicker from affecting them.

It is only circumstance that the wall wart and bus is 3.5 volt. I had it and I did not want to screw around with various resistor needs. Just wanted to light everything from a single power source and I added a switched plug so I can switch them all off together.

[peteski]

I am only running the flicker on a battery in the enginhouse office and I will only turn it on manually by lifting the structure, which is on magnets.

I do not want that making all of the rest of the lights on the layout flicker, I unless there is some way to prevent the flicker from affecting them.

It is only circumstance that the wall wart and bus is 3.5 volt. I had it and I did not want to screw around with various resistor needs. Just wanted to light everything from a single power source and I added a switched plug so I can switch them all off together.

First of all, if you use your 3.5V supply to power the flickering LED, it should not affect any other LEDs because it should be capable of supplying enough current (has its internal resistance low enough) for the voltage to stay steady.  Give it a try and see what happens. Connect the flickering LED to your main power bus.  If it does make other LEDs flicker slightly, add one of those 500 ohm resistors in series with the flickering LED. That will make it operate on smaller current.  Actually you can add that 500 ohm series resistor to the flickering LED even when using the battery. It should reduce the flicker of the other LEDs

Second (and it is too late now): using resistors with LEDs is basically the correct way to power them.  It should be a second nature for anybody using LEDs to include those resistors.  Sometimes actually "screwing around" really makes sense for best results.  You could likely get by with just few different resistor values. The way you have the LED powered now, you don't have any control over their individual brightness.

[Lemosteam]

Again, all of the LEDs I have soldered to the bus have the 500ohm resistor and I am happy with that. You can see all of them in my layout thread.

Again, the only led not on the bus circuit currently (pun intended) is the flicker led, so it’s already independent.