Skip to main content

Hello Everyone,

I am working on my control panel.  The panel controls block power and supports power districts via DPDT on-off-on toggles.  It also provides switch control via SPDT momentary toggles connected to Atlas 2964 non derail boards.

I am using bicolor LEDs for block power indication and switch position.

I tested the bicolor LEDs on a DPDT on-off-on toggle.  It flipped from red to off to blue as expected.  When I connected two of the toggles in parallel (see the diagrams) they did not behave as I expected.

If both toggles are in the red positdion, each associated LED is red.  If both toggles are in the blue position, each associated LED is blue.

 If I have one toggle set to red and one set to blue, the LED associated to the red setting is red, the LED associated to the Blue setting goes dark. If I reset the red toggle to the off position, the blue LED illuminates.

I have not used LEDs or bicolor LEDs before. I don’t know why this is happening. 

I will greatly appreciate any help

Thanks

BiColor LEDs

Attachments

Images (1)
  • BiColor LEDs
Last edited by JVM-51
Original Post

Replies sorted oldest to newest

What I think is happening here is that your red LED is shunting the voltage to the blue LED. As I recall, the colored LEDs all trip at slightly different voltages (I can't recall if red or green is the lowest, but I know white and blue have a higher trigger voltage). The spec sheet on the diodes would tell you.

 

Off the top of my head, I don't have a work-around for the wiring situation presented.

Last edited by AGHRMatt
Originally Posted by JVM-51:
...If both toggles are in the blue position, each associated LED is red...

Did you mean blue?

 

I think Matt nailed it.  Red LEDs typically run at about 2V, Blue LEDs at more than 3V.  As wired, when a Red and Blue are both ON the Red effectively steals all the current thru the common 470 resistor.  The Blue LED only sees 2V which is not enough.

 

I don't know how many of these you already wired up in terms of what's expedient, but a workaround could be to give each bicolor LED its own 470 resistor from the common/center pin to what you label "Out".

You may want to try a reverse polarity hook-up with a double pole/double throw cross over switch. Diodes (LED's) are just that only pass current one way.  Connect the diodes so that Red works polarity one way, and Blue polarity the other way

These marker lights on a Weaver C630 are connected that way.  Two lead, LED's

Red in reverse

Green in forward.

 

Last edited by Mike CT
Originally Posted by Moonman:

Aren't most bi-color led's common anode?

In this situation the OP would have the same problem; that is, wired as he shows when one of each color is called into action thru the common/shared 470 resistor the Red will starve the Blue of voltage.

 

OTOH if you're making an observation about bi-color LEDs, I think you're right.  It's common to find circuits and components with "active-low" outputs to turn something on by pulling down the voltage. 

There is a difference between a Bi-Color and Bi-Polar LED... one has a common leg, the other doesn't. I learned that the hard way when I built our control panel  I choose the Bi-Polar for ease of wiring with the Tortoise switch machines.

 

It seems I always learn electrical lessons the hard way.

Last edited by Former Member

Placing resistors on each color pin lets you independently adjust the brightness of Red and Blue.  With just 1 resistor per LED common, the relative brightness might not be to your liking - for example red might be just right, but blue may be too dim.  As GRJ says, if you end up using 470 ohm resistors in all 4 places anyway, you'd use twice as many resistors to accomplish the same electrical function.

 

Corrected%20wiring%20for%20BiColor%20LEDs%20and%20SPDT%20toggle

Attachments

Images (1)
  • Corrected%20wiring%20for%20BiColor%20LEDs%20and%20SPDT%20toggle

I was assuming there might be intensity adjustment desired, but if they're "close enough", just one resistor gets it done.

 

As to the other comment about common anode, most of these bi-color LEDs actually have a common cathode configuration.  I buy common anode ones specifically for TMCC use as the lighting outputs are negative in respect to frame ground.  To have directional markers, I need the positive side to be common.

 

Carl, you're drawn a common cathode configuration with negative on the common pin.

 

Thank you to everyone for helping me get a quick solution and for guidance using the LEDs.  I haven't decided which solution to use, since I didn't know the brightness of the red and blue LEDs might be and issue until Stan and GRJ discussed that wrinkle.  Anyway, I can now check out which way works best for me.

 

 

 

Panel

Attachments

Images (1)
  • Panel
Last edited by JVM-51

Out of curiosity how did you settled on red/blue as opposed to red/green?

 

I am no expert on control panels, but it occurs to me that if you occasionally operate in darkened room your panel might be overwhelmingly bright!  That is, it appears you have say 50 LEDs.  With a 12VDC supply and 470 ohm resistors you are driving each active LED at nominal full brightness. I realize you don't have all on at any given time but even so curious if you've given thought to a high/low brightness switch to tone down the entire panel - could be as simple as a few 25 cent diodes to lower the 12V to, say, 10V or whatever.

Actually, I wanted to use blue/white or blue/yellow for block power and power district indication.  The only bi-color LEDs I could find were red/green and red/blue.  Since I planned to use red/green for switch route indication, I settled on red/blue for power.

 

I am somewhat concerned regarding LED brightness.  I think most of the LEDs will be on all the time and may blind me.  I guess there is a few ways to control it. I am using an HO transformer that I can set the voltage from zero through 12. When I reduce the power the LEDs do dim somewhat before they shut off.  I would rather use a small fixed voltage wall plug-in unit of some appropriate voltage.

 

Because I don't fully understand the relation of voltage, resistance and amperage to LEDs, I don't know if I should use different resistors for red/blue and red/green (need to solder 100 resistors to the color pins) or if the brightness is close enough on red/blue and red/green to use a single resistor on each common pin (50 resistors).  I will experiment using using different resistors on the color leads and with lower voltages.  I am not even sure if the 470 ohm resistors are appropriate.

I tried 2200 ohm resisters on the common legs. They dimmed the red nicely. The blue, however remained significantly brighter. I think I will have to use GRJ's solution, placing a lower value resistor on the red leg and a higher value resistor on the blue leg. I will try a few combinations until I get the red and blue at equivalent brightness. 

 

Btw, I am using diffused LEDs. 

 

I see I need to learn more about electronics fast. 

 

 

With 2200 instead of 470, Red is operating at ~5 milliamps (mA) of current instead of ~20 mA.  LED brightness is exactly proportional to current so you dropped the brightness to about 25% of original.  We can get into equations if you're really interested.

 

Earlier you mentioned you'd rather use a wall-plug fixed output adapter than your HO transformer.  Now's a good time to bring this into the conversation.  Presently, you start with a 12V DC supply and use resistors to limit the current into the LEDs.  Since the operating voltage of the Red and Blue are only about 2V and 3V this means most of the 12V must be dropped by the resistors which means wasted as heat.  So some 75%-80% of your power is wasted by using such a large starting voltage.  Granted we are only talking a handful of Watts of wasted power so you won't see it in your electric bill.

 

But you can get reasonably priced ($2-3 on eBay free shipping) DC output wall-plug supplies that put out a fixed 5V with enough power to easily drive your 50 or so LEDs.  If you are running each LED at ~5mA, then 50 LEDs would use ~250mA which is 1/4 Amp which is very modest for such wall-plug supplies.  You might even have an unused cellphone charger that fits the bill.

 

Whatever resistor values you come up with for 12V operation would have to be adjusted for 5V.  We can get into specifics if you're interested.

 

OK, 12 VDC with a 2200 ohm resistor produced an acceptable level of illumination for the red.  I added the 470 and 2200 in series to the blue and got the illumination levels of the red and blue look pretty close

 

That was 12v/2200 to get to ~54mA for red and 12v/2670 to get to get to 44mA for blue. Using R=V/I, at 5 VDC I would need ~920 Ohm for red and ~1140 Ohm for blue.

Assuming that is right, I will order the appropriate resistors (if they exist). I will also look to get the wall units you described.

Again, thanks.

Getting warmer!

 

Red LED takes ~2V to operate...so rather than 12V/2200, it's 10V/2200 or 0.0045 Amps or 4.5mA.  Likewise, the Blue LED takes ~3V to operate...so it's 9V/2670 or 3.4mA.  In other words, the 12V voltage is split between the resistor and LED, and the voltage across the resistors are 10V and 9V respectively.

 

Note that you can get a 12V DC fixed-output wall-plug supply with suitable current for essentially the same price as a 5V one.  So if you're already stocked up on 2200, then 2700 resistors are a common value and you're done (except for a boat-load of soldering!).  Since running at about 5mA per LED, you need about 50 x 5mA = 250mA of current.  A 12V, 1000mA (1 Amp) fixed-output adapter for about $2 (free shipping) from Asia from eBay.  If you don't want to cut/splice the supplied coax-barrel plug you can get an coax-to-screw terminal adapter for about 50 cents.  I recently posted an example in this thread:

 

https://ogrforum.com/t...-relay-recomendation

 

Note that at 1/4 Amp with all LEDs on, the 12V supply is delivering only 3 Watts (Power = Voltage x Current).  So as unpatriotic as it sounds, even if all of that was wasted as heat it's not that bad.  A typical nite-lite uses I think a 7 Watt bulb.

 

But we can re-do the numbers for 5V too.

 

In any case, confirm the resistor values on a few LEDs.  Resistors will be spot-on with what they are marked.  But, depending on your LED source, LEDs can show some variance even within a single batch.

You are right, I forgot  to subtract the LED voltages. (I did see it in an equation last night, but I forgot by this morning)

 

I like the 12V adapter, they are everywhere and cheap. I thought I might have a problem finding a 2700 ohm resistor. Since I don't have much room with everything jammed in the panel I really can't set them in series. I will check for 2700 resistors. 

 

 

 

 

 

 

 

 

 

 

I was thinking i would use the 12volt plugins, but then I thought the J3 connector on my 6924 boards puts out 5 volts to the panel LEDs so maybe I should be consistent feeding power all of the LEDs in the panel.

Anyway I continued tinkering, I tried the red/blue for the blocks at 5 v with red at 1000 ohm and blue at 2700 ohm.  The brightness was very close in a darkened room

Next I connected all 3 wires of the J3 connector on the 6924 to a red/green LED.  The green looked equivalent in brightness to the red/blue on the block circuit but the 6924's red was much brighter. After looking at the instructions for the 6924 I see they use a 390 ohm resistor on the black wire. Since I’d like to have all the lights at the same intensity, I think I may not use the black wire on the J3. I will use a ~390 ohm resistor on the green leg, a 1000 ohm resistor on the red leg and connect the center leg of the LED to ZW common.  I tested that with a 270 ohm on green and 100 ohm on red and it looked good.

So I do have an extra 100 solder connections, but I don't have to run 25 black wires from the 6924 to the LEDs.  Oddly enough, this is turning out to be another fun part of playing with trains.

Add Reply

Post
×
×
×
×
Link copied to your clipboard.
×
×