Adding Marker Lights to a PS 3.0 Railking

the red and green marker lights on my gp 38's and 40's are very appealing. Both of my green  front leds were out on my GM+O GP ps2 and i found one bad one and when replaced activated both. I thought these would be wired in parallel but appear in series..one dropping the voltage to the other. Does this sound right? I want to add marker lights to 7 GP35's ps3's.Are the ps3"s the same?

Yes, MTH uses series LED for markers for the PS-2.  Early PS-1 and some early PS-2 5V may have had parallel marker LEDs.

Jason, I can give you pin outs for the markers, but the wire and connectors for these 20 and 40 pin connectors are not available via MTH and you run the risk of damaging the connector wires and or pins if not careful. 

Your best bet may be wiring in LEDs via a CV source powered from track power.  MTH LED harness are available and I have some 6V CV boards if you need them.  G

If you're running from track power, using a CV board is overkill IMO.  You can wire the LED's back to back in reverse polarity and just use a resistor for current limiting.  That connects directly to track power, job done.

John, Don't you need a diode to rectify the AC to DC?

Jason, Yes you can tap AC inputs to power your CV circuit and the LEDs.  This is the way MTH handled it in the PS-1 engines.   G

When you wire LED's back to back with opposing polarity, the LED that is lit provides the reverse voltage protection for the unlit diode, hence no additional diode needed.

I thought there was a limit to what the LED could handle in reverse and it wasnot as much as regular diodes.

When I work on Atlas handling the half AC wave without diodes on TMCC units, about 1 in 4 of the LEDs is not working.  G

There is a limit, but remember we have a current limiting resistor, so the most they'll get is 20ma, not a problem.  I assure you, it's a valid solution.

I never put single LED's in without a protection diode, as I had the same issue you had, they'd eventually fail.  Not all of them, but enough to get my attention.

GunrunnerJohn,

Any chance you can walk me through what you mean by a protection diode, I just the circuit you laid out on the LED strip lighting for passenger cars, it is easy to follow,and works great. 

Best,

Jason

Also, my deepest thanks to you both, I am an electronically challenged on this type of stuff but having read multiple post by both of you, I am at least willing to try it now and in general it is starting to make some sense, whereas a year ago it was all unknown.

Thank you, both for your continued posts and assistance.

Best,

Jason

On a ps3 engine [premier] where can i tap in to for marker lights and would these be wired in parallel? I have a LMK soft key.

For a single LED running off track power, the "protection diode" is simply a diode in series with the power to the LED.  If you put it on the positive leg of the LED circuit, the band would face the LED, if it's on the negative leg of the LED circuit, the band would face away from the LED.

Thank you.

Can I use the 1N4003 or 1N4001 ones that I have on hand? 

Thank you,

Jason

So you do use a diode?  I think it is prudent for track power and not relying on an LED.

Yes Jason those diodes will work.  G

Awesome.

THANK YOU

There parallel, but your going to have an issue getting wires with contacts small enough for the connector.  Your talking tender correct? G

GGG, please read what I write!  If you have two back-to-back LED's, you need no diode, they provide the reverse voltage protection for themselves.  For a single LED, yes I use a diode.

OK, But again LEDs are not as reliable as protection diodes.

If he is going to run this off AC track power he is safer using a diode.  If he wants to run this conventionally too it is better to use a CV and always have the correct ma going to the LEDs.  Were only talking a few extra nickels for the diode or a $1 for the CV components.

Wrong GGG.  LED's will certainly be every bit as effective as the protection diode, think about what an LED is, it's a diode that happens to emit light as a byproduct of conduction.  The maximum forward voltage across an operating while LED is around 3 volts, red or green is around 1.5 volts.  The typical reverse voltage specification for LED's is in the 5-7 volt range.

As for the "correct" current going to the LED, any current under the maximum operating current is fine, in point of fact they'll probably last longer running at lower than the 20ma maximum of a typical LED.  There is no increase in reliability using a current or voltage regulator, AAMOF one could argue the additional component count can only make it less reliable.  The MTBF of a typical film resistor running well below it's power rating is millions of hours.

If you have back to back LED's running on AC power and wired with reverse polarity with a proper value current limiting resistor, there is NO benefit to adding a protection diode.  There will be no increase in LED life adding the diode.

John, If you use a limiting resistor for 18-20V Command voltage at 20ma, than run in conventional at 10-12 you may wind up with dim or out markers.

I have seen it with atlas when playing around with the resistors.

To each his own, but I am sure MTH and Lionel would rather go the cheaper route with lighting, but both use CV boards to power markers in the conventional MTH, and TMCC for Lionel. Or they are using diodes on the marker lighting boards, not relying on one LED to protect the other.

So once your film resistor fails you loose all the LEDs. 5-7V reverse voltage is not a high margin of safety with 18V-24V available, especially if using series LEDs.  That is all I am saying.

Yes your method is the easier/cheaper method.   G

After the millions of hours of MTBF of the film resistor's life, I can live with no lights.  I'm also not sure how the markers are damaged, if I'm around in that million hours to check.  By far, the most common failure of a resistor is an open.  I replace the 2 cent part and I'm back in business.  Of course, I'm guessing that I'll have to come back from the grave to do the repair, because I suspect it'll be awhile before I can see the failure.

I'm just curious how you figure your CV circuit with more components is more reliable, that's certainly a mystery to anyone that's ever done an circuit MTBF analysis.

If you can tell the difference between the brightness of an LED at 20ma vs. 10ma, you have a really good eye, as the apparent brightness chances very little.

I'm amused by your reservation about using one LED to protect the other one.  The premise here is in order to lose the protection, one of the LED's has to fail.  So, we already have lost the marker, right?

I'm not sure why you're working so hard to prove me wrong, I suggest we agree to disagree on this point.  You do it your way, and I'll do it mine.

John,  You stated the failure life of the resistor not me, I only used it to explain my point.

If you want to use an LED with a 5V reverse voltage limit to protect your engine lighting, please go for it.  Your already with in about 2.5V of the limit depending on your resistor.

I would rather use a 50V diode and ensure I have a better margin of safety for voltage spikes and tolerance issues.

If you can't see the difference, don't knock me because I can.  I have run them from 25ma down to about 5-7ma and you can definitely see the difference.

If self protecting LED was so assured why would MTH and Lionel and even newer Atlas add the cost of diodes and CV boards.  Clearly they are in the profit mode.

The same MTF really applies to the CV board too.  It is only a few diodes, regulator and a cap.  It probably will die before your resistor, and still past both our lives, but I am sure you will be replacing LEDs before you die.   G