Headlight/BU Light Troubles on PS-2 MP15AC Switcher

PS/1 diesel directional lights are 1.5V bulbs, if they're the originals, they toast after being connected to the PS/2 lighting outputs.  The bulbs were not LED's for a PS/1 vintage locomotive.

Thanks John; these are definitely leds, so maybe they got switched out with the PS-2 upgrade. I guess I'll play with a few resistor values and see what I need inline with the replacement leds.

Rod

Are the LED's working?  You need a 220 ohm resistor for LED's with a PS/2 or PS32 board.

No the leds are both shot. I'll try an inline 220 ohm reisistor and see how that works. What does a PS-2 board put out to its lighting circuits? I thought I remembered about 6 vdc.

Rod

If they were connected directly to the PS/2 output, the life expectancy is about 15 seconds.  Don't ask me how I know this.

The 220 ohm series resistor is what I use for PS/2 to upgrade to LED lighting, I've done a ton of them.  The lighting output is PWM, but it's "effective" voltage is 6 volts, hence the 6V bulbs for PS/2 equipment.  I've never gotten one of my LED lighting upgrades back, so the 220 must be a good choice.

OK thanks John. Based on dropping about 4 v and with I=10 ma for a 3mm led, I was going to use about 400 ohms inline for starters. But the 220 ohm don't make the leds unrealistically bright to you?

Rod

Just noticed that the Vf for white leds is about 3.2v. So dropping 2.8v with a 220 ohm gives I=13ma; which should be fine I am thinking.  Sound about right?

Rod

That's what I was shooting for. It's well below the actual rating of the LED, but gives you nice light.

As for brightness, take a look into a locomotive headlight, those suckers are BRIGHT!  I love running in the dark and seeing the headlights look like actual locomotive headlights as they come down the track.  IMO, the old incandescent bulbs look nothing like real headlights.

Got 'er done. All fixed and working just great. Thanks John for the advice and help! All good.

Rod

Rod Stewart posted:

Just noticed that the Vf for white leds is about 3.2v. So dropping 2.8v with a 220 ohm gives I=13ma; which should be fine I am thinking.  Sound about right? 

A 220 Ohm will yield an average LED current of about 7 mA.  It's a somewhat tedious explanation but has to do with the pulse-modulation used in PS2 to drive incandescent bulbs.  With command track voltage, you might have an internal PS2 DC bus voltage (aka PV) of, say, 20V.  To drive 6V bulbs, PS2 electronics drives 20V pulses but only ~9% of the time.  (6/20)^2 = 9% because lamp power goes as the square of the voltage for constant brightness.  With the LED and resistor, when the pulse is ON, 20V thru 220 Ohms into the ~3.2V LED results in current pulses about 75 mA.  But it's 75 mA only 9% of the time so the average LED current is about 7 mA (75 mA x 9%).  

As an aside, if you measure the DC voltage at a PS2 lamp, it will always read much less than 6V.  In the example above, with a PV of 20V, the lamp only sees this 9% of the time.  A DC meter indicates the average voltage so it will measure 20V x 9% = 1.8V.  Kind of confusing if you're expecting to read 6V...but it is what it is. 

My experience has been that the 220 ohm resistor gives me a nice bright headlight, the object of the exercise.  With that value, I'm sure I'm not abusing the LED, so life is good.

stan2004 posted:

Rod Stewart posted:

Just noticed that the Vf for white leds is about 3.2v. So dropping 2.8v with a 220 ohm gives I=13ma; which should be fine I am thinking.  Sound about right? 

A 220 Ohm will yield an average LED current of about 7 mA.  It's a somewhat tedious explanation but has to do with the pulse-modulation used in PS2 to drive incandescent bulbs.  With command track voltage, you might have an internal PS2 DC bus voltage (aka PV) of, say, 20V.  To drive 6V bulbs, PS2 electronics drives 20V pulses but only ~9% of the time.  (6/20)^2 = 9% because lamp power goes as the square of the voltage for constant brightness.  With the LED and resistor, when the pulse is ON, 20V thru 220 Ohms into the ~3.2V LED results in current pulses about 75 mA.  But it's 75 mA only 9% of the time so the average LED current is about 7 mA (75 mA x 9%).  

As an aside, if you measure the DC voltage at a PS2 lamp, it will always read much less than 6V.  In the example above, with a PV of 20V, the lamp only sees this 9% of the time.  A DC meter indicates the average voltage so it will measure 20V x 9% = 1.8V.  Kind of confusing if you're expecting to read 6V...but it is what it is. 

Thanks Stan for that analysis; makes sense. Based on your calculated 7 ma led current, one might be tempted to use a lower value resistor, perhaps a 120 ohm. The led would be brighter I suppose. But the 220 ohm suggested by John does look quite bright, so I think it's OK. And it looks pretty realistic, at least to my eye.

I ran a successful headlight/ BU light function test with the bare engine using a standard GOW bulb in a compatible harness, before even replacing the leds with the added resistors. I was quite happy to find out that the light drivers on the PS-2 board had not been toasted by the earlier misadventure. That would have required a sendoff to GGG for repairs, if they are even repairable.

Rod

The light FETs on a PS/2 board are pretty easy to replace.  Don't toast the PS/3 light outputs, those are difficult to impossible to repair.

gunrunnerjohn posted:

The light FETs on a PS/2 board are pretty easy to replace.  Don't toast the PS/3 light outputs, those are difficult to impossible to repair.

Good to know John. Actually I have just noticed that the newly replaced BU led occasionally flickers on and off when the engine is moving forward. No pattern, just random flickering. When in reverse it is on solid. My initial thought was some anomaly in the wiring perhaps. But maybe the FET is flakey?

Rod

I think I'd start with a replacement of the LED. 

Well actually the led is getting power somehow when it shouldn't be!

Rod

Ah ha, didn't read carefully enough.  Looks like electrical noise.  Not sure exactly what is happening there, I don't see that effect when I put LED's in.

Yes there is something weird happening. Better pop the hood again and re-check for any wiring interference.

Rod

Well opened her up and sure enough, a minor wiring snafu. Pinched wire. Got that fixed up and now all is good. It runs quite well and the forward and reverse lights look great. Thanks for all the help everyone!

Rod

You're lucky the pinched wire didn't take out the FET on the PS/2 board, that's a frequently issue.

gunrunnerjohn posted:

You're lucky the pinched wire didn't take out the FET on the PS/2 board, that's a frequently issue.

Think I dodged a bullet there alright. Might be because it was only an intermittent flicker, and I didn't run it long that way.

Rod