Voltage of stock Fox Valley GEVO LED?

[MetroRedLine]

Does anyone know the rating/voltage of the stock dark amber LEDs in the Fox Valley Models ES44 GEVOs?

Mine is too dim and doesn't bend low enough to illuminate the ditch light piping correctly. I plan to replace it with a bright golden white SMD, but I want it to work. I believe I see a resistor on the circuitboard already, so I just need to remove the stock bulb and solder in the SMD with short wires.

[Greg Elmassian]

You also, and actually more importantly, get the current also. LEDs are controlled by current, not voltage, almost all white/amber LEDs are 3.7 volt drop.

Greg

[peteski]

You also, and actually more importantly, get the current also. LEDs are controlled by current, not voltage, almost all white/amber LEDs are 3.7 volt drop.

Greg

Red, amber, and standard green LEDs usually have forward voltage of 2.1V or less.  Blue, white, true-green, and UV LEDs use different chemistry and those have forward voltage of around 2.6 to over slightly 3V. But I have never encountered any discrete LED with a forward voltage of 3.7V  Maybe some high-power white LED (used in lighting fixtures) can come close to 3.7V while consuming 1A or more of current, but not the small SMD LEDs we used in N scale.

[wmcbride]

Does anyone know the rating/voltage of the stock dark amber LEDs in the Fox Valley Models ES44 GEVOs?

Mine is too dim and doesn't bend low enough to illuminate the ditch light piping correctly. I plan to replace it with a bright golden white SMD, but I want it to work. I believe I see a resistor on the circuitboard already, so I just need to remove the stock bulb and solder in the SMD with short wires.

For the early-issue FVM GEVOs with the dim lights, I just used a white LED off a stock Kato circuit board from an engine into which I had installed a decoder. I also used some white LEDs I had purchased off eBay in others. The LED leads were long enough to allow it to be positioned to better illuminate the light tube and the difference was quite noticeable. Of course, SMDs would be great; I just had the larger LEDs lying around.

[mmagliaro]

As others are already telling you, this is about current, not voltage, with LEDs.  You might put a white LED in there, but it might still be dim (even though you may be able to position it better so you get more light into the light tube).

If it's still dim, you'll need to change that resistor you see in there.  Can you tell us what value it is?

[peteski]

In my experience, simply replacing a yellow LED with a white one (without changing the value of the current limiting resistor) will result in much brighter light output anyway. Yes, the current going through the white LED will be smaller (because of its larger voltage drop), but white LEDs are way more efficient than yellow LEDs which more than makes up for the lower current.  The white LED will appear much brighter than the yellow one it replaced.

[Greg Elmassian]

Is Texas Instruments a good enough reference? http://www.ti.com/lit/wp/slva130/slva130.pdf

I meant to put 3.1, not 3.6 on the white led.

The difference between red forward drop and even white forward drop is not a lot if you are running from 12 volts, but makes a difference in the current limiting resistor if you are running from 5 volts for example.

The bottom line is get the specs on the LED you are replacing or measure it in situ.

Greg

[peteski]

Is Texas Instruments a good enough reference? http://www.ti.com/lit/wp/slva130/slva130.pdf

I meant to put 3.1, not 3.6 on the white led.

The difference between red forward drop and even white forward drop is not a lot if you are running from 12 volts, but makes a difference in the current limiting resistor if you are running from 5 volts for example.

The bottom line is get the specs on the LED you are replacing or measure it in situ.

Greg

No  references needed. 3.6V is way too high, and 3.1V is perfectly reasonable value for a white LED.

You are 100% correct that for precise biasing of an LED, knowing the specs is vital.  However, even the technical specs show a range of the Vf (MIN-TYPICAL-HIGH), so there are expected variation of the Vf within the same LEDs

I'm also approaching this from a model railroader's perspective (not from electrical engineer's perspective). Not only I don't expect an average model railroader to be super-exact when calculating the resistor value, there are also many variables in play.  The decoder's common positive voltage is not a fixed or well-defined exact value, the desired LED brightness is also very subjective to each individual.   Also, the LED installation plays a role on the choice of brightness. If the LED is directly in the headlight housing, it doesn't have to be as bright as one that is remote to the headlight and uses a light-pipe to direct the light to the headlight.

That is why instead of performing of accurate voltage measurements, looking up technical specs for the LED, and then doing accurate calculations, my advice is to simply start with any white LED and 1k resistor, then go from there (until its brightness is acceptable to the modeler).  For this specific application, it really isn't rocket science (but more of an experiment in visual brightness tuning).

[Greg Elmassian]

The hole in your procedure is "until brightness is acceptable to the modeler".... I've seen too many people go over 20 ma on a led to get the brightness they wanted, or stress an LED so much the color changed...

Not all modelers have expectations that can be accomodated

Greg

[peteski]

The hole in your procedure is "until brightness is acceptable to the modeler".... I've seen too many people go over 20 ma on a led to get the brightness they wanted, or stress an LED so much the color changed...

Not all modelers have expectations that can be accomodated

Greg

I agree, but also it seems that our experiences seems totally opposite. In N scale modeling (which is basically what most people on this forum model) and and starting with 1k resistor (which in the worst case will allow around 10mA through a generic white LED) modelers usually think that is too bright and increase the resistance.   I'm sure you know just how bright and efficient white LEDs are.  Some decoders have 680 resistors installed at the factory.  That will result in worst case current of around 16mA. Still well below 20mA. 

And as you said, once the LED color changes, you are right at the point of releasing the magic smoke and the LED is not longer usable.