I was going through a drawer in a cabinet today where I store decoder manuals and some old decoders.
Some of these decoders were removed when I converted locos back to DC from DCC before selling them. They are still perfectly good and available for reuse.
A few are marked "fried" from stupid mistakes that I made as a newbie to scale model railroading. I should just toss them.
However, I also have a few decoders that work, but the lighting function output is marked as "inoperative". And this raises a question in my mind.
I understand that low voltage incandescent bulbs and unresistored LEDs will blow immediately when connected to function outputs producing voltage in excess of the current capacity of the bulb or LED. But, what causes a lighting function output to blow?
Wouldn't a higher voltage incandescent bulb, for example, simply not light if the lighting function output voltage were too low?
What am I not understanding here?
Rich
Alton Junction
RichThe Functions are a simple solid state switch that switch the low or negative side of the circuit. The blue wire being the high side or positive, normally around 12 volts.This is a simple block diagram of a typical decoder.The Function output is a simple switch. The solid state switch current is typically 100ma. Mel My Model Railroad http://melvineperry.blogspot.com/ Bakersfield, California I'm beginning to realize that aging is not for wimps.
Thanks, Mel, but what causes a lighting function output to blow?
Too much current or a momentary short. Mel My Model Railroad http://melvineperry.blogspot.com/ Bakersfield, California I'm beginning to realize that aging is not for wimps.
I use the #4 function output to control the 1.5 volt micro bulbs in the marker lamps in my articulateds. I use the mini DC to DC Buck Converter (17mm x 11mm) between the blue wire and the function to adjust the voltage to 1.4 volts at 30ma. Mel My Model Railroad http://melvineperry.blogspot.com/ Bakersfield, California I'm beginning to realize that aging is not for wimps.
If you do something like hook an LED up without a resistor, when it blow,s it can go to a short condition - if you look inside one of the clear types you can see the fine wire fromt he one leg to the top of the die where the actual diode junction is - if it blow from over current, it can easily form a short and then stays a short until the current melts that fine bond wire. Which can be long enough to fry the driver on the function.
Incandescent bulbs, even ones rated for the full voltage, can have a high iunrush current, which is why a lto of times it is recommended to use a low value resistor even with bulbs rated at 14 volts, to reduce the inrush current. The resistence of the filament is lower when cool, so when first turned on, more that the rated current is drawn until the bulb heats up. The rated running current of the bulb may be within the limits of the function output, but the inrush could exceed wat the decoder can deliver. This may work fine at first, but with repeated cycling, it stresses those tranistors and eventually they could fail.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Thanks so much, Randy, for that thorough explanation of the problem. So, the bulb or LED itself can cause the lighting function output to fry. That clears up a mystery which has been with me for so long a time, wondering how a lighting function output gets fried - - - and without frying the entire decoder.
I mentioned in my original post that I was looking through a drawer full of decoder manuals that I have held onto over the years.
Nowadays if I buy a locomotive, it is almost always factory equipped DCC with sound. But back in my early days of scale modeling, I bought a lot of DCC Ready locomotives and installed my own non-sound decoders.
However, sometimes I still buy an occasional DCC Ready locomotive and install a non-sound decoder. These locomotives are often factory equipped with incandescent bulbs.
What I find so frustrating is the lack of information on these incandescent bulbs and on the capacity of the lighting function output on the decoders. I acknowledge that part of the problem is me and my limited understanding of electrical current.
Here is one example. In the manual for the NCE D13SRP, it reads in part:
Recommended Lighting: If you use LEDs we recommend a 1K ohm 1/4 Watt series resistor in each function lead:
NCE #524-218 Golden White LEDs.
Miniatronics #12-310-05 Yellow Glo White LEDs.
Miniatronics #18-712-10 (12v) or #18-014-10 (14v) incandescent bulbs
Function output ratings: Due to the high in-rush current of incandescent grain-of-wheat type bulbs (about 10 times their normal operating current) function outputs are rated at 40mA each if used with incandescent bulbs. We recommend the Miniatronics part number mentioned above. If you wish to use 50-100mA rated lamps we recommend a 22 ohm 1/4 Watt series resistor in function leads with each bulb (this will also greatly extend bulb life).
So, in this example, function output ratings are expressed in milliamps.
In another example, the instruction sheet for a Proto 2000 GP38-2 reads in part:
Conversion to DCC: Because DCC systems operate at different voltages than standard electrical systems for HO scale trains, you will need to replace the existing light bulbs with light bulbs rated at 12 volts/40 milliamps.
To me, that leaves open the question of the capacity of the lighting function output on the selected decoder. Is it 40 milliamps?
One last point of confusion for me. If I consult the table on the Miniatronics website for incandescent bulbs, for example, there are several different milliamps for each voltage (i.e., 1.5 volts, 12 volts, etc.)
Mel's diagram shows why you can blow a light function without blowing the whole decoder - conversely you can blow the motor drive on a decoder but still have the functions work, so a decoder with a blown motor control could stiull be usable as a function only decoder.
Incandescent bulbs are resistive loads, so they come in varying current ratings. There's a voltage, by design - 1.5, 3, 12, 14, etc. ANd then there is a characteristic resistence when lit, from which the current is determined. This all depends on the material, length, and thickness of the filament used. Those NCE instructions illustrate the inrush for incandescent bulbs. 40ma I don't think is the function rating for absolute maximum load you can put on one - rather it's the maximum light bulb where the inrush will stay below the meximum that the decoder can handle. Specific to that NCE decoder. Elsewhere in the instructions it may state the actual limit.
Most manufacturers state a current limit in their spec. Note that the total may be less than the number of functions x the max per output. A decoder may list 100ma as the max per output, have 6 total functions, but then say max 500ma for all functions together. That's a limit of either the driver chip containing the 6 transistors, or the recifier circuit feeding the blue common wire.
And this is why LEDs are so useful, especially if you really want to light up a modern loco with all the lights they typically have - a white LED with a 1K resistor is under 10ma, and the high brightness SMD ones often need much larger resistor values to not blind you, reducing the current per LED even more. So you might end up with 10 LEDs all controlled by the same function wire consuming only 50ma total. You can pretty much go crazy with adding SMD LEDs and not overload the decoder. You'll run out of places to stick LEDs, or room for all the wires, before you overload the functions.
Thanks again, Randy. All of this information is so helpful to my understanding.
Another question for you on lighting function outputs. When the lighting function output blows, is it typically the Common (+) tab or the Output (-) tab that blows or both? I think I have stated the positive and negative sides correctly. For wiring purposes, blue is typically the Common and white or yellow is typically the Output.
Depends on which way you want to say current flows - there's the electrical convention and the physics convention.
But yes, the blue common is +, each function 'output' is the - side of the circuit. The easiest way to think of the function wires is that each one is a switch between the light and ground.
The feed for the blue wire is capable of handling the full current of all the functions, so unless you overload a function by a lot, the weak point is the individual driver (one of the colored wires), not the blue wire. If you get an internal short because of a wire touching the hot chassis or something, you can blow the blue one, but that will probably also blow most of the decoder. The blue wire connects not just to the FUNCTION positive common, it's the common for the entire decoder, which is why it makes a good attachment point for the + side of a keep alive if a dedicated socket isn't present on the decoder.
Once again, thank you, Randy. I appreciate all of your comments and advice.
We'll make EEs out of everyone of the forum if it kills us!
rrinker We'll make EEs out of everyone of the forum if it kills us!
The more that I convert locomotives from DC to DCC, the more I subscribe to your approach, Randy.
1. Take out the DC lighting board and hardwire the decoder.
2. Remove the factory installed incandescent bulbs and install LEDs.
If it still has a Life Like box - don;t forget just how old it is. Walthers buying LL happened 16 years ago.
They used low current bulbs to prevent the shell from melting around them, and low voltage to work with their directional constant lighting. Digitrax was the first I think with a true drop in decoder for the P2K locos that worked with the stock bulbs, no extra resistors needed. There are others for the form factor, but they require swapping to higher voltage bulbs or LEDs. 20+ years ago, white LEDs weren;t all that, getting ones the right shade of white for a first gen diesel was a lot harder than getting 100 for $5 on eBay. Almost everyone used incandescent bulbs. Some Stewart locos had LEDs - but they were ORANGE.
We've come a long way in the past 20 years! But mainly I want to knwo where did those years go - my older son turned 30 a few weeks ago!
rrinker If it still has a Life Like box - don;t forget just how old it is. Walthers buying LL happened 16 years ago.
I have quite a few of those - basically all the first gen power they produced factory lettered for Reading - 4 GP7s, 2 S-1s, and a few AB sets of FAs. I don't think I paid more than $45 for any of them, most were still sealed - I guess people would rather spend $100+ instead of $40 plus another $2 in Athearn parts and 10 minutes of work to replace the cracked gears, which has only happened on my GP7s.
One of the S1s I got more than 16 years ago (so they were fairly new) for a whole $20. The eBay listing had all the correct keywords, they didn't misspell anything or have it in the wrong category, just no one bid on it. It was NOS from a hobby shop that was closing down and selling their inventory on eBay, too. Seller even said, you got us on that one. Bad week to sell Reading I guess.
Now if I could find a few undec ones - I really don't need any more small switchers and it's a tough race between Baldwin and Alco there for me. I do need a few more GP7s to properly round out the class, but the myriad RS3s I have will haul much of the road freight.