I have just installed a digitDig dh166 decoder in my ten year old Rivarossi Allegheny. I wired up an LED with two 680 ohm resistors in series for a flickering firebox. The "flicker" is set up on cv 51 for F1. By pressing F1, I can turn the firebox on and off... UNTIL I set CV 49 (headlight) to 104 to enable "rule 17 dimming (F4) and headlight dimming in reverse." Now, F1 no longer will turn off the firebox. It stays on all the time! If I reset cv 51to anything OTHER than the "random flicker", I can turn it on and off with F1. WHAT? If I reset the headlight back to zero (no dimming) I can then turn the flicker off with f1! Is there a way to get F4 dimming for headlight AND be able to turn off the firebox flicker? ANOTHER QUESTION: There's a zener diode, a diode, and a 150ohm, .5 watt resistor on the onboard circuit board to limit the current going to the (incandescent) headlight. When I turn the headlight ON, the resistor gets really HOT! It's been doing this for a while, as evidenced by marks on the board. Why is it getting so hot? Should it be doing that? As soon as I turn the lights off, it immediately gets cool...! Should I replace the resistor? I checked it with my ohm meter, and it read 148.4 ohms. That's ok, right? This is the main reason I want to dim the headlight! Do I need a higher wattage resistor, or "how hot is too hot?" PS: This is exactly what the previous decoder (DN136) was doing, right before it BLEW UP...!!
Mike C.
A half watt resistor is almost certainly too small for what is probably a 30ma or maybe even higher current incadescent bulb. Between it being probably high current (relatively - an LED with resistor is 9ma or so!) and the inrush current an incandescent bulb has - that's probably stressing the decoder function output.
The problem of rule 17 conflicting with the firebox flicker is that there aren't enough effect generators for each function to have their own, so if you use rule 17 then the flicker won;t work on F1 at least. You might try remapping and using the violet wire instead of green. Make CV36=4 will allow F1 to control the violet wire, and hook that to the firebox LED and use CV52 for the firebox effect.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Would you recommend a 1 watt resistor, or more? (I don't have much room). This is a weird setup anyway. The resistor is in a parallel configuration, bridging the zener and the diode, on TOP of the circuit board! Looks kinda messed up to me. It has been this way for ten years, but two decoders have burned up during that time! I keep the headlight dimmed most of the time during operation, which seems to cool the resistor a bit. Is a 1watt resistor much larger than a .5? I have to rely on Amazon for my components (250 per bag!). If I parallel another resistor along side it (a value I might already have), could that cool things down, and put less stress on the decoder?
Yes, a 1 watt resistor will be physically larger than a .5 watt one. ANd of course you want to keep the heat away from plastic bits that might melt.
Might be easier to replace the headlight with an LED. Remove that board and use a 1K resistor.
What I think is going on there is that circuit is great for constant lighting on DC power where you seldom run continuously at a full 12V. The decoder is putting out at least 12V on the function output whenever the headlight is turned on. Whatever the zener rating is, all the extra voltage is being handled by the resistor. If it's a 3V bulb/zener, then you have 9+ volts at whatever the bulb current is, whereas running the loco on DC at half throttle, you only have 3 volts. That's why it probably never got hot on DC.
I'm not quite ready to do the LED thing yet. I will one day, after the bub blows (I need "motivation"!!) If you think the 150ohm-1watt resistor will help with the heat and not "stress" my new decoder, I want to try it out. This would be quick and easy! I'm pretty sure I can make room, and if the one watt resistor is cooler, shouldn't be a problem... right? Even the 1/2watt hasn't melted anything...yet.
The 1 watt resistor will be giving off just as much heat as the .5 watt one, it will just be spread over a larger area. The current through it will be exactly the same - just the bigger size is capable of handling more heat without melting. If the .5 watt is hot, a 1 watt will at least be warm. If the 1 watt gets hot, it would be amazing that the .5 watt even survived.
Aside from removing the board altogether at this time, and hardwiring the decoder, (this board has the socket), what would be the best way to reduce the heat on the resistor? Would a 220ohm resistor work better without dimming the light too much, or would the 150/1watt be good enough? After all, it's just around 16 years old, and has pretty much a new, well, rebuilt motor that will draw barely more than .5 amps under stall. (The motor out of a BLI A class is exactly the same). 1. 1watt resistor. 2. 220ohm resistor, or 3. Leave the darned thing alone? It has been running for the last eight years with a teeny, cheap decoder in it, and a HOT resistor. I just didn't know it...
Well, I found out if I set the headlight as a "Mars Light", the resistor cools down considerably! Did any Allegheny locos have a Mars light for a headlight...? Sure....
Like the similar question on F unit diesels - not as the ONLY headlight. The regualr headlight has to shine steady, you can ADD additional warning lights, but the main headlight would not oscillate.
The resistor is cooler because when using the Mars light effect, the bulb is not running at full voltage all the time, just part of the flash cycle.
Yeah, I kinda knew that locos didn't "just" run a mars light for a headlight. Much less the old H8's! (I was being sarcastic...) I am running rule 17 dimming, so I will keep the headlight dimmed most of the time, which seems to keep the resistor "comfortably cool" until I can find a different resistor, maybe try a 1 watt. The thing is probably ok as it is, but hot components on a circuit board just make me nervous...