I suppose it is possible that the wya the function is damaged it's showing voltage, but can't support any current flow at all. Even a cheap meter had a very high resistence, so very little current flows through it when meauring voltage - this is by design so as not to affect the circuit under test. That fellow Heisenberg said that when we make an observation, we alter that which we are observing, and measuring an electronic circuit is no different. It would be extremely odd for this to be the case, based on the way function outputs are typically deisgned - it should either fail such that it never turns on, or fail such that it is always on, like the controlling transistor is effectively replaced with a wire. On but providing insufficient current for the bulb, that's just wierd. Being that it's already broken, there are some slightly out there things I would try, just to see what is happening, like taking a 1K resistor and putting it across the function output and measuring the voltage across it - a 1K resistor at 2.5 volts will only be 2.5ma which is well below the function capcity and much less than the original light bulb draws, and thus will cause no harm. If the function is working properly, the voltage should still be near the original 2.5V. If it drops to 0, or below 1.5V, then indeed, we have a function stuck on but incapable of providing any significant amount of current. If it still is around 2.5V, then I'd try a smaller resistor, say 470 ohm. That should be 5.3ma, STILL way below what it needs to put out to light the bulb - those Athearn bulbs are at least 15ma, or maybe 20-30ma, no idea if the specs are actually listed anywhere. Also hooking up the bulb and then measuring voltage across it - has to be well below 1.5V if it doesn;t light up. And you cna try putting the volt meter in series with the bulb and see if there is any voltage.
Though about now I would probably be ordering up a Loksound Select Direct and some LEDs and just rebuilding the loco the way it should have been to start with. Depends on how bored I was, and how much time I had on my hands. Doing all those meaurements won;t fix the existing decoder, I'm just slightly curious as to the odd failure mode.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Up until about an hour ago, all of my testing was on the rear function outputs. But, I decided to go all the way. Here are my findings.
14 volt Connector: It works but remains constantly ON. This connector is in the middle of the decoder and provides a bypass to the 1.5 voltage regulators for the Athearn bulbs, permitting LED installation.
Rear Function Output: It measures 2.651 volts but does not light a 1.5 volt test lamp.
Front Function Output: It measures zero volts. However the front headlamp works on a 1.5 volt battery, so the front Athearn lamps are good. The rear Athearn bulb does not light on a 1.5 volt battery.
F5 Function Output: It works, toggles ON and OFF and Mars light operates as intended.
F6 Function Output: It works, toggles ON and OFF and it lights the 1.5 volt test bulb. The F6 oscillates the light, just like F5. Is this a CV control to kill the oscillation?
It is interesting to note that the F5 and F6 functions cannot be lit at the same time. It is one or the other. Is this a CV control that can be defeated so that F6 can control the headlight?
The wiring is a mess. Whatever the source of the heat, wires are "fused" together where insulation melted and there are bare spots where the stranded copper wire is visible.
Also, ever since I took the shell off and "released" the wires where they were pinched, the Mars light is much brighter than before.
On the other A unit without any problems, the Mars light is dim. Eventually, I will remove that shell to see what the wiring looks like. It probably isn't going to be pretty.
Rich
Alton Junction
If the rear bulb is actually dead (does not light on a 1.5V battery), then I'll bet the bulb blew and took the function output with it - that's not uncommon, expecially with incandescent bulbs.
Since there is also a full voltage output for the function, if you have a spare 1K resistor and LED you could try it on the 14V connections and see if it lights. Since it never shuts off, the function is blown and uncontrollable and the only way to fix that would be to replace the decoder. Or use one of the unused ones for the rear light.
rrinker If the rear bulb is actually dead (does not light on a 1.5V battery), then I'll bet the bulb blew and took the function output with it - that's not uncommon, expecially with incandescent bulbs.
Randy, that is a troublesome suggestion that you made, particularly because you just may be right.
Before purchasing the Athearn Genesis Monons and Eries, my only other Athearn Genesis locos were my infamous F7 ABBA Warbonnets. I used to blame my NCE DA-SR decoders with their blown function outputs on the A units, but it was long thought by others to be the Athearn 1.5 volt incandescent bulbs. The answer seems to be to replace the incandescents with LEDs. Once I did that on the Warbonnets, my problems ended.
But, in the case of the Monons, I have to believe that the shoddy wiring job contributed to the problem.
Ever since I took the shell off the problem A unit and "released" the wires where they were pinched, the Mars light is much brighter than before. On the other A unit without any problems, however, the Mars light is dim.
So, this morning, I removed that shell on the problem free unit to see what the wiring looks like. It looks fine. No pinched or bare wires. But, the Mars light remains dim. Now, the question is, which light bulb is in trouble, the dim bulb or the super bright bulb?
All of the CV values on the two A units are indentical, eliminating the possibility of a dimmable light setting.
What is it about those bulbs? Until I replaced the incandescents with LEDs on my Athearn Genesis Santa Fe F7A units, I had nothing but problems, blaming the NCE DA-SR decoders for blown function outputs.
Why would an incandescent bulb blow in the first place?
And when it blows, why does it take the decoder function output with it?
Does the entire decoder light circuit get fried?
Or is it more likely to be the positive common side of the circuit?
Or is it more likely to be the negative supply?
Incandescent bulbs just do burn out - surely uyoi've had light bulbs burn out in your house. And 'burn out' is pretty literal - the filament is glowing because it is nearly burning. The vacuum in the glass bulb keeps it from instantly vaporizing like a flashbulb, but there is no way to create aperfect vacuum so the whole time the light is on, atoms of the filament material are combining with what little oxygen is in the bulb. Eventually a spot will wear too thin to support itself and a break will occur.
Now, if it happens cleanly and the entire thing opens up - no problem, it's an open circuit. But instead what happens is that the loose ends still can touch each other, which is a a complete circuit, but less filament than before - which is lower resistence and more current. Doesn't last long, the higher current quickly melts what's left of the filament and it opens up again, but that momentary high current surge can easily exceed the rating on the function output. On decoder funtion outputs, the controlling transistor 'switches' are on the negative lead - the various color funtion 'outputs', the blue is the posistive supply common, This is called a 'current sink', the function colors are actually the ground returns and when the function is on, it is like a switch connecting the wire to ground. The main advantages are that you can connect the same light to more than one function without harm - so it can turn on when wither or both are turned on, and that the transistor is either fully on or fully off, which is when it is most efficient, with less loss internally and thus less heating. But that momentary surge from an incandescent bulb burning out can be multiple times the current rating of that transistor and away it goes as well as the bulb. They can fail open, and the function never turns on, or they can fail closed and the function never turns off. The positive supply is seldom affected, since it comes directly off the diode rectifier bridge which are often the heftiest components on the decoder. Some decoders use a transistor array chip instead of individual transistors, in which case more than just the connected function could be affected. Rarely does it get back tot he decoder's processor and actually mess something up there, although that has been known to happen if a function wire touches the track power, the function transistor blows but with the voltage and currentup tot he booster maximum going through it, it can also feed back through the link to the processor that normally tells that transistor to turn on and off.
Randy, thanks for that reply. That is insane, if you ask me. What is the answer? To avoid incandescent bulbs like the plague? Especially Athearn bulbs?
Those Athearn incandescents were blowing up on my Athearn Genesis Santa Fe F7As and frying the function outputs on my NCE DA-SR decoders. I kept asking myself if I was doing something wrong with the decoder installations. But this decoder was factory installed by Athearn. This is nuts!
It used to be common to put a small, say 50 ohm, resistor in the light circuit even with bulbs rated for the full 14V, to keep the surge current under control - that's the other piece of the puzzle, incandescents bulbs have an initial surge whent hey turn on - the cold resistence is lower than the hot resistence - often by 10x or more. So when power is initially applied, before the filament glows, it draws much more current then when it is continuously on. That surge can easily exceed the limits ont he function output, hence the addition of the resistor even for full voltage bulbs, to limit that surge current. It's also why an 1156 bulb can be used as a 'circuit breaker' - when the current draw is too low to light up the bulb, the resistence is very low, so they don;t drop much voltage. Increase the current and the bulb lights up, increasing resistence and eating up a good cunk of the power before it gets to the short.
The best solution is, indeed, to avoid incandescent bulbs like the plague. LEDs don;t have these turn on surges, and unless overdriven, generally last effectively forever. If you connect and LED to a full voltage function output and forget the resistor, it will flash a sudden death, and can still take out the function on the decoder, but properly limited, the LED will last forever. Less handling, less opening of the shell - do it once, and never have to touch it again.
Randy, I mentioned at the outset of this thread that the loco had sat idle on my layout for several months.
Could that layoff contribute to the cold resistance of the initial surge when I turned the headlight on?
No, the cold resistence of the filment returns within minutes if not seconds of turning the bulb off. Sitting ont he shelf for months is no different than sitting parked ont he layout for 5 minutes with the light turned off. Or, since this was the rear light, running around going forward for a half hour pulling a train.
Now, if it was subject to lots of vibration, on or off - like working on the layout nearby with saws and hammers, that can weaken the filament and cause it to burn out the next time power is applied.
Rich, My problem too with new Athearn DCC/Sound Diesels. In 2011 I bought an F7A&B Super Chief Set,and a GP15. All three were faulty out of the box,and went back to the U.S. for warranty repair. Due to my "Dale Carnegie" address to the Romans at Horizon,they paid for the return shipping. A friend could not find a decoder in the F7B unit on the JMRI . The F7 A unit had no pick up connectivity on the front truck. The Gp15 had a faulty decoder. All three came back and worked correctly. Athearn is not alone here,as at the same time a new Proto 2000 EMD Switcher ran uncotrolled crazy,and it took the trip across the Pacific.I might add Walthers back up personal service was most appreciated,as their man messaged me several times to try to rectify the problem. A Hornby "Thomas" new with DCC silent decoder went straight nback this year,thankfully I bought it from a local state dealer,anf he found that Hornby had a batch of bad decoders. After replyi to your post,I am going down to my !0x5 US Small Town layout,and enjoy running two good old Athearn DC Diesels,a pair of second hand one owner GE-U30B ,s whichj I have further updated with LED,s etc.,and watch them go around "quietly" with that old Athearn Power,and great wire handrails . I must have removed these bodies twenty times yesterday,fitting the new stuff in them. Hope your equipment runs better for you,we could do without the inbuilt problems. Owen Sunshine Express
Before I put this thread to bed, I thought a final update might be informative.
The good folks at Athearn replaced the faulty decoder and sent a replacement bulb for the one that was burned out. I rewired the loco myself including the bulbs, and everything now works great.
I put the plastic clips back on to secure the wires in place. I decided not to solder the wires yet, because if those Athearn incandescent bulbs go out, I plan to replace them with fiber optic/LED assemblies. Then, I will solder everything in place.