Hi, hoping someone might know if I damaged a decoder... So I've been converting a Kato RDC to DCC sound, and really, everything was going super well, motor works great, speaker works great as well, even added 440 uf of capacitance. Then came the lights. I unfortunately followed a youtube video of what I thought was a good source (TSG Multimedia) and when I rewired the light boards for DCC, I ended up connecting wires wrong. That led me to burning out three of the four LEDs. Then mistake after mistake was made trying to sort out what was wrong in debugging! Okay, so now, all is sorted out, and I decided to bypass the lightboards entirely for testing, and I connected the rear headlight (yellow wire) and red rear marker light (purple) and common (blue) with some spare LEDs, resistors and test leads. Now, the rear red marker light only lights up when the loco is forward and F0 is turned on. perfect! But, the rear headlight is always on. I mean always, it never turns off, even if I turn off everything in the function mapping. At this point I tried everything I could, reuploading everything, but whatever I do, the headlight is always on (it even turns on while programming!) Is it possible I damaged the Rear Light output on the decoder (LokSound Nano) in such a way that it now always on and can never be turned off? Any ideas how I could fix this issue?
The blue wire is positive and the decoder turns the LEDs on by grounding the white, yellow or whatever color the other wire going to the LED is. Sounds like the one that is on all the time is always grounded. Is there any posibiltiy that the LED is ground through another source that the wire going to the decoder? How may function outputs does the decoder have? Sounds like you are using 3 at the moment. If you have another one that you aren't using you might be able to use it for the rear headlight and then remap the new function you are using to turn the rear headlight on and off. Get the decoder to not see the function and wire that currently is being used for the rear light (remap it also). Since the decoder is ESU (probably others also) I'm about sure you can do this as I did something similar on one install. It was complicated using ESU's manual but got it done. Then found out I could of done it much easier using JMRI and DecoderPro. Sumner
Thanks Sumner! I'm not great at electronics, so I'm not sure if the LED could somehow be grounded in some other way. At one point I did connect a bare 3mm white LED to the decoder directly, blue to LED+, and yellow to LED- (through a 470 ohm resistor) and it was still staying on all the time; unless the wire is frayed somewhere that I didn't see, I just don't know. I wish I had a nice decoder tester like you built lol, I may have to revisit that thread of yours and make myself one of these! Though without a filament printer I'll have to make the box out of resin printer. My wife gave me that same idea this morning of just simply bypassing the "Rear light" output entirely, just like you mention! I really should have thought of this too! I do have Aux 5 and 6 that are unused, so if somehow "Rear light" is "permanently grounded", at least the decoder will still be usable. As for programming, it's super easy, barely an inconvenience with the Lokprogrammer interface. Today after work I'm going setup a bread board of LEDs and resistors and hook up each output independently and see which work and which behave strangely. I'd be happy to drive the highlights off any other unused output if that means I can still use this rather expensive decoder! As for the lightboards on the RDC, they're in a rather pitiful state after soldering and unsoldering wires and changing LEDs, I may just do without and use LEDs straight to the decoder with resistors. at least there's plenty of space in the RDC Bonus is that I won't use the awful amber/orange light they put in there for a headlight. In messing around with the wires to get those silly lightboards to work in DCC I'm pretty sure I ended up connecting the yellow and violet wires together and trying to turn on the headlight. It was a dumb rushing thing after working on it all day, I really should have stopped and really traced the lightboard connections and figured out what was happening instead of blindly trying things. Didn't help that the TSG Multimedia youtube video was telling me to connect the common to the wrong part of the lightboard, bypassing the resistor entirely.
Just a normal installation of a decoder scares me. I have a new one coming today I will have to install, and always hope I am doing it correctly, even though I have only done a few, two to be exact, not counting the new one arriving with it's locomotive. I can't even imagine doing what you are doing. Actually soldering wires and such. For instance this one requires Kapton tape for installation. How many brand NEW people to the hobby you think know what it is, let alone have it? I'm also curious if there is a Huge warning about it with the Decoder! And that I apply it correctly either way. Good luck. Hope you get yours working! In some manner that pleases you!
I would think that wouldn't harm anything as the decoder takes those wires to ground when activated completing the circuit from the blue positive thru the resistor and LED to ground. So with them connected together it should of been like connecting two ground wires together which shouldn't hurt anything. Also that would of only happened if both functions were activated at the same time. Now running the LED's with no resistors would probably quickly burn them out and connecting the positive blue to the wrong part of the lightboard might of caused a dead short so I feel that might of been when the problem to the decoder, if it has one, might of occurred. That sounds like a good test and the results aren't good so I'd move onto using one of the other function wires. The decoder tester could be put together without a printer using other items such as styrene for anyone interested in that. Sumner
@in2tech, I watched a lot of videos on disassembly of n scale locomotives by "Watchtrains Now" on YouTube, and that got out the fear of opening these things up and working on them! As for decoder installs, this is only my second sound install after the Kato Es44ac, and I wouldn't have attempted it without a great guide on therailwire forum. As for kapton tape, you only added way too much if the shell doesn't fit afterwards . I'm sure after you've dcc'd a few locos you'll wonder why you thought it so daunting afterwards. Sumner, I did unspeakable things to those wires and those lightboards, I wouldn't rule out causing dead shorts! The angry buzzing I got when I'd turn on the lights (and the lokprogrammer shutting down the test cab) should have told me to stop and get a better look
Warning, LEDs were absolutely murdered in the making of this test So, the test setup.... Ye olde mess of wires, I just connected, in order, from left to right: Front headlight AUX 1: Front red marker lights Rear headlight AUX 2: Rear red marker lights AUX 5: Alternate front headlight AUX 6: Alternate rear headlight So, the above picture shows the track powered, and F0 (lights) is currently OFF. Not stopping Front and Rear headlight outputs to shine bright! In Lokprogrammer, I have it setup such: Basically, in forward direction and with F0 on, I should have Front headlight, rear marker lights, and the alternate headlight on: Ignoring the Front/Rear headlights which is always on, it works, rear marker lights and alternate front headlight is ON, the rest is off. And when in reverse: The front marker lights are on, and the rear alternate is on. Okay, so I really did a number on the Front and Rear headlight outputs, must have fried something in the decoder. Still, it runs forwards and backwards, and the sound output is fine! So I'll just carry on with using AUX 5 and 6 for the headlights, ignoring the original outputs. I also learned that Front and Rear headlight outputs appear to have an onboard resistor. But NOT for AUX 1,2, 5 and 6. Poor murdered LEDs, they died for science. Well, I can move forward at least, and not worry too much about having lost a decoder; thankfully I don't actually need the additional outputs for this RDC, I'm not adding any other lights.
I have a few of those in that condition also. Maybe we need to setup an 'International Final Resting' place for dead LED's and other electrical components that bite the dust during the learning period. Nice job on getting it all figured out. I haven't used my LokProgramer yet. Will on some of the sound installs coming up. Sumner
Sumner, I think you've got the same symptoms as me, you buy stuff then never use it (resin printer, lokprogrammer )! My last such purchase was all the stuff to make etched metal parts. Still sitting in a drawer! Tonight I replaced the LEDs on the circuit boards, added wires, and put them in place. With more alligator clips, I tested it out and the lights are behaving normally. Tomorrow I connect all the wires back together, stuff everything in place and should be done! One thing about these RDCs, the lightboards circuit only has a 271 ohm resistor, not enough for the LokSound. I measured it with a multimeter and they were being driven at 36 mA. I solved it by dimming the LEDs to a setting of 14 for the white LEDs and 12 for the reds (full power setting for those CVs is 31). Now they're roughly measured to 19 or so mA.
Yep, pretty sad, but I have built two different work stations for the resin printer and I did get one sound decoder with no sound file loaded so will eventually (probably) use the LokProgrammer to put a sound file in it. I tell Dottie that if things got tough financially I've now got enough to keep me busy for this life time but wonder if I maybe need more cork . Sumner
I'm (almost) done! I did most of the programming to slow it down, get the right horn and bell and slow speed performance. Then made crappy video and got it here for you all! Really sorry for the quality and background noise, I'm not a videographer, hehe I'm almost done because I just noticed in the video the pilot seems crooked, hmmm. Going to open it up, take the picture I forgot to take earlier, and get whatever is not snapped in place back to it's normal snapped position! I'll upload a picture of the innards afterwards.
Turns out the crooked thing was a non-issue, it snapped back in place. Here are the insides! 3d printed enclosure (with longer "wings" to hide the wiring), the decoder and it's mess of wires, the 440 uF capacitor, and the new LEDs! When I paint it in Via Rail scheme and add decals, I'm also going to get a small U-shaped channel of styrene that can fit between the seats and hide the wires going to the back. I think it will look better to have a grey painted channel of plastic instead of the colourful wires. Though they can barely be seen anyway, the interior isn't lighted, and the seats and passenger area is kinda nothing to look at anyway!
I first got into DCC almost 10 years ago - in late January 2013. One of my main reasons for making the jump was to have sound. So I bought a Digitrax sound decoder for my Kato Amtrak P-42 - It's a safe subject since there's a lot of real estate inside the shell to work with (I just stuck the speaker underneath the radiator fan). The only thing I had to solder was the speaker wire and the capacitor, but the stock capacitor was too large for any place to tuck it inside, so for several months I had a P42 with the nose sticking higher off the ground than the rear. Then someone told me about tiny tantalum capacitors, so I bought some, soldered them in and could finally shut that shell closed. But then later on I wanted flashing ditchlights, so I bought SMD LEDs with resistors and soldered them to the decoder and installed them behind the ditchlight housings... I always say that DCC is a hobby in itself! The moral of my story is that learning it is always incremental. In November I did my very first wired sound decoder install - an ESU LokSound 5 Nano inside an Athearn F59PHI. Wiring the decoder was the easy part - the tough part was the hedlights, ditchlights and reverse marker lights! The whole project took about 2 weeks to do, but the lights came out perfectly and I want to run the F59PHI more than my P-42s now!
