Randy; to confirm, yes I got + 13.1V on blue and white wires at the 8-pin decoder when connected in and - .308V at the connector where the LED plugs into. So it sounds like there's no juice from 8-pin to connector where LED plugs in.....maybe something in pcb circuit route going from 8-pin to connector where LED plugs in?
Steve.
Can you reverse the 2 wires going to the front LED?
Pete
I pray every day I break even, Cause I can really use the money!
I started with nothing and still have most of it left!
When I get home today I'll give it a try but wouldn't reversing the two wires just give me + .308V at opposed to - .308V that I'm getting now? I just thought the LEDs took more power to run....it almost sounds like I'm not getting juice at this switch. In any event, I'll give it a go and hope for the best. Steve
That's why I suggested testing the voltage at the pins for the rear LED which works. If it's also just .308 volts, the low number is caused by the meter interacting with the circuit design and the probem indeed is the polarity on the front LED. Or just reverse the front LED and try it, it won;t hurt anything.
Although looking at pictures of the insides, it seems that the front LED attaches right to the board, and the rear LED is on a seperate circuit board that attachs to the main board, so testing at the connector where the rear LED plugs in may not be an apples to apples comparison.
The wiring from the DCC plug to the front LED is 'fine' since the front light works on DC. There really isn;t much else that could be wrong other than polarity - unless there is another jumper you need to move for DCC operation. But none of the installation pictures shows anythign like this. You won't be able to just flip around the plug for the LED - the shell for the plug is keyed. But if you look carefully you will see tha tthe metal pins int he connector have a tiny tab on one side that keeps them from being pulled out the wire side. You need to dpress this tab and pull the pin out from the wire side, one at a time, then put them back in the opposite way (red ont eh side that was black) to reverse the polarity.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
I'll test the rear plug in point from the rear headlight to board. As well, I'll plug the front headlight into the rear connection point on the pcb and see what happens.
Both the front and rear lights have their own boards which are screwed vertically into the aluminum framing. If I need to reverse the polarity of the front lamp wires I can use a couple of small test prods and stick them in the front headlight plug and reverse the polarity and touch them to the main pcb connection points, to see if this works.
I'm at work so I'll attempt this when I get home in about 3 hrs. Thanks!
Before attempting to switch front and rear light boards, make sure htey both have resistors on them, the pictures I've seen of other Atlas Silver locos show the rear light with a resistor ont he board and the front LED just plugs directly to the main board which has the resistor. And anyway that won't accomplish anything - you already know neither LED is blown out. If there is a polarity problem it would be caused by the diodes ont he circuit board used for directional lighting in DC mode.
Both the front and rear headlights have a resistor. Something must be different though because I tried the front lamp in the rear headlamp socket just to see and it didn't work there.....but both bulbs originally worked.
I just tried the front headlight point again and checked voltage. I must of not had the test prods on the prongs that well. I am now getting - 12.9 V for the rear headlight (that works) and zero volts for the front. Sounds like no juice from decoder circuit to the front headlight point (J1). Sounds like something beyond me...I"m assuming it's a fault in the train's pcb?
Was this link to the Atlas web site of any help?
http://forum.atlasrr.com/forum/topic.asp?ARCHIVE=true&TOPIC_ID=37773
Rich
Alton Junction
Yes, it is the loco PCB if you get full voltage at the blue and white wires but nothing at the pins where the LED board plugs in to. Unplug the decoder and front LED and test continuity from the pin that the blue wire goes to to each pin of the LED socket, and from the pin that the white wire goes in to each of the LED pins. Test with the leads both ways - it's commin for a meter to have the internal battery for continuity have the negative hooked to the red probe. On one wire you should have a dead short regardles sof polarity - that pin has a direct trace from the decoder socket to the LED plug. If the resistor for the LED is on the LED board and not the main board, you can clip whichever LED wire DOESN'T have continuity in both directions and solder it to the appropriate wire ont he 8 pin plug. Or trace the lead ont he circuit board that does NOT have contunuity in both directions - it probably leads to a diode. You can solder the LED board wire to the side of the didoe that is towards the 8 pin DCC socket, that way you can still pull the decoder from the socket if need be.
Oh yeah, sanity check - the ORANGE wire in the 8 pin plug shoudl eb in the hole marked with a small triangle on the circuit board.
Randy, I've got a continuity testor and I'll test, lots of info to absorb for someone who's not too smart with electronics...especially small pcbs. I'm off tomorrow for New Years so I'll have some time in the morning to play with it.
Rich: I went to the Atlas forum, lots of questions though no one had an answer, sounds like their problem was a bit different, more w/ function lights (i.e. mars, strobes..etc). Sounds like my problem is with the pcb. Anyhow great info in case I ever run into those issues with the Digitrax decoders.
Hope you all have a great and safe New Years Eve...I'll touch base when I get some results, thanks for all the pointers, Steve
Ok here's what I got with the continuity test;
Blue Wire = with negative continuity lead on blue pin and red continuity lead on LED board plug both LED pins cause continuity tester to light up. When continuity leads are reversed, LED pin closest to J1 (pin wire color is black -) symbol causes tester to light up, on other pin, continuity tester doesn't light up (pin wire color is red +)
White Wire = Exact same results as blue wire except using white pin
The orange wire is in right location small triangle on circuit board.
I was kind of confused toward the end of your suggestions. Let me know what you think. Thanks, Steve
Hmm, that doesn;t sound right at all. This was witht he decoder unplugged, right? Check also with the LED board unplugged, so just between the 8 pin socket and the 2 pins where the front LED plugs in.
Also try using your other meter set to the ohms setting instead of one of those check lights - dead short or say a piece of wire should show close to 0 ohms, open circuit would show infinite ohms. The check light type of contonuitity tester is usually higher voltage than a multimeter. Either way when doing continuity checks be careful of what might be in the circuit since you will be effectively applying power when doing a continuity check.
My suggest was, because the LED board itself has the resistor, to just bypass the junk on the circuit board and connect the LED essentially right to the decoder by soldering one of the wires to the appropriate decoder wire on the 8 pin plug. Or so you don;t have to solder the wire tot he decoder plug, makign it harder to unplug, I'm willing to bet the circuit trace from eithe r the blue pin or the white pin goes through a diode before it gets to the 2 pins for the front LED. So instead of soldering the wire to the 8 pin plug, you could solder it to the circuit board on the side fo the diode that connects to the 8 pin socket.
By this time I'm pretty sure I would haven given up, ripped up the stock circuit board, adn just soldered the decoder in place. And with all these back and forths, you can see why I usually do this when installign decoders - if I solder the decoder in place I KNOW the white wire is getting ot my front light, the red wire is getting to the right side truck pickups, the orange is going to the motor, etc. Using a factory supplied plug you are at the mercy of hoping there isn;t a mistake on the factory board, which it so far seems to be.
The continuity test was with the decoder out and the LED plug unplugged.Again checked blue and white and got same results as mentioned in the earlier post. I"ll try a by-pass and go directly to either the diode point or directly to the decoder. Thanks
See if this makes sense. In looking at the pcb on the Trainmaster there are two diodes, let's call them DIODE A and DIODE B. i saw something that just looked out of place on the board. On DIODE A let's say one end is 1A and the other end is 2A and on DIODE B it's ends are 1B and 2B. Bear with me and see if this makes sense.
On the rear headlight that works 1A (DIODE A) and 1B (DIODE B) on the diodes are connected together and go to the negative pins on both headlights (front and rear). The positive pin (headlight plug) is connected to 2B on DIODE B. Here's the catcher.......
On the front headlight that doesn't work 1A (DIODE A) and 1B (DIODE B) on the diodes are connected and go to the negative pins on both headlights (just like the working headlight) but here's the kicker, the positive pin (headlight plug) on the non-working light is not connected to nothing, absolutely nothing.........shouldn't this connect to 2A on DIODE A?
When I test for continuity in the working scenario, I get continuity. When I test for continuity in the non-working scenario and touch one end to 1B on DIODE B and 2A on DIODE A there is continuity. It sounds like it was left as maybe a ground for DC (when it worked) and needs to make a complete circuit in order to satisfy DCC (?) My best guess.
It could go in a through-hole and be routed on the other side of the circuit board. Somewhere all that mess is also probably connected to the motor (when the dummy plug is installed) so that the motor gets a coupel of volts less than the LEDs (so that the LEDs cna light up before the loco actually moves.
If the lights are directionel when run on DC, I'd expect the two diode setups to be opposite one another, not exactly the same, so that one gets juice in forward and the other in reverse. It could be that removing the dummy plug and installing a decoder doesn't properly bypass this
One other check, didn't you have problem with oen of the wires coming off the 8 pin plug? Make sure they are all in the proper place. Starint as the orange wire and goign across, it shoudl be orange, yellow, blank, and black. Across from the orange should be red, blue, white, grey.. If the white is in the wrong palce that's the only other thing that would give you voltage at the white wire but not at the LED connection.
I still say clip the LED wires and solder directly to the 8 pin plug, the red shoudl go to blue and black to white, but test it first in case the LED polarity isn;t the obvious red for positive, black for negative.
I checked both sides of that pin and it's definitely not routed on either side....just interesting. I'll solder like you said but isn't the diode needed...? Not sure is it preventing backflow to the decoder or is it a rectifier dioder changing ac to dc...so I can go directly from blue and white to LED pins? I'll check polarity as well. Thanks
As long as there is a resistor, the diode is not needed. The diodes are to block power to a given LED when the loco runs the opposite way on DC. Sinc ethe decoder controls which output is turned on for a given direction, there's no need for diodes to control anything. By default the white wire will be on only when the decoder is going forward.
Blue wire is always +, the white or yellow or other function wire is always the -.
I wish there was a good picture of what all this looked like - the TCS web site has an install for a Silver TrainMaster but the page shows a B23-7 not the TM, and it has a circuit board for one LED but not for the other. ALlt he other SIlver series locos seem to have the same circuit board though, next to the 8 pin plug are 3 diodes, D1 and D2 are round barrel types, D3 is a flat surface mount type. In some of them I can see the wya the traces go, it's D1 that appears to not be connected to anything (because the traces are on the other side). I also see int he pictures that there is a diode on the dummy plug? Also, for the resistors on the LED boards, what are the colors?
There are (2) resistors that appear to be labeled "681" for both R1 and R2 on the front non-working headlight pcb.
On the working rear LED pcb, there are (4) LEDs and which appears to be (4) of these "681" resistors inline.
These resistors have no colors associated with them. I'm finally starting to understand what's gong on. I'm going to solder directly to the 8 pin, I'll let you know how it turns out. Thanks!
OK, once again the pictures of OTHER Atlas locos shows standard resistors on the LED board, yours has surface mount. 681 is 680 ohm which should be fine for full DCC power.
Did the soldering. The polarity is reversed on the wires, so black is +, red is - on the LED. On eight pin I soldered red to white and black to blue.
With this I tested. In the forward position things look good, front headlight is finally working! But when switching to reverse direction, both front and rear headlights light up. It doesn't appear that two wires nothing is accidentally touching on the eight pin. Any ideas.....getting very close now finally!
If there are no wires touching, it could be the decoder programming. Do a proper reset, NOT the recovery macro in the PowerCab - just program CV8=8. See if it works then, since the decoder default is directional lights, only the correct one should be on.
I reset the decoder, no go. Both front/rear still go on in reverse direction. In forward direction, only front headlight works (which is the main thing). I re-checked soldered wires, nothing touching to cause both lights to go on. I went to program track then selected option 2 which was program CV then it asks for CV number and I put in 08. I went back to check engine and it doesn't respond so I assumed that this clears to decoder, removed loco from track then put back on and then reprogrammed loco and lights still no functioning correctly, but I do have a front light which was what I striving for. If someone has an NCE starter system can you let me know if I did the right thing to clear the decoder........thanks.
That's the right thing to clear the decoder, it shoudl respond on 3 instead of the address you had in it. What do you mean by "reprogram the lights"? If you just want to have directional lights, there is nothign to reprogram after the rest other than the desired address. Those lighting settings are, once again, for NCE decoders only, the settings they do will either have no effect or modify something that shouldn't be modified on a Digitrax decoder.
Did you only solder the forward LED? There may still be some sort of sneak path in the circuit. Try it with the rear light unplugged - if the front light now goes out in reverse, that's the problem,a nd you'll probably have to solder the wires from the rear LED as well, only this time using yellow and blue.
For another check, to make sure the decoder has the correct settings, program CV49 and CV 50 both to 0. Also set CV33 to 1 and CV34 to 2. This is all the default function control in a Digitrax decoder.
This thread has turned boring.
HAZMAT, you are lucky that Randy has endured and hung in there.
As I re-read this entire thread, it seems that there is nothing wrong with the NCE DCC sysyem, and there is nothing wrong with the Digitrax decoders.
The problem seems to rest squarely on the OP. The decoder chips were not initially installed tightly and the wires were not soldered to the decoder resulting in loose connections.
Correct me if I am wrong.
Rich.
You are basically correct. This lighting problem should have been a separate thread. The OP needs a better understanding of basic electronics and to understand that he can not use NCE effects on a Digitrax decoder. I don't know how it can be put across any clearer than what has been posted. Randy was very patient with him and probably getting a little tense near the middle of the sixth page. DCC has a bigger learning curve for some.
HAZMAT9 I reset the decoder, no go. Both front/rear still go on in reverse direction. In forward direction, only front headlight works (which is the main thing). I re-checked soldered wires, nothing touching to cause both lights to go on. I went to program track then selected option 2 which was program CV then it asks for CV number and I put in 08. I went back to check engine and it doesn't respond so I assumed that this clears to decoder, removed loco from track then put back on and then reprogrammed loco and lights still no functioning correctly, but I do have a front light which was what I striving for. If someone has an NCE starter system can you let me know if I did the right thing to clear the decoder........thanks.
What you did is not clear to me, maybe this is what you did but here it goes.
In program track mode select option 2. Then input 8 for the CV number and enter. Then input 8 for the value to enter into CV8 and enter. Done, get out of programming mode.
Jack W.
Well I do apologize for any I have brought to some but I attempted to read as much as I possibly could on DCC and as it turns out, this problem hasn't really anything to do with DCC but with a defect in the board. You are correct, maybe it should have been in a separate post somewhere else but just receiving a new DCC system I didn't know what to expect and I did learn an immense amount just being here and being able to run through these troubleshooting techniques. Yes my electronic knowledge is weak but it's not so much a purpose of getting the train to have "perfect headlights" I just wanted to understand the complete picture and have so far learned much.
I do appreciate all of you help especially Randy's. And if he's ever in Phoenix look me up I owe him a steak dinner! And again for those "bored" with my post I do apologize but you need not read. I've always thought of this place as a place of knowledge and this really is the first time I've ever asked any question of substance. And since 2004 I've given my two cents and have been very patient even answering the same question numerous times to "newbies." And yes, the problem is finally fixed, both front and rear lights are operational and independent.
Yes this is the first time I've ventured into the DCC area, sorry for the intrusion and many thanks for all the help and you can rest easy that this thread will now sleep. Best Regards Steve
The soldering wires thing is somethign that should NOT have to be done - there appears to be some sort of problem with this locos "DCC Ready" circuit. It wouldn't be the first time. I think this shows a great example of why I typically, even on a loco that is supposedly "DCC Ready", I rip out the factory electronics and wire the decoder wires directly to the truck pickups, motor, and lights. Learn to solder with a light touch, use heat shrink tube for insulation, and you'll know things will work because you know you connected the decoder inputs directly tot he track pickups, the decoder output directly to the motor, and the light functions directly to the lights. Just remember to put a piece of shrink tubing on the wires BEFORE you solder them together.
I appreciate your help Randy, thanks for your patience. I did have to solder the rear headlight to yellow and blue, now everything is correct and working good. And I definitely did insulate, but I used electrical insulating goop....just a pin head's worth stuff works great, though I didn't know I could get heat shrink with that small of a diameter. Email me if you're ever in Phoenix, the steak is on me Best Regards, Steve
Steve,
First of all, congrats on solving your problems and getting everything fully functional. And, hats off to Randy for staying the course and lending invaluable assistance.
Second of all, I am going to retract the word "boring" as. perhaps, unduly harsh. My point in using that word was that the thread had become bogged down in a dialog between you and Randy on what had obviously, or at least apparently, been nothing more than a loose wiring issue. We have all been there, though, so I don't fault you. My first attempts at installing decoders led to gently pressed chips that didn't make full contact and wires that were simply wrapped around the decoder leads because soldering seems too final and, as a newbie to DCC, I was intimidated by the finalization aspect of soldering. But, like you, that led to false readings on the programming track.
Third of all, instead of buying Randy that steak, which he does deserve, use that money to buy a decoder tester. That would have spared you a lot of grief. All of your problems would have been favorably resolved through the capabilities of a decoder tester, leads, wiring connection, lights, everything.
Lastly, you are right. Characters like me could simply stop following the thread if we lose interest. However, this thread interested me from the outset, so I wanted to stick around for the resolution phase. As an owner of an NCE DCC system, it concerned me that the "NCE Doesn't Run" isue was raised, since NCE has a pretty good quality control system. Potentially faulty decoders are another issue of concern and interest. And CV problems are of interest to all DCC users. So, it wasn't a matter of simply abandoning interest in this thread. It made for a good detective story. Once we realized that it was not Colonel Mustard in the Dining Room with the Candlestick but, rather, that Miss Scarlet simply fainted at dinner, we grew frustrated.