Hi all My so far, not succesfull(?) efforts to set up a Dcc++ base station with JMRI for H0 scalet. Any suggestions? Because so far, the engine does not move. Hardware in test #1 The Arduino Uno (r.3 genuine) + Arduino Motor Shield (r3 genuine) combo. Uno powered only via USB, 5 V and the power connection from the MS cut as instructed on the Github pages and verified with a multimeter. Also cut are both the brake connections on the MS. Vin to the MS was 15V DC, 4.0 A. All tests as described at https://github.com/DccPlusPlus/BaseStation/wiki/Diagnostics---D---Command were succesfull. Using my only and old bougtht second hand H0 Roco engine article no 63414 with a Lenz LE130 v.4 decoder installed via an 8 pin. plug (NEM652) . Track voltage, measured with a Digital Multimeter (AC settings used) reads to about 13,4 V without the loco on the track. Same on both tracks. With the loco on track the voltage dropped with around 1 volt. I can read the same voltage inside the engine on the left and right rail pins on the 8 pin plug! So the engine receives its voltage! The problem In short, the engine does not run on the Ops track. I tried different settings via JMRI especially with the speed steps. (14 vs 128). The decoder could be ”read” and also written to and this small engine makes very small forward steps, almost not visible when reading to the chip. I have to put my finger on it to gently press it against the rails. The rail is new, cleaned anyway and so are also the wheels and the power pick-ups on the inside of the wheels. When the ”analogue” function is enabled the loco moves forward and backwards when the track is powered with the traditional DC signal from an old Roco transformer. Albeit a little bit slower than the normal rushes when full speed is set. When the analogue function is "OFF" in the decoder, it does not run. So at least the decoder is working.. Changes during test 1 In the file CurrentMonitor.h I changed the: #define CURRENT_SAMPLE_MAX from 300 to 600 and also changed PSU to the MS and tried with an old laptop PSU that gives 19V DC out. Still no movement when tried to ”run” but the engine seemed to be a little bit more ”powerfull” when reading the whole CV with a little bit more powerful forward steps.. My conclusion up till now was, after reading in this forum that H0 requires more voltage so I tried with 19V DC in which according to the spec. of the MS is 1V to high. But the MS apparently has survived. The function tests are still OK but no movement. Hardware in test #2 now using a Pololu motor shield MC33926 Same as above but now just a more compentent MS, set up accordingly as described on the pdf at Github (cut traces underneath at 3 positions, soldered two wires etc. Jumper wires. No power connection to the Arduino.) I also changed the settings in the config.h from 0 to 1 to tell the system about the change and also se the CURRENT_SAMPLE_MAX up to 900, saved/compiled and uploaded it. Tried again, now with the 19V DC power adapter. Run the same tests as above= OK. Track voltage now is about 17,5V, AC measured which also could be measured inside the engine before the decoder. The forward steps of the engine when reading the whole decoder CV:s are even more powerfull than before but still NO movement on OPS! At this point I must assume the problem is in the decoder, OR are there some more parameters inside the software to change? Any hints? Bst rgds Leif Jakobsson p { margin-bottom: 0.25cm; line-height: 115%; background: transparent none repeat scroll 0% 0%; }a:link { color: rgb(0, 0, 128); text-decoration: underline; }
If the "D' command pulses all the output LEDs as expected, and you measure 13.4V AC across the tracks, I would guess that the DCC++ system is working as expected. Your description under "The Problem", capable of "reading" CVs and the loco pulsing forward, suggests your have it on the Programming/Maintenance track - not the Ops track. DCC, by default, cannot "read" CVs on the Main/Operations track. Also you cannot run the train on the Programming/Maintenance track - it will not respond to throttle commands. Make sure you have the correct DCC signal connected to the "Ops" rails. Hope that helps. Willem.
Thanks Willem. I do have a separate prog. track, and a separate ops. track. And it is on the later, the operations one, I can not get it to move. The readings and the pulsing are on the prog. track. Leif Jakobsson
Understood. I missed that in your initial post. The station does seem to work, being able to read and write CVs and seeing in-range power on the tracks. Other than maybe a problem loco I do not have any further suggestions right now. I use a breadboard mounted decoder to check for correct behaviour from the DCC++ station. Willem.
You mentioned JMRI. The first thing you need to check is the JMRI preferences all set for DCC++ and not another command station or "Internal"
I am a total layman when it comes to JMRI. Cannot advise on that. However, from the opening post it seems that reading CVs work. That implies the base station received the correct instruction and passed it on to the mobile decoder which pulsed the loco motor (showing that works). The base station was seeing/reading the current pulses, read and replied back with the, I presume, correct CV value. So, can the JMRI settings be correct for the programming track but wrong for the operations track? Unless a known correctly working DCC loco is used, or the DCC track signal analysed, we are pretty much in the dark. EDIT: Question - does issuing a throttle command directly from the IDE Serial Monitor also have no effect? Willem.
The ops and program tracks are separate in JMRI, both have to be checked in the Files/Preferences/defaults window. Check to make sure all are checked for DCC++
one other note... DCC++ (AVR) doesn't support running a train on the PROG track, it is dedicated *ONLY* to read/write of CVs. For testing the loco you need to move it to the OPS track.
Thanks. So we can have a settings problem effecting only the Operations track. I think we have cleared that one. The Programming track seems to work OK. The problem is the loco not moving on the Operations track. Do not yet know if issuing a throttle command outside of JMRI, like from the IDE Serial Monitor, also exhibits the problem. Willem
I will check that, I am sure I had that checked. Anyway, on operations track using the Dcc++ base station, either in the serial console, or in the Dcc++ Controller I can not get the engine to run. I have renumbered it to 2004 as the first engine in Gregs sample file. Yesterday evening I also ruined the Pololu. I do not know what happened, I know I dropped it on the concrete floor from a desktop height. Anyway a new one plus an Arduino Mega have been ordered. As Willem have mentioned I think the problem well could be in the engine, or in its decoder..
Yes, it does not work there either, neither in the Dcc++ Command GUI. For that purpose I "renumbered" my own and only engine to 2004 as per Greg:s first example..
The above pre-empted my next question. (Are you sure you are using the correct decoder address). Does the loco have lights? If so, can you turn them on (Function 0)? Have you tried resetting the loco decoder to factory defaults and then using address 3? As you can guess, I am basically out of ideas. Willem.
Willem, thanks for your support. No it has no lights at all. I have made a reset and also tried in that basic config. As number 3 that is. For the moment, because the Pololu gave up after a fall to the floor yesterday, I am stuck. I will first look for another decoder, if not that, another engine to try with when I receive my replacement Pololu. About 2-3 weeks because it was currently out of stock. Found them in Spain, Electan. The most logic explanation is the decoder. It seems to work (reads and writes), but could it not be faulty in some functions anyway? It is an old engine with what I think is a factory installed Lenz decoder. The manual to the engine is printed in 2001! After all the engine runs in analogue mode and the power is there at the pins before the decoder. Until then I will recheck the Pololu to see if there are some broken soldering, due to my clumsiness when it fell to the floor. I can no longer switch on power to the tracks. Thanks again, I will inform more later on.
Thanks for the update. While waiting for your replacement Pololu, your Arduino motor shield giving 13.4V on the rails should be ample to run a HO loco. I do think there is something else amiss. All currently points to the loco decoder. I have a discontinued Digitrax DZ125 decoder. It has a heavy 4x2 wired plug (I understand the plug is the reason they replaced it with the DZ126 - the shell would not go down all the way when plugged into the intended loco). I cut the plug down the middle to give two 1x4 sections which I plug into a breadboard. Wire up some LEDs, with resistors, to the light outputs and a 10 Ohm 10W resistor (5x 50 Ohm 2W resistors in parallel - it still heats up rather quickly) to the motor outs and you can test your base station output voltages and current. I also use it to run some of my older DC only locos by hooking the motor outs to the layout (see picture). One relatively cheap and easy test option. Hope you get it sorted soon. Willem.
Willem During the upcoming week I hope to receive a new decoder which has been ordered today. Bst rgds
A new decoder, Digitrax seems to have solved it.... Today I received a new decoder and the thing moves! AC measured at track is 13,1 V, fluctuates a little. But movement a little bit funny. Like a ketchup bottle, first there is nothing, then after some seconds when I have gone into full throttle the engine runs away and due to the short track onto my desktop in an designated crash area. The decoder manual that is.. I have tested different speed table settings etc. but I suspect that 15V DC input is to low. There are posts, Scott Eric Catalano, is a name I remember, that somewhere here says that for an H0 loco more than 15V is needed. I will not risk the Arduino MS with my 19V DC PSU. As soon as the new Pololu arrives, due in about 2 weeks perhaps I will assemble that fellow and make more tests. Until then happy modelling to all
Glad you had some success. That is not quite the behaviour I would have expected. Even with 13V at the tracks the loco should move smoothly. The lower voltage should basically only reduce your maximum speed. Maybe the loco and motor needs a good clean and lubrication? Hope you get it all sorted soon. Willem.
I see your point Willem, but I still have the fact that the loco didn't worked at all with the old decoder. But now it does. I will dive into the motor and check it. I do not recall from where, but I am sure I have read a lot of postings "everywhere" about WHAT to lubricate with but I also recall that there was not much of concensus in that matter. Just any type of any not to fluid oil vs. those heafty prized model railway making companys products? I.e the easy to find oil for my wifes sewing machine? No good? Or will it be the perfect solution. I am qutie clear about that it has to be done very sparsely and only in the right spots. To much oil could that spreads uncontrolled I am sure is not good. Thanks again for your suggestions, Leif
I see more "don't use sewing machine oil" warnings than recommendations. It seems the overall consensus is that is not really plastic friendly. My experience with 3 in 1 is that it does get sticky over time - I would not use that. I have seen no warnings using Labelle, light oil (108) for bearings, Grease (106) for gears. Seems the most regularly recommended lubricants. I do not know where to get that over here. What I can get is the Woodland Scenics lube kit. I do not know how good that is. I also see some modelers use Dexron III transmission oil. Do not know whether that should be recommended. Most important is that it must be plastic compatible. I think I will leave recommendation for lubricating locos to other more experienced modellers. I could use the advice as well. Willem.
You're absolutely right in prioritizing finding a plastic-compatible lubricant for your model trains. Here's a breakdown of your options and some additional advice: Labelle: Labelle is widely considered a top choice for model train lubrication, with their #108 oil and #106 grease being popular options. Their products are specifically formulated for plastic and metal parts, so they won't cause damage or gumming up over time. However, finding Labelle outside the US can be challenging. Woodland Scenics Lube Kit: Woodland Scenics lube kits are decent alternatives, especially their Moly Grease (HL656). It's plastic-compatible and provides good lubrication for gears and bearings. However, some users find it a bit thick for lighter applications like axles. Dexron III transmission fluid: This is a controversial choice. While it can work as a lubricant in a pinch, it's not specifically designed for model trains and could potentially harm certain plastics over time. Some modelers swear by it, while others strongly advise against it.
