Hi!
I just finished setting up a dedicated/isolated programming track on my layout. It’s working nicely, and if I connect it to my command station JMRI can read and program engines.
Is there a reason to hook the track up to the PR3 instead? Does one offer a better &q
It has to do with what command station you have. I have an older, actually old, (2004) DB150, and I do not have a seperate program track except through the PR4 that I recently bought to set up JMRI.
Newer command stations have a connection for a seperate programing track. Now which is best, the PR4 or the command station? I’m thinking that the command station would be the way to go, but, I don’t know for sure.
Hang on, Randy will find this, and confirm what you need to do.
Mike.
EDIT: Wanted to add this, I’m sitting here in my office, looking at my laptop, PR4, and a section of programming track, and just thought, the layout and command station is down in the basement, but I can sit up here, not cennected to the layout, and program a loco. Might be something to consider, and I’m sure you could go both ways, connected to the command station, or stand-alone.
Mike is correct. Depending on what command station you have, the PR3 is arguably the better choice.
The DCS100 and DCS200 command stations were designed way before power-hungry sound decoders hit the market, and their programming circuitry sometimes has problems with those decoders.
However, the PR3, when used with an appropriate power supply**, will easily program those same power-hungry sound decoders.
The newer DCS240 basically has a PR3 built in so it doesn’t have that programming issue. Not sure about the DCS210, but since it’s also a new design it should be okay as well.
**To use the PR3’s programming outputs, you need to supply the proper DC power. Your PR3Xtra came with a PS14, which may or may not be robust enough for some sound decoders.
I use an 18VDC regulated power supply for my PR3 (same internally as the Xtra) and it reliably reads and writes to any sound decoder I’ve ever tried it with - QSI, BLI BlueLine, Loksound, TCS WOW, Soundtraxx LC, Tsunami, & Tsunami 2, and of course Digitrax. I haven’t used my DCS100’s programming outputs since I got my PR3. They aren’t even hooked up to anything any more.
I have a DCS51. It’s never given me trouble programming sound engines on the main for what it’s worth. And in my (albiet very limited testing with a Digitrax sound equipped engine) it seems to work for reading and writing of CVs on the programming track.
Programming on the main is a different animal. It gives you the full power of the command station. That’s good for programming power-hungry sound decoders, but bad for testing a new decoder installation. And of course you can’t read back CV’s on the main unless you use transponding, Railcom, etc.
As for the Digitrax sound decoders on the programming track, they were designed to work with the programming ability of Digitrax command stations. Try a TCS WOW with a keep-alive motherboard, or a Soundtraxx LC and you might not get as good of a result (Although the DCS51 is a newer design and might do okay.)
I dunno, my original Zephyr also works fine on the program track with the supposedly hard to program QSI decoders. But only after using a DT40x series throttle - the same loco has issues using the Zephyr console. JMRI can have issues - and all the JMRI people will say “but it only does what a throttle does” Yes and no. Yes, it controls the system as if it were just a throttle plugged in. However, it “presses those buttons” far faster than an human could.
I have my PR3 set up completely off the layout, wired to the same test track I built to set coupler heights. Using that with JMRI has worked fine with every brand of sound or non sound decoder I’ve thrown at it - and there is a wide mix at the club.
–Randy
JMRI can have issue but the point should be, if it doesn’t work on a throttle then it won’t work using JMRI.
Peter
Well, the DCS50 is also a newer design than the DCS100 or DCS200, so that could well be.
The “JMRI people” are correct. No matter how fast JMRI “assembles” the programming commands, they are still subject to the timing requirements that DCC requires.
So JMRI still has to wait for the command station to send the command to the decoder, wait for the decoder to reply to the command station, and wait for the command station to pass that reply back to JMRI.
That’s why even when using JMRI it can take an hour or more to read all sheets on some decoders.
As I mentioned earlier, that’s the same setup I have, except that I personally own the whole mix of sound decoders.
FWIW I have my ‘on the layout’ programming track connected directly to the DCC system, and use a PR3 to connect a separate programming track to my PC. I do most programming using the DCC system, but can use the PC/PR3 with JMRI Decoder Pro, and to download different sound files to Digitrax SDH / SDX decoders using their Soundloader software.
That’s right, it the throttle won’t work, JMRI won’t work. But the opposite most definitely is not true.
The DCS50 is a prime example. ANd not just with JMRI, but with the internal throttle. Specifically with QSI decoders - even if you turn off the verbal response, it still seems to take some amount of time (when if using verbal response, it would be telling you the value) when setting a long address. On the program track, the system automatically figures out the values for CV17, 18, and 29 based on the address and address type. So you want the loco to be 1234, that doesn;t mean CV17 is 12 and CV18 is 34, it’s a bit more complicated than that. But no worries, the system, all modern systems, calculate this for you, so all you need to know is you want the address to be 1234.
The system writes the proper values to CV17, 18, and 290. On the Zephyr, using the built in throttle, it does exactly that. Write, write, write. Fails 100% of the time with a QSI decoder, at least the original ones in BLI and Atlas locos.
However - plug in a DT400 throttle, and use that - now the programming control is coming through the throttle, even though the actual signals still come from the Zephyr’s program track terminals. The difference is, in the sequence on a DT400, it writes CV17 and CV18, then prompts you and asks if you want to use a 4 digit address or not. This introduces a dealy between CV17 and 18, and CV29 getting set - it takes you some finite time to say Yes. Works all the time, even with a QSI decoder. No program track booster, no “blast modes”. Just that extra delay allows the QSI decoder to properly accept. JMRI can have this problem too, if you just hit “Write sheet” for the address page.
JMRI is great, but it’s not perfect. For my own locos I don;t bother any more, it takes longer to set up a new loco than to just hit a few buttons on my throttle. I use all the same brand decoders in all my locos - well wor