I have 3 BLI Paragon 2 diesels (SD40-s, NW-2 and SD-7). I was putting them in my DecoderPro roster today and realized I may not understand what to do and not do.
With decoderPro, I have been putting a loco on the program track, reading the decoder type (in these cases BLI Paragon2 diesel) and adding the long address, turning off DC mode and writing (to loco) & saving (to roster). Then I select program on main mode and set speed curve and sound levels, etc, so I can observe the effects. then I save to roster, which is a combo of the DecoderPro values incorporating the changes I've made ("written" to the loco).
What I'm wondering is whether the 3 type locos out of the box (or with a CV8 reset) have the same values and how that relates to DecoderPro's values for the "BLI Paragon2 Diesel"that come up for the identified decoder. What makes me wonder is the DecoderPro "reset to defaults" button. Not sure what happens if I click that and/or then 'write all sheets". If the locos come with different settings should I "read" all values before making changes, or can I trust DecoderPro to have the correct (for that loco) CV8 default values built in?
Some related observations:
- when DecoderPro "reads type from decoder", there is no value shown in the Mfg ID box
- when I wrote all the DecoderPro sheets to the NW2 it ran goofy at higher speeds (something was wrong). When I did a CV8 = 8 reset, it again ran fine and I could adjust the speed curve and save, etc.
So, I'm quite confused (easy to do). Don't know whether to ID the decoder, then read all values from the loco, then make changes, etc. Of course, if I save to roster the DecoderPro values without reading the loco, if they are different I won't have a file with the real loco values, only the changes I've made.
I hope I've explained this clearly. Help!
Paul
Modeling HO with a transition era UP bent
For most of the sound decoders that are available with multiple sound sets, it makes no difference what decoder you pick. They all have the same CVs, just different sounds. Picking the Paragon II for all of them is fine. You probably don;t want to pick a Paragon II Steam decoder, because it will have CVs for controlling the smoke unit. But even then, it probbaly will do no harm, the diesel decoders probbaly have those CVs, they just don't control an output.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Paul,
The "Set to default" in DecoderPro isn't necessarily the same as a decoder reset (such as writing 8 to CV8 for Digitrax decoders). Instead, it's whatever values the person who wrote the decoder definition, put in that definition when they wrote it. So any given CV value could be the same as a factory reset, or it could be different.
For that reason, the first thing I do with a new decoder/loco is create the roster entry, read all sheets, and then save that as my baseline. That way I have exactly what was in MY decoder when I read it. I'm not depending on what someone else thinks the defaults for that decoder/loco SHOULD be.
After that, I make my adjustments. But I keep that baseline decoder file in case I ever need to go back to the "real" ground zero for that decoder/loco.
Stevert, I can see that overall logic of first reading the decoder's values into DecoderPro, making adjustments from there. Maybe that's why the NW2 behaved funny when I wrote the DecoderPro "reset to defaults" values to it. Though I looked quickly at the BLI Paragon2 Diesel Tech Reference Manual and it would imply that all such diesels would have the same settings (though some would do nothing, I suppose, such as on a loco without ditch lights).
I have a NCE PowerPro and when I told it to read all sheets and it was running thru the CVs I think it said it could not read some of them. Do I likely need to add a reader (program track) booster for the program track? Why would it read most but not some? Maybe it was trying to read some unused CVs on the diesel decoder? I may need to tinker back and forth to see what's actually happening with some specific CVs.
On another note, I see that DecoderPro programmer box offers a "reset menu" item in the top bar that will initiate partial or full decoder (loco) resets directly. But as you note that's different than the "reset to defaults" button that resets the DP boxes to THEIR default values (not necessarily the real loco reset values), after which they MAY (required additional action) be written to the loco and/or the roster entry.
Seems important to understand how this all works to know what's actually happening!
Maybe I'll re-do the NW-2 with this knowledge, read some CVs and compare to the DecoderPro defaults, compare to the Paragon2 tech manual, see wht CVs can't seem to be read, etc. With that info, maybe talk with BLI tech folks.
I don't believe that you will be able to reliably read back the CV values from a sound decoder with the PowerPro. At least I'm not able to.
As I think was mentioned, the default values in decoderpro are those that whoever did the compilation used. Resetting to the default values will reset the decoder to those values.
As far as I know, the instructions that come with the BLI locos list all the CVs and their default values. In decoderpro you can look at one of the tabs and see what all the CV values are. If you go to that tab and go down the CV list you should be able to compare all the values against what the instructions say they were. You will have to do this manually which is not as convenient as allowing decoderpro to read back the values, but if you can't reliably read back I see no other option.
There are two things that may affect the ability to read back all the CV values in DecoderPro:
One is the physical ability of your programming setup. For example, some command stations don't have the programming track power required required for sound decoders, others may need a resistor across the rails, still others may require a programming track booster, etc.
The other factor, again, is what's written in the decoder definition you're using. If you tell it to "read all sheets", but the author has only defined half of that decoder's CV's, well, that's all it's going to read. You have to go on blind faith that they included them all, or else check each definition against the manufacturer's doc. And even that's no guarantee; the doc can be wrong, too.
In my case, I use a Digitrax PR3 as a stand-alone programmer with an 18VDC regulated power supply. It reliably reads/writes to any sound decoder I've ever tried it with, so I have that part covered. But I'm still at the mercy of the authors of the decoder definitions.
By and large, though, I tend to think they're pretty accurate. I regularly follow the JMRI group on Yahoo!, and there's usually a good amount of conversation regarding getting the decoder definitions (and everything else!) right.
Edit: Fixed punctuation.
I also do what Stevert does.
" For that reason, the first thing I do with a new decoder/loco is create the roster entry, read all sheets, and then save that as my baseline. That way I have exactly what was in MY decoder when I read it. I'm not depending on what someone else thinks the defaults for that decoder/loco SHOULD be."
I have an NCE Power Pro and it needs an increase in programming track current or a booster to read sound decoders.
When I get a new loco or do a new decoder install, I read the decoders mfg and version info first, then set the address of the loco, and select DCC mode only, all using the hammerhead throttle first. Then I create a roster entry in JMRI using the mfg and ver numbers to manually select the decoder type. Then I select program track mode, and do a read all sheets. I still get some "can not read sheet" errors, but I keep going. Once the read is finished, I have all the info in the roster for that loco and do a save. Then I make any adjustments I want, usually on the main with decoder pro, and save them. I don't know why I get the "can not read sheet" errors, but it doesn't seem to affect things and I have never read all sheets without getting some of those errors.
Elmer.
The above is my opinion, from an active and experienced Model Railroader in N scale and HO since 1961.
(Modeling Freelance, Eastern US, HO scale, in 1962, with NCE DCC for locomotive control and a stand alone LocoNet for block detection and signals.) http://waynes-trains.com/ at home, and N scale at the Club.
Elmer,
You get the errors reading because your PowerPro (like my DCS100) was designed before power-hungry sound decoders became the norm. The "old" NMRA programming specs those command stations were designed to meet just don't cut it with those decoders.
In your case, a programming track booster (the PTB100 is often recommended) would probably eliminate the errors.
I opted for a PR3 instead of a programming track booster because it gives me other, Digitrax-specific benefits as well as the ability to reliably read from and write to sound decoders.
I did some tinkering with two of the Paragon2 diesels today and got a bit more clear.
If I ask DecoderPro to read all sheets it goes through so many CVs so fast that it's hard to see what's happening when it can't read some. So I went through (reading) sheet by sheet so I could watch the CV blocks turn color, etc. On both locos it failed to read the same 7 sound CVs (145-151). Odd ones like air filling volume, etc, all set at 128 default (per the tech info, and the DecoderPro defaults; i.e., no discrepancy). I have a hunch it is not lack of a program track booster but something else unique. Or is it harder to read some CVs than others??
I had no big problems playing with each loco. I tended to do a CV8 = 8 reset, then change the address and turn off DC mode. Then read the sheets of interest (rather than all at once), looking for any key differences from the DecoderPro defaults. I did not bother with consist sheet at this point (can do with the NCE). I did note it could read the manuf ID and number. I then went to Ops mode and set the speed curve to my liking and tweaked the master volume, bell and horn, selected a preferred horn, etc. Made sure I wrote all changes to the decoder and saved the file to the roster.
I balled it up once, but just went back to a CV8 reset and started over.
So, I understand more than I did, but not all of it. Enough for now.
Stevert Elmer, You get the errors reading because your PowerPro (like my DCS100) was designed before power-hungry sound decoders became the norm. The "old" NMRA programming specs those command stations were designed to meet just don't cut it with those decoders. In your case, a programming track booster (the PTB100 is often recommended) would probably eliminate the errors. I opted for a PR3 instead of a programming track booster because it gives me other, Digitrax-specific benefits as well as the ability to reliably read from and write to sound decoders.
I don't need a programming track booster. I modified my command station to output more current to the programming track and it works just fine. I have no problem when programming sound or non sound decoders with the hammerhead. I just get the errors when using JMRI, and always have, and that includes programming non sound decoders as well.
gandydancer19I modified my command station to output more current to the programming track and it works just fine.
Any hint as to how you did that?
Elmer's comment got me curious about the 7 Paragon2 CVs that DecoderPro could not read with my NCE program track setup. I just tried reading those CVs on my NW2 on the program track via the NCE PowerCab throttle, choosing program track mode, and the NCE was able to read all 7 CVs that DecoderPro could not. Good to know, as I think I can now manage around that as needed. I'll not be adding a program track booster until another issue identifies the need.
Try just reading the single page where the CVs are that it missed. Part of the problem is that some locos can actually creep along the track when CVs are read or written,. and if contact is less than perfect, it can miss on a few CVs when trying to read the entire decoder in one big shot.
Even some of those MRC sound decoders that supposedly didn;t even have readback - I've been able to read those, by letting DecoderPro only read a couple of CVs at a time, not a full sheet or all sheets.
gandydancer19 Stevert Elmer, You get the errors reading because your PowerPro (like my DCS100) was designed before power-hungry sound decoders became the norm. The "old" NMRA programming specs those command stations were designed to meet just don't cut it with those decoders. In your case, a programming track booster (the PTB100 is often recommended) would probably eliminate the errors. I opted for a PR3 instead of a programming track booster because it gives me other, Digitrax-specific benefits as well as the ability to reliably read from and write to sound decoders. I don't need a programming track booster. I modified my command station to output more current to the programming track and it works just fine. I have no problem when programming sound or non sound decoders with the hammerhead. I just get the errors when using JMRI, and always have, and that includes programming non sound decoders as well.
JMRI sends the same commands to the CS as the hammerhead does. But it's a lot faster, so the issue is probably that JMRI is overrunning either the CS, or the decoder's ability to accept and respond. In that case, I agree with Randy - Just manually read the individual CV's.
With my PR3, the original problem was that the power supply (they were still selling the PS12 at the time) simply didn't provide enough "oomph" for some sound decoders. I switched to an 18 VDC regulated supply, and although a "Read all sheets" may take a while, especially for LokSound decoders, it always completes without any errors.
Mine works just fine witht he PS14 power supply...
Be careful with which version of JMRI is used with Loksound decoders. They are still working on perfecting things. The V4 and Select series have over a thousand CVs - nearly 1000 in the function remapping alone, since they allow you to map any function to any sound slot and/or any function wire and any F key. Not sure on the exact count of sound slots, but say it's 10. So you have 29 possible F keys x 10 sounds slots x 6 function wires - 1740 combinations. A given CV can store 255 possible values, so that's 8 CVs even if you just do it in list order, which really isn't possible, it would kill the processor to brute force all those entries every time you pressed a function button.
So an all sheets read of a Loksound - yeah, that's going to take a while,a nd there definitely is a chance to miss a few. The times I do use JMRI, I rarely if ever read all sheets - definitely not on any sound decoder. No real reason to. I generally reset a new decoder before initial programming, to make sure there aren;t any goofy artifacts from testing or manufacturing. Then I change what I want, and write those changes, also saving that as the roster entry. Everything else should be the default then anyway - and even if it's not, if something goes wrong, the first thing to do on the decoder is reset to defaults, then reapply the changes you made - which is exactly what's in the roster entry. In the days before decoders had factory reset options, I guess it made sense - the only way to 'reset' the decoder was to re-write every supported CV with known values.
Most of the time, I don;t use JMRI - my non sound locos all have some TCS decoder in them, and my railroad uses no special light features or effects, just a headlight and a rear light, if the loco has one. ALl I really set is an address, and I can do that far quicker witht he throttle than going through all the setup in JMRI. And for the sound ones, the most often changed thing is the volume, and I usually do that in Ops Mode so I can immediately hear the level. Also easier via the throttle. Someday I may even memorize the various CVs for each sub-volume as well as the main volume, since I'm using only Loksound now. I use JMRI more at club shows when i set up my program track and configure locos for people, since they could have any brand of decoder, sometimes the loco owner doesn't even know what kind it is.
I always use the most current test release of JMRI, sometimes even an unreleased Jenkins build if it has something I need or want, so I'm not concerned with being behind on decoder definitions.
But when I get a new loco/decoder, I always read all sheets, especially with ESU decoders. The ESU decoders for resale (that is, the non-OEM ones) are produced blank, and the reseller loads the sound files. There is a final step in that process that resets some of the default CV values based on the sound file that's been loaded. If that step is missed, and it sometimes is, those defaults don't get set and a decoder reset could cause the decoder to lose all or part of that sound scheme.
Sure, I could get a LocSound Programmer and avoid that, but why bother? I only have a few ESU decoders, and my JMRI/PR3 combination reliably reads them, even when using "Read all sheets". The money would be better spent on yet another loco...
That should be a non-issue now, ESU worked with the vendors who were forgetting that important but easily overlooked checkbox when saving the sound sets.
I DO have a Lokprogrammer, because I won;t be buying any other sound decoders for the foreseeable future. Lets me find the lowest price on the decoders and not worry about if the seller provides a sound set loaded, plus I've already updated my older PCM steam locos with Loksound 3.5 decoders to improved actual recordings, and updated 3 others for people at the club. As soon as I get everything set up and working again, I have an updated Baldwin switcher sound set to load in my DS4-4-1000. I bought it before ESU released the new updated recordings they got from (I presume) SMS's Baldwins, some of the few running with the original Baldwin diesels (many were re-engined with EMD prime moves, making for interesting hybrids with a Baldwin cab but EMD switcher long hood and stacks). Then they just announced a new horn for their library, a proper M3RT1. Just what I need for my RS-3's. Right now the one RS-3 sound set I programmed has a regular M3, closest they had. Now it will be the exact M3RT1 with the reversed and angled (tilted) bell They mounted on the front of the cab and two bells faced forwards (long hood) and one leaned back over the cab roof.
All very worth it to me to standardize, I can keep a couple of decoders on hand so I never have to wait to finish a project, and just install the latest firmware plus whichever sounds I need. Just say no to hard-coded decoders :)