Any remaining enthusiasm with DCC is pretty much gone

[randgust]

It would be a lot easier to flame the DCS50 if on the first epic fail with the Z2's wasn't validated by Lee Weldon, who does DCC installs, and tested it himself.  He got the same erratic results out of it I did.  And TCS support made repeated attempts to both replace it and suggest CV changes, finally told me they gave up, and 'decoders won't work in a locomotive that small', which even I regarded as a cop-out after watching a Hornby SW1 work flawlessly at Altoona.

I can see where the JMRI interface could certainly address the decoder settings and diagnosis, their documentation is pretty indeterminate about the connection between a computer and a base station, lots of tips, but as said above, kind of sketchy.   And that doesn't really address the base station transmitter that transmitting the packets and it's own issues here, at least not from what I read.  That's part of the problem, I don't know what it's doing.  In computer terms, I need a packet sniffer/IP trace.   Is there an interface at that level?

I've wired my own serial interface cables before when pins were reversed, that's going back to the 'everybody has their own standard' era in equipment when machines had proprietary serial interfaces and you had to figure it out.  No fear there.

I'd only found two decoders that were small enough to fit in my kits, the Z-2 and the Dz126, if there's a Lenz in that size range, and the  Lenz stations are that much better (old Bachmann v. new Bachmann?) then I'd look at it.   I'd think there was a good entry-level system that has interface like this.   And you'd think that some systems would simply include JMRI as an integrated feature...

Fired up the DCS50 this morning, only the headlight came on, no response to throttle.  Worked yesterday.  Three more factory resets, unplug from wall, nothing.   Check on analog, running OK so decoder isn't toast.  Reset again, hard force address 03 and try again.  Damn, it worked.  How I hate this thing.

Wait a minute, I could be running trains on the DC system that works....  And remember, I have walkaround transistor throttles with momentum, no problem with MU setups at all, constant lighting installed in all lead units with strobes, beacons and mars lights, working markers and lit passenger cars.  And a working, functional signal system integrated with power and switch position.  I also have digital sound installed in several locations.  So I keep wondering what I'm missing on what is really a one-person layout 95% of the time.  I do like the  authentic sound I've seen particularly with steam, but other than that, this is a lot to go through.

[Scottl]

The ESU Nano decoders are tiny, perhaps they might fit your need.  They work with all compliant systems and are only a few bucks more.

[jagged ben]

...

Wait a minute, I could be running trains on the DC system that works....  And remember, I have walkaround transistor throttles with momentum, no problem with MU setups at all, constant lighting installed in all lead units with strobes, beacons and mars lights, working markers and lit passenger cars.  And a working, functional signal system integrated with power and switch position.  I also have digital sound installed in several locations.  So I keep wondering what I'm missing on what is really a one-person layout 95% of the time.  I do like the  authentic sound I've seen particularly with steam, but other than that, this is a lot to go through.

And surely it took lot of work and some frustrating moments to get all that working, too. Not many people have accomplished that, and nobody's says you should perform a redo.  To each his own, I just bristle a little when the implication seems to be that something just sucks, despite that a lot of the rest of us find it usable and get enjoyment out of it.

[peteski]

I wonder if the decoder needs to be status edited ..

The Digitrax DCS50 Zephyr is set at the factory to send 128 speed step commands to all mobile decoders.

If you have a non-Digitrax decoder that can’t run in 128 speed step mode, you can change the number of speed steps the DCS50 sends to that decoder so that you will be able to control the decoder. This is called status editing the decoder. When you status edit a decoder, the DCS50 will send a different format to that specific decoder address without affecting the 128 speed step operation of Digitrax decoders and other 128 speed step decoders on the railroad.

That would be a good possibility, but both the Digitrax and TCS decoders Randy used recognize 128 speed steps.  At least I'm 99.99% sure that TCS is 128 step capable. I just can't find the manuals on the TCS website (call me stupid).   

But then Z2 had a well known (at least to me) bug in its firmware where if the 3-speed speed table was enabled (CV2, 5, and 6)< the speed steps woudl not be increasing incrementally.  For example, step 6 would be slower than step 5 and then speed 7 would be faster than the speed step 6 was supposed to be.  That could have contributed to the strange behavior.

[peteski]

It would be a lot easier to flame the DCS50 if on the first epic fail with the Z2's wasn't validated by Lee Weldon, who does DCC installs, and tested it himself.  He got the same erratic results out of it I did.  And TCS support made repeated attempts to both replace it and suggest CV changes, finally told me they gave up, and 'decoders won't work in a locomotive that small', which even I regarded as a cop-out after watching a Hornby SW1 work flawlessly at Altoona.

TCS decoders have fallen out of favor with me.  Mainly due to the poor motor control (even if the lighting effects are way better thana Digitrax).  But once I opened my eyes and moved onto European brands decoders, I have never looked back at Digitrax or TCS.  I still use a few Digitrax decoders in non coreless motor locos (and install them for friends), but I have become a fanboy of ESU and ZIMO.  There is also Lenz, but I  don't work much with those.  I think they have also fallen behind the other 2 manufacturers.

And as mentioned, there are some really small decoders available from ESU and also CT Electronik.  I have not worked with the latter, but I have read (here) that they are very good decoders.

Let me ask you again, is the latest decoder fiasco in the same model that had the problem with that Z2 decoder?

I'd only found two decoders that were small enough to fit in my kits, the Z-2 and the Dz126, if there's a Lenz in that size range, and the  Lenz stations are that much better (old Bachmann v. new Bachmann?) then I'd look at it.   I'd think there was a good entry-level system that has interface like this.   And you'd think that some systems would simply include JMRI as an integrated feature...

Not familiar with Bachmann, but I would recommend the NCE Power Cab as a good (and IMO more intuitive than many others) starter system for a person with a small layout  There is also MRC.  As far as integrating JMRI into a DCC system package, I'm not sure how that could be done. JMRI is an application developed and maintained by hobbyists (freeware) using Java programming and running on a personal computer.  Not sure if DCC manufacturers would want to make it an integral part of their DCC systems. That would be a tech support headache.  Manufacturers already work with JMRI by equipping their command stations with a serial interface which "talks" with a PC running JMRI.

I can see where the JMRI interface could certainly address the decoder settings and diagnosis, their documentation is pretty indeterminate about the connection between a computer and a base station, lots of tips, but as said above, kind of sketchy.   And that doesn't really address the base station transmitter that transmitting the packets and it's own issues here, at least not from what I read.  That's part of the problem, I don't know what it's doing.  In computer terms, I need a packet sniffer/IP trace.   Is there an interface at that level?

Fired up the DCS50 this morning, only the headlight came on, no response to throttle.  Worked yesterday.  Three more factory resets, unplug from wall, nothing.   Check on analog, running OK so decoder isn't toast.  Reset again, hard force address 03 and try again.  Damn, it worked.  How I hate this thing.

Man, you do have some serious issues! I have seen many DCC related issues and you seem to really unlucky.  But the only way I can see troubleshooting this maddening issue would be to have another DCC loco or another DCC setup (command station and a loop of track - or someone's working DCC equipped layout) to isolate the problem to either the loco, or the command station and/or track or wiring.  There is really something weird going on in your case. Maybe you have some sort of electric field around your body which seems to mess with the decoders non-volatile Flash memory?  I'm actually half-serious here as I have never seen decoders blowing their programming so much, especially on a simple track plan.

I have decoders which have not been programmed for years and they run fine. The only times I have seen decoders "blow their brains" is when a short somewhere on the layout causes a voltage spike, being zapped by few thousands volts of static electricity in a dry environment, or by accidental main-track programming to the broadcast address "0".  A faulty decoder hardware would also make them misbehave of course.

There is a DCC sniffer out there from Pricom Design (I own one) but they are out of production.  You really need to find a way to isolate the problem to your model or the other half of the  equation.  That is why I offered to test your loco.  Or if you have a friend with a known good DCC-equipped model (hopefully with the same brand and model decoder), that would also be a good thing to try on your layout.

Wait a minute, I could be running trains on the DC system that works....  And remember, I have walkaround transistor throttles with momentum, no problem with MU setups at all, constant lighting installed in all lead units with strobes, beacons and mars lights, working markers and lit passenger cars.  And a working, functional signal system integrated with power and switch position.  I also have digital sound installed in several locations.  So I keep wondering what I'm missing on what is really a one-person layout 95% of the time.  I do like the  authentic sound I've seen particularly with steam, but other than that, this is a lot to go through.

Well, you are correct - if you are happy with your reliable DC layout then staying with DC is a valid option. Unless you are pining for constant lighting with all sorts of fancy light effects, sound, or running multiple models in a consist or in a single electrical block, then there really is not reason to give you self another source of headaches,. You have a reliable layout - you could just continue using it.  Evne with new locos, you can always convert them to DC.  And I should also mention that the DC compatibility mode of decoders will always result in inferior motor control when compared to a model with the decoder removed.

EDIT:  As I understand, you have a fairly small and simple track setup with your DCC system.  If all the DCC hardware is in working order then you should  have absolutely no problems running your models reliably (and not ever see the crazy stuff you are seeing happening). You are not having a typical DCC experience.

[Doug G.]

Oh come on.
Not if you set the operating rules on your layout for them not to do that. There could be a rule book and a protocol for them to follow (just like with real trains). Real engineers also don't fumble with block control switches while the train is running. 

I was being facetious. I used the words "To me" and "like" for a reason.

Doug

[peteski]

I was being facetious. I used the words "To me" and "like" for a reason.

Doug

And I ended my post with a smiley.  It's all in good fun (until a decoder blows  up)! 

[atsf_arizona]

NCE PowerCab.  I spent 2004-2012 in Silicon Valley with some of the best/brightest minds in tech who also happened to be (obsessive) model railroaders.

General consensus was:  NCE + JMRI is the best for us North American model RR'ers.  There was a lot of reasons.  But most of all, the more intuitive interface.  Give DCC another chance if/when you're willing and able.  Peteski has good comments.

My Silicon Valley friends didn't have anything against Digitrax, Lenz, Zimo, etc.  There is a uber-huge Digitrax layout  in Los Altos that is among the densest concentration of EMF and multiple boosters in one model RR, it's often a "test bed" for Digitrax.  So they could make all of them work, and did.  But sometimes it took a "meeting of the right minds" to debug hard-to-shoot issues, and find the answers.

If you ever run NTrak, that's a good reason to be in Digitrax, which is the predominant standard there.

(Disclaimer:  I still run DC for the same reasons that Randgust does.  I like DCC especially with sound, but for the collection I have and the small RR I have, there's not a pressing need.  I have DCC NCE PowerCab for testing things out and for sound locos).

[learmoia]

I've been messing with DCC and Decoders for several years and I've never gotten a 'RESET' CV08=08 to work properly..   I've always had to force change the address to 03..

I learned DCC on a Digitrax and have a Super Chief, and Zephyr.. I learned manual programming on the DT400.. and I can't for the life of me program on a Zephyr.. I just plug in the DT400 or JMRI to do my programming..

It can be very frustrating from a technical stand point.. and when you get into a multi-person operating situation - it's pretty cool.

~Ian

[peteski]

I've been messing with DCC and Decoders for several years and I've never gotten a 'RESET' CV08=08 to work properly..   I've always had to force change the address to 03..

Interesting. Assuming that you are doing this on the programming track and it is a Digitrax decoder, then the reset procedure should work as described in the manual.  Other brand decoders might use alternate reset process - it is always good to check their manuals.  I don't ever recall having the decoder reset procedure not work properly (including changing to the short address"3"). Som decoders have to be powered down and "rebooted" (lifted off the track) to complete the reset process.  That should also be mentioned in the manual.

[rrjim1]

If I read Randy's post correctly he has had several decoders (TCS support made repeated attempts to both replace it), and two different brands. I use both brands of decoders and in fact I have 37 TCS Z2 decoders and 6 of them are in small Life Like SW switchers. Common sense tells me it isn't the decoders. 

[peteski]

If I read Randy's post correctly he has had several decoders (TCS support made repeated attempts to both replace it), and two different brands. I use both brands of decoders and in fact I have 37 TCS Z2 decoders and 6 of them are in small Life Like SW switchers. Common sense tells me it isn't the decoders.

That is exactly why I mentioned that additional troubleshooting steps should be taken to isolate the problem.  Quoting my earlier post:
But the only way I can see troubleshooting this maddening issue would be to have another DCC loco or another DCC setup (command station and a loop of track - or someone's working DCC equipped layout) to isolate the problem to either the loco, or the command station and/or track or wiring.  There is really something weird going on in your case. Maybe you have some sort of electric field around your body which seems to mess with the decoders non-volatile Flash memory?  I'm actually half-serious here as I have never seen decoders blowing their programming so much, especially on a simple track plan.

[randgust]

To clarify a couple things here.

First it is two different locomotives.   The one with the Z2 was unceremoniously converted back to DC and sold on Ebay, new owner very happy with it.  The unsolvable issue with it was no low speed control, it was stalling and resetting constantly.   Hiccupping.  And I sent it to Lee Weldon to see if he could reconfigure any CV's, BEMF, whatever, to fix it, he gave up.  TCS replaced it twice and they gave up.  I gave up.  Took over a year.  I still have the last decoder and have threatened to install it to anyone that crosses me.

Second one, the Digitrax Dz126, by the way, 'bench tested' the chassis with the headlights and controls and the decoder OK, although I had a rough time with the F0 headlights as one diode was reversed, but  that worked OK.  THEN I assembled the shell on the body, ran it on analog DC, both headlights worked directionally, and it ran sorta-kinda OK (minor motor surging), and THEN I took it back to DCC and everything went south.

Current status is that the decoder is working, one headlight is 'stuck' on with no F0 control, and no directional headlights.  But motor control is actually good, on the one of four times I have to repeatedly factory reset the DCS50 to even get it to communicate with the decoder at all.   I could attempt to reset the decoder if I had any confidence at all in the DCS50, which I do not now.   That may be all it is, but random testing without an elimination process has never been my bag and there are too many variables here.

The only thing the two installs have in common is the control station, and that both mechanisms were using Kato 11-105 pager-size 12v motors. 

I'm only using DCC to set up some equipment to run on the show TTRAK modules and 'maybe' the HVRR down the road, although that's highly unlikely now.   I had to just run trains on the DC layout for a couple hours to get my sanity back.

Without another control station or another decoder to play with I'm pretty much stuck, walking away from it, but the JMRI features are enough of a hook to maybe get me to reinvest in a different control station entirely.   I basically hate the thing and how it works to do anything more sophisticated than 'on/off/reverse' anyway.    So I think the way that I'm rephrasing the question is to find another entry-level system that has a reasonably tested interface relationship with JMRI likely running under Win7, and still has a real serial port although USB should be a standard.

Oh, and after working in IT for years, I have no detectable electronic field around me, although I do have a co-worker that did.  He worked in industrial electronics and a locomotive paint shop, and got enough mercury in his body that he had to go through a multi-year process to leach it back out.  During that time, no kidding, he could ruin an electronic watch, and he trashed three different laptops assigned to him.   Anything digital he came near was impacted.   He joked that he had enough metal in his body that that he literally could feel a ground fault in a locomotive when he would work on it.   So I'm a believer in that, but apparently not a carrier.

[peteski]

If you are willing to try DCC again then (as I and few others have already mentioned) get the NCE Power Cab.  The entire DCC system (throttle/command station/booster) is contained in the handheld throttle (which can be also used as a standard NCE throttle on a fully-featured NCE  setup).  NCE throttle has a 2-line 16-character alphanumeric display and all the programming prompts are in English language.  JMRI is not really needed (unless you are planning on owning a larger fleet of locos or programming sound equipped models).  Compared to Digitrax, Power Cab is really intuitive and easy to use.  It can also program sound decoders without additional booster.

I own the Power Cab using it for bench-testing locos I service, and I also take it with me to NTRAK shows to have a standalone DCC system for programming and testing our locos during the show.  Actually for a simple setup like this I think that JMRI would actually make things overly complex. And you would need a computer attached to your DCC system, making it no longer standalone. I"m not against JMRI, but for simple DCC setups it isn't really needed.

If you are interested in sound decoders (especially ESU or Zimo brands), both companies sell special programmers for downloading the sound files to their decoders. Those programmers are also capable of becoming a basic DCC system, so you can run your models using those programmers.

[railnerd]

There is a uber-huge Digitrax layout  in Los Altos that is among the densest concentration of EMF and multiple boosters in one model RR, it's often a "test bed" for Digitrax.  So they could make all of them work, and did.  But sometimes it took a "meeting of the right minds" to debug hard-to-shoot issues, and find the answers.

David and AJ had a bit of a falling out, and he has since switched to NCE for train control, and is now the largest test bed for LCC for his detection and signaling.

-Dave