That KATO RDC problem....? Well, funny thing ....

[OldEastRR]

Actually not so funny. Using a Digitrax DCS50 Zrphyr, with the same TCS Z2 decoder on the RDC w/ the light boards removed, on a layout that other decoder locos (most of them) have run on smoothly, makes the RDC, running normally on straight and level track far from any rail joints of any kind  SUDDENLY REVERSE DIRECTION AND SHOOT OFF AT TOP SPEED until the layout power is cut.
Ehh, I don't know about this DCC stuff... seems like it has potential but if you need a long long LONG learning curve to get a handle on just basic operation, I'll go back to DC. At least with that when you turn the throttle off the loco stops running.
Oh, BTW, this same berserk condition attacked my drop-in sound unit GG-1 which now,surprise! is burned out. From just running it slowly along a straight section of track.
ANd for fun, this Zephyr refuses to acknowledge there's a programing track, even when it's the only thing hooked to it. Joys of DCC?

[mmyers]

If it is running away at full speed with no control then there is most likely a short in the decoder installation. Track power is getting to the motor brushes and off it goes. Unfortunately, track power also feeds back into the decoder's motor output at the same time and fries it.
Not sure about the Zephyr's program outputs. Inadvertently allowing track power to make contact with the Program A or Program B outputs will fry the programming outputs. That is possible if one sets up a program siding that allows an engine to run into it from the main line. When it crosses the gap, the engine will momentarily jumper track power to the program track. Momentarily is all it takes. You might also check to make sure you are not in Operation Mode programming (Po on the display). That works only on the main line.

Martin Myers

[MK]

Correct Martin!  That's why on my program track (I too use a Zephyr) that is built into my layout a have a straight piece of track that is longer than nay of my locos.  Switching between Program and Track power is via a 4PDT toggle switch.

http://www.dccwiki.com/Programming_Track#Isolating_the_Programming_Track (after reading, scroll down for the 4PDT diagram).

[OldEastRR]

The programming track is completely separate from the layout track. Even using a separate Zephyr to only power the separate programming track still no reading the decoder.
The RDC does go back to normal after the racing incidents (happened twice so far). The decoder then works fine, not burned out. Sorry to say the GG-1 decoder was destroyed. Because of manufacturer mismatch, apparently. But no reimbursement from either.
A friend told me this happens when the power system is made by a different company than the decoder is. Everything works fine until the loco hits a dead spot or a dirty track and then it takes off like my RDC does. It happens randomly all the time, he says.
So this DCC thing is pretty much a con job, eh? "Just hook up 2 wires and you're set to go!"  is how they sell it. Go crazy, I suppose. Obviously there are many many important details to learn to avoid any grief. For those of you who don't mind shelling out $$$$ each year to replace decoders burned up by the glitches built into the way it's made and sold, I guess it's OK for you, too.
As for another arcane and mysterious detail of DCC, the front/rear lighting circuits on the RDC seem to have moods. Switching directions, sometimes the rear LED comes on, sometimes it doesn't.
The wonders of DCC are great for those with EE and Computer Programming degrees (Wow, I can't even imagine the amount of expertise it must take to use that JMRI stuff!!). I never got past just changing the loco address, much less all the dozens and dozens of complex and confusing settings for anything else. Now I hope at least SOME of my decoders don't explode or burn or just get mad at a competitor's control system and go nuts so I can run them on my layout, at least for a few years.

[mmyers]

Whoever told you mismatched equipment burns up decoders fed you a pile of  bull...t. I run just about every major type of decoder on my Digitrax sytems as well as my club's sytem. Haven't burned one up yet that wasn't my own fault. I've got over a 150 decoder installs . Burned up exactly 3. Every one because I rushed it and didn't properly isolate the motor or function outputs from the track.
There is more than two wires to hook up. For motor you will need 4. two for track\, two for the motor. Decoder goes between them and should be the only thing that is aver between them. BTW, I'm a carpenter with a 12th grade edumacation.

Martin Myers

[peteski]

The programming track is completely separate from the layout track. Even using a separate Zephyr to only power the separate programming track still no reading the decoder.
The RDC does go back to normal after the racing incidents (happened twice so far). The decoder then works fine, not burned out. Sorry to say the GG-1 decoder was destroyed. Because of manufacturer mismatch, apparently. But no reimbursement from either.
A friend told me this happens when the power system is made by a different company than the decoder is. Everything works fine until the loco hits a dead spot or a dirty track and then it takes off like my RDC does. It happens randomly all the time, he says.
So this DCC thing is pretty much a con job, eh? "Just hook up 2 wires and you're set to go!"  is how they sell it. Go crazy, I suppose. Obviously there are many many important details to learn to avoid any grief. For those of you who don't mind shelling out $$$$ each year to replace decoders burned up by the glitches built into the way it's made and sold, I guess it's OK for you, too.
As for another arcane and mysterious detail of DCC, the front/rear lighting circuits on the RDC seem to have moods. Switching directions, sometimes the rear LED comes on, sometimes it doesn't.
The wonders of DCC are great for those with EE and Computer Programming degrees (Wow, I can't even imagine the amount of expertise it must take to use that JMRI stuff!!). I never got past just changing the loco address, much less all the dozens and dozens of complex and confusing settings for anything else. Now I hope at least SOME of my decoders don't explode or burn or just get mad at a competitor's control system and go nuts so I can run them on my layout, at least for a few years.

If DCC was as bad or unreliable as you think then this section of the forum would be chock-full of complaints about how bad DCC is.  Yes, decoders are delicate (but still fairly robust) complex computer devices - not just a motor and a light bulb like DC locos have inside.  So, yes, then can sometimes lose their "mind". But many people (who are not computer engineers) think that the added functionality of simpler wiring and lighting and sound effects are worth the extra hassle.

If properly wired, DCC is simple and reliable. But I will never say that it is just as simple or robust as DC operation. Again, it is a tradeoff.   For a simple layout DC makes more sense. But if you have a larger layout and want to have operating sessions with several operators, DCC makes the wiring and control of the trains much simpler. I can't guarantee that it will be glitch and dead-reliable operation, but believe me, it is so much easier to have that type of operating sessions with DCC.

As fa as compatibility goes, that is pretty well worked out. In general, all brands of decoders work well with all brandsof command stations. 

[lyled1117]

Did you disable the DC functionality of the decoder?  The symptoms you describe many times are a result of not doing so. When a decoder becomes alive and aware of its own existence it looks for DCC commands. If it does not see them in a certain period of time it decides it is on DC track. It turns on its motor output to pass track voltage to the motor. That voltage is full voltage so the locomotive speed is full throttle. In addition the decoder thinking it is on DC track will no longer listen for DCC commands. The hand of gawd must come done and lift the locomotive or track power must be dropped to stop the locomotive.  This same runaway can be triggered by shorts and other flaky conditions. The reset circuits of decoders are better than they use to be, it happens less frequently than it use to, but it does happen.

If you are not going to run the locos on a DC layout, turn off DC capability. Most, but not all, decoders come with the option turned ON. Digitrax's Zephyr turns it ON if you use the LOCO2 or LOCO4 method of installing an address. You can over ride that by changing CV29. For a short address make CV = 2 (DC on = 6). For a long address make CV29 = 34 (DC on = 38). If you need to turn the direction of travel, add ONE to the value in CV29  (values of 3 and 35 respectively). Disabling DC is a step that most people ignore, but it can be an important one. I wish the manufacturers took the opposite viewpoint in that turning on DC capability should be a deliberate step rather than default.

Lyle Dowell

[lyled1117]

<snip>The RDC does go back to normal after the racing incidents (happened twice so far). The decoder then works fine, not burned out. Sorry to say the GG-1 decoder was destroyed. Because of manufacturer mismatch, apparently. But no reimbursement from either. A friend told me this happens when the power system is made by a different company than the decoder is.

This is totally untrue. A decoder has no idea of what system is powering it. I know of no system that isn't NMRA compliant. Being compliant means they all speak the same language in the same fashion. From the decoder's viewpoint the system is a black box. Your friend may have been experiencing the runaway issues described in my previous post but came up with his perceived solution.

Just as an FYI, I have been installing decoders professionally and privately for close to 20 years. Also, in a previous life as an electrical engineer I designed a DCC system (it never made it to market but it worked).

Lyle Dowell

[MK]

BTW, I'm a carpenter with a 12th grade edumacation.

Martin Myers

...with the experience of a PhD!  Don't sell yourself short Martin, I've been following your posts on the various groups and forums since I got into the hobby in the early 2000's.  You are always full of good information! 

[peteski]

...with the experience of a PhD!  Don't sell yourself short Martin, I've been following your posts on the various groups and forums since I got into the hobby in the early 2000's.  You are always full of good information! 

True, while higher education is helpful, experience usually trumps that, especially if the individual happens to be smart, curious, and willing to learn on their own.

[OldEastRR]

Did you disable the DC functionality of the decoder?  The symptoms you describe many times are a result of not doing so. When a decoder becomes alive and aware of its own existence it looks for DCC commands. If it does not see them in a certain period of time it decides it is on DC track. It turns on its motor output to pass track voltage to the motor. That voltage is full voltage so the locomotive speed is full throttle. In addition the decoder thinking it is on DC track will no longer listen for DCC commands. The hand of gawd must come done and lift the locomotive or track power must be dropped to stop the locomotive.  This same runaway can be triggered by shorts and other flaky conditions. The reset circuits of decoders are better than they use to be, it happens less frequently than it use to, but it does happen.

If you are not going to run the locos on a DC layout, turn off DC capability. Most, but not all, decoders come with the option turned ON. Digitrax's Zephyr turns it ON if you use the LOCO2 or LOCO4 method of installing an address. You can over ride that by changing CV29. For a short address make CV = 2 (DC on = 6). For a long address make CV29 = 34 (DC on = 38). If you need to turn the direction of travel, add ONE to the value in CV29  (values of 3 and 35 respectively). Disabling DC is a step that most people ignore, but it can be an important one. I wish the manufacturers took the opposite viewpoint in that turning on DC capability should be a deliberate step rather than default.

Lyle Dowell

I was talking with a friend about this problem and he said just what you did: check the DC operation control. The losing DCC signal and switching to DC racing off problem sounds exactly what is happening. Still not sure why the MRC decoder also blew up, but oh well ...
My question is, why isn't this "DC Off" feature listed prominently in decoder instructions? Seems to have a pretty big effect on operation. The CV setting section should start off with the question "Are you going to be running this loco ONLY on DCC power?"

[peteski]

I was talking with a friend about this problem and he said just what you did: check the DC operation control. The losing DCC signal and switching to DC racing off problem sounds exactly what is happening. Still not sure why the MRC decoder also blew up, but oh well ...
My question is, why isn't this "DC Off" feature listed prominently in decoder instructions? Seems to have a pretty big effect on operation. The CV setting section should start off with the question "Are you going to be running this loco ONLY on DCC power?"

Why not? Because it is just another feature of the DCC decoders (and most manufacturers leave it enabled, making them operational with DC too) and because generally it is not a big problem.  I have operated on many DCC layouts where the DC compatibility was left turned on in the decoders (as it is by default on most brands) and runaways were not happening.

Personally I also think that leaving DC-mode on is very handy for troubleshooting problems. When  loco acts like it is dead, I have a 9V battery handy. With DC mode enabled I can touch the battery terminals to the wheels and if the loco runs it is a good chance that  the decoder is ok (but the programming got scrambled). That to me is a very good basic decoder sanity check.  Plus it seems like a good idea to have a decoder-equipped model compatible with both DC and DCC by default.

As far as MRC decoders go, they don't have very good track record in general. When Athearn used them in the first batch of Big Boys, they had a tendency to heat up and go up in smoke.

[OldEastRR]

I may be getting close to an answer about the RDC, anyway. What happens when, on a DCC-equipped loco, a contact from the motor touches a pick-up strip with the loco running?

Had thing on the club layout again, it ran fine for a while, then did the running off full speed, stopped it, then ran nice, then fullspeed, stopped it etc etc. -- my friend disabled the DC option so that's not the answer.

But to end on a happy note (sort of) I finally got the lights to function properly in the right directions -- by scavenging the boards from another RDC. In which I discovered KATO made two kinds of LED colors for these headlights -- one a weak yellow and the other a bright white. So I need to swap out the LEDs on the "broken" boards. More fun microsoldering .....

[peteski]

Thee is probably zero possibility of you posting a clear close-up photo of each of those LED boards (both sides of each board)?  I would like to see them and I have feelings others would too.

As far as micro-soldering goes, I don't know if you are again being sarcastic, but I would not consider a 3mm through-hole-leads soldering as "micro". Well maybe if all you've done in the past was solder copper water pipes then yes.    If those LEDs were really small surface mount type like 0603 or 0402 (0.060 x 0.030" or 0.040 x 0.020" respectively) then that would be micro-soldering.    Maybe N scale is not your calling? 

[OldEastRR]

Thee is probably zero possibility of you posting a clear close-up photo of each of those LED boards (both sides of each board)?  I would like to see them and I have feelings others would too.

As far as micro-soldering goes, I don't know if you are again being sarcastic, but I would not consider a 3mm through-hole-leads soldering as "micro". Well maybe if all you've done in the past was solder copper water pipes then yes.    If those LEDs were really small surface mount type like 0603 or 0402 (0.060 x 0.030" or 0.040 x 0.020" respectively) then that would be micro-soldering.    Maybe N scale is not your calling? 

I'm pretty good at all aspects of N (including doing articles) but running DCC is spooky magic to me. Anything other than a factory-installed or KATO drop-in loco decoder I can't fathom. Things happen to me that no one else has even seen or heard of happening before. I doubt even the DCC gurus on this forum could figure out the problems I have.
The RDC lightboards turned out OK -- tho I had to reverse the + and - wire connections on one.