is digitrax loconet truly redundant?

greenspun.com : LUSENET : Wiring for DCC : One Thread

I'm having a problem with my Digitrax Chief system. I've contacted Digitrax, but they claim I don't have a problem. Frankly, I think they're trying to avoid a warranty repair. I'd appreciate your opinion and any advice you may have.

As I understand it, the loconet network system is supposed to be redundant. Which should mean that cutting any one wire in the net should not cause it to fail. Well, when the red wire is cut in my network, the system stops functioning. (Specifically, DT100's out on the net willnot clear off idle.) I've checked all the wires in the network for shorting and breaks with a continuity tester, and everything checks out O.K. I returned the DCS100 booster to Digitrax, and they claim there is nothing wrong with it. But I can't believe it.

More details: When Digitrax returned my booster, they included a net tester with it. They said if my net was working properly, all 4 lights on the tester would come on when it is plugged into any network port. Well, only 3 lights come on. But when I plug the tester directly into the booster, only 3 lights come on, too. When I told Digitrax the results, they changed their position to that only 3 lights need to come on for the system to be O.K! (By the way, when I plug an extra throttle into one port in the booster and plug the tester into the other port, *then* all 4 lights come on. The system also functions fine with the red wire cut, *if* I have an extra throttle plugged into the unused port, but stops functioning if the extra throttle is unplugged. This leads me to think there's some kind of break inside the booster, which winds up getting closed when an extra throttle is plugged in.)

Do you agree that cutting just one wire in the network shouldn't make it fail? Any ideas on what might be happening, or on what I might say to Digitrax to get them to repair my booster?

Has anyone else tried cutting the red wire to see whether the DT100's out on the loconet will fail to clear off idle?

-- jean c. piquette (jpiquette@code20nl.npt.nuwc.navy.mil), April 13, 1998

Answers

Jean,

Hmm, I am having a problem here... First it would really suprise me if Digitrax is trying to duck a repair job.. Heck they repair product in a blink. They have an exellent reputation here.. I sent in two boosters that I blew on my own :-( and they were returned in 24hrs ready to go..

Next I still can not figure why you would want to cut a LocoNet wire and then call it a Digitrax problem.. The LocoNet is redundant as far as polarity, and can actually function just fine as a three wire system. That is Black and Yellow tied together, Red and Green tied together, and Blue and White tied together. But have never heard of anyone trying to cut out wires??.

As far as the tester they sent you I know nothing about that.. Would have to look at it to understand it...

Now your idle not comming off. This is very understandable. The Red and Green LocoNet wires are the bus power leads. And are used just for that. Infact if you tie the Red and Green together, then tie them to either/both the Black an Yellow (LocoNet common) this removes power and will put all throttle on the bus in IDLE.

As far as agreeing with you well IMHO no I do not.. As what to say to Digitrax. I would suggest you either call or write and make sure you talk to AJ. No matter what the answer is, you can be assured he will give you the correct one..

I have never tried cutting any wire on the LocoNet, but I have tied the six into three as above and it works fine. Also back in the Big Boy days, I used a rely and tied em together as above to IDLE the trottles and apply power from a 12vdc wallwart to keep them alive and off the internal batteries. Ie turned the entire LocoNet into a battery save. Kill the power the Throttle IDLE automatically and 12vdc was applied. Kind of like the Chief is today..

I still after all this can not figure why you want to cut the RED wire, if you do not want to use it, just tie it to the green wire??

Don

-- Don Crano (donc@sssnet.com), April 13, 1998.


Just one more after thought..

You say when you add an extra DT100 to the booster all works fine.. Well if you feel there is a problem with your booster. Unhook all LocoNet wires from the booster, and just plug in one DT100. If it comes off IDLE and works like it should. Then I would have to say the DCS100 is fine and the problem is in your LocoNet wiring..

Don

-- Don Crano (donc@sssnet.com), April 13, 1998.


Jean,

I talked with AJ. It appears the green and red are not really both carrying the Loconet power. It goes out on the red. The green is dead until something bridges it. DT100 throttles bridge(short) the red and green together. At that point, red and green are carrying the power.

-- Allan Gartner (wire4dcc@aol.com), April 14, 1998.


Okay. Just to bring closure to this matter, below I'm copying the email I was sent by Digitrax when I asked them whether cutting the red wire in the loconet would make it fail. Below that is my answer to Don, which I hope expalins how I came to cut the red wire in the first place.

In light of Allan's posted reply to my question, I am now satisfied neither my booster nor my loconet have a problem. But I do wish Digitrax had told me in the first place what AJ told Allan. I also wish they'd clear up the material they have posted on their web page under "loconet wiring." That material definitely gives me the impression that no problem should come up if only the red wire is broken. (Also see the email from Digitrax to me copied below.)

Here's Digitrax' email to me on my question about cutting the red wire:

Glad to read that you are up and running. The red and green wires are the loconet signal wires and the system should run if one is down, however if for some reason the other wire is not continuous then the system will not talk to all the components. Please advise if you are still having problems. (770) 441 7992 Fax (770) 441-0759 P.O. Box 1424 Norcross, GA 30091 Ship to: 450 Cemetery St. Suite 206 Norcross, GA 30071

Here's my answer to Don:

Don,

To answer your question on why anyone would cut the red wire, I did *not* do this intentionally. What happened was that one day my DT100's out on the net would not clear off idle. So, I did a continuity check of all wires in the net, and discovered the red-wire connection on the male jack going into the DCS100 was bad. So I replaced the connector, and then the throttles cleared off idle, and everything worked fine.

But then I started to worry that just a single faulty connection should not cause a problem, because of the designed-in redundancy of the network. When I posted my question to Allan's Q&A section, I didn't remember where I got that idea. But if you go to Digitrax web page, and look under the hypertext "loconet wiring," it's all in there. In fact, right at the beginning of that material they claim that the network should work with *just 2 or 3 wires*! So, I hope you can understand why I went ahead and cut the red wire, to see what would happen. From what they say on their web page, I would expect things to still work. But they don't.

Why do I think Digitrax is trying to dodge a warranty repair? Well, at first when I contacted them, they agreed that cutting one wire should not cause a problem. So I returned my DCS100. They sent it back with the net tester, and said everything is all right. They also said that all four lights on the tester should come on if my net was wired correctly. But, when I found that plugging their tester directly into the booster gave only 3 lights, *then* they changed their position to only 3 (or 4) lights need to come on. (3 *or* 4? How can it be both? Does a software switch toggle between the 2 conditions? They haven't explained that, and so far have not answered my email asking about it.)

When I contacted them subsequently, they said the results I got with the tester show there is also no problem with my net! This means that Digitrax position is that: 1) The net should work with 2 or 3 wires. 2) My DCS100 is fine 3) My loconet is fine.

We have an inconsistency here. All 3 statements can't be right.

I disagree with you that plugging the DT100 directly into the booster demonstrates that the booster is fine. As I said in my posting, I think there's an internal break that gets closed when a DT100 is plugged in. The evidence for this is also in my posting: Only three lights come on in the tester when it's plugged into the booster alone, but all 4 come on when a DT100 is plugged into the booster along with the tester. (This at least mimics the closing of a break.) There is also the evidence that, with the red wire cut, a DT100 out on the net will not clear off idle *unless another DT100 is also plugged directly into the booster*. Plugging a second DT100 into the booster causes both itself, and a DT100 out on the net (with the red wire cut) to clear off idle. Unplugging the DT100 from the booster will cause both itself *and the DT100 out on the net* to go back to idle.

If you're right and the "2 or 3" wire system really only means that pairs of wires need to be connected together (which I would still consider a 6-wire system), they should have explained that to me, instead of stating their position as indicated in 1-3 above. But they didn't, and gave me every indication that cutting just one wire in the net should *not* make it fail. In fact, instead of explaining any of that, they have simply started to ignore my emails to them, and have in fact ignored 3 such communications.

Lest you think they misunderstood, I included a detailed list of instructions on what to do to reproduce my problem when I returned the booster. These instructions clearly said "cut the red wire." If it's simply obvious that cutting the red wire will cause the failure I'm getting, they should have said so. (And my problem would have ended there.) But they completely ignored those instructions, and simply returned the booster to me with the tester.

So, thanks for your answer. But I'm still not satisfied that everything is as it should be.

Sincerely, Jean.

-- jean c. piquette (piquettejc@code20nl.npt.nuwc.navy.mil), April 14, 1998.


Moderation questions? read the FAQ