2018 WGH Charlotte Show Report
As a continuation of Kato USA’s increasing support for T-TRAK due to T-TRAK using Kato Unitrack, NRMRC was invited to set up and operate a T-TRAK layout as part of the Kato Booth at the World’s Greatest Hobby on Tour Show in Charlotte this past weekend. Our layout was U-shaped and was about 15’ x 8.5’.
DC & DCC
This layout is the first time that we have included DC power as well as DCC. The two inner (yellow track) loops were DC powered, while the red track was DCC powered. Kato provided the DC power packs for the two inner loops, as well as the trains operated on those loops — Olympian Hiawatha passenger set, and a freight train of BNSF Protein hoppers. Note: One of the two Kato DC power packs failed on Saturday morning in the full on mode, with the train taking off at 12 VDC speed. It was replaced.
The Hiawatha ran perfectly all weekend except for the one stray elbow that derailed it while the elbow’s owner was cleaning locomotive wheels on the adjacent track. The freight train was more problematic until its direction was reversed and a module with track issues was replaced on Sunday morning.
Various club members ran trains on the red track all weekend. Few problems were encountered.
This layout was clearly the Club’s best-looking T-TRAK layout to date. All modules were scenicked, and the view of the layout was excellent from all directions. There was a good crowd around the layout all weekend. At least two club members were present at all times to speak with the viewers; sometimes we had four members present.
Larry and Sue Hicks brought a new module displayed for the first time — a working drive-in theater. The module looked great and ran great, with lots of illuminations, including the screen itself, the projector, lighted vehicles, etc. An excellent first module for Larry and Sue.
There is an apparent issue with the yellow track on Flatsburg and Conhocton River Valley. The train on the yellow track would slow when traversing these modules. The modules themselves had no power feeds, but there was power feed at the adjacent modules at each end. No apparent slowdown occurred on red. Two possible considerations — the red track had a constant 14.5 volts DCC power (DCS51 output voltage) while the voltage on yellow was probably around 5 – 6 volts, thus resulting in voltage drop. The other consideration could be dirt in the UniJoiners.
A great layout and a good time.
For the NTRAK display we set up a layout about 80' x 34' with Eastern N Lines partners Central Carolina (CCNS) and Lynchburg (LANS). The Central Carolina folks erected a closed loop which connected to Red Rock Junction by way of Central Carolina Junction and a three foot bridge. From each end of Red Rock corner, a run of modules extended in a manner almost creating a loop. The West leg terminated with Lynchburg's Roundhouse turn-around module while the North leg terminated with Lollipop Farm module. This layout arrangement worked well and created a very spacious pit area.
Setup started about 9:30 AM with the arrival of the Carolina Central modules. Their loop was setup by early afternoon. We decided to power up their side using [Frank Fezzie's] command station substituting for the NRMRC equipment that had not yet arrived. They initially had some problems with both of their boosters indicating short circuits. The problem then appeared to resolve itself even though it was never clearly determined what was wrong. They proceeded to test and clean their loop.
A mid-afternoon arrival of the final NRMRC pieces allowed Lynchburg and NRMRC legs to be set up and command station, the club's DCS240, put into place. Some testing was done and the layout was declared as basically operational by 8:00 PM closing time.Don Wenzel brought his new Programming Track set up which was used heavily during the show. Thanks to Linwood for setting up Don's computer to work with JMRI and the Locobuffer, including downloading new drivers by way of his cell phone.
Red Rock Junction had a turnout problem, dead frog, in the straight through direction of the Red-to-Yellow turnout. Apparently both sides of the insulfrog were not soldered well to the frog power wire. We tried cleaning the turnout points to no avail as the turnout was just not conducting. While I resumed working on other things Joel went shopping for some solution to use on the turnout to improve conductivity and found a booth staffed by Deluxe Materials from the U.K. The nice young man from the booth came over to the module and brought a bottle of Track Magic, said to "clean and protect track and contacts for maximum response and reliability". After a few moments of attention the problem cleared and, I think, did not re-appear. Seems like pretty good stuff. While we were packing up the same young man from Deluxe brought over a bottle of Track Magic for the club. I will put it with the rest of our cleaning cars, Etc. Track Magic: "contains Naptha (petroleum), hydrotreated light" . I am familiar with naptha but have no idea what hydrotreated light is. I'm guessing that is the magic part.
Beyond that the layout had a fair number of raised rails and dirty track.
Red Rock junction and Central Carolina junction both have in-built reverser circuits on the RR Blue line which matches the CC Red line. Normally, such use of adjacent reversing circuits is not advised but in this situation, where we knew the two tracks would always be in phase, We deemed the risk acceptable. It turned out that Murphy one again proved us wrong and somehow one of the reverser circuits got set and a short was created. This was easily cured by disconnecting the reverser from the blue line on RR, and plugging in the mating track connectors so that the joined track section is powered only from the CC reverser. RR is designed so that such re-configurations can be made easily, usually by Power Poll connectors.
It should be noted that there was no official NRMRC show superintendent. Gil Brauch of CCNS was the overall show superintendent and designed the layout with input from John Wallis, Tim Hammock and Dave T.
N. O. Body, NRMRC Show Superintendent
Basic control setup was a DCS240 command station running a single Loconet to all components, 5 boosters supplying track power, 3 radio towers, 2 Digitrax LNWI Loconet Wi-Fi units supplied by Central Carolina, and numerous throttle jacks around the layout. The NRMRC club computers did not get packed so a substitute laptop was used to run JMRI monitoring the layout and to support the NRMRC router supporting smart phone throttles. Thanks to Linwood for the use of his computer and to Frank who volunteered his computer when we thought that there might be a problem with Linwood's. A channel scan revealed nearly 20 active Associate member routers in the arena and that channel 4 was the quietest available channel. The network was set to Duplex channel 14, which corresponds to Wi-Fi channel 4.
Work resumed at 8:00 AM Saturday morning and trains were rolling. Two problems became apparent.
First, duplex throttles were having significant problems getting radio messages to the system, so train control was spotty, with users frequently having to plug in to regain control of their trains. In some cases this was made more difficult because some throttle jacks around the layout were covered over by the skirting.
Frank passed word around that the throttles should be plugged in when acquiring, dispatching, MU'ing or breaking MU'ing on engines. Only actual running of trains should be done while unplugged.
The second problem was that the computer and router supporting smart phones through JMRI would fail after a few minutes of operation. We tried a second computer and had the same result.
We had a Raspberry Pi / JMRI / router package available so we hooked it up. It would also operate fine for a few minutes and then fail.
Our frustrations were exacerbated by the JMRI problem where, when you start JMRI and the LocoNet connection is not correct (USB port assignment for example), the defaults page of the preferences must be corrected. With all the restarts, computer changes, Etc. this problem repeated many times.
By mid-afternoon on Saturday, having tried everything that we could think of, we decided to switch the layout command station from the DCS240 to the DCS100 on Carolina Central's side of the layout. This was done about 5:00 PM. We got the layout powered up, but ran into an unusual problem. A computer running JMRI and connected to the Loconet on the NRMRC side could not read any Loconet traffic. The CCNS part of the layout appeared to be functioning well at this time. We ran out of time as the show closed up at 6:00 PM.
Work on the problem resumed at 9:00 AM Sunday morning. We installed a Loconet repeater (LNRP) between the Carolina Central side and the NRMRC / Lynchburg side, and quickly noted that the LNRP indicated a problem with the Loconet signal coming from the Carolina Central side. We starting rechecking the Loconet on that side and discovered that the Loconet was connected to the rear jacks of their LNRP and that it was not powered. We removed the unpowered LNRP from the Loconet and the error indication from our LNRP disappeared. Our computer running JMRI immediately began reading Loconet traffic.
Central Carolina had interpreted the Digitrax manual as saying that it was okay to have the Loconet connected to an unpowered LNRP, but this is clearly not the case. The unpowered LNRP was causing a problem with the Loconet signal.
We hoped that this was also causing our other communication problems but this was not the case. The computer and NRMRC router as well as the Raspberry Pi package both continued to operate for periods of time ranging from a few minutes to maybe an hour before they would fail. Duplex throttles like the UT4D and D402D were still subject to a lot of radio interference.
The two LNWI units of Carolina Central operated well and we had no problems *reported* about them. However, since each one can only support four phones at a time, both of their units were fully tied up most of the time. Frank had his personal LNWI with him so he hooked it up. It quickly had its maximum of four phones connected to it as well, but at least we could support 12 smart phone throttles running trains on the layout. A drawback of the LNWI is that there is no way to easily see how many phones are connected to it or who the users are. A user has to remember to disconnect from the LNWI when they quit running trains. I suspect that some users forgot to do this and therefore they keep their port in the LNWI tied up, even when not running a train.
Frank tested the simplex radio system used by UT4R and DT400R throttles. It was subject to radio interference from the NCE powered layout next to us, but it basically worked and he had minimal problems controlling a train with his simplex throttle.
Frank also tested infrared control. This option is rarely used but is actually built into all Digitrax throttles. Frank's radio tower was intentionally built with infrared receivers providing 360 degree coverage. He designed this after our experience at the NMRA show in Atlanta where we also encountered significant communication issues. The tower was located near the center of the layout at Charlotte to support this test. The infrared control worked properly and Frank was able to get his signal to the tower from any location around the layout, but he did note that he was at the extreme limit of the range of the technology at some points. Two such towers located near the middle of each half of the layout would have provided excellent coverage. One must point the throttle at the tower when using infrared, but otherwise the operation is very similar to the simplex radio system. Frank had also conducted this test at our last meeting in Danville with similar results.
Here is a summary of Frank's observations from this show regarding Loconet and communications:
Large shows held in large arenas with many routers running, such as this show and the one held in Atlanta, are going to present problems for Duplex radios and for computer / JMRI / routers used to support smart phone throttles. We do not understand why the routers are crashing. I also suspect that this problem can only get worse as more and more wireless devices are used. Possible solutions are;
- acquiring more powerful routers that can operate in a crowed radio environment without failing
- depend more on LNWI's to support smart phones. One Loconet can have up to 8 LNWI's which in turn would support 32 phones.
- Determine if duplex channel can be set to a different channel, say 25 or 26, while LNWI's are run on a lower channel. This would get duplex radios away from the router channels and make the duplex radios less subject to interference. Digitrax currently says no to this after problems we encountered at Danville last fall.
- Restructure radios towers to support infrared and then run throttles in infrared mode if necessary.
- All routers and LNWI's should require a password to connect to the network. This would prevent unwanted or accidental connections to the layout by phones that might be set to automatically connect to open networks.
Always connect the Loconet from the command station to the side jack of an LNRP. The LNRP warns of certain problems with the Loconet that are not always immediately apparent to us.
If the layout is going to be run on a single Loconet, then just continue the wiring as normal and do not use the rear jacks of the LNRP. You still get the benefit of the LNRP monitoring the Loconet signals. If it is desired to have separate boosternet and throttlenet on the layout then the Loconet off of the side jack and command station jacks becomes the boosternet and is run to the boosters. The loconet off of the rear jacks is the throttlenet and is run to the UP5's. It is preferable for radios, LNWI's, and computers to be connected to the boosternet, but they can be connected to the throttlenet if necessary.
Large layouts or multi-club layouts are probably best operated with separate throttlenet and boosternet wiring. An acceptable alternative would be to use a separate LNRP for each club to supply loconet to its section of the layout.
The LNWI's seem to be a good way to support smart phone throttles, even in tough radio environments. Because of limitations of the LNWI, a manual system should be devised and used to assign users to LNWI units and keep track of who is using which one.
I do not like using the DCS240 as a command station because JMRI is not fully compatible with it. JMRI cannot display the entire slot table of the DCS240 so you cannot see all of the engine numbers in use. This make the slot monitor in JMRI pretty much worthless for telling what is running on the layout. The JMRI features to release and clear slots appears to not work properly on the DCS240, and of course can only work on the slots that you can see, and cannot be used on the extended slots. JMRI cannot correctly interpret many of the messages that appear on the Loconet when the DCS240 is being used. This raises the possibility that JMRI might misunderstand a Loconet message and take some action that is unintended, perhaps even harmful. We thought that this might have been part of the cause for our computer routers to keep crashing. I believe the use of a DCS100 or DCS200 for the command station is a much better choice for our operations. Thus far, Digitrax has failed to provide JMRI with the information and permission needed so that JMRI can become fully compatible with the DCS240. We can at least send Digitrax a message by refusing to use the DCS240 until they provide the necessary information to JMRI.
Recognize that large layouts and multi-club layouts require more time to assemble the modules, run the wiring, test and troubleshoot the electronics, and test and clean the track. And people need to eat meals and check into hotels. Clamp-down time, the time when all modules must be present and ready to clamp together must be set early enough to allow sufficient time to perform these tasks prior to show opening. This was a problem at Charlotte. We did not have enough time to get the layout fully operational Friday, and we were troubleshooting control problems during show hours on Saturday.
We had the usual track problems with the layout such as joiner tracks that were misaligned, electrically dead rails on joiner tracks due to loose rail joiners, dead rails on turnouts due to bad power feeds or dirty points, and humps and dips in track causing derailments and causing cars to uncouple. The Green line was nearly impossible to run without derailments caused by track problems in the Lynchburg section, and as a result I think very few trains were run on the Green line. I ran it twice and then gave up on it. I don't think I ever ran the Yellow Line on the NRMRC / Lynchburg legs and I never ran the Yellow or Blue lines on the Carolina Central loop. This was partially due to the fact that I was absorbed in trying to solve the control and throttle problems we were having. As a result I have fewer comments on track work than I might otherwise have had.
Nobody wants control problems, throttle problems, or track problems. A layout that has these issues is much less fun to run. This layout had all of these problems. Frank had a lot less fun running on this layout than others we have done and Dave T had none. we suspect this may be true for a lot of the other members present this weekend. As part of the DCC team we offer apologies to everyone for the control issues. We *will* do better the next time.
Dave Thompson, DCC Committee Chair
Flickr photos from the group Eastern N Lines
Copyright © 1997–2018 North Raleigh Model Railroad Club. All rights reserved. The NRMRC is a Not-For-Profit Corporation incorporated in North Carolina. The various logos and heralds shown here are the property of their respective organizations.