Is it just a coincidence, or do Protosound 3 equipped locomotives running with DCS Version 4.3 on a Revision G TIU require better wiring than running with earlier DCS Versions or Protosound 2 equipped locomotives???
A background, and then the long diagnostics and results...
We've been working on a layout since 2005, expanding it a little bit more each year. Trackwork is entirely Gargraves Phantom Line Flexible track with a mixture of Gargraves and Ross switches. Track 1 is about 115' long, with two long passing sidings, two reverse loops, and a six track passenger yard. Track 2 is about 90' long and is essentially a figure-eight, over-under, bent into an "L" with one leg being more gradual and the other steeper. Track 3 parallels Track 2, and also features a six track freight yard and the lead to an engine facillity under construction. All three tracks share a tunnel constructed with OSB board as a roof, and fanfold insulation sheathing and the track is unballasted. The six tracks in the passenger yard off Track 1 all have the center power rails insulated, originally powered through SPDT toggle switches at one end of the yard, however they have recently been rewired using three-way household lighting switches allowing power to be controlled at either end of the yard. All the freight yard tracks off Track 3 are uninsulated. There is a set of crossovers which allows movement between Track 1 and Track 2, Track 1 and Track 3, and a set of double crossovers between Track 2 and Track 3. Each crossover is insulated on all three rails.
We have used the Atlas Signal System consisting of four block signals on both Track 1 and Track 3, with the outside rail being unpowered and used for signal communication, and the inside being the track ground. Track 2 remains the "plain" track with no signals or switches other than the crossovers. As mentioned above, DCS is the primary form of control for the layout, with a Z-4000 powering Track 1 through Fixed Voltage 1 and Track 3 through Variable Voltage 2 in fixed mode, and a Z-1000 brick powering Track 2 through Variable Voltage 1 in fixed mode. A Z-750 brick provides auxillary power to the TIU at all times. Fixed Voltage 2 is reserved for a future G-gauge layout that will be running around the base of the O-gauge.
There is nothing fancy about the wiring scheme. No bulbs, no filters, just straight 14 gauge wire from Transformer to TIU to Track. Both Track 2 and Track 3 each have two wire drops, at equal spacing apart at the top and bottom of the figure-eight. Up until this past October, Track 1 featured only three wire drops equally spaced on the mainline, one on the longer passing siding, and then one for the six-insulated passenger tracks mentioned above. Again, nothing fancy here, each output on the TIU has a length of 14 gauge wire to the center of the layout where each drop is joined together at a hub with a wire nut. There are no insulating gaps between the wire drops as recommended in the star wiring pattern. For nearly nine years this method of wiring has worked PERFECTLY, running conventional, DCS (Revision G with Versions 2.3 and later 3.10) with dozens of PS2 engines and TMCC through DCS.
With more and more PS3 locomotives hitting the market, and the quilible whistle being a feature of many of these, it became apparent that the TIU/Remote would need to be upgraded to a Version 4+ to access these features. The TIU and Remote were upgraded at the Fall 2014 TCA meet in York by MTH to Version 4.3. Back home and installed on the layout, a delimma soon appeared. On Track 1 when trying to start a PS2 engine, ENGINE NOT ON TRACK would appear. On Track 2, the engine would start up with no issue, and on Track 3 it would have the same ENGINE NOT ON TRACK message, however would start up immidately afterwards. Curiously, an engine on either Track 2 or 3 would move on to Track 1, and once on Track 1 the only function the remote had control of was throttle and direction-no bell, whistle, headlight, smoke etc. Interestingly, once the train was brought to a stop, even the direction and throttle ceased working, either nothing would happen on the remote, or ENGINE NOT ON TRACK would appear. I tried several times while on Track 1 to remove the existing ENGINE NOT ON TRACK and re-add, or add a new engine never before in the remote only to receive NO ENGINE TO ADD. On Track 3 it was hit-or-miss, sometimes I could add an engine in with no issue, other times I would get ENGINE ERROR only to have it add, other times would be the classic MTC ghost engine-but each time I would get something other than NO ENGINE TO ADD. One really bizarre attempt was after trying to add an engine on Track 1 with both Track 2 and Track 3 powered down, I kept Track 1 powered up while starting a train on Track 3. As the train on track 3 started to move I noticed the engine on Track 1 which I was not able to add into the remote had started moving-still in an unstarted (no sound/lights) state. Regardless of what I did with the remote to the locomotive on Track 3, the haunted engine on Track 1 kept crawling along at the same speed until I finally interrupted the power. Track 2 would always allow me to add an engine the first time, no issues-and to rule out the battery in these engines the process was repeated on the other two tracks generating my previous results. Multiple engines were tried multiple times, the inputs and outputs were all swapped about at the TIU, two seperate remotes and finally alternating with a teather after each failed attempt; yet each test proved that something on Track 1 was preventing engines to be added, and something on Track 3 was messing up the engine adding process, while Track 2 was functioning perfectly.
I did a lot of late night research and its my understanding that the Revision L TIU's have a much more powerful signal processor than the previous revisions. One benefit I have read about is there is no longer the need to install a bulb or filter to improve signal strength (neither of which was done on the layout previously). With DCS version 4+ being released following the release of the Revision L TIU's, I couldn't help but wonder if there was a connection here-does the version 4+ software actually retard the DCS signal being sent out of the TIU assuming it is a powerful Revision L? If this is the case, with the software having no way to distinguish between a Revision G and L, once my remote and TIU were upgraded to 4.3, the system retards the already weaker DCS signal in my Revision G TIU. In otherwords, where I used to have 9's and 10's on the layout with version 3.10, would I have a lower signal with version 4.3?
With this theory fresh in my mind, I started out a locomotive on Track 2, put it in signal strength mode, and ran it around Track 2. I was shocked to see the highest reading I received was an 8 and that was right above one of the wire drops-I was averaging 6's with an occasional 5 far from the wire drops. Intrigued by this, I kept the engine in signal mode and threw the crossover onto Track 1. Not surprisingly the highest signal I received was a 7, also on a wire drop. Amazingly most of Track 1 was giving me 4's, with a couple of stretches receiving 1's. I added two more wire drops spaced equally on the mainline on Track 1, and much to my delight I was now able to add PS2 locomotives into the remote, and had full control of them. Remembering I had some issues with adding engines on Track 3, I added another wire drop to the far end of the yard near the engine facillity to help improve the signal through the yard, and this helped improved the signal on the mainline, also allowing me to add engines into the remote the first time. I felt that I proved my theory with this test, and was pretty happy with the results.
With Christmas this past week came the arrival of two new PS3 locomotives-the first O-gauge equipped to make it. Previously we hooked up a short stretch of track to test out a Standard gauge tinplate locomotive with PS3, funnily enough powered by using jumper wires off of Track 1 right at a wire drop, and with no issues at all. Christmas afternoon the first engine came out of the box and was placed on Track 1 right above a wire drop in anticipation of a quick and easy add to the remote. Power was applied and the headlight turned on immediately and after several seconds the engine came to life. Knowing this is how the engine operates in conventional mode, I tried adding the engine into the remote regardless-NO ENGINE TO ADD. I reset the power, this time the headlight remained off and the engine did not start up. Again I tried adding the engine-NO ENGINE TO ADD. Remembering that the PS3 engines have a capacitor that needs to charge, I let the engine sit for well over a minute before trying again to no avail. Next was trying to add the engine with the phone cord teather, again with the same negative results. Thinking I somehow fried the engine, I hooked the transformer directly to the track bypassing the TIU and the engine not only started up in conventional, but it ran beautifully. With the track wires already out of the TIU, I swapped them around to different outputs on the TIU, and to the different Transformer outputs. With no positive results, and thinking maybe something got knocked out of place while cleaning the day before, I took the engine off the track and tried a PS2 engine on Track 1. I was able to add it into the remote and operate it with no problems at all. I performed a signal test on Track 1, with the highest spot being an 8 at only one wire drop, averaging 5's, and a good amount of 3's (this is since adding the two wire drops mentioned above) Going under the assumption it must be a problem with this engine, I tried the other new PS3 engine only to receive the same results. Thinking I had exhausted all my options, out of desperation I placed the first PS3 engine onto Track 2. Much to my surprise the locomotive added right away to the remote! The locomotive ran beautifully on Track 2, and I was able to cross it over to Track 3 and finally on to Track 1. However, once on Track 1, like with the PS2 issues I mentioned above, I lost control of the engine of all features except the throttle and direction until it was brought to a stop, at which point all I received was ENGINE NOT ON TRACK, while the engine was standing still with lights, sound and smoke on. I then tried to add the second PS3 engine into the remote while on Track 3 which worked as it should, and again lost control of it once it was on Track 1. I tried physically moving the engine to different spots along Track 1 thinking it was just a bad spot where it would not work with no positive results at all. Remembering the poor signal readings I got on Track 1 versus Track 2 got me thinking, and I placed the engine on a spot on Track 2 where I remember receiving a 5 reading. ENGINE NOT ON TRACK. I moved the engine back to the wire drop where I got an 8 and the engine responded normally. With that in mind I ran the engine around on Track 2 and used every function I could while it was moving and where it dropped down to 5 I got no response, however elsewhere it performed normally. To add to the fun, I tried three different PS2 engines on each track, each of which not only added in the remote, but responded normally on the entire length of each track.
From all of this I have a few theories...
Revision G TIU's operating with DCS v4.3 will have a weaker signal strength than with earlier DCS versions
PS3 equipped locomotives will not add into a remote with a signal weaker than a 8
PS3 equipped locomotives will not respond to any DCS command with a signal weaker than a 6
Unfortunately, I am not capable of backdating my DCS system to an earlier version of DCS to test out these theories. So:
1) Has anyone run into a Pre-Revision L TIU having a weaker DCS signal with Version 4.3 than earlier versions?
2) Short of obtaining a new Revision L TIU, is there a preferred method for improving DCS signal strength through wiring SPECIFICALLY for Version 4.3? (cutting rails, adding more/fewer drops?)
Original Post
