How do I improve DCS signal?

Hi Train Doctor,

Try the simple stuff first.  Try running the middle loop off a different TIU channel.  I'd suggest swapping channels between your middle and outside loops.  If the problem moves to the outside loop you know the TIU has a bad channel.  If the problem stays on the middle loop you know the TIU is OK, and you can move to the layout.

It is possible that the coils in the relays that trigger the signals or TMCC engines are interfering with the DCS signal.  Start by disconnecting the relays and re-test.  If the signal improves, install inductors in the wires to the relay coils.  Remove all TMCC engines from the middle loop and re-test.  If signal improves add the TMCC engines back to the middle loop one at a time.  When you pinpoint the TMCC engine(s) causing the signal interference install inductors between the pick-up rollers and TMCC board.

Try those things and report back.

Dave

do you have enough feeders going to the track? The more feeders you have the better the reception . Also, what size wire are you using? If you are using too small a diameter for the main power supply you could have a resistence issue. Sort of like trying to get a golf ball to go through a garden hose.

Frank

Hi fgroza,

Actually, that's not always true.  You can spread the DCS signal too thin by splitting it up into too many feeders.  To prevent signal interference we split the center rail so that each length of track has a single pair of feeder wires.  If we have five feeders running off a single TIU channel, each section will get roughly 20% of the available DCS signal power.  If we have ten feeders, each section only gets 10% of the signal power.  The best method is to divide the layout into a fewest number of blocks that will provide good signal coverage.  If a layout has a single loop of track no more than 100 feet around the best results may be with a single pair of wires.  It's better to focus on good continuity across track joints so you can use larger blocks.

Thank you all, will take a whack at it tomorrow night, and see what happens. Will update. I don't think the TIU is at fault, as when we used the one TIU, the middle loop was on variable voltage, and running DCS locos was "iffy".

Hi fgroza,

DCS uses a completely different technology from DCC to send commands.  DCC takes DC current and swaps the polarity.  If the polarity swaps quickly that's read as a 1 and if the polarity swaps more slowly it's read as a 0.  The track power itself is the command signal.  This makes it easy for most people to understand.  If you have good electrical continuity you will have good command signal.  The down side is the bandwidth for transmitting command codes is very limited and commands can only be sent one way.

DCS uses spread spectrum technology that sends the command signal separate from track power.  This allows you to add DCS to any layout running AC or DC track power.  It also allows two-way communications between the DCS system and Protosound 2 and 3 locomotives.  When you add a PS2 or PS3 engine to your DCS remote you don't have to program a bunch of CV's.  You also don't have to learn a new list of function keys.  With DCS the engine talks back to the system and tells it which functions it has.  They are displayed above the remote's "soft keys" with alphanumeric abbreviations that are consistent among all PS2 and PS3 engines.  Essentially, DCS uses the track and power wires like a USB cable.  However, some wiring methods, such as bus wiring, don't play well with this type of signal technology.  Most layouts up to about 200 sq. ft. are pretty plug and play.  Larger layouts like Train Doctors's club can be challenging to get DCS to work well.  This is especially true of large existing layouts.  I actually got started as a layout builder by helping people re-wire their layouts to work properly with DCS.

Thanks for the info. This explaines a lot more than I realized. Again, I have to read up before I get started.

Frank

Thanks Russ, we can load a loco into the DCS remote, and switch to the middle track, it will run but not respond to the remote, and using the remote to check for signal strength shows no signal at most of the middle loop, and an occasional 1 here and there. Sheesh.