Maine Central, Lamoille Valley

Maine Central, Lamoille Valley
Click image to link to my web site, nekrailroad.com

Thursday, July 09, 2015

Short Circuit Issues and possible fix

The Northeast Kingdom uses Digitrax, with a DCS100 and DB150 in booster mode handling DCC duties. I route the output from these 2 units into 6 separate PowerShield circuit breakers to create distinct power districts. The goal is to prevent a short circuit in one location from taking out the whole layout. In general this has worked fine. A derailment or engine running into a thrown switch in Groveton would not impact operations in St. Johnsbury yard, or Morrisville yard. All was fine except for one issue.

If too many sound locos were in one power district (and especially if this included QSI sound decoder equipped locos) a short circuit would trip the breaker, which was good, but it was unable to reset itself. The rest of the layout was good, but the power district would remain down until enough locomotives were removed (or at least tipped off one rail) to reduce the power draw and allow a reset of the circuit board controlling that district.

In research and asking questions, I found out that I could use jumper wires on the PowerShield boards to up the threshold for current draw before tripping, allowing for a reset even when a large number of locos are drawing power on the attempted restart (the sound decoders were drawing a larger amount of power on startup then settling back down a bit).

This basically worked, but then I started to see an occasional issue where a short would take down the whole layout, essentially rendering the PowerShields useless in providing power districts. I contemplated what to do and thought about removing the jumpers that upped the threshold to at least get back to what was working before, and the undesireable issue within just one power district to deal with.

During the Op Session, just such a short occurred and took out the railroad. A quick fix is to cycle the track power on and off via to the DT400 throttle. But I started to discuss the issue with Bruce Barrett and he asked to see how I had my DCC system set up and located, physically, which I though was interesting.

I have the wire outputs from the command station and booster running a short distance to inputs for the PowerShields, maybe 8 inches of wire. Bruce said "Aha!", but I was not getting it.

He explained that what was happening is that the PowerShields were letting the short occur long enough before tripping that the command station was seeing it and shutting down, thereby impacting the whole layout. Interesting I thought - but how to fix it...

Bruce said some people have their circuit breakers out on the layout, further away from the command stations, and the running of the power bus out to the boards was enough to prevent a short from getting to the command station before the PowerShield tripped.

In my case, Bruce suggested routing the existing power outputs out one leg of the layout and back before attaching to the PowerShields. Maybe creating a wire distance of 30 feet or so between the command station and the PowerShields should do the trick he advised.

It does sound logical given what I have seen occurring. I will give this solution a try tonight and test out short circuit impacts within each district. I'll report back the results, in case anyone else has experienced similar issues. And if you have tried this solution, please comment and let me know your results.

This is an older look at my power center, before expanding to a DCS100 and another PowerShield board. But it helps show the relatively short distance between the command station and the circuit boards.

No comments: