Product Finder
Click Here To Subscribe

PSX Tripping at The Reverse Loop Boundary

Posted by on Jul 8, 2015 in Blog, Tips DCC Components, Tony's Tips |

Buy PSX & PSX-AR

psx_2255

PSX-1 Shown

Hi Erik & Larry,

I am having the following problem with PSX boards. I Bought 2 PSX-4 boards from you within the last month, so they are new. I installed one PSX along with two PSX-AR boards on reverse loops, on my customer’s layout. It’s a NCE system with two 5 Amp boosters, boosters do not show short.

Testing with a single loco (no rolling stock)  when the loco crosses from the reverse loop (needing a polarity switch) into the PSX district, the PSX goes to a short, when it recycles to check for short it comes back up and has no problem. I can cause the PSX to short just by placing test light on the track, again when it cycles the light comes on.

Tried a lot of tests:

  • Normal draw about 0.5 a, he has lighted bumpers and some lighted cars, and other locomotives on the layout. Not near 3.8 amps.
  • Replaced with a second PSX-4 and the light bulb test showed the problem with all 8 PSX
  • Programmed a reset on all boards (CV63 to 42) with no change
  • Tried other locomotives, same problem. Actually have the problem with a lighted caboose crossing the gap.

This layout has been working for years with original Power Shields and On-Gaurd reversers, but one locomotive was causing the On-Gaurd to cycle power, so I suggested to the customer that he upgrade to the latest PSX design.

I tested the PSX’s (all 8) without the layout attached to remove other wire, and the light bulb test still shows the problem.  Immediate short and in about 1 -2 seconds power restored.  The PSX-ARs do not do this.

The gaps are not staggered, they are on the natural end of the track sections, but a light bulb causes the short and I have been using this tester for a long time.  Have installed lots of these boards and not I have not encountered this before.  All boards will go to a short condition when a real short is placed on the track, as expected.

I am starting Monday to install a very similar set up on a G scale layout with even more reverse loops. Your help is much appreciated.
Thank You, Rob Quimby

 

Hi Rob,

First, a light bulb is not the best choice for a test load because its cold resistance is 5 to 10 times less than when it is hot (i.e. lit). Thus a 2 amp light bulb (typical automotive tail light) will draw 10 to 20 amperes for a short time while turning on. The PSX interprets this sudden change in load as a short and turns off. When it turns on, it knows it is turning on and that there may be some non-linear loads (typically input capacitors on a sound decoder, but a light bulb behaves the same way). Hence, it monitors the current profile to determine if the load is turning on or there is just a short. This is why the PSX trips when you apply the light bulb but then turns on.

Now, for the real issue: the PSX tripping at the reverse loop boundary. When the loco crosses the gap, there is a short. It then becomes a contest between the PSX and the PSX-AR as to which one will operate first (e.g. trip or reverse). The fix is simple. The PSX series is designed to allow you to insert a programmable delay in the PSX between the over-current condition and the breaker tripping. This delay may also prevent the insertion of the light bulb from tripping the breaker. You enable the delay by setting CV55=1. This will enable a default delay of about 3ms, which is enough for most layouts. If you still have an issue, you can increase the delay by adjusting CV65. The default value is 24 (3 ms) each increment of 8 will increase the delay by 1 ms (e.g. 32=4ms, 40=5ms, etc.). Set CV65 just high enough to solve the problem. Unless there are very long wire runs, you should not need more than CV65=80, but you can go as high as 255 (about 30ms).

As an aside, if you are wiring large layouts with significant wire runs in DCC, you should use “speaker wire”. This is 2 wires molded in parallel to each other. This wire configuration cancels series inductance and prevents distortion of the DCC waveform. Waveform distortion from long runs can cause a significant delay in the operation of the PSX-AR. Also check Tony’s website for the article on Wiring Long Reverse Loops. The appropriate wire can be purchased from Parts Express. Part number SKRL-12-100 is 100 feet of 12 gauge “speaker wire” at a reasonable price, and this should work well for most layouts.