Hello, I need help with a pair of Atlas O 7.5 switched configured in a crossover and controlled by a single atlas 6924 circuit board. I have 2 other sets of Atlas O switches (two 7.5s and two #5s) configured in crossovers and operating correctly with non-derailing and power routing. The problem switches are wired to their 6924 the same way the other crossovers sets are wired. The problem presents when I feed track power through the 3 amp fuses to the power routing connectors on the 6924. As soon as I add track power the board begins throwing the 2 switches at rapid random intervals. The behavior continues until I remove track power. I have been working on this problem for days with no success. I have swapped the board with another previously working 6924 this board also behaved the same way (so maybe it’s not the board). I took the switches and the board off the layout and connected everything using and old ZW. They worked correctly for non-derail and power routing. I did notice that the older 7.5 switch showed a much lower voltage on the other side of the frog. I thought this might have something to do with this switch because it is an early model 7.5 with ineffective wiring that reportedly does not power all sides of the switch reliably. I added a connection to the unpowered side and the power was consistent throughout the switch.
Next I put switches and the board back on the layout back connecting and testing one function at a time before connecting other functions:
- I connected Board Power (14V). The board lit up.
- I connected SW PWR IN (14V). IN A and IN B show 9 Volts (DC I think).
- I connected the switch machines to Out A and Out B. I used a jumper to ground to trigger IN A and then IN B. Both switch machines threw each switch correctly.
- I connected track power through the 3 amp fuse to COM2. It showed correct track voltage on COM2 and correct voltage on C1-2 and C2-2 depending on which direction the switch was set.
- I connected track power through the 3 amp fuse to COM1. It showed correct track voltage on COM1 and correct voltage on C1-1 and C2-1 depending on which direction the switch was set.
- I connected C1-2, C2-2, C1-1 and C1-2 to their associated closure rails. Each rail powered correctly depending on the switch position. (So far so good).
- I did note an odd thing on the old 7.5 switch. There was a voltage on a closure rail that should have been dead. The voltage was low, 4V. I disconnected the feed to the closure rail, but the 4V was still there so it must be leaking from the track power going through the center rails of the switch. I expect this will cause a short when I run a wheel set over it.
- Another issue with the old 7.5 switch is there is no way to insulate the switch for non-derailing. The two entry rails that are normally insulated for non-derailing are hardwired to each other and to the frog, so these rails cannot be used for the non-derailing control. I had to use an insulated rail on an approach track to the straight side of the switch. The curved side does not need non-derailing because both switches are automatically aligned for crossover using the switch controller button.
- I connected the insulated rails of the modern 7.5 switch and the approach rail of the old 7.5 to IN A and IN B. I applied track power. The board began its erratic behavior again. I thought it was the connection of the insulated rails to IN A and IN B so I powered down and disconnected them. When I power up and apply track power, the problem still persists.
10. It seems that the non-derail connections triggers the behavior, but once started it continues even if they are disconnected. I am baffled.
Thanks for reading this convoluted post.
