Inconsistent frog switching with Cobalt IP Analogue point motors

[doug22150]

I used SEEP point motors for the visible part of my layout but in the fiddle yard, decided to try Cobalts.  Having installed 10 of them and completed the wiring of a control panel,  I was disappointed to find that 3 of them don't switch the frog polarity reliably.  What happens is that each of them, when switching in one direction only, seems to cause a short circuit. As the point blades begin to move, DCC power cuts out (as shown by the neon on my NCE Powercab) for several seconds.  After this, there is a click and the power is restored.  In each case, switching the points in the other direction, power is maintained all of the time.  I don't know whether it's a coincidence but all of the points affected have been fitted with Cobalt 90 degree adaptors. Has anyone else experienced problems like this? I've tried using the alternative switch provided on the point motors but it makes no difference. I have removed all of the over-centre springs on the points but made no other modifications. 

I have now installed a Hex Frog Juicer to overcome the frog switching problems as changing the position of the motors by the every small amount necessary has proved impossible in the congested space under the baseboard in n gauge.  All points are now switching perfectly.  This does make for an expensive solution but at least I can now run trains without interruption!