Hi all
In another group i stumble across some guys who use a booster to protect the command station. They connect the layout to the booster not using the command station to power the layout.
I havent heard of that before, has some logic, but, is really necesary this kind of protection today with almost all stations having galvanic isolation?
Any thoughts?
Regards

Hello Tom
My doubt finally is if a short circuit in the layout can really kill a command station with galvanic isolation.
If so, to feed the layout trough a booster seems not a bad idea at all.
I had a 60214 fried years ago, but was an early unit, with no galvanic isolation
Regards

Galvanic isolation has nothing to do with this, it is purely about destroying the outputs from the control unit with a short circuit (or more likely a low impedance circuit that doesn't cause the overload protection to kick in). The continuous current draw at a level that is causing the drivers to have a lot of voltage across them and therefore dissipate a lot of heat causes them to burn out.

It is easy to get to this stage with a derailment (say on a point) where the combined resistance of the feeder wires and the imperfect connection of the wheels or whatever that is causing the short will limit the current below the overload current that would cut out the output stage, but the output drivers cannot maintain the full output voltage, but in trying to do so will burn out. so as mentioned, some people prefer to use a booster as a sacrificial device to keep the control unit intact.

This is how I've wired my layout - it is feed from two boosters but not the CS2 controller. It's cheaper to replace a booster than it is a controller.



You can connect whatever you like to the CS3's track output as long as you follow the rules around only having one device with no galvanic isolation providing track power - the CS3 does not have galvanic isolation, the CS3+ does (a really daft decision by Marklin IMHO). But doing so kind of negates the reason for not using the controller's track output in the first place.

Hello again,

I should have been more precise ... I actually have a CS3+ and so don't have a Link-88 with it's own power supply. My S88s plug directly into the CS3 plus's RJ45 port.

So my question needs to be re-framed: Is there any protective value for my CS3+ if I isolate my M83s, M84s and signals to a booster, in lieu of powering them directly from the Control Station? Since my M83s and M84s all have their own power supplies, the only thing drawing power from the CS3+ should be the S88s and the handful of signals that have built in decoders.

Thanks and sorry for the confusion,

Thank you for your response Minok.

As I transition from last RR build season's Tisch Bahn of "Test Ovals" to the first phase(s) of a permeant layout, I will follow everyone's advice and plan for the following:

• Start with a single Booster to power my track and control my M83s and M84s.
  
• Other than the booster, the only components connected to my CS3+ will be the S88s.
  
• Each block of M83s and M84s will receive their own power supply.
  
• I will provide independent digital feeds for each M83 / M84 block, so that in the future they can be moved to new booster regions.
  
• As the layout grows, I will add a terminal and additional boosters as needed.

Since I've built modular layout furniture (in anticipation of room and or house moves), it should not cause a huge amount of pain to add the new power regions. I do have one important question though ... should I keep a common ground ("O" / Brown) across the CS3+ and all of the booster regions? My thought is yes. But I remember reading a recent post suggesting that this is not a good practice for very large layouts.

Thank you everyone for bringing up this very important topic at an extremely timely moment, for my layout's final design.

Regards,