Hi Alan,

It seems to me as during the switching the signal absorbs more than 0.5 A and that is the reason why the plastic housing of the reed relay has fused and consequently the reed contact is permanently in closed position.

If I remember well, during the switching the solenoid consumption is about 10 VA for a very short time (it means: 10 VA : 16 V = 0.625 A and this value exceeds the Switching current (max.) of 0.5 A). Therefore the reed relay is not able to open the contact, the solenoid of the signal is permanently energized and also the reed heats up till the fusion of the plastic.

Maybe an higher value of 1 A could make the difference.

Even better solution: the reed relay is connected to an auxiliary relay which is used to swith the signal.

Hope this explanation is clear and helpful for you.

Cheers and Happy New Year

Renato

Alan,

Regardless of the current rating, the general recommendation is not to switch accessories such as signals and turnouts directly from a reed switch due to the arcing and burning of contacts, as much from the very small size of the contacts as anything else.

You will achieve much longer life by going through an intermediate solid-state device.

A simple example would be feeding the Reed Switch to a driver transistor, and using that device in turn to actually switch the magnet.

I'm sure I've seen a really simple circuit for doing this, and it's like only 2 or 3 actual components. I will have seen it either in Tom Catherall's Digital Newsletter, or one of Rutger Friberg's Model Railroad Electronics books

Or even possibly posted here previously by one of our electronic boffins? While I'm comfortable working with electronics, I simply don't do enough of it to recommend the specific components required - but possibly someone else will come to the rescue. Clappers ( clapcott ), you have something in mind maybe??

To give you an idea, here's a really good and simple explanation with a neat little animation of how a transistor works and can be used to do exactly what you want

In the animation on the above site, the transistor 'Base' is where the Reed Switch would be connected to (low current ie; 5mA), the Collector is your higher current (500mA) from the Trafo, and the Emitter goes off to the Solenoid. This is somewhat simplified, you will need a couple of other components to make it actually work smoothly and reliably and need to get the correct transistor rating but I suggest it's worth having a play with, electronics can be a heap of fun to play with especially in this hobby because you can make all sorts of useful gadgets which you would otherwise pay a lot more to purchase a ready-made module to do the same thing.

Instructables is a great site for all sorts of info.

Good luck, let us know how you go.

Regards

Cookee
NZ

Resistors.

Somewhere on the forum is the diagram I posted of the circuit using triacs that was used on Bergtalbahn for reed switch load switching, before they were replaced with K83's. I'll see if I can find it.

With direct analog wiring, the reed needs to handle the power to switch eg a turnout just as egt a momentary contact track would do...

With full digital, it's enough that the reed indicates "closed or not" to an s88.
So using "any" reeds with digital is easier, they just have to close or not by magnetic influence when trains pass...

Hi Alan,

If I may take a slightly different slant on your situation. I am not sure you are asking the question with a full perspective.

Personally I consider the use of reed switches as the best of the "easy" train detection systems and when operated within there design specs the most reliable of any of the mechanical electronic devices.

I do echo the observation of others that suggest moving towards the use of the reed_switch as a "control" and not the driver itself
I do not adhere to the "buy a bigger" idea! Aside from the cosmetic issue, you will start to have issues with the loco shoe being physically affected.

However expanding the "reed as a control" tangent, the direction of a "Triac type" circuit is WRONG for someone like yourself (stated not electrician) or anyone who doesn't want to invest in something that only they can understand and fix (assuming component documentation has been kept). If you want an electronic solution (maybe because of cost) then keep it modular and standard.

Now, the real point I am heading towards is that at some time in the future you may be able to make a jump to a holistic control system (In the Marklin world this is an s88's/m83's and supporting controller - 6021/CS1/CS2 , with OR without a PC ). And most of us are horrified to realize that the many and varied bespoke and RIGID electronic solutions we have added have been unnecessarily time consuming AND , in there sum, expensive (usually more expensive than even a s88 and controller - but here the time frame does come into play). Even if cost is important, the versatility offered by a controller managed system is wonderful by not needing to "get under the layout" if you want to change your sequencing.

You have not mentioned your existing controller type - If it is analogue then that slows thing down, however I need to emphasize that you can get started with digital by working on the accessories (signal and point control) and maybe never change the locomotives. Up to a couple of years ago I would have suggested getting a second hand 6021 and 6043 memory (instead of a CS1/CS2) but I have to say that, for someone starting out, I think these days are over and a CS1/CS2 would be the target - unless you NEVER foresee using an advanced controller.

In closing, I am not not sure what your timeline and investment schedule is, but even if you do not see a controller based solution in the immediate future, plan your wiring for easy upgrade - i.e. do not run a wire via the "shortest possible route" between switch and signal, instead identify a nexus area under the layout where, in future, a m83 can be installed using the existing wires and a "bundle" of sensor/reed wires can convirge into a single cable to head for your s88.

Hi Brian. I am just looking for using the reed relays to extend the automated railroad grade crossing. I somehow dont like marklins idea of cutting the track in the manual . So any thats a diversion . I did post my question here before i read this article so sorry if it is a repeat.

https://www.marklin-users.net/fo...act-track-vs-reed-relays

so as there my question is if i use the neodymium magnets will is have issues with the digital decoders in the locomotive and points and crossing. It looks like you have used the NON MARKLIN magnet.

Thank you

Khirad