Decoder issues with Coaching Stock.

[ModelHawk216]

I have a Hornby Railmaster set-up with an Elink controller which I love using.

I have a Revolution Trains OO-SGMIS-003 Inspection Saloon 975025 "Caroline" which I fitted with a ESU 59029 LokPilot 5 Basic Decorder and all the lighting functions worked perfectly. But then I noticed they did not operate.

Also the lighting functions are now not working on my Bachmann 39-737DC BR Mk2F DBSO Network Rail which was on the layout also.

Occasionally I get short circuits, due to derailments etc, but usually everything still works when I correct the issue.

I was wondering if anyone in the past has had this issue and can give me advise and help.

I have tried reprogramming them but to no avail. It seems odd to me that if there was an electrical issue of sorts, why did it affect only these two coaches. My Bachmann DRS DVT with DCC lighting was not affected and still works perfectly.

May I in advance thank anyone who replies to my posting. Thank You.

[96RAF]

It is likely the derail short has reset the decoders back to default address 3, so try it on that address and see if it responds.

You say you have tried reprogramming them without success. I suggest you reset the decoder first then try to readdress it. Check your decoder spec sheet / paperwork to find out what the reset value is and if it applies to CV8 the normal reset CV.

[P-Henny]

Try factory resetting the errant decoders by writing 8 to CV8 on the programming track. This is potentially opening a 'can of worms', but despite what others might say, 'short circuits' can corrupt decoders. I speak from personal experience. It is a lottery as to which decoders may or may not be affected by this issue.

Your decoders can be protected, from corruptions in the future (assuming that is the cause of your issue) by fitting a 'snubber' across your track once you have fixed your issue. The fact that 'snubbers' are documented as a protection for decoders somewhat proves, in my view, that decoders are indeed affected in some scenarios by 'shorts'.

Want to know more about 'snubbers'. What they are. How to make one etc. Then open the pinned sticky 'Useful Links' thread at the top of the 'General Discussion' forum and scroll down to the links to 'Mark Gurries' pages where the 'snubber' link is located. Mark was a member of NMRA and knows what he is talking about.

Once factory reset, you will of course have to reinstate your original configurations. I suggest testing them first on the factory default DCC Address 3 first to prove the reset is successful and the decoder recovered. If there has been a very severe corruption, then a factory reset may not make any difference and the decoder may be damaged beyond repair and need replacing.

P.S Although you use Railmaster, this is not really a Railmaster software question. Your posted thread would have been better placed in the 'Hornby DCC Forum'. If you lose sight of your thread here in this forum, it may have been moved to 'Hornby DCC' by a mod.

EDIT: 96RAF posted whilst I was still typing mine, but we both seem to be singing from the same hymn book.

[96RAF]

I am sure I read recently that unfortunately Mark Gurries had passed away, but his wealth of knowledge lives on thanks to the internet.

[P-Henny]

Google confirms that Mark passed very recently on 25th November 2023 aged 62.

https://www.trains.com/mrr/news-reviews/news/in-memoriam-mark-gurries/

The article in the linked page says that there is work in hand to save his website info if they can. So there is still a risk that it could be lost.

[Rog RJ]

That is very sad.

[ColinB]

Short circuits shouldn't blow up decoders, the decoder input is sitting the other side of the short circuit. I have had issues with ESU decoders just dying, surprising seeing as they have a high reputation. The only way a short could kill a decoder is when the controller reapplies the DCC signal after the fault has cleared but again I would be surprised if it did. You can do the usual like trying a reset but it sounds like it has died. The other thing to check is that there isn't a short on the lighting circuit.

[Fishmanoz]

Colin, I agree shorts shouldn’t blow up decoders but in practice they can, and the references above to Mark Gurries are the definitive source of why and how to fix.

The problem is stray reactance leading to high voltage transients, presumably stressing the DCC signal processing input of the decoder causing resets in milder cases and complete failure in others. Hence the snubber to dampen them out and, on large layouts, using twisted bus wires to minimise stray reactance in the first place.

[96RAF]

The late Chrissaf reported long, long ago that he had seen his Hornby acc decoders reset due to a derail short, so he moved them to a separate bus. I have followed that practice ever since along with snubbers to each bus.

[P-Henny]

Short circuits shouldn't blow up decoders, the decoder input is sitting the other side of the short circuit.

 

 

The Decoder INPUT is in effect directly across the track rails, so any changes in track voltage due to the short is also applied to the decoder input.

Yes, it is true that the short circuit current is not passing through the decoder, but in this documented scenario that is not the root cause of the problem.

What follows is my interpretated 'precis' of the technical discussions on the Mark Gurries site, so if anyone disagrees with it, hold a seance to ask Mark to justify them.

He says that when a track short circuit starts, there is a rise in current. If there is also a relatively high level of capacitance and/or inductance due to either the size of layout or the wiring methodology of it i.e lack of twisted pairs, the sudden rise in current creates 'ringing' on the leading edge of the DCC waveform. This ringing occurs BEFORE the controller short circuit protection completely kicks in. The 'ringing' generates voltages that exceed the maximum DCC voltage by a factor of 2 or 3 times. These high voltages can break down capacitors & components causing configuration corruption or if severe, complete destruction. He goes on to explain how the 'snubber' filters subdue these ringing voltages providing decoder protection.

[96RAF]

A lot depends upon the characteristics of the PSU powering the controller.

For instance the Hornby PSUs generally work on a constant power basis… i.e. as the load increases above the rated power level, the voltage drops thus maintaining constant power while increasing amps. This continues until the controller cries enough and cuts power to the track. Thus you can see the increased current could also toast a decoder.

It also explains why a Select controller screen flashes under 1-amp PSU overload/short conditions, as the voltage drops the controller reboots. With a 4-amp PSU the Select self protection kicks in before the PSU.

If you have a ‘scope watch the waveforms as a derail is simulated.

[ModelHawk216]

Hello All,

 

I am very pleased to report that I, with great support and advice from Hornby RailMaster Help, managed to sort out my issue I had with my layout since a electrical short I had in early December.

Therefore I am a very happy Railway Modeller again.

It turned out to be a Hornby Track Connector on my layout that must have been corrupted when the short happened, because when this was isolated everything worked as it should.

Thanks again to everyone who took the time to read my posting, especially those who replied with advice which was followed and ruling out certain issues that it could have been. This was much appreciated.

Happy Modelling Everyone.