King TTS decoder

[ColinB]

I bought a King TTS decoder to go in my new Hornby King. Anyway put it into the loco put it on my DCC railway and as soon as I pressed the "run" button the loco sped off on its own for half a circuit, by which time it stopped and started doing all the DCC stuff (sound etc). I have two DCC controllers an Elite and a Fleishmann twin track (I bought both as scrap and fixed them). On the Fleishman which is on the layout as I said before, soon as I press "run" which puts the DCC signal on the track it just shoots off.  On the Elite the loco behaves normally. Now having read many posts, I remembered Chrissaf mentioning if it does this, switch off the DC bit by setting the relevant CV, which does fix it. Thing that I find funny is it does not do this with any of my other TTS decoders and I have quite a few, plus it doesn't do it with the Hornby. I suppose they developed the TTS with an Elite so I suppose that is why it doesn't happen with the Elite.

[Chrissaf]

If you have access to an Oscilloscope I would be looking at the quality of your DCC track signal waveform. it needs to be a perfect Bi-polar squarewave without any spikes or sloping rises and falls.

.

Decoders (any decoder) assumes that it needs to invoke DC Running if it can't establish a stable synchronisation with the DCC signal timing. True, TTS decoders are more prone to this than other decoders, but even so, a good quality DCC waveform that has no distortion should be read correctly by all decoders including TTS ones.

.

Either your Fleishman is outputting a distorted waveform, or there is something about your track and DCC wiring method that is adding distortion. You may need to add a 'snubber' to your DCC power distribution to condition it.

[96RAF]

Rewire your bus with double-knitting yarn Colin you know it makes sense.

Seriously have you tried a reset of the twin-track.

[ColinB]

It only does it with this one, all the other TTS decoders worked perfectly out of the box. I must admit I have never checked to see if their default was DC off. This is a different generation TTS decoder to all the others I have, so I suspect it is not as tolerant to bad signals as the newer ones. Amazingly, I do have a really good digital oscilloscope I bought it when I retired, but my bet is with the error software or the initialisation in the TTS. I have the best part of 15 plus TTS decoder, 4 Zimo sound decoders and I dread to think how many ordinary decoders of which none misbehave, so my bet is with the error handling software in this decoder. I used to write comms software, admittadly it was CAN, but you will find they all do funny things if the initilisation code is not correct. It seems to be taking about 10 seconds to process the DCC signal, and doesn't switch to DCC mode until it has. I doubt the Fleishmann is giving a distorted signal, I may moan about German software being clunky, but generally it is very robust, anyway it would do it with at least one of the others. Even the LaisDCC decoders don't do it. It would be interesting to see if anyone else has had similar issues with this flavour of TTS.

[96RAF]

DC running is OFF by default on the latest batches Hornby decoders. This caused a few complaints as it prevented Analogue users from running DCC-Fitted locos as they had no way of re-enabling it without going back to the shop, or asking a friend or club.

it was introduced to preclude DC runaways. Ditto latest firmware in Hornby controllers boots them into DC running disabled for the same reason.

[ColinB]

When I was thinking about it all Hornbys latest TTS locos have DC disabled, so perhaps they noticed the same issue. In software terms  it is an easy fix if you see a high frequency pulse, disable DC, if after 5 seconds you don't get a valid DCC signal revert back to DC, but obviously they couldn't do that. I used to write engine control software, we used to do this sort of thing all the time, especialy for the flywheel sensor.  As I said, I have never tried all my other TTS decoders on DC so I wouldn't know. This one seems to have a lag of 5 to 10 seconds before doing anything, initially I thought it was inertia simulation but I am suspecting it is just the way it works. I had difficulty getting this one as Hornby don't sell it anymore and I doubt if another one would be any different, so I will put up with it. I think Hornby just adds the sound signatures, so the base software they have no control over, so they did the British thing, solve it by other means.

[96RAF]

I think Hornby just adds the sound signatures, so the base software they have no control over, so they did the British thing, solve it by other means.

 

Wrong - if you think back to the TTS uni-directional running and programming batch problem, this was fixed in-house by Hornby and decoders were returned to works to be re-programmed to correct it. The problem was in the decoder MCU programming not in the sound chip element. In a production environment of course the factor uses blank chips and loads the Hornby files provided under the contract.

 

All this speculation does not help and can even generate a false news urban myth about a presumed fault, e.g. the Select is a rubbish controller  - all based on the original firmware, long since sorted and the Select is now a pretty decent basic controller. Please stick to the facts or pose your concerns as a question and you should get the correct answer as per my TTS example above.

[ColinB]

I am not going to dwell on this. I originally said that my TTS controller was an earlier generation, meaning that they had fixed the issue on the newer ones. In fact none of the newer TTS decoders show the same issues, so if it was a fault they have fixed it, which is the normal process in software. You don't need to tell me about MCU reprogramming. There is a huge difference between MCU reprogramming and the actual software that runs the module. I assumed that they just added the sound signatures, which is what I expected them to do, implying the fault was down to the firm that wrote the base software, which means it was not their fault. From what you are saying, if they did write the base software then it is their fault. The reason I suspect that they didn't write the base software is that they would have fixed the "run away" issue, rather than have all that agro when a Dealer phones up about a disgruntled customer complaining he cannot get their latest loco to work on DC, not including the postage on all those returned ones. As for the Select I cannot understand why they don't use the same software as the Elite, perhaps they now do. Trouble is, it appears from all the posts on this site Hornby didn't realise that they would sell more Selects than Elites, so didn't sort all their issues out, until now.