Random start ups of TTS fitted loco's

[JJGraphics]

2 hours ago, Michael Edge said:

We've had this on Carlisle but not just with TTS decoders - did they yours all go in the same direction? The problem seems to be with locos that are DC enabled (ours are not supposed to be but there are a few in the 200+ on the layout) when a loco somewhere else runs into points set the wrong way.

 

I had that happen with a visitor's loco fitted with a Zimo decoder which needed a "hand of God" grab to stop it hitting the end of my traverser at full speed. The loco owner was fairly new to DCC at the time and was not aware of the "analogue-enabled" issue. He had also experienced the runaway problem on his own layout with his Lenz controller but did not realise what was needed to fix it. I changed CV29 for him and explained what I was doing all has been well since.

 

John

[cravensdmufan]

4 hours ago, Michael Edge said:

We've had this on Carlisle but not just with TTS decoders - did they yours all go in the same direction? The problem seems to be with locos that are DC enabled (ours are not supposed to be but there are a few in the 200+ on the layout) when a loco somewhere else runs into points set the wrong way.

Thanks for that information Michael. I have had a few locos all take off suddenly (not just TTS ones) as soon as a loco hits a turnout set against it. Always bugged me! I will now turn off the DC enablement on the decoders. After all, I'd prefer a minor derailment than a loco flying full speed off the end of the layout!

[RedgateModels]

Fairly common issue with DC running enabled. TCS decoders would do it all the time with the Hornby Select controller. A extreme example of a less than perfect waveform being interpreted by a decoder as not DCC so assuming it's DC and running off at full track voltage (Well full 12v at least)

[RAF96]

Probably why many controllers actively do not support analogue running. If yours does then make sure you have not accidently selected loco zero and applied throttle as the stretche zero bit methodology will also distort the DCC signal even of no analogue locos are on track to use it.

[Michael Edge]

1 hour ago, cravensdmufan said:

Thanks for that information Michael. I have had a few locos all take off suddenly (not just TTS ones) as soon as a loco hits a turnout set against it. Always bugged me! I will now turn off the DC enablement on the decoders. After all, I'd prefer a minor derailment than a loco flying full speed off the end of the layout!

Glad to be of help, it puzzled me for a long time and nearly resulted in one expensive loco taking a 5ft dive to the tiled floor. When working on or operating Carlisle I always carry a largish coin in my pocket to drop on the track in case of emergency.

[Izzy]

I believe it is for the same basic reason that when stay-alive capacitor units are fitted that it is advised/required that DC running is disabled. That it is not mis-interperated as a DC signal. Certainly advised/required with TTS. It may also explain why some decoders won't play nice with stay-alive units, either don't work at all or give erractic responses.

 

Izzy

[Crosland]

On 14/03/2019 at 12:35, RAF96 said:

Probably why many controllers actively do not support analogue running. If yours does then make sure you have not accidently selected loco zero and applied throttle as the stretche zero bit methodology will also distort the DCC signal even of no analogue locos are on track to use it.

 

Stretched zeroes are part of ther NMRA spec, and have been for a very long time. They do not "distort" the DCC waveform. Decoders are designed to interpret it correctly.

 

The reason that running a DC loco on DCC is not widely supported is that it can fry the motors, especially expensive coreless types.

[cravensdmufan]

On 14/03/2019 at 13:10, Michael Edge said:

Glad to be of help, it puzzled me for a long time and nearly resulted in one expensive loco taking a 5ft dive to the tiled floor. When working on or operating Carlisle I always carry a largish coin in my pocket to drop on the track in case of emergency.

Update - I disabled DC operation on all 12 locos on the layout last night (all were set as DC enabled).  Then deliberately ran one against a turnout set against it.  Not one loco moved!  Nor did I get any unplanned start up sounds! 

 

As an added bonus, a few of the locos now run smoother.

 

Very pleased!  Thanks again Michael and other posters.

Edited March 15, 2019 by cravensdmufan
To add a further comment

[RAF96]

18 hours ago, Crosland said:

 

Stretched zeroes are part of ther NMRA spec, and have been for a very long time. They do not "distort" the DCC waveform. Decoders are designed to interpret it correctly.

 

 

OK it is a valid wavefrom modification but it certainly does not help if you have no analogue loco to run but one is selected. It just slows things down and can affect reliability dependant upon the decoder ‘quality’.

[Phil S]

If 'this problem' occurs whenever a loco 'hits' / passes over a point .... then surely the correct remedy is to investigate why the point interaction is causing the problem and replace the point with a better design or the wheel profile if it is steamroller wheel sets of the past.  Better that than removing a useful diagnostic feature - the capability of testing a loco with a 9V battery.  

I don't recall having a runaway condition since changing from ZTC ( whose instructions advised including a switch to delay powering the track until after the ZTC511 had started up).  

(Or the early Hornby decoders  which should all have been recalled in my view - I did send 6 back for replacement with ones which were 'fit for purpose' once they had modified the design  )   [ They had also claimed on the packaging that they worked on addresses 1- 256  instead of only 1-127.  ... as well as using Expressnet 8-)  ]

 

 

[cravensdmufan]

2 hours ago, Phil S said:

If 'this problem' occurs whenever a loco 'hits' / passes over a point .... then surely the correct remedy is to investigate why the point interaction is causing the problem and replace the point with a better design or the wheel profile if it is steamroller wheel sets of the past.  Better that than removing a useful diagnostic feature - the capability of testing a loco with a 9V battery.  

I don't recall having a runaway condition since changing from ZTC ( whose instructions advised including a switch to delay powering the track until after the ZTC511 had started up).  

(Or the early Hornby decoders  which should all have been recalled in my view - I did send 6 back for replacement with ones which were 'fit for purpose' once they had modified the design  )   [ They had also claimed on the packaging that they worked on addresses 1- 256  instead of only 1-127.  ... as well as using Expressnet 8-)  ]

 

 

The 'problem' was when a loco hit a wrongly set turnout (approaching from the heel end).  I have mostly live frogs, hence the short circit which caused other locos to suddenly take off at full speed.

 

So nothing to do with decoders as such, just turning off the DC capability has completely cured the sudden take offs.

 

But the signalman (me!) still makes errors so the short circuits still happen - it's just that I don't have to worry now that locos will suddenly fly off the end of the layout.  A Bachmann 47 hit the laminate floor once!  But thankfully survived without injury!

Edited March 16, 2019 by cravensdmufan
words added for clarification

[Phil S]

Do you have any protective circuits  such as PSX intelligent breakers  ....as I do....  When I have shorts from also running into wrongly set live frog points  .. and the PSX is attempting to auto reset every 2 seconds ....Reapplying power each time, albeit momentarily, I do not recall any runaways... then or after I have cleared the problem.

It might therefore be worth investigating whether the PSX  inhibits the problem  ( I no longer have a ZTC511 to see if I can re- introduce the problem with or without PSX.

 

Otherwise the only time I recall runaways/uncontrolled locos is when the user has accidentally set the Multimaus to an adjacent loco number 8-(    in which case the emergency stop button is usually useful ... but we have some of our multimauses set to stop ONLY the selected loco.... which does not help when the wrong loco is selected.

( They are set like this mainly for use on our shunting puzzles ...the user can stop precisely at the uncoupled without affecting the trains running in the background )

[cravensdmufan]

1 hour ago, Phil S said:

Do you have any protective circuits  such as PSX intelligent breakers  ....as I do....  When I have shorts from also running into wrongly set live frog points  .. and the PSX is attempting to auto reset every 2 seconds ....Reapplying power each time, albeit momentarily, I do not recall any runaways... then or after I have cleared the problem.

It might therefore be worth investigating whether the PSX  inhibits the problem  ( I no longer have a ZTC511 to see if I can re- introduce the problem with or without PSX.

 

Otherwise the only time I recall runaways/uncontrolled locos is when the user has accidentally set the Multimaus to an adjacent loco number 8-(    in which case the emergency stop button is usually useful ... but we have some of our multimauses set to stop ONLY the selected loco.... which does not help when the wrong loco is selected.

( They are set like this mainly for use on our shunting puzzles ...the user can stop precisely at the uncoupled without affecting the trains running in the background )

Thanks for the information Phil.  No I don't have protective circuits, but since incorporating a 5amp Tam Valley booster into the system, that shuts the juice  off when there is a short circuit until the problem is cleared.  This is an improvement to when I was using the just NCE Powercab on it's own which I believe only has a very limited short circuit protection.

 

Anyway, I'm very happy with the operation of the layout at present.  One day I may incorporate signals with interlocking with the turnouts etc - but the more I think about that, the more I think it will be a feature of my next layout!