DCC Dilemma

[Chrissaf]

The first thing to do is disable dc running in all of the affected loco decoders by adjusting the value of CV29.

[Chrissaf]

To do what Rog(RJ) is suggesting (a good idea by the way) you will probably need to read this web site information on CV29. Loco uncontrolled 'runaway' is nearly always attributable to 'DC Operation' being enabled in CV29. Disable it, you don't really need it enabled on a DCC controlled layout.

.

http://www.2mm.org.uk/articles/cv29%20calculator.htm

[Chrissaf]

Firstly a bit of background info: I have built over the last six months a semicircular layout that is 25ft long and 2 of 4ft wide on either side. The layout is controlled with Railmaster with 5 amps of power. There is no DC present. The sidings contain 55 locos with assorted coaches and wagons.

Now to the nubb of the problem, I can run I loco all day long with no problem, But if I run 2 or 3 then occasionally a random loco in the sidings will start at full speed and  ram the one parked in front of it, or one would come out of the sidings and derail at the point. The runaway loco will not respond to the stop button and the only way to stop it is to cut the power or remove it from the track. The last event was the loco Cairngorm ramming a scarce Pullman observation car. If anyone has any ideas, they will be welcomed with deep gratitude, Regards Tom

[Bigskybirds]

I have experienced the runaway problems especially when I had an Elite controller. I put a switch between the track and controller. I use DPDT switches to switch between DC and DCC but centre is off so if I get a runaway I cut the power to the track by switching to centre off.

 

I made a video for a previous forum poster in which you can see the switch panel at 1 minute 37 seconds in to video.

 

https://youtu.be/3TvXFpu5qEo

[Ard Lochan]

As stated in my original post, There is no dc current present, all points are ac and all signal lights tat are dc have been unplugged. This problem has been ongoing for weeks now and I previously removed run on DC from cv 28 on all locos a while ago. I have also installed a kill switch but often when 2 or3 locos are running it is sometimes not possible to hear another start up of its own accord, Regards, Tom

[Bigskybirds]

 If you read my post although I use DC I am referring to using a switch to cut the DCC power in different sections when more than one loco is running. Just forget about the DC bit.

[Rog RJ]

OK.  So with dc running disabled (CV29 BTW not CV28), the next thing I would try is fitting snubbers, (often, wrongly, referred to as terminators), to reduce any voltage spikes.

[Howbi]

@ARD Lochan..........Tom, you said  "As stated in my original post, There is no dc current present"......that's not strictly true because every decoder has a Bridge Rectifier built in so the motor, lights, sound etc are all run on DC......but we know what you meant......On a large layout it is always possible that a rogue spike can upset the decoder, as Rog has mentioned..........with so many locos on track presumably in sidings or fiddle yard, it may be necessary to fit some isolating switches so that not so much track/locos are live all the time........HB

[Chrissaf]

You also state that the controller is RailMaster with 5 amps of power. Given that RailMaster only works with a Hornby track interface (Elite and/or eLink) and that the maximum power supply current that Hornby support is 4 amps. Then that infers that you are using a non Hornby power supply for your 5 amp supply. With so many locos drawing power (even a stationary DCC loco will be consuming a small amount of current) who knows what an Oscilloscope image of your track signal would look like, probably not a very clean DCC signal at all, particularly with the unknown impact of using a non Hornby certified power supply. A dirty (spikey) DCC signal can confuse a DCC decoder.

.

I would suspect ( particularly that you say you have already disabled DC operation in CV29 [not CV28] ) that your layout would benefit from a complete review of your power distribution design. Fitting 'Snubbers' again as Rog(RJ) has suggested, could be beneficial as well as possibly creating isolated 'power districts' with power boosters to even out the distributed power loads.

.

Don't know what a 'Snubber' is.....then review these pages:

https://sites.google.com/site/markgurries/home/dcc-general-best-practices/wiring-planing/snubbers-rc-filter

.

Note that Snubbers consume power, about 100 mA per Snubber fitted, so use sparingly.

[Ard Lochan]

Fixed [Hopefully].

First a bit of background info: My job before I retired last year was as a motor vehicle diagnostic technichian. There is a similarity between can-bus on a motor vehicle where a digital signal is sent to the consumer down a twisted pair, for example your steering wheel and column where you would have a mass of wires to operate various bits and pieces, with can-bus you have a lot less wiring as the switches send a signal in digital format to the consumer. There is similarity between DCC control and can-bus where a signal is sent down the track to a consumer, eg the loco.

I have split the track into two power zones with a hornby 4 amp power supply on the main circuit and a hornby  4 amp power supply conected to a hornby power and signal booster [R8239] to the other circuit. Using the 4channel scope set to auto, one channel on one circuit and one on  the other  I see a nice clean AC waveform on both circuits.

Running a loco on both circuits I see the signal sent to the locos to make them run, After about 10 mins I start a second loco on the second circuit and all is well. Stop one on the second circuit and start another, repeat several times and then all hell is let loose on the second circuit where 4 or 5 locos take off at high speed. I cut power to both tracks and remove affected locos and then playback recorded waveforms on scope, channel 2 is showing a mess of signals. Switch popwer back on and all waveforms are still shown but graduallly fade. Repeat process a few times and the same thing happens again, always on circuit 2. Having by-passed and removed the booster all seems to be well. Time will tell.

[96RAF]

Couple of things Tom

Which method have you used to connect your booster to Elite as the only totally reliable method is track to track (Diagram 4 in the manual). Other methods described can have their little foibles as it were.

 

You may also like to have a look here where Hornby worked with PicoTech initially to decode basic DCC signals...

https://www.picotech.com/library/oscilloscopes/digital-command-control-dcc-protocol-decoding

then I wrote a few words for them about 'scoping the Elite outputs ...

https://www.picotech.com/library/application-note/dcc-demonstrating-the-hornby-elite-using-picoscope

 

Rob

[Ard Lochan]

I did wire the booster as in diagram number 4. Fingers crossed all seems to be ok now. My next concern is that I am about to install a helix so that I can park a lot of my locos underneath, which would be zone 3. The question is this: If I isolate this zone can I use a completely seperate power supply? Regards, Tom

[Chrissaf]

Ard, by the term "Power Supply" if you mean a different "DCC Controller" then the answer is no, not even if you use insulating rail joiners between the different zones. To use a different (additional) power supply, it has to go via a "Booster" type device.

[96RAF]

All 'boosted' zones are totally isolated from each other in any case.

 

 

The DCC signal from the main district (and only controller as Chris says) is purely amplified and passed along to the other districts.

 

There is no direct electrical connection between power districts.

 

Don't ask me how the booster device mixes the DCC signal into its own district power - beyond my pay grade of understanding.

Rob

[Chrissaf]

Don't ask me how the booster device mixes the DCC signal into its own district power.

.

Most likely an opto-isolator chip (Opto.....optical.....light based)