Cobalt IP digital point motors losing address

[ITG]

Hi

building my first layout for 50 years (ouch!) as a retirement present to myself. 

I am using a mixture of Cobalt IP and Gaugemaster PM point motors, controlled via a Prodigy Advance 2. Very happy with progress on what to me is a big project from scratch. Apart from the Cobalt motors occasionally keep forgetting their address, and then need re-addressing which involves scrabbling under the baseboard. The Gaugemasters are fine.

i think it occurs if there’s an overload (probably caused by a couple of pre-owned Hornby locos which don’t seem to like Peco Code 100 points, so the locos will probably be exchanged for Bachmanns which seem to run perfectly).

 

its been suggested to me that the Prodigy causes spikes which impact on the Cobalts, thus inferring a different DCC system would not have this issue. It’s been recommended to fit Power Bus Terminators/ Suppressors at the end of the bus (DCC Concepts make them) but early indication is this hasn’t worked.

 

any advice or experience most welcome.

ian 

[Edward]

Hi as a user of Cobalts but not having had the problem you describe,  could I suggest that you contact DCC Concepts direct?  They are very helpful in my experience.  ( I had a problem with an AEU - Alpha Encoder Unit and they tested it and promptly replaced it).  

[Paul80]

Hi

 

I have had a number of odd problems with mine, fortunately only got 12 of them

 

I have had them refuse to work after self centring was re-enabled to move a motor to a different place on the layout

 

I have one that works if controlled by the DCC bus but not of I try to use buttons

 

I also have to odd one that for no reason will just move.

 

I had one with bits missing which they and the dealer refused to accept was possible

 

I will not be using them on my next layout, I am going over to servo controlled points on that one, £2 a point instead of £20 was a no brainer and wish I had heard of using them first.

 

Paul

[Crosland]

19 hours ago, ITG said:

its been suggested to me that the Prodigy causes spikes which impact on the Cobalts, thus inferring a different DCC system would not have this issue. It’s been recommended to fit Power Bus Terminators/ Suppressors at the end of the bus (DCC Concepts make them) but early indication is this hasn’t worked.

 

Any system can cause spikes during a short circuit due to the inductance of the wiring.

 

Is your bus wiring of sufficient thickness?

Can you get by with a lower track voltage? This may help minimise the peaks.

Is there any part of the layout where the Cobalts are more prone to lose their address?

[WIMorrison]

I found that creating 2 busses with the track bus protected with a NCE EB1 cut out and the accessory bus direct from my controller solved the issues of a couple of Cobalt IPs loosing their address occasionally. They only ever lost their address after a serious short e.g. placing metal ruler or scalpel handle across live track. 

 

I still place stuff across the track (sometime deliberately) but since introducing the EB1 which trips 'instantly' I haven't had the issue. 

 

Some people have also found that issuing a 198 command to the Cobalt IP in 'learn' mode, disconnecting and reconnecting has also cured this issue.

[imt]

On 05/04/2019 at 19:17, WIMorrison said:

I found that creating 2 busses with the track bus protected with a NCE EB1 cut out and the accessory bus direct from my controller solved the issues

 

Mine have behaved oddly from time to time along the lines of Paul above.  I have had some replaced over time.  The thing that "solved" it all is Iain's solution as above.

[rob D2]

I’m also having problems. Gaugemaster prodigy operating the cobalts. They were fine right up until Bachmann 47346 caused an “ SVDA “. Seems to have issues with its wheels as it’s shorting on the frogs.

 

after it did it twice , all the three cobalts have lost their address ! Damn annoying as I have to fiddle around with the set switch again, which involves raising the boards.

 

would the EB1 fix this and then have a separate bus for the feeds as opposed to an accessories ? Bearing in mind there is only one option of feeding power out the prodigy so if it goes” svda “ again , I need to keep the points working .

[Nigelcliffe]

answering "rob D2",  yes a breaker would help if installed on the main track feed, it will stop the "SVDA" message appearing by isolating the short, and should also stop the noisy pulse caused by the short getting onto the accessory bus and thus causing the Cobalt to reset.  
As a cheaper alternative to the EB1, a car indicator bulb (21W traditional bulb from the days of bulbs in cars:-) ) wired in series with one track feed will also achieve similar results: in the event of high current (a short), the indicator bulb will light up and limit the current through the track to around 1.8A.  (The current continues to flow, but is limited by the bulb).

 

 

BUT, note that if you've got one of the Cobalts with a single "frog" output, your short involving the frog will now be connecting the track bus to the accessory bus, so whether the breaker would help then is a massive "well it depends how things trip out".   The solution if you have this type of Cobalt is to use the other set of switch contacts; bringing in the track feeds and one out to the frog. 

 

 

- Nigel

[rob D2]

23 hours ago, Nigelcliffe said:

answering "rob D2",  yes a breaker would help if installed on the main track feed, it will stop the "SVDA" message appearing by isolating the short, and should also stop the noisy pulse caused by the short getting onto the accessory bus and thus causing the Cobalt to reset.  
As a cheaper alternative to the EB1, a car indicator bulb (21W traditional bulb from the days of bulbs in cars:-) ) wired in series with one track feed will also achieve similar results: in the event of high current (a short), the indicator bulb will light up and limit the current through the track to around 1.8A.  (The current continues to flow, but is limited by the bulb).

 

 

BUT, note that if you've got one of the Cobalts with a single "frog" output, your short involving the frog will now be connecting the track bus to the accessory bus, so whether the breaker would help then is a massive "well it depends how things trip out".   The solution if you have this type of Cobalt is to use the other set of switch contacts; bringing in the track feeds and one out to the frog. 

 

 

- Nigel

Thanks for that , I shall look into it . The offending loco 47346 has been withdrawn from service , as it’s the only one that seems to” bridge the gap “ and cause the short 

[reded23]

FYI. DCCConcepts knows about this issue (losing the address) there are some motors that have the wrong components in them and they will fix them for you.