Jump to content

ITG

DCC - how many sections? (For fault finding)

Recommended Posts

Just about to start (when baseboards arrive) new OO layout, approx 11’ x 7’. It will be double track circuit (including reverse loop on inner circuit with auto-reverse module) with two-way fiddle yard, with long single track branch gradient up to 3 platform mid-size terminus with goods/loco facilities. I’ve been using DCC on previous layout and quite happy with how I’ve approached it, except that I have read plenty of comment about creating separate power districts, which I didn’t do on the earlier layout. I realise the main benefit of so doing is all about isolating track sections when fault finding etc. Or is there any other reason?

 

so the questions I have are around Pros and cons of how many different districts?  All point motors are DCC.

1. Keeping the point motors fed by a separate bus? If so, is that two separate sections, being one for terminus station points, and another for Mainline and FY? Or one as a whole.
2. Each main line circular track separate? Or combine the two?

3. The branch and terminus separate? If so, where should the break best come - immediately outside station, or immediately where branch junctions from main line.

 

I’m conscious that some of the above permutations would result in a significant increase in the length and number of bus wires under the boards. So want to get the best compromise.

 

Finally, on how to do this. Am I right in assuming it’s a single + & - feed from power unit, into a set of switches, with outputs then forming each bus? Any recommendations on type of switches?

 

thanks

Ian

Share this post


Link to post
Share on other sites

I think the point about all this is not switches, but multiple circuit breakers. Fix the fault and the layout runs again. Having a separate bus for the point motors ensures that if you get a short on the track, those motors will still work. So a feed direct from the command station is best.

 

Between the command station and the track you need a few breakers, each protecting a section of the layout. Sidings are a very vulnerable area for shorts - we forget to change a pair of points. So having siding areas on a CB is a good idea - it means trains still run on the main line. You might wish to have separate CBs for up and down main lines - but my feeling is that if you have a stalled or derailed train on the up line, you may not want down trains rushing by while you fix it. Nevertheless, breaking the mains into two or more sections may make sense. Branch and terminus as separate section makes sense too.

 

A preferred brand of breaker is Powershield PSX, which comes in several different multiples or as singles. 

Share this post


Link to post
Share on other sites

you separate the Accessory bus and the track bus at the command station with the track bus 'protected' by one or more circuit breakers. This then allows you to change turnouts when you overrun a closed turnout. Immediately upon changing the turnout the circuit breaker will reset and the train(s) will start again. you can protect the accessory bus with another circuit breaker however this is generally not required. 

 

You can get as complex as you need - but be careful of over complicating this as the headache from wiring it up could be worse than the headache of finding faults - essentially the minimum is 2 busses, one for the track and the other for the accessories e.g. turnout motors. Finally, be careful of using the built in frog switching feed from many DCC point motors using this arrangement because this will be fed from the DCC input to the motor which is incorrect if you haveseperated the busses - you would then be feeding the accessory bus into the track negating the benefits of separating them.

Share this post


Link to post
Share on other sites
Posted (edited)

Dividing the layout into multiple sections and feeding each through a separate circuit breaker means that if there's a short in one section, only that section will stop and any trains in other sections will continue running. Is this desirable for a home layout where you are the only operator? I'm in the same situation being the only operator, and if there's a short anywhere on the layout I want all trains to stop immediately - I can't both deal with the short and keep running the other locos. Obviously it could be different for a large club or exhibition layout where there are multiple operators so keeping other sections running is desirable.

 

On my layout therefore there are several sections but feed through suitable on-off switches. The location of most shorts is usually obvious (derailment, point not set correctly etc) and it's rare to encounter a short whose location is not obvious. I've had a couple of these where it was a failed decoder. In this situation you can disable each section via the switches until you locate which one has the fault. 

 

But do feed the accessory bus for turnouts in a way that allows that to continue to operate when there's a track short. A single PSX circuit breaker on the track bus feed (ie before the switches) will do that, with the accessory bus bypassing the CB. This way you can still change turnouts when there's a short. 

Edited by RFS

Share this post


Link to post
Share on other sites
1 hour ago, Oldddudders said:

I think the point about all this is not switches, but multiple circuit breakers. Fix the fault and the layout runs again. Having a separate bus for the point motors ensures that if you get a short on the track, those motors will still work. So a feed direct from the command station is best.

 

 

You can do it either way - switches or circuit breakers.   They behave differently, but for different situations either works.    Circuit breakers work for temporary faults, like driving a train into an incorrect turnout, or a derailment.  For a single-operator layout, a layout without multiple breakers might be "good" - having everything stop on a fault may be the best outcome, it depends on the layout.   

 

Switches are more useful for infrastructure faults; a wire has come adrift and touching where it shouldn't, a turnout frog changeover not working, an accessory decoder failed. In the event of a fault, switch out sections until the fault is isolated and thus is in a known smaller area, then track down the fault.   Such switches could be "easy to separate connectors" rather than actual toggle switches.

 

( RFS posts much the same, a few seconds before I was about to hit "submit")

 

Nigel

 

Share this post


Link to post
Share on other sites

Thanks all. Very helpful.

 

 Finally, be careful of using the built in frog switching feed from many DCC point motors using this arrangement because this will be fed from the DCC input to the motor which is incorrect if you haveseperated the busses - you would then be feeding the accessory bus into the track negating the benefits of separating them.
 

re the comment above. I have a mixture of DCC Concepts Cobalts and Gaugemaster digital point motors. The former, I understand, have built in frog feed switchers and the latter (on current layout and planning to do similar on future one) I have added Gaugemaster frog juicers. For each of these cases, if I separated the track and accessory buses, how would I then wire things to get the benefit of the separate circuits? Or couldn’t I?

ian

Share this post


Link to post
Share on other sites

In addition to the built in Frog switch the cobalts have a completely separate 'accessory' switch.

You would use that to switch the frog by wiring the track bus to it and the frog to its common contact pin.

 

Regards,

 

John P

Share this post


Link to post
Share on other sites
Posted (edited)

For the Cobalts you need to wire the frogs using their separate built-in switch using feeds from the track bus, rather than the built-on frog terminal, as the latter you will provide power the frog from the accessory bus. A short on the frog could therefore short the accessory bus.

 

EDIT: beaten to it by John P!

Edited by RFS

Share this post


Link to post
Share on other sites

Ah great, thanks RFS and jpendle. 
I didn’t have the Cobalt instructions in front of me when I posted. Presumably they would have indicated the different terminals.

any idea about the Gaugemasters? Although when I root them out, maybe that is on instructions.

Share this post


Link to post
Share on other sites
29 minutes ago, ITG said:

re the comment above. I have a mixture of DCC Concepts Cobalts and Gaugemaster digital point motors. The former, I understand, have built in frog feed switchers and the latter (on current layout and planning to do similar on future one) I have added Gaugemaster frog juicers. For each of these cases, if I separated the track and accessory buses, how would I then wire things to get the benefit of the separate circuits? Or couldn’t I?

ian

The drawing should be self explanatory 

The cobalt on the left is powered by the accessory bus & the frog is powered by the track bus. This is the preferred option

The cobalt on the right & the frog are both powered by the track bus command

 

 

1562019472_Cobalttkaccbus.png.f6b946c723ed5cea557ce8bd10a14edf.png

John 

PS RFS & John P answered while I was drawing, as for the gaugemaster's I don't know

  • Like 2

Share this post


Link to post
Share on other sites

Because I have an extensive "set track" layout in the form of Lego 9V, I have a bit of a combination of approaches:  1.  Feeding a seperate accessory bus is a good idea, running all stationary decoders off it.  2:  Use some 5A or so DPST switchs on the track bus, to divide the track bus into 2-3 large sections (50-200 sq ft) to allow for fault finding.  Because of how modular the lego wiring is, it can be possible to end up with shorts in all kinds of interesting places- not to mention that the 20 year old wires are somewhat prone to having insulation fall off now !

I have no DCC'd turnouts on the Lego, they are all at present hand thrown.

Long Marton is mostly wired without an accessory bus, and runs into the problem of turnout shorts resulting in hand shunting of trains to clear the short.  However, that can mostly be avoided by using the software run approach to avoid running trains into turnouts set against- it doesn't happen very often in "auto", though you do have to remember to change all the turnouts after operating without the computer.  

So, 2 bus's, and then divide the track bus into sections that make sense- if you have a large yard area, a seperate breaker feeding that "branch" is a good idea as articulated above.

 

Share this post


Link to post
Share on other sites

Don't know what DCC system you have, but I ran into problems trying to daisy-chain circuit breakers for sub-districts with my Prodigy PA2. They just conflicted and trashed each other. I now think doing this when the command station has it's own 'smart' CB is questionable. Unless you have lots of problems all the time with faults - which raises the question of why? - then I would follow the advice of RFS & Nigel Cliffe and just section using switches.

 

Izzy

Share this post


Link to post
Share on other sites
29 minutes ago, Izzy said:

Don't know what DCC system you have, but I ran into problems trying to daisy-chain circuit breakers for sub-districts with my Prodigy PA2. They just conflicted and trashed each other.

 

That's exactly why you shouldn't daisy chain them. Feed them all in parallel from a singe bus.

 

If you were able to daisy chain them then it would be like my comment about daisy chaining boosters in another topic. A fault in one section will take out all subsequent sections if they are daisy chained.

  • Agree 1

Share this post


Link to post
Share on other sites
1 hour ago, Crosland said:

 

That's exactly why you shouldn't daisy chain them. Feed them all in parallel from a singe bus.

 

If you were able to daisy chain them then it would be like my comment about daisy chaining boosters in another topic. A fault in one section will take out all subsequent sections if they are daisy chained.

 

I'm sorry, I didn't really make myself very clear, I meant daisy chain in relation to the command station. To have two CB's in a serial connection. The command station CB followed by one or more sub-section CB's.

 

If the base station doesn't have a CB, or, if both the CB's were what could be termed 'dumb', they just react to any short and cut off, bang, then I don't think there would be an issue. But, one of the newer fast 'smart' CB's, designed to keep testing the short to determine if it's long term, and also attempting to discover if it's just high draw from a sound decoder instead, followed by another doing exactly the same, seems to me a recipe for trouble, which I had in spades.

 

Izzy

 

Share this post


Link to post
Share on other sites

Izzy & Crosland,

 

I do indeed have a Prodigy Advance 2, and reading Izzy’s 2 posts alongside Crosland’s has made me think somewhat. The earlier posts in this thread were making me lean more towards CBs than switches. Now I’m not sure.

 

Could anyone enlighten me further, particularly re the use of the PA2 in this situation, and whether that means switches are better? Or maybe someone could post a wiring diagram of what the distinction is (if any) between daisy-chaining and what I assume to be in parallel. 
thank you all.

Share this post


Link to post
Share on other sites

I'll explain what I encountered. Others may be able to offer solutions for you.

 

I bought a PSX-1. To use between the PA2 and the track. My idea was that it would trip before the PA2's breaker and stop it shutting down power. They seems to conflict. At first the PSX-1 didn't seem to work. Then it did and all seemed fine. Then I discovered the PA2 had been damaged, it's CB no longer worked. Then the PSX-1 no longer worked, again either. The PA2 had to go back to Gaugemaster to be fixed.

 

The makers of the PSX-1 could not work out why all this had happened. They thought the unit I had bought was faulty. They sent a replacement to me direct from the  USA. They could not have been more helpful. The same issues arose. I ceased attempting to use the PSX-1 in case the PA2 suffered the same damage again. They tried to work it out, try this, try that, but could not find an answer. I just called it a day eventually, and came to the conclusions I did. Perhaps others have more luck than me.

 

Izzy

 

 

Share this post


Link to post
Share on other sites
22 hours ago, Izzy said:

 

I'm sorry, I didn't really make myself very clear, I meant daisy chain in relation to the command station. To have two CB's in a serial connection. The command station CB followed by one or more sub-section CB's.

 

OK, I thought you meant multiple CBs in a daisy chain, e.g., booster feeding section 1 and CB1. CBi feeds section 2 and CB2, CB2 feeds section 3, and so on...

 

22 hours ago, Izzy said:

I bought a PSX-1. To use between the PA2 and the track. My idea was that it would trip before the PA2's breaker and stop it shutting down power. They seems to conflict. At first the PSX-1 didn't seem to work. Then it did and all seemed fine. Then I discovered the PA2 had been damaged, it's CB no longer worked. Then the PSX-1 no longer worked, again either. The PA2 had to go back to Gaugemaster to be fixed.

 

Unless you can be sure of the relative trip current and reaction times , it's a common issue that they fight each other, but it shouldn't result in failure. The SPROG, for example cuts out very quick and works with a limited range of CBs, e.g. Tam Valley. I think the Dynamis can also switch too quickly for some. I seem to recall issues with MERG's cut out and Bachmann. 

 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.