System architecture advice please

[ngaugenic]

Hi, I'm building a new n gauge layout with a focus on shunting operations and therefore many points and slips.

 

I'd like to operate the trains via dcc (I have a zephyr for this purpose) but control the point motors via a mimic/control panel with switches.  Cobalt point motors look almost ideal for this but the complication comes from powering electrofrogs.

 

The ip digital looks like it would do the job, can I combine these with a adfx decoder to simplify the wiring?  I think this would limit the longer wiring to just two wires with led and switch wiring kept at the control board with the adfx.    

 

Can I save money by using the ip analog?  I don't think so as it seems you can't run dcc power in/out of the analog, and therefore the frog connection is not going to be dcc either?  I think that's right.

 

This probably makes little sense, but I'm trying to get my head around the entire scheme from scratch and there are so many variations and overlaps that it becomes very difficult to draw up a plan from scratch!

 

So, what components would you employ to have train control on dcc and point control on manual switching please?

[WIMorrison]

Which ever way you go you need to create a separate accessory bus and a track bus. The track bus needs to be fed through a cut out device that will cut the track power when you have a short from driving to a turnout set against a loco or wagon (inevitable).

 

On the turnout motor front if you use IP Digitals then you do not drive them through anything because they have the decoder built in. I also suggest that you keep your eyes on https://www.dcctrainautomation.co.uk/ and look for MTB turnout motors. They will soon be selling a motor that as the MTB MP4 but with an inbuilt decoder that will be better than the IP Digital 😉

[Nigelcliffe]

I think you've got "wrong end of stick"

 

The switch contacts on the Cobalt motors are just "switches",  they can switch anything.   So, any of the Cobalt's switches can switch your frog current(*)

 

The IP_Digitals have an internal digital decoder,  so you don't need any other devices.   (However, if you power/control them from a separate accessory bus of DCC, then don't use the "frog" terminal" as that connects your track to the accessory bus, which you were trying to separate.  Instead use the regular contacts). 

 

The IP Analogue needs a decoder if you are operating it from DCC signals.   The ADFX is one option.  

 

 

However, before you jump, you say "switches on a mimic panel".   

To make these switches control accessory decoders, you need more electronics, to connect the switches to the LocoNet input to the Zephyr.    There are various boards which can do this, including some low-cost DIY things.   One of the long-established finished boards is the DTM-30 from Signatrak   (its a good board with a lot of powerful features, hard to beat unless you DIY your own).  

 

BUT, for a layout which is fairly compact,  do you need to use DCC to operate the turnouts ?   It does reduce panel to layout wiring, but adds a pile of cost (electronics to encode the switches, decoders to decode to the motor).     Conventional switch wiring would be cheaper (removing two sets of components), though requires more wires travelling from switch panel to layout.  

 

 

Or, staying with the DCC accessory decoders on the turnout motors, one could use a software mimic control panel (eg. running on an iPad or Android tablet), which is a negligble wiring, and may be a low-cost solution.   

 

 

 

There are other motors around, notably MTB which Iain mentions above.   

 

 

 

(* there are issues around how electrofrogs in N gauge are wired, that's a whole separate essay  ). 

 

 

- Nigel

[WIMorrison]

The model numbers for the MTB motors with inbuilt decoders will be DP1 DP4 and DP10.

[Harlequin]

For simple analog manual control of points from a panel look at the existing MTB MP1 point motors.

 

They just need a suitable power supply, a standard SPDT switch for each motor on the control panel and each one will switch the "polarity" of the relevant frog, whether that's DC or DCC.

 

[Ron Ron Ron]

There's a very simple, non-complicated solution to combine the Cobalt iP digital motors with a physical mimic panel.

 

DCC Concepts Cobalt Alpha.

 

You can power and control your point motors either via the main power (track) bus, via a dedicated power (accessory) bus, or on a stand alone DCC bus, not connected too the main DCC system.

 

There are a few videos online, demonstrating how these options work.

 

(p.s. I have no connection with DCC Concepts)

 

 

 

 

 

 

[Nigelcliffe]

1 hour ago, Ron Ron Ron said:

There's a very simple, non-complicated solution to combine the Cobalt iP digital motors with a physical mimic panel.

 

DCC Concepts Cobalt Alpha.

 

 

You missed out "expensive" from the attributes.    A very expensive system compared to using a DTM30.     

 

 

[WIMorrison]

There is also the Lenz LW150 which will operate a mimic panel and integrate fully with any digital control system used.l that has XPressnet

 

https://www.coastaldcc.co.uk/products/lenz/lw150-mimic-panel-module

Edited January 16 by WIMorrison

[Ron Ron Ron]

32 minutes ago, Nigelcliffe said:

 

You missed out "expensive" from the attributes.    A very expensive system compared to using a DTM30.     

 

 

 

24 minutes ago, WIMorrison said:

There is also the Lenz LW150 which will operate a mimic panel and integrate fully with any digital control system used.l that has XPressnet

 

https://www.coastaldcc.co.uk/products/lenz/lw150-mimic-panel-module

Both the DTM-30 and LW150 are good options, but neither offer a full range of additional items (switches, plug & play cables etc,) to allow a convenient and relatively simple installation.

The Cobalt Alpha kit offers all that and is much closer to a plug & play solution.

 

To answer Nigel’s point about price, the DTM-30 isn’t cheap either, at £71 just for the module.

Cobalt Alpha can work out fairly pricey if more components to the system are needed for a large number of points.

However, for a stand-alone arrangement, with a separate DCC accessory system, the £29.95 Power Bus driver and Sniffer adapter is the cheapest “DCC Command Station” you can buy to do the job.

 

So why would you want a stand-alone DCC accessory bus, not connected to the main DCC system?

 

If you never need, or want to set routes or points through your DCC handset or console, this may be an option.

 

It allows continued use of your DCC control of points, when running the layout under DC.

 

.

 

 

Edited January 16 by Ron Ron Ron

[WIMorrison]

There is also the option of a glass mimic panel which can be very effective - JMRI would give you it for free and iTrain Lite will provide the functionality for around £100.

[Ron Ron Ron]

13 minutes ago, WIMorrison said:

There is also the option of a glass mimic panel which can be very effective - JMRI would give you it for free and iTrain Lite will provide the functionality for around £100.

Glass screens being the most flexible solution, but not everyone wants a computer screen, or to have to run a computer, just to run their layout.

(Kudos to the ECoS and Z21 here…)

 

Others prefer the idea of tactile, physical buttons or switches, rather than touch screens, using a mouse or touchpad.

 

Incidentally, Bachmann’s new wireless Kinesis DCC system comes ready bundled with their RailController layout control software, all for a lower price than most, if not all, other wireless DCC systems on the market.

 

 

 

.

Edited January 16 by Ron Ron Ron

[WIMorrison]

27 minutes ago, Ron Ron Ron said:

Kinesis DCC system comes ready bundled with their RailController

 

But have you seen it and played with it?

 

Looks and feels like something out of the 1980s 😒

[AndrueC]

3 hours ago, Nigelcliffe said:

(* there are issues around how electrofrogs in N gauge are wired, that's a whole separate essay  ). 

The only electrical issue I encountered is that Peco don't provide a pigtail so you have to either solder to the wires on the underside of the frog or direct to the frog rails and in both cases be careful of the plastic.

Edited January 16 by AndrueC

[Nigelcliffe]

11 minutes ago, AndrueC said:

3 hours ago, Nigelcliffe said:

(* there are issues around how electrofrogs in N gauge are wired, that's a whole separate essay  ). 

 

The only electrical issue I encountered is that Peco don't provide a pigtail so you have to either solder to the wires on the underside of the frog or direct to the frog rails and in both cases be careful of the plastic.

 

OK,  short version of the Essay.  

 

Out of the box, a Peco N Electrofrog has frog switching provided by the blades touching the stock rails.   This works fine until someone does something to stop it working - dirt, paint, ballast, distortion from how its laid, etc..   

If someone adds a wire to the frog, and a switch (such as using the switch in the Cobalt mentioned above), then there are now two switches altering the polarity of the frog.   

 

If one of the switches changes over and makes contact whilst the other is still connected the "original way", then there is a short circuit.   This causes the track power on a DCC system to turn off.   (Its also a short on a DC layout, but those are usually less drastic, and if the track power in the vacinity of the turnout was turned down to zero when the turnouts move, there is no recorded short circuit.)  

 

With some motor installations, the switches (blades and motor switch) move together;  the internal switches in the motor change whilst the blades are mid-travel (not connected to anything) and no short occurs.   But that's not really a good solution as a slight delay in a switch breaking to "nothing" will still have a short.  

 

There are numerous proposed fixes,  all with various positive and negative consequences.  

 

The Unifrog is different to the Electrofrog.

 

 

-  Nigel  

[AndrueC]

34 minutes ago, Nigelcliffe said:

If one of the switches changes over and makes contact whilst the other is still connected the "original way", then there is a short circuit.   This causes the track power on a DCC system to turn off.   (Its also a short on a DC layout, but those are usually less drastic, and if the track power in the vacinity of the turnout was turned down to zero when the turnouts move, there is no recorded short circuit.)  

 

With some motor installations, the switches (blades and motor switch) move together;  the internal switches in the motor change whilst the blades are mid-travel (not connected to anything) and no short occurs.   But that's not really a good solution as a slight delay in a switch breaking to "nothing" will still have a short.  

Why is that specific to N gauge though?

 

I use DCC Concepts Cobalt-SS and haven't had the problem you describe. These apparently switch polarity in the mid-position so if you have the throw set incorrectly there could be a problem I imagine although surely no-one would leave the controller configured for such an overrun in the first place.

 

IPDigital might be more susceptible because it's possible to have them installed asymetrically (ie; pulling more to one side) whereas the Cobalt-SS are self correcting.

 

Either way all we're saying is that if you install the motor or configure the controller incorrectly you might have problems.

Edited January 16 by AndrueC

[Nigelcliffe]

52 minutes ago, AndrueC said:

Why is that specific to N gauge though?

 

 

Peco's OO turnouts are different.  The OO turnouts provide gaps between the blades and the frog.   To create those in the N-gauge electrofrogs requires cutting rails, and connecting wires to rails.    Can be done well,  but from the pictures I see posted, often ends up as a messy massive gap.  

 

Or there are other ways.    

 

 

 

 

[AndrueC]

57 minutes ago, Nigelcliffe said:

 

Peco's OO turnouts are different.  The OO turnouts provide gaps between the blades and the frog.   To create those in the N-gauge electrofrogs requires cutting rails, and connecting wires to rails.    Can be done well,  but from the pictures I see posted, often ends up as a messy massive gap.  

 

Or there are other ways.    

Oh right. Well as long as the motor and/or controller allow the throw to be reduced to suit the scale and as long as the modeller is reasonably careful when installing there shouldn't be a problem. If either of the switch blades is touching the stock rail when the motor is at its mid-travel point then the motor has not been installed correctly. I don't see how the turnout can be blamed.

 

I do wonder though why the controllers aren't designed to cut the frog power then move the blades then restore (switched) power. I wouldn't have thought it would complicate the design or increase the cost significantly.

Edited January 16 by AndrueC

[ngaugenic]

Thanks for the input everyone, I think the MTB mp1s are what I'm looking for.  The running method I want to employ is driving a train by the zephyr and flicking switches for the points, so DCC alpha is beyond the level of automation I need, anything screen based is out.

 

Does anyone a wiring schematic for the MTBs in combination with a physical switch and led display of which way the point is set please?

[Nigelcliffe]

11 minutes ago, ngaugenic said:

Thanks for the input everyone, I think the MTB mp1s are what I'm looking for.  The running method I want to employ is driving a train by the zephyr and flicking switches for the points, so DCC alpha is beyond the level of automation I need, anything screen based is out.

 

Does anyone a wiring schematic for the MTBs in combination with a physical switch and led display of which way the point is set please?

 

Why do you need LEDs ?   A decent switch (toggle or rotary) shows its position clearly when moved.  Rotary switches(*) with "pointer" style knobs are particularly clear on mimic diagrams.  

 

Wiring for a MP1 requires a power source and a single-pole change-over switch to decide between the two terminals on the MP1 (one for each direction).   
If wanting LED's, you connect them between the switch outputs and the common wire returning from the MP1 to the power supply (ie. parallel to the MP1 motor).  Every LED will require a current limiting resistor.   

 

When wiring the layout, the common wires from many MP1's can be combined.   

 

Maker's wiring diagram:    https://www.mtb-model.com/files/produkty/MP1-setup_CZ_EN_DE.pdf

 

(* make sure you buy "break before make" rotary switches.   The "make before break" type can cause issues in the middle of their change-over. ). 

 

 

 

[Ron Ron Ron]

33 minutes ago, ngaugenic said:

......The running method I want to employ is driving a train by the zephyr and flicking switches for the points, so DCC alpha is beyond the level of automation I need, anything screen based is out.....

 

Cobalt Alpha, as well as the DTM-30,  LW150 and (for NCE users) the NCE MiniPanel, all do what you want.

 

Physical switches or buttons with LED's.....even point levers with Alpha.... No screens and no automation.

 

 

 

.

[Harlequin]

Quick diagram without indicator LEDs:

 

Edited January 17 by Harlequin

[Nigelcliffe]

Adding to Phil's diagram....   if LEDs are wanted then within the rectangle on left which is the control panel,  connect LED+Resistor between the Red supply wire and the relevant black turnout direction output wire.   

 

 

 

 

[Harlequin]

1 hour ago, Nigelcliffe said:

Adding to Phil's diagram....   if LEDs are wanted then within the rectangle on left which is the control panel,  connect LED+Resistor between the Red supply wire and the relevant black turnout direction output wire.   

 

 

 

 

Now with indicator LEDs:

 

 

[ngaugenic]

That's great, thanks

 

And this works with dcc power on red and black?

[ngaugenic]

@Ron Ron Ron  As stated at the top, I'm looking to mix and match dcc train control with point switching under manual control.  which I understand is a half way house but it's also what I would find fun to do!

Edited January 17 by ngaugenic