NCE Mini Panel non-momentary inputs

[St Enodoc]

I use an NCE Power Pro system both for track and accessory operation. Specifically, in my storage loops I use DCC Concepts ADS-8fx decoders to operate the points, with route setting through NCE Mini Panels activated by momentary push buttons. On the other hand, in the station areas I will be using Modratec lever frames and microswitches to operate the points. I would like to use ADS-8fx decoders again, as I find the frog switching capability very convenient, but they are only configured for momentary inputs, not a continuous input from a microswitch.

 

In a separate topic, Ray H (Virney Junction) described how he uses a Mini Panel with non-momentary inputs by using two Mini Panels inputs per lever - one for normal and one for reverse. However, Ray alluded to a caution from NCE that the resulting "continuous polling" might slow down the performance of the NCE system.

 

Has anyone else had any experience of using a Mini Panel in this way, and if so was there any degradation in performance? Alternatively, is there another way to operate ADS-8fx decoders with non-momentary inputs (other than the separate "fire" button described in DCC Concepts' literature)?

[Ray H]

Although NCE warned of a slow down it wasn't something that I found in practice when using the Mini-Panel conventionally despite having two mini-panels connected to the system. There was a delay however when I tried to set the Mini-Panel up for continuous memory with some sneaky Skip commands - e.g. Skip next command if Input XX Normal. From memory that seemed to make the Mini-Panel poll continuously with only the occasional break-out to test other inputs.

 

In the end I found that I'd need far more than the 30 inputs that the mini-panel provides - and I didn't have space for a further mini-panel - so I removed them and changed to CBUS instead.

[Ray H]

I may be missing something but . . . . .

 

The ADS-8fx is a decoder with a nominated decoder address. Reading the User Guide it appears to take power from the DCC track or accessory bus whereby, I presume, it also receives its DCC commands. So the Mini-Panel will issue that instruction without the need for an intervening switch.

 

Or . . .

 

Are you looking for ways to activate the Mini-Panel inputs using conventional On-Off switches?

[St Enodoc]

Although NCE warned of a slow down it wasn't something that I found in practice when using the Mini-Panel conventionally despite having two mini-panels connected to the system. There was a delay however when I tried to set the Mini-Panel up for continuous memory with some sneaky Skip commands - e.g. Skip next command if Input XX Normal. From memory that seemed to make the Mini-Panel poll continuously with only the occasional break-out to test other inputs.

 

In the end I found that I'd need far more than the 30 inputs that the mini-panel provides - and I didn't have space for a further mini-panel - so I removed them and changed to CBUS instead.

Thanks Ray, that's very helpful. With my planned station layouts I would need two Mini Panels, both using all 30 inputs (15 point levers x 2 inputs, one each for Normal and Reverse, on each panel). Each input would only need one or two steps to drive the respective point motors and I wouldn't need any Skip, Wait, Link or other advanced commands, so it sounds as though it might work. This is in addition to the two existing Mini Panels with push-button activation, so a total of four Mini Panels.

 

By the way, I've posed the question direct to NCE via their website and also to two local gurus, Marcus Ammann (DCC for Novices) and Gary Spencer-Salt (owner of The Model Railroad Craftsman, the Australian agents for NCE). I will also pose the question on the NCE Yahoo group, which I forgot to do earlier. I'll post any feedback in this topic for the benefit of others.

[St Enodoc]

I may be missing something but . . . . .

 

The ADS-8fx is a decoder with a nominated decoder address. Reading the User Guide it appears to take power from the DCC track or accessory bus whereby, I presume, it also receives its DCC commands. So the Mini-Panel will issue that instruction without the need for an intervening switch.

 

Or . . .

 

Are you looking for ways to activate the Mini-Panel inputs using conventional On-Off switches?

Ray, this came in just as I sent my last reply. You are correct about the ADS-8fx. It takes both power and commands from the DCC track or accessory bus in the same way as other accessory decoders. It does, however, require a momentary not a continuous input.

 

I'm looking to operate the Mini Panels with SPDT microswitches, using the Mini Panel to convert that continuous signal to a momentary one for the ADS-8fx.

 

Hope that's clearer.

[tender]

As Ray H has suggested, having closed contacts on a mini panel input will result in bus commands being sent every time the input is polled. If you have several inputs closed at the same time there is a danger of 'flooding" the cab bus with information.

 

Have you seen the DCC concepts Cobalt Alpha control panel (DCD-AEU). This has 12, 3 wire inputs and although the manual says preferably 'momentary' it doesn't say change over type switches cannot be used. Might be worth giving Richard a call. The unit connects directly to the NCE cab bus giving control of 12 accessory decoders. It is a bit more expensive than a mini panel (£52). I have one on order and will probably use the Alpha S lever which has momentary contacts but will give it a try with micro switches when it arrives and report back.

[Ray H]

One other possibility is a small electronic circuit that will convert the operation of an On-Off switch to a pulse, a little like a CDU. This could reduce the number of inputs that are required to be used on the Min-Panel.

 

It wouldn't need to handle any great power levels, just provide a path for a negligible momentary current flow to ground the Mini-Panel input. Unfortunately, whilst I'm sure it is possible, my less than basic knowledge of electronics can't help with a circuit but I'd be surprised if several couldn't be built on a single piece of strip-board with minimal components.

[St Enodoc]

As Ray H has suggested, having closed contacts on a mini panel input will result in bus commands being sent every time the input is polled. If you have several inputs closed at the same time there is a danger of 'flooding" the cab bus with information.

 

Have you seen the DCC concepts Cobalt Alpha control panel (DCD-AEU). This has 12, 3 wire inputs and although the manual says preferably 'momentary' it doesn't say change over type switches cannot be used. Might be worth giving Richard a call. The unit connects directly to the NCE cab bus giving control of 12 accessory decoders. It is a bit more expensive than a mini panel (£52). I have one on order and will probably use the Alpha S lever which has momentary contacts but will give it a try with micro switches when it arrives and report back.

Thanks tender. I'm aware of the Cobalt Alpha range but hadn't looked at it in detail up to now. I'd be very interested to hear your experience with the DCD-AEU in due course, although at one-and-a-half times the price of a Mini Panel for in effect 24 outputs rather than 30 it isn't a very economical solution.

[St Enodoc]

One other possibility is a small electronic circuit that will convert the operation of an On-Off switch to a pulse, a little like a CDU. This could reduce the number of inputs that are required to be used on the Min-Panel.

 

It wouldn't need to handle any great power levels, just provide a path for a negligible momentary current flow to ground the Mini-Panel input. Unfortunately, whilst I'm sure it is possible, my less than basic knowledge of electronics can't help with a circuit but I'd be surprised if several couldn't be built on a single piece of strip-board with minimal components.

Ray, my knowledge of electronics is on a par with yours, at the black box level I fear, so I don't know how to do that either!

[Oakydoke]

Would a change from Modratec levers to DCC Concepts Cobalt S levers, solve this problem?

They have momentary switching, as well as on/off switching built-in and can be formed into a lever frame assembly.

[St Enodoc]

Would a change from Modratec levers to DCC Concepts Cobalt S levers, solve this problem?

They have momentary switching, as well as on/off switching built-in and can be formed into a lever frame assembly.

Yes it would Oakydoke (as would Peco passing-contact switches come to that), except that the Modratec levers form part of a fully-interlocked mechanical lever frame. That is a must-have.

[tender]

Thanks tender. I'm aware of the Cobalt Alpha range but hadn't looked at it in detail up to now. I'd be very interested to hear your experience with the DCD-AEU in due course, although at one-and-a-half times the price of a Mini Panel for in effect 24 outputs rather than 30 it isn't a very economical solution.

Yes, it certainly isn't cheap, and I didn't realise how many mini panels you were going to be using, so probably not a practical solution in terms of cost alone.

[Nigelcliffe]

The electronics for this problem (constant switch input to a single pulse output) isn't very complicated.   The most efficient implementation depends on the number of contacts required.

 

Two fairly simple approaches, off the top of my head:

a - a simple timer device (the 555 timer chip) with a few components around it, and that produces a single pulse until it is reset.    But, if needing a dozen or more of these circuits for a lever frame, the time to assemble could get tedious. 

 

b - a cheap programmable device designed for many inputs and outputs.  An Arduino microprocessor for example, costs a few pounds to purchase.  Requires a minimal amount of coding to connect a change in an input pin with a short duration pulse on an output pin.  The advantage of this approach is that a single Arduino could handle a fairly large lever frame, the "uno" has 14 in/out pins, the "mega" has 54.  For the "simple" implementation, its one input pin and one output pin per lever (so a Uno can handle 7 levers, a Mega 27 levers).  Slightly more complex multiplexed arrangements are possible which reduces the number of pins required per lever on the lever frame.  

[tender]

I'm not familiar with the Modratec lever frame but could you add a wiping (ground) contact to the end of the lever that passes over a contact pad that terminates just before each end of the throw? This would give the required momentary contact for the mini panel.

[Nigelcliffe]

I'm not familiar with the Modratec lever frame but could you add a wiping (ground) contact to the end of the lever that passes over a contact pad that terminates just before each end of the throw? This would give the required momentary contact for the mini panel.

 

Its possibly more complex than that though - the contact would also wipe when pushing the lever in the opposite direction.  So, for any lever throw, two contacts are required, and a means to ignore one of them (because its in the "wrong" direction), or accept that a "wrong" instruction is issued first (which sets an already set route), followed very shortly afterwards by the "right" instruction.  

 

A passing-contact switch deals with this by its internal mechanical arrangements.  

 

 

- Nigel

[tender]

Its possibly more complex than that though - the contact would also wipe when pushing the lever in the opposite direction.  So, for any lever throw, two contacts are required, and a means to ignore one of them (because its in the "wrong" direction), or accept that a "wrong" instruction is issued first (which sets an already set route), followed very shortly afterwards by the "right" instruction.  

 

A passing-contact switch deals with this by its internal mechanical arrangements.  

 

 

- Nigel

In my experience an accessory decoder will ignore a command if already set in that direction so having the lever re-trigger on the way back shouldn't be a problem.

[Nigelcliffe]

In my experience an accessory decoder will ignore a command if already set in that direction so having the lever re-trigger on the way back shouldn't be a problem.

 

In which case, your solution could work (I suggest a bit of testing on the decoders in question before building lots of identical parts !).  

A related solution would be two microswitches and an extension to the leverframe lever which trips the microswitches towards the end of the lever throw, releasing the switch at then end of the throw. 

[St Enodoc]

In which case, your solution could work (I suggest a bit of testing on the decoders in question before building lots of identical parts !).  

A related solution would be two microswitches and an extension to the leverframe lever which trips the microswitches towards the end of the lever throw, releasing the switch at then end of the throw. 

 

In my experience an accessory decoder will ignore a command if already set in that direction so having the lever re-trigger on the way back shouldn't be a problem.

 

In which case, your solution could work (I suggest a bit of testing on the decoders in question before building lots of identical parts !).  

A related solution would be two microswitches and an extension to the leverframe lever which trips the microswitches towards the end of the lever throw, releasing the switch at then end of the throw. 

Nigel/tender, I might try the "double wipe" idea but I suspect that the problem will be that the decoder's on-board CDU will discharge at the first contact and not have time to recharge for the second contact.

 

Two microswitches might work if there is room to fit them!

 

I've had some feedback through the NCE-DCC Yahoo group, which looks promising - I'll probably post something on that later.

[Ron Ron Ron]

From what I can see, Modratec have added new electrical switch options for their lever frame assembly, resulting in several options.

However a momentary contact option is still not available.

 

Considering the increase in the number of DCC devices that would benefit from such an option (NCE mini panel, DCC Concepts Cobalt Alpha range, Lenz LW150 etc, etc,), it seems to me it would make good sense (and business) for Modratec to add another switch option to their product.

Has anyone thought to contact them, or use their forum, to suggest or push the idea?

 

If intending to use DCC to electrically switch points and operate signals.....and if you don't need the mechanical interlocking provided by the Modratec lever frame, DCC Concepts Cobalt S levers offer a very good alternative IMHO.

 

 

.

[tender]

I've had some feedback through the NCE-DCC Yahoo group, which looks promising - I'll probably post something on that later.

Just read Mark's reply on the the NCE forum. The way I read that is the mini panel will only respond to a change of state on an input so having a changeover switch wired to two inputs as was your original question should not present a problem. I.e. The command steps for an input will only be excited when there is a change of state on the input (high to low) not if the input remains at a steady (low) state.

Must file that little snippet of information away for future reference.

[St Enodoc]

From what I can see, Modratec have added new electrical switch options for their lever frame assembly, resulting in several options.

However a momentary contact option is still not available.

 

Considering the increase in the number of DCC devices that would benefit from such an option (NCE mini panel, DCC Concepts Cobalt Alpha range, Lenz LW150 etc, etc,), it seems to me it would make good sense (and business) for Modratec to add another switch option to their product.

Has anyone thought to contact them, or use their forum, to suggest or push the idea?

 

If intending to use DCC to electrically switch points and operate signals.....and if you don't need the mechanical interlocking provided by the Modratec lever frame, DCC Concepts Cobalt S levers offer a very good alternative IMHO.

 

 

.

Thanks Ron. The interlocked lever frame is very important as far as I am concerned and so the Cobalt levers - nice as they are - are not what I need.

[St Enodoc]

Just read Mark's reply on the the NCE forum. The way I read that is the mini panel will only respond to a change of state on an input so having a changeover switch wired to two inputs as was your original question should not present a problem. I.e. The command steps for an input will only be excited when there is a change of state on the input (high to low) not if the input remains at a steady (low) state.

Must file that little snippet of information away for future reference.

Thanks tender, that's what I understood too. I've had some great feedback here and on the NCE-DCC Yahoo group both by way of explanations and suggestions. My conclusion is that what I am proposing will work, but being a belt-and-braces sort of chap I will also add the little R-C blocker/limiter described there as well.

 

This consists of a 47 kilo-ohm resistor and a 10 microfarad capacitor in parallel between each Mini Panel input and its respective switch connection. Probably not absolutely necessary but good practice nevertheless.

[St Enodoc]

Thanks tender, that's what I understood too. I've had some great feedback here and on the NCE-DCC Yahoo group both by way of explanations and suggestions. My conclusion is that what I am proposing will work, but being a belt-and-braces sort of chap I will also add the little R-C blocker/limiter described there as well.

 

This consists of a 47 kilo-ohm resistor and a 10 microfarad capacitor in parallel between each Mini Panel input and its respective switch connection. Probably not absolutely necessary but good practice nevertheless.

I've been advised that the R-C circuit is known technically as an "RC differentiator", which produces output only when the input is changing.

[tender]

I've been advised that the R-C circuit is known technically as an "RC differentiator", which produces output only when the input is changing.

With the ground contact open the capacitor will discharge through the parallel 47k resistor. When you close the ground contact the mimipanel input briefly will see the voltage across the discharged capacitor I.e. 0v. The capacitor will then charge up slowly through the internal pull up resistor in the mini panel input returning the input voltage to a high level even though the contact is still closed. When the contact is opened again the capacitor will again discharge through the 47k resistor ready for the next cycle. The down side is if the contact is retriggered before the capacitor has discharged sufficiently the mini panel input might miss the trigger. However with the values given that shouldn't be a problem.

[St Enodoc]

With the ground contact open the capacitor will discharge through the parallel 47k resistor. When you close the ground contact the mimipanel input briefly will see the voltage across the discharged capacitor I.e. 0v. The capacitor will then charge up slowly through the internal pull up resistor in the mini panel input returning the input voltage to a high level even though the contact is still closed. When the contact is opened again the capacitor will again discharge through the 47k resistor ready for the next cycle. The down side is if the contact is retriggered before the capacitor has discharged sufficiently the mini panel input might miss the trigger. However with the values given that shouldn't be a problem.

Thanks again tender, that's very helpful. I can understand what's happening very clearly now. I don't think that retriggering would be a problem as that would involve the unlikely event of waggling the point lever to and fro. Presumably the currents involved are tiny so there would be no particular power requirement for the resistor and a small ceramic capacitor would do?