DCC issues (and solution?)

[martin goodall]

I am looking at the possibility of using DCC on my layout, but I have some unanswered questions about whether and to what extent DCC in itself could cope with any current collection problems that might be encountered, for example due to dirty track. I appreciate that a stay-alive capacitor should be able to cope with any momentary interruption of the power supply to the locomotive when it is in motion, but I am wondering what would happen if the locomotive were to be brought to a controlled stand on piece of track that happened to have a dead spot (e.g. due to dirt, etc.) so that power would not then be supplied to the loco from the track when it attempts to start off again.

Assuming that the stay-alive capacitor was fully charged up when the loco was brought to a stand, so that it has not been partly discharged by having to supply current to the loco before it was brought to a stand, would there be a sufficient charge left in the stay-alive capacitor to get the loco moving on receiving a command to the loco to start after a minute or two? Assuming there is dirt on the track, would the command to start actually reach the loco in these circumstances? If not, then presumably the loco could not be started, even if there was a sufficient charge in the stay-alive capacitor to get it moving. If I am wrong about this, how would the command to start reach the loco’s chip in the absence of power reaching the loco through the rails?

If, as I suspect, the answer is that nothing could be done in this scenario to start the loco, other than to give it a nudge to move it on to a cleaner part of the rail, then I am wondering whether (short of installing radio controlled rechargeable batteries in one’s locomotives) there might possibly be an alternative, whereby a constant 16v power supply would be fed to the track (as it is with DCC), but commands to the onboard loco chip could be transmitted by radio control. This assumes that a stay-alive capacitor could be relied upon to start the loco on a piece of dirty track upon the ‘start’ command reaching the loco’s chip through the radio receiver.

I appreciate that Ted Scannell would justifiably query why one should not wish to fit a battery, but please bear with me on this. If I am wrong in thinking that a stay-alive capacitor would not help in getting a DCC-fitted loco moving if it is standing on a piece of dirty track, then I would prefer not to have to replace all my existing motors with lower voltage motors for use with radio control (and to be perfectly honest, I am not confident as to the safety of rechargeable batteries). What I have in mind is exploring the practicability of a compromise that delivers DCC by radio control, but where power is still supplied through the track (relying on stay-alive capacitors to deal with any temporary interruptions in the power, such as dirty track).

Any thoughts on this?

[barhamd]

I've operated a layout with DCC for several years and fitted stay-alive capacitors to several items of stock. Personally I have found that these can be transformative particularly with 4-wheel locos. Generally I've found that DCC power helps running as the full voltage is on the rails all the time and, particularly with a stay-alive locomotives tend to stop when I want them too rather than when they lose the voltage from the track.

David

[davebradwell]

Martin, I would encourage you to just switch to dcc as it's unlikely that you'd regret it. It's worth it just for the improved control and the convenience is a big plus. I've been using it for a long time and would find cab control very irritating.

If you look at the latest Zimo decoders they inhibit stopping until they find a clean bit of track so there is a much greater chance of a re-start.

Sounds to me you're really just proposing a battery system with a much smaller battery. As you say, one for Ted.

DaveB

[grovenor-2685]

I am looking at the possibility of using DCC on my layout, but I have some unanswered questions about whether and to what extent DCC in itself could cope with any current collection problems that might be encountered, for example due to dirty track. I appreciate that a stay-alive capacitor should be able to cope with any momentary interruption of the power supply to the locomotive when it is in motion, but I am wondering what would happen if the locomotive were to be brought to a controlled stand on piece of track that happened to have a dead spot (e.g. due to dirt, etc.) so that power would not then be supplied to the loco from the track when it attempts to start off again.

As Dave has said some decoders will try not to stop on a dead spot.

Assuming that the stay-alive capacitor was fully charged up when the loco was brought to a stand, so that it has not been partly discharged by having to supply current to the loco before it was brought to a stand, would there be a sufficient charge left in the stay-alive capacitor to get the loco moving on receiving a command to the loco to start after a minute or two?

That's really a length of string question, stay alives come in different sizes and may last anything from a second or less up to 30 seconds or more, also very dependent on the residual load, is it just the decoder or are there also lights and/or sound.

Assuming there is dirt on the track, would the command to start actually reach the loco in these circumstances?

Maybe, some decoders, Lenz?, can pick up the control signal in such circunstances, most won't.

If not, then presumably the loco could not be started, even if there was a sufficient charge in the stay-alive capacitor to get it moving. If I am wrong about this, how would the command to start reach the loco’s chip in the absence of power reaching the loco through the rails?

If, as I suspect, the answer is that nothing could be done in this scenario to start the loco, other than to give it a nudge to move it on to a cleaner part of the rail, then I am wondering whether (short of installing radio controlled rechargeable batteries in one’s locomotives) there might possibly be an alternative, whereby a constant 16v power supply would be fed to the track (as it is with DCC), but commands to the onboard loco chip could be transmitted by radio control. This assumes that a stay-alive capacitor could be relied upon to start the loco on a piece of dirty track upon the ‘start’ command reaching the loco’s chip through the radio receiver.

Hornby have just introduced their HM7000 series decoders that do just this, and the radio control uses an app on your phone so you need only buy one decoder to test it, no need for a DCC system initially. Currently the app is only available for i-phones but Android versions are expected soon.

[davebradwell]

In the end, I suggest the factor that most influences choice of control, and indeed couplings, is what other layout owners in your group are using.

My worry with the Hornby system would be whether the actual motor control is up to the standard available from the best of a wide choice of dcc decoders. Where are you when the world updates to the next version of Bluetooth (isn't that used here)? The dcc standard is controlled by the model railway industry - I suggest this is an important point.

DaveB

[Tim V]

Dirt has less effect on DCC than on DC, don't forget there is around 16v (DCC square wave) on the track, not miniscule DC volts for slow running.

Just convert, you won't regret it.

Plus the quality of control is far superior to any DC control - and you can even get some money back on your Pentrollers. I did!

[DougN]

I have found that converting to DCC increases the level of enjoyment of "driving trains". A local layout of Bath Green park was the one that convinced me. Not form the point of view of a tail chaser, however the entire action of driving a loco backwards and forwards over a complex throat to the station with out every having to change a section switch. Yes to changing points but no other switching. The interesting thing about this was the loco was also fitting in between the other trains entering and exiting the station!

As Tim V says the difference is there is 16v on the track at all times which results in less interruptions of dirty track. Yes everything runs nicely with clean track but I have found I clean my OO layout once in a blue moon rather than every running session. The layout gets run intermittently... Its nice to have a couple of trains circulating while I am building some of my kits in P4.

My recommendation would be to have a hour or 3 running trains on DCC on a larger layout to get your head around the possibilities and also which systems suit your operating style. My system is getting quite old at 15yrs old, Digitrax DCS100 Super Chief. The systems I have used are NCE (powercab and Procab) and my Digitrax Zephyr DCS50. I know the Digitrax systems have been upgraded and the current Zephyr with a hand held unit would be large enough to run my current layout. The NCE power cab is the best starting point with that system again a stand alone unit.

One thing to be aware of is any decoder will work with any system. However the more technical items such as block detection etc may be stuck to throttle network used. Digitrax uses Loconet, I am not sure what system NCE uses.

[Mark Tatlow]

Martin

No learned answer from me as I can't add anything technical to the response above.

What I can say is that the number of sticking locos that I had once I had converted to DCC dropped by at least 80%. At at an exhibition, so long as we clean the track properly before the show opens on each day and the wheels once on the Friday night/Saturday morning, the finger might come out once per day and sometimes not even that.

It does not completely take away the need for an occasional track/wheel cleaning but it does improve running significantly. So as others have said, you won't regret it.

Sound is a more emotive point but we find it increases the viewers experience a lot and our stresses at an exhibition. This is because things happen, from a sound perspective, before movement occurs. It means that there is both anticipation of activity and also the beginnings of activity before any movement actually occurs. This slows things down for us and is altogether more believable as a result.

[nigelcliffe]

I am looking at the possibility of using DCC on my layout, but I have some unanswered questions about whether and to what extent DCC in itself could cope with any current collection problems that might be encountered, for example due to dirty track. I appreciate that a stay-alive capacitor should be able to cope with any momentary interruption of the power supply to the locomotive when it is in motion, but I am wondering what would happen if the locomotive were to be brought to a controlled stand on piece of track that happened to have a dead spot (e.g. due to dirt, etc.) so that power would not then be supplied to the loco from the track when it attempts to start off again.

Assuming that the stay-alive capacitor was fully charged up when the loco was brought to a stand, so that it has not been partly discharged by having to supply current to the loco before it was brought to a stand, would there be a sufficient charge left in the stay-alive capacitor to get the loco moving on receiving a command to the loco to start after a minute or two? Assuming there is dirt on the track, would the command to start actually reach the loco in these circumstances? If not, then presumably the loco could not be started, even if there was a sufficient charge in the stay-alive capacitor to get it moving. If I am wrong about this, how would the command to start reach the loco’s chip in the absence of power reaching the loco through the rails?
..........

There are already two somewhat hidden solutions to the problem you outline mentioned by others above.

1) Lenz. Their signal coupling has a capacitive effect, rather than contact current flow. So commands will flow over a thin insulator (such as a piece of paper). There's a comedy demo available in a few places - drive a loco with a Lenz decoder and massive stay-alive to stop on paper, then reverse off the paper. The commands got through.

2) Zimo. (I prefer this method). If the loco is brought to a stop, and the decoder detects "no track power", the decoder will use the stay-alive power to nudge the loco along until track power is restored. For this to work, a substantial stay-alive is needed (gold caps, say 0.05F or much larger greater) as it needs a lot of power to get the motor moving.



However, as others have indicated, the addition of relatively small stay-alives to smaller locos with limited pickup is transformative. The biggest change is in the operator; previously it was constant monitoring of "will this loco stall on that cross-over (finger poised to prod it)", now it is "run the loco, it will run (and near zero concern about stalling)".
There are a number of small tank engines used for shunting duties on Burntisland-1883. That's been DCC since the outset (20 years?). The shunting tanks were always problematic, stalling when crossing complex pointwork, but their use was essential to the layout. About 4 years ago (just prior covid) I started to fit stay-alives to these locos. The biggest gain is the attitude of operators who are either DCC-skeptical or have no interest in the technology - they're the ones who are saying how reliable these locos now are. The lack of stalling also means they run more smoothly, as there are no longer the microscopic temporary stalls when pickup is lost, the mechanism relaxes slightly and pickup restored.
A few Tantalum capacitors was all I could fit, perhaps 1000uF, but varies with each loco. Most done with a home-brew charging circuit consisting of: charging resistor, discharge path diode, and a zener diode to limit the maximum voltage over the capacitors (prevents blowing up capacitors from over-voltage).




Battery and radio control is a whole different area. Has its possibilities, but also means changing the control approach for those locos. The battery is usually the limiting factor on space in the loco.


- Nigel

[martin goodall]

Many thanks for these very helpful responses. They do answer most of my concerns, and they have convinced me that most of my reservations about DCC were groundless, and that trying to start on a dead spot (due to dirty track) need not be such a problem as I had feared.

I have done some reading around the subject, and was relieved to learn that the way I have wired my layout, and its track and points, makes it completely compatible with DCC without the need for any re-wiring. Admittedly my main power bus could not handle much more than about 1 amp (and certainly not the 5A that is often the quoted current demand which should be catered for with DCC), but I expect only to have one fairly small lightly loaded loco moving about the layout at a comparatively low speed, and perhaps occasionally a second, with no lights, sound or other bells and whistles. Nor shall I be powering points, signals or any other ‘accessories’ from the DCC circuit. I am firmly wedded to mechanical operation of points and signals. The electro-magnets I use for uncoupling (which can draw up to 3A in some cases) are fed via an entirely separate circuit from a different transformer.

So far as controllers (‘cabs’) are concerned, I like the look of the NCE PowerPro with its control knob for fine speed control when shunting. Paul Ash uses one of these for shunting on his ‘Hembourne’ layout (which I saw and admired at Expo-EM last weekend). The MERG DCC cab kit has a similar configuration (and I see they describe it as “Probably the best DCC cab kit in the world!”), but being a kit, presumably the purchaser has to solder the various parts and components in order to assemble it. So the NCE product would probably be a better bet for a beginner, although I don’t know who supplies these.

I would also be happy to follow the example of Bernie Baker in S4N 223 in using Zimo decoders (the MX600 and MX617 series), but I believe there have been supply problems with these, and I have no idea where to obtain them. I am not confident about ‘hard wiring’ the decoders into locos, and so I suppose I would need 8-pin sockets into which the decoders can be plugged. (None of my locos is ‘DCC ready’.) I am also uncertain about the specification and availability of stay-alive capacitors.

I am completely at sea when it comes to other parts of the system. I have seen references to a “Command Station”. Assuming this is not built in to the commercially available cab controllers, presumably a command station would also be needed and would have to be accommodated somewhere under the layout or the control panel (or am I wrong about this?) I have no idea as to the make and type that might be required. I have similarly seen references to “Boosters”, but really don’t know whether these are necessary, or whether just one booster would suffice for my fairly simple branch terminus. Again, if I do need one, what is the required specification and make?

Most important, as I have indicated above, I have no idea where to get this stuff. Adverts in the model press are surprisingly uncommunicative about suppliers and supplies (not to mention prices).

So any further help that members can offer to a complete DCC ignoramus such as myself would be much appreciated.

[jim s-w]

I am not confident about ‘hard wiring’ the decoders into locos, and so I suppose I would need 8-pin sockets into which the decoders can be plugged. (None of my locos is ‘DCC ready’.) I am also uncertain about the specification and availability of stay-alive capacitors.

If youre not having any lights or sound then it's just 4 wires.

Simply remember, red and black to the track, orange and grey the other way

Hth

Jim

[nigelcliffe]

Answering a few questions mentioned above.


DCC decoders. There is a shortage, its related to the global shortage of semiconductors. DCC manufacturers are at the bottom of queue for components. It may ease this year, slowly.
Zimo are in the process of replacing their decoder range. The MX600 and MX617 are no more. The replacement MN-series is slowly trickling through, but it seems that EU stockists are getting them sooner than UK. I doubt the very cheap prices of MX600/617 will be seen again (they were unique to the UK, the UK importer negotiated a very good price, they were much more expensive in EU shops), expect to be paying £30-£35.


"Command station". The "central processor" in a DCC system. There is only ever one on a layout.
"Throttle". The speed control and other controls, usually a handset one holds, but could be built into a desk-system. Can have one or many on a layout.
"Booster". That's an amplifier, to provide current to the layout. Require one (usually built into a box with command station), can use many on large layouts.

In the PowerCab the Command Station, and sufficient booster power for track, is inside the PowerCab handset, so it's a combined "Command Station, small Booster, and Throttle".
In the MERG system, the Command Station and small booster is one box. The Throttle is another box (handset).

Most systems will allow multiple throttles to be connected (certainly all mentioned so far do), should you need more than one. Such as multiple operators, or just convenience.


The 5A wiring often recommended is to do with two things.
Firstly, a lot of commercial middle/large systems put out about 5A max to the track. But the PowerCab is nearer 2.5A max, the MERG system (without extra booster boxes) is under 2A. For one loco moving at a time 1A is enough(*).
Second is voltage drop, related to short circuit tripping correctly. Poor wiring will lead to excessive voltage drop over the layout. For small layouts this is rarely an issue, unless the wiring was extremely bad.


(* though Burntisland-1883 has a 5A system in the middle of a layout which can have six or seven locos moving simultaneously, the track cutouts (extra bits of electronics which split the layout into sections so a derailment/short in one section doesn't short out everyone else) are all set to 1.5A. Nowhere on the layout will draw more than 1.5A before it shuts down. We regularly run multiple locos on the sections, so that's "in practise", 1.5A is plenty for a couple of locos.
There are a few RTR exceptions, some Heljan diesels have motors which seem to draw excessive amounts of current, and 1A won't be enough for those.
For kit-built things, it depends on the motor fitted; if it was built 40 years ago, with some old open-framed motor it may be drawing a lot of current, the decoder and system needs to be chosen to suit. But anything with a Mashima or a coreless motor, or a Portescap, will be drawing a fraction of an amp and won't be a problem. ).


Sockets in locos, vs "hard wiring". It isn't difficult to wire a decoder into a loco. There are four wires needed for a typical loco (no lights, no sound), they are: two to the pickups, and two to the motor. A decoder will come with reasonable lengths of wire pre-attached, so its just four wires to connect.
If fitting decoder sockets, then I'd future-proof the loco, and not bother with the antique 8-pin. Its big, and doesn't carry many wires. If really stuck for space, use a 6-pin. Both of the above still leave a problem with "stay alive" units - you've got to add wires to solder pads on the decoder in many cases. The better manufacturers detail where to add these wires.

So, my recommendation: if there's room, fit a Next18 socket, and use Next18 decoders which clip onto these. As Next18 is a standard, you can look up in advance how much space to leave for the decoder and be totally sure about space in the model (excluding Hornby's new "bluetooth" decoders which ignored the standard for dimensions, but that's Hornby for you).
As well as the usual motor and lights, it has pins for sound speakers and pins for stay-alive devices. PCBs with a socket for these are available, with solder pads on them, just connect the solder pads to the four places on the loco, and if required, add the stay alive module to the relevant pads. If you mess up the wiring, its a few pounds for a new PCB (but its simple stuff). Zimo have in the pipeline (later this summer?) a Next18 PCB for self-install which includes all the circuitry for a stay-alive, so even simpler.


I'd recommend thinking before jumping on the PowerCab. Its good, but its a very dated design. It has become expensive (exchange rates, semiconductor prices, and has anyone got stock). The biggest downside is how it handles functions. It requires that all functions, except F2, are "latching", this means: one push for "on" and another push for "off". With sound equipped locos this gets really tedious; press button for whistle to play, wait for it to play out, then press button again to "turn it off". In contrast, a throttle which allows "non latching" or "momentary" buttons, it would be "press and hold button, whistle plays, release button, whistle stops". If operating locos without sounds and lights, its got a lot of buttons on the handset which will never be used, but just get in the way. Several additional features, notably RailCom, are not possible on an NCE system (I could explain why "not possible" rather than "can be retrofitted"). RailCom can identify locos automatically, as well as other things, its surprisingly useful, though not widely used.
Most European designs of throttle allow the end user to determine which function keys are "latching", and usually on a per-loco basis. Most US throttle designs (which includes Gaugemaster, its a badge on a US product), have no user control of latching behaviour - everything latches, except F2. The exception from the US is the new TCS throttle, and the spec of the new TCS entry system makes the PowerCab look even more antique, but I don't think its reached any UK shops yet.


DCC retailers, there are a few I can think of who I'd class as competent, including:

Coastal DCC
DCC Train Automation
Digitrains
DCC Supplies
Sunningwell Command and Control (tends to specialise in US outline stuff, both systems and layouts)
YouChoos (mostly a sound supplier, but, in normal times, would have a good stock of Zimo non-sound decoders).

[davebradwell]

After all that tech stuff I'll throw-in that as Coastal DCC have supported our exhibitions then I've supported them and Kevin has been very helpful over the years despite some rather trying requests.

If you go for sound then you can blow the whistle before starting to check all is making contact, including the cassette. As for current, my little NCE system can display this and with the layout "leaking" 0.16 amps, this rises to only 0.35 amps with a Q6 spinning its wheels (16mm Mashima) and to 0.5 amps with a Pacific doing the same despite 1833 Canon motor. You really don't need much power if you're not driving accessories off the bus. The big systems just give possibility for more controllers and perhaps wireless operation.

I've come across a cheapie handset with a knob that doesn't give very good control - someone will know what this is.

DaveB

[Tim V]

Control is in the decoder, not the handset.

It is this finesse that sets DCC apart from DC control, you can refine each engine to the N'th degree inside the decoder. Impossible with DC.

[davebradwell]

I know that. This handset had a limited number of facilities but the speed control was coarse so perhaps using an encoder with less bits and lacking the least significant bits from the speed steps - i'm trying to make informed guesses here. Seems I should have been more careful with my words in the first place.

DaveB

[nigelcliffe]

Control is in the decoder, not the handset.

It is this finesse that sets DCC apart from DC control, you can refine each engine to the N'th degree inside the decoder. Impossible with DC.

Yes I agree that's how DCC works, but its a yes-but....

The design of handset has a huge impact on how well control is perceived by an individual. It is massively subjective, and depends how the decoder is programmed. If you hand a range of different controllers to someone and ask them to drive a loco, they'll find some which work well for them, and some which are not so good. Same loco, same layout, same command station. Change the loco's programming and a different handset may work better and the original less well. The user's experience of the knobs/buttons and the reaction the user perceives to be happening at the loco are linked.

Chuck in a faulty handset (where the clicky-wheel isn't running as smoothly as it should) and the user interface is now erratic, particularly as the user is likely to be controlling by feel (rather than watching numbers change on a screen). Or a handset where the speed granularity has been changed without the user expecting it, so one-click has a different effect to that previously, and the user gets confused by it all.


I'm currently designing a DCC handset (wireless hand-held DIY for well under £50), the basics are all working. It will be what I want.


- Nigel

[Geoff]

Re DCC, I've used it on a couple of exhibition layouts for over twenty years and never regretted changing. If your layout is conventionally wired, just leave all the section switches on.
Re systems, I tried a couple and eventually plumped for Lenz, as like a Ford car, there are dealers everywhere. However, I also use the ROCO Multimaus handset with my Lenz command station as the Multimaus has a facility to enter the name/class of the loco on the display so that your scroll to a loco by name rather than having to remember an address. It is also ergonomically designed and comfortable to use, and your can operate points and signals from it, too.
Finally, the greatest advantage to me is to be able to set a speed curve and tame any fast-running motors. For this, I use a SPROG interface via a small, refurbished laptop; I know nothing about CVs etc, the SPROG does it all.
Hope that helps.
P.S. Some decoders feature a 'shunt speed' facility, very useful on a small layout.