Consider the Problem, then select the Solution
Dear Ken,
As the first (and possibly only) witness for the defense:
(Full Disclosure: I am not anti or pro _either_ of the 2 control system options here in court today,
but I am very PRO the concept of "selecting the best solution for a given problem").
1 - Let's first ask the question that _really_ matters
How many _simultaneous_independent_human_operators_ is the layout intended to handle?
(Put another way, how many people will have a throttle in their hand, directly driving a unique given loco?)
If the truthful "in practise" answer is> 1
("in practise" covers layouts that are _designed_ to be operated by more than 1 person, but rarely if ever actually _are_ operated as such)
then DCC has a potential benefit right out of the blocks.
However, as the case before us is stated as being about a rank Newcomer, a layout the size of which needs/requires/could host a "team of operators" is unlikely to be involved.
therefore rendering DCC back to "equal footing" in _this_particular_case_
2 - Cost
Again, keeping clearly in mind that we are considering a rank Newcomer,
- 1x loco
- 1x analog throttle
= "put the loco on the track and it can be controlled" solution
will very likely be cheaper when directly compared to
- 1x loco
- 1x DCC Control system (assuming choosing a "Starter" system such as the Zephyr or Power-Cab)
- 1x decoder if required (If the loco comes with a decoder installed, the price of the loco will likely reflect that. compared to it's "non-decoder equipped" cousin, so you pays either way...)
= "put the loco on the track and it can be controlled" solution
Additional "1x loco + 1x throttle (+ 1x decoder if required)" locos multiply the cost.
(a basic NCE "Cab04" at approx US$70 street cost is still more than the equivalent basic analog throttle).
SO, we conclude that as far as
- the equipment required to be mounted in the loco
(IE decoder, whether "factory equipped" or "user installed")
- and the "Control system"
DCC is likely a bigger "up-front expenditure"
3 - Wiring "complexity"
Again, _context_ is required.
Let us first consider again that this is a rank Newcomer we are advising.
They have no background or bias in the "bad old days" of multiple-block switching, poor-user-interface "central control panels" for layouts,and other such holdovers from circa 1950 - 80's model RRing.
For those of us who put even a modicum of thought into such things, elimination of "single central control panels" and their attendent tying-down of the model RR operator to one fixed position was and is _easy_ to achieve, _without_ loosing any of the ability to "control train A with throttle A over _there_, while switching train B with throttle B over _here_". Do not confuse "additional non-proto human-interface block electrical switches and the elimination thereof" as being a feature _only_ possible with DCC...
("Order in the court, calm yourself sir"
"yes, m'lud, sorry m'lud")
3a - for a given "length of track" or "simple circuit"
(which is often where a rank Newcomer is encouraged to start)
both analog DC and DCC provide the oft-touted "2 wires to the track" solution.
Oops, sorry, what was that? DCC can't rely on rail joiners, and needs a feeder to each length of rail?
That sounds awfully like _more_ wiring to be installed.
3b - Add a passing siding/loop to that "circle/oval", and _both_ DCC and DC will likely require some gaps to be cut.
(The selection of plastic or metal-frog turnouts is _not_ what we are debating here today, but directly affects how each control system is deployed. In either case, DCC will likely still require gaps, whereas plastic frog turnouts will allow analog DC to get away without them).
3c - NB that should a metal-frog turnout we wired _correctly_ with a suitable frog-switching micro-switch (think about manual or ground-throw-operated turnouts) or the AUX switch on a point-motor, then _both_ analog DC _and_ DCC should not face "frog polarity' issues. As far as our discussion today is concerned, this represents a _common_ piece of work required for either system, and thus does not represent a benefit or loss for either option in question.
(NB that both of the above switching solutions do _NOT_ require any "manual controls" for the operator to interface with, the movement of the turnout will perform the required traction-switching without human intervention).
IF however the rank Newcomer starts with a DCC system, and elects to use the (admitedly brilliant) auto-frog-switching options such as the "Frog Juicer", the "solution" to this particular "problem" has just taken DCC as a "starter option" up in the "$$$" stakes, and represents an additional time/$$$/wiring load which analog DC simply does not have to deal with.
3d - At this point, having added a passing loop and maybe some spurs,
the rank Newcomer will probably accidentally run a loco thru a turnout which is not set for them,
(although, in the case of analog DC, a very specific set of conditions would have to occur in order for this to be possible)
In the case of analog, the fault condition is temporary, and both loco and throttle will very likely shrug off the issue and live to fight another day.
In the case of DCC, the booster will very likely take extreme umberage to such behaviour and shutdown. At the worst, possibly loco, decoder, and booster damage may occur. At the least, the "fun" of running a train has been severely affected.
Logically and entirely correctly, in steps the concept of "short management", addition of "breaker units" or "tail-light bulbs". However, again, this is a uniquely-DCC issue, and thus represents _more_ wiring and "complexity" that the rank Newcomer must learn, face, and master.
3e - It is open to interpretation if the rank Newcomer in question would want to take on a reversing loop, wye, or turntable as part of their "first effort" layout.
Assuming the rank Newcomer does, (and with the hope that, should they consult any "experienced modeller" at this point, the experienced modeller would see and call attention to the risk as a matter of urgency...), then both Analog DC and DCC systems will face significant challenges in handling the resulting track arrangement.
DCC has the upper hand in that various _automatic_ solutions exist which will take care of any polarity issues without human intervention.
However, equally, analog DC can handle such situations with a much more $$$-effective manual switch,
which is completely compatible with DCC (Think about it...) The downside is that such solutions require the "detection and software" part of the solution to be played by the human.
Some cases, such as staging-only reversing loops can be handled "automatically" as shown HERE
http://www.zelmeroz.com/album_model/members/klyzlr/DynamiteCanyon.pdf
However, in summary
- DCC allows handling of reversing loops and turntables automatically, at the cost of $$$$
- analog DC handles reversing loops and turntables cheaply, at the cost of the operator having to _think_ about what they are doing, and _possibly_ having to manipulate some (arguably) unprototypical controls/switchgear
4 - Compatibility
Stated simply, at the point of "excited purchase", most any loco a rank Newcomer is likely to choose will run on analog DC. (I understand that there are some DCC-equipped locos which do not "understand" what a analog DC signal is, but IIRC the NMRA spec requires such "compatible" behaviour).
The inverse is not true, a DCC control system _may_ or _may_not_ successfully run an analog loco, but will only do so to the detriment of other (both DCC and analog) locos on the system. Adding a decoder to an analog loco adds cost and potentially effort (see Point 1 above).
Considering the case of adding a "factory DCC loco" to a "existing DCC layout"
(which potentially takes us a long way away from our starting positions of considering a rank Newcomer),
the new DCC loco will likely need at least an address CV change if it is to "co-exist happily" with existing locos on the layout. Failure to do this will result in multiple locos, with a common address, all reacting tot he same DCC commands, which sounds remarkably like "analog DC" behaviour (IE DCC requires more effort to achieve "better than analog DC" benefits).
5 - Performance
This is a complex one, but we must keep in mind the stated parameters of dealing with a rank Newcomer.
This is assumed to set the following criteria
- the Newcomer will not want, or be able to tweak the loco mechanism
(Logical, few newcomers will want to "take the shell off and 'get under the hood'")
- the Newcomer will not want or be able to tweak decoder CVs
(programming CVs "direct" requires knowledge of mechanical, electrical, and DCC technology.
Replacing this knowledge with "just program it with DecoderPro" adds $$$ and equipment.
SO, given this baseline, where does that leave analog VS DCC?
5a - even Decoder manufacturers agree that "...a loco that has some form of mechanical issue, and thus does not run smoothly on analog DC, is unlikely to run any better with the addition of DCC..." (check the documentation and install instructions which come with any user-install decoder).
Ergo, if the loco has a mechanical fault, there is no gain or loss with either system.
(use of BEMF requires CV tweaking knowledge, and knowledge of electro-mechanical behaviour of mechs.
See Baseline principles above).
Equally, a loco with limited or poorly-configured pickups is going to operate poorly on analog DC, and even worse on DCC. It would be easy to say "get the pickups optimised, and both control systems become about even in behaviour". However, refer to Point 5 above, the rank Newcomer is unlikely to want or be able to adjust such things. On a "out of box" loco with "questionable" pickup config, analog has a slight advantage, as it simply needs _some_ volts to move, it does not require perfect transmission/reception of high-timing-accuracy digital signals or "power distribution".
5b - Given a mechanically smooth-running mech, all but the most "toy trainset" throttles will give a smooth 0-12VDC sweep of voltages, with reasonably knob-rotation<> output voltage predictability.
I'm not saying the rank Newcomer must have a voltmeter connected to their analog DC throttle. However, if a given loco needs 2.7volts to "start visibly creeping",
even a rank Newcomer can turn the throttle knob,
stop when they see the loco start moving,
and surprise surprise if they haven't hit 2.7Volts...
5c - In contrast, at the common "default" 28step format of most DCC systems + "DCC factory equipped" locos,
accurately hitting the _optimum_starting_voltage_ at the motor of the loco,
let alone optimising the "human interface <> loco response"
(having that "optimal starting voltage" match Speed Step 1),
is nigh on impossible.
(Again, tweaking "Starting Voltage" and Speed Tables is above and beyond "rank Newcomer" capabilities, see above).
As far as the _critical_ issues are concerned, the above covers a lot of ground. However, in addressing the issues as set out in the thread:
- Sound: Agreed that the current drop of DCC sound decoders makes "onboard sound" a reality for many more modellers than the previous generation of ModelTronics, Circuitron, or even PFM/PBL systems. If sound is a _Must_have_ for the rank Newcomer, then DCC and some rather large $$$ purchases are virtually un-avoidable.
The question which therefore needs to be asked is not "analog or DCC", but rather "is the cost of the required DCC equipment worth the want for Sound?"
- Control: See Points 1x and 3x above
- Simplicity in Wiring: See Point 3x above
- Clean Appearance: See Point 3c above. With intelligent use of such switching, many "block switching" situations can be "automated" via the motion/throwing of turnouts. Ergo, as mentioned, the only "User Interface" switchgear the operator has to manipulate is to "throw the turnouts", all _traction_ switching is effectively "invisible" to the operator, and does whatever needs doing _without_ requiring manual operator action.
As noted, deployment of such systems falls within "good wiring practise" for _both_ analog and DCC layouts, and as such does not represent either a cost or effort "load" to either control system option.
(I have used such techniques on my last 5 layouts, which were intended as primarily analog-controlled systems.
However, as the wiring was _correct_ from the outset, simply "disconnecting the analog throttles, plug in the DCC system, swap the locos, and keep operating" was and is very much do-able...)
- Adjustability: Agreed that with CV tweaking, a DCC decoder can be optmised to match a given loco mech. However, please refer back to our "baseline", the rank Newcomer. refer to Point 5 above.
- handheld throttles: This is a "personal preference" issue, and is not a directly analog-VS-DCC issue.
- Better Operation: No-one, let alone me, can debate the experience you have had with your layout and equipment. However, I would submit the contents of Point 5x for consideration. I would also note the sheer number of "my DCC-controlled loco is not working right" posts on the MRH forum, (from all "skill levels" or modeller, not just rank Newcomers),
let alone the various MRR forums online, and the average "bull session" at most MRR clubs worldwide.
Compared to the equivalent number of "my analog loco is not running right" posts, we might infer either
- DCC is more prevalent than analog, and thus is statistically likely to have more related posting
(IIRC Joe's last "MRH reader survey" does not support this theory)
- DCC introduces compatibility issues which were unheard of under analog
- DCC is more "finicky" both internally (firmware/software) and externally (dirty track, momentary shorts, wiring impedance/capacitance/configuration), and therefore suffers "inexplicable failures" more often.
- Constant Power: See "Better Operation" comments above. It has also been noted that DCC relies on close/equal-voltage +/- signals being transmitted/recieved thru the rails. One risk of using "direct from track" lighting on DCC is that it may unduly load one "polarity" of the DCC signal. The results can vary, but in worst case can cause very oddball (unwanted!) loco behaviour.
Defense rests...
Happy Modelling,
Aim to Improve,
Prof Klyzlr
PS There is no "right" or "wrong" choice when it comes to control system choice. However, there _is_ a "wrong" choice if One fails to match the _Solution_ to the stated "Givens and Druthurs" of the layout in question...