Trains.com

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Yet another "what were they thinking" from someone who wants to be a DCC alternative Locked

9876 views
155 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Monday, August 10, 2020 6:33 PM

Lastspikemike

In that spirit I finally did what the original responder to the OP might have done, check with LocoFi

https://www.wifimodelrailroad.com/faq

The last two FAQ answer the questions posed, express and implied.

The design is deliberate and intended to capture the DC only market as well as DCC "ready". 

 

 Which is the product I started this thread about - the 8 pin compatible with DCC ready locos plug is wired backwards with regards to functions. It has the white and tyellow wires +, and the blue -. DCC specifications for the 8 pin connector have the blue wire as + and the white and yellow are -.

 ANd the other huge goof so far - they do not have an app fo Apple IOS. Only Android phones.

                                        --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Monday, August 10, 2020 6:51 PM

SeeYou190

Going around in circles.

We are so far from the original question that we might need to engage the hyper-drive to get back on course.

-Kevin

 

 Even better - the original wasn't a question, it was merely my observation that the vendor got the wiring backwards for a product that is supposed to plug right in to a standard 8 pin DCC socket.

                                      --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Tuesday, August 11, 2020 7:26 AM

rrinker
the original wasn't a question, it was merely my observation that the vendor got the wiring backwards for a product that is supposed to plug right in to a standard 8 pin DCC socket.

CVP makes a similar product, in the Airwire900 series

http://cvpusa.com/doc_center/r2_M15_User_Guide_web.pdf

and they see no need to reverse standard NMRA wiring even though their decoders don't come hardwired.  They also provide a simple LED test that, in my opinion, shows up LocoFi's supposed "FAQscuse" as, at best, lame.

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Tuesday, August 11, 2020 8:49 AM

 CVP has been at this a LONG time - starting with CTC-16 in the 70's. Railcommand was an expanded version that added headlight controls, right before DCC. They make DCC systems, and AirWire itself has been around a long time, one fo the first commercial direct radio control systems, very popular with garden railroads. The throttles are even compatible between their DCC system and AirWire.

                                     --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

Moderator
  • Member since
    June 2003
  • From: Northeast OH
  • 17,249 posts
Posted by tstage on Tuesday, August 11, 2020 2:57 PM

DC...it's soooooo primative...<tongue firmly placed in cheek>

https://tstage9.wixsite.com/nyc-modeling

Time...It marches on...without ever turning around to see if anyone is even keeping in step.

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Tuesday, August 11, 2020 3:47 PM

Lastspikemike
Ahem, yes, like those electrons that orbit the nucleus in Bohr's "model" of the atom. And wander through the airwaves or the "ether" remember that stuff?

Yes, essentially point carriers of very well-defined charge, with very well defined virtual mass.  I'm sure you'll whine they are 'theoretical constructs' but they have been just as good as other subatomic particles as engineering working assumptions for many years, first in electricity and then in electric power.  I think it is safe to continue working with the assumption that "they" behave as described as there is so much interesting nuclear physics associated with them and their 'heavier' versions.

Incidentally the stuff that 'wanders through the airwaves' isn't electrons.  You really should look stuff like that up before shaming yourself further in postings.

You know current isn't electrons actually flowing or transistors couldn't work.

And you think current is... what, exactly?  ExB?  Magic displacement?  "Waves" in the "luminiferous aether"?  Hint: what is a coulomb, and why is it important to amperage?...

Again there is so much physics on conduction that relates to electronics that it is comical to hear a discourse 'debunking' it as if you knew something secret the rest of us don't.  You probably think nerve conduction is electronic, too.  What's next, that holes in semiconductors don't exist either, because you can't have charge less than zero?

And so on.

And it's certainly on and on: this is something like the sixth field I've seen you discourse on that you haven't really understood, but that doesn't stop you from browbeating those on here who I think actually might.

  • Member since
    September 2004
  • From: Dearborn Station
  • 24,281 posts
Posted by richhotrain on Tuesday, August 11, 2020 3:55 PM

Overmod
 
Lastspikemike
And so on. 

And it's certainly on and on: this is something like the sixth field I've seen you discourse on that you haven't really understood, but that doesn't stop you from browbeating those on here who I think actually might. 

Here is something that every 19 year old law student (hmm) would be familiar with - - that old evidentiary adage, He knew that he did not know whether or not he knew.

Rich

Alton Junction

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Tuesday, August 11, 2020 4:50 PM

richhotrain
He knew that he did not know whether or not he knew.

I think the issue here is different.  He is quite sure he knows whether or not he knows.  It's whether he does know that is at issue.

Now to cut him some slack, there is that aspect of 'knowledge' my father would expound after the 5th bottle of Old Dutch Philosopher -- the part you can't know.  No one has ever 'seen' an electron and it does not behave like a true particle (other than statistically).  Likewise no one can make a 'perfect' square waveform in a number of respects.  And you can certainly have 'currents' with, for example, protons in fluids, or holes in some semiconductors.  Perhaps my favorite is phlogiston, the career-wrecking politicized ego-ridden global-warming-style consternation of the 18th Century.

The problem is that if you're going to disparage that sort of thing in practical use, you have to know the degree to which the theory applies or is a 'good predictor' -- certainly for technology, and in some cases for science.  As Copernicus pointed out (long before Fourier) there's no reason why epicycles sufficiently superposed won't generate any orbit ... just that treating the Earth's motion as if the orbit were heliocentric gives you easier calculations for accurate feast days.

  • Member since
    September 2004
  • From: Dearborn Station
  • 24,281 posts
Posted by richhotrain on Tuesday, August 11, 2020 5:07 PM

Overmod
 
richhotrain
He knew that he did not know whether or not he knew. 

I think the issue here is different.  He is quite sure he knows whether or not he knows.  It's whether he does know that is at issue. 

Either way...

Alton Junction

  • Member since
    March 2017
  • 8,173 posts
Posted by Track fiddler on Tuesday, August 11, 2020 6:53 PM

ConfusedConfusedIndifferentZip it!

 

 

TF

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Tuesday, August 11, 2020 8:52 PM

 Yes, many of us EEs also had to take enough physics so we know the conventional way circuits are discussed is sort of backwards from how it really works in the physics world. But it works fine as a model to describe the behavior of all common electronic components.

 That's a pretty lame excuse, that WiFiTraxx isn;t making a decoder. No, they aren;t. But they ARE making a device to be used IN PLACE OF a DCC decoder and even outfit it with the standard connection plug that is provided in many locos so you don;t have to solder wires in - AND THEN WIRE IT BACKWARDS. The AirWire board is NOT a DCC decoder, either. It doesn;t respond to DCC signals on the track, it responds only to radio signals sent from the AirWire throttle. It connects between the power source and the motor, in place of a decoder. They just happened to not do their lighting circuits backwards, so if you are adding AirWire to a loco that already had an NMRA 8 pin DCC decoder socket, you can just plug it in, no changes, and everything works, unlike the WiFiTraxx board.

 AirWire at least also has features people have come to expect in a controller - a knob, which you can click to change direction, so no hunting for a button, toggle, or a part of the screen to tap on, and a screen to display info or walk you through configuration settings. Best of all, the same throttle works with their DCC system. so you could use their DCC system on your basement HO layout, then take the throttle outside and use it to run your dead rail, battery pack on boards, G scale garden layout. Same throttle, no need to learn two different control systems. 

                                  --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Wednesday, August 12, 2020 8:54 AM

Lastspikemike
I appreciate the concerns expressed about my personality defects but it really is not necessary.

It is not the 'personality defects' at issue -- we all have them, me more than most.  It's the making of claims of superior expertise or experience that contain things that ain't so.  Those need to be corrected before the susceptible jury of modelers reading for technical advice are led into rendering an unjustified verdict...

We all love the sinners, but that won't stop us from hating sin...

  • Member since
    January 2017
  • From: Southern Florida Gulf Coast
  • 18,255 posts
Posted by SeeYou190 on Wednesday, August 12, 2020 10:55 AM

Dots - Sign

-Kevin

Living the dream.

  • Member since
    July 2009
  • From: lavale, md
  • 4,678 posts
Posted by gregc on Wednesday, August 12, 2020 11:05 AM

Lastspikemike
I deduce that one DCC throttle can control only up to 6 trains simutaneously operating and wonder if the up to 5 that are cut loose from the human operator at any one time are controlled by some sort of looping within the decoder or by simultaneous transmission of up to 6 streams of packets by the DCC control computer. I infer the latter as it is hard for me to see how a decoder could preserve the signal internally.

not looping within the decoder; the decoder is in the loco.

a single DCC command station communciates with all cabs (user interface) and generates the DCC signal to one or more boosters that provide power to the track.   the command station will sequentially generate a DCC frame for each active loco.  less than 8 msec/packet considering that the max bit is 200 usec and there are ~40 bits in a typcal frame (see S9.2 or all standards, S9)

i'm not sure what the max # of locos is for any one system but 100 would be conceivable.   i think NCE will poll up to 64 cabs.

 

Lastspikemike
smug superiority

errant information is confusing to other readers.   i think it better to state things as your understanding (e.g. "deduce") rather than facts

 

greg - Philadelphia & Reading / Reading

  • Member since
    July 2006
  • 33 posts
Posted by stevetx on Wednesday, August 12, 2020 11:39 AM

I am confused about the subject of this thread.  I'm DC so most of this thread is over my 79 year old brain but I keep trying to learn.  I started out reading rrinker's starting post about LocoFi.

rrinker on 7/27/20 wrote:

". . . review the LocoFi system. . . . . . But they blew it - on their decoder, the function wires are positive and the common is negative - exactly opposite the DCC standard!" 

So, I went to the LocoFi web site and found in thier FAQ that the wiring was in fact not DCC compatible for 8 pin.  You do need to change the wiring in DCC ready locos to get LEDs to work.  Thus, rrinker is correct.

I thought this thread was all about LocoFi but then on page 5 rrinker brought up WifiTrax and said it was backwards.  

rrinker on 8/11/20 Wrote:

"That's a pretty lame excuse, that WiFiTraxx isn't making a decoder. No, they aren't. But they ARE making a device to be used IN PLACE OF a DCC decoder and even outfit it with the standard connection plug that is provided in many locos so you don;t have to solder wires in - AND THEN WIRE IT BACKWARDS." 

So, I went to the WifiTrax web site but I cannot find that a "used in place of a DCC decoder" product by WifiTrax.  They do make an 8 pin NMRA DCC standard compatible wireless interface to a DCC decoder that does process the incoming two track wires. I found this document for the inerface product "WDMI-35 Wi-Fi/DCC Loco Interface Module Getting Started" but it shows to be compatible to the 8 pin NMRA.  There is nothing that indicated that the common was backwards. 

What am I missing?  Yes, most of the commentary in this thread I do not understand but I am missing something in going to the LocoFi and WifiTrax web sites. 

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Wednesday, August 12, 2020 11:41 AM

Lastspikemike
What number of trains can one operator reasonably expect to control on a home sized layout?  We have three DC controllers and running three trains simultaneously seems very challenging for one operator.

This will be an interesting question to ask the various modelers here.  I have generally thought that most of the 'multiple train' control involved the 'club night lite' operating sessions, where people get together to run on a private layout.  If I were going to run multiple trains at once it would be mostly to 'watch' (meaning only intermediate use of any controls, with some form of automatic otherwise) or to run just one at a time.  For DC this would immediately imply complex power management and ATS-enabled signaling; the operation would also benefit from 'interlocked' route selection.

We have two continuous loops and one yard so in theory operating three trains should be simple.

I would agree, and it would be comparatively simple to assign a 'cab' to each and have the trains merrily go when no one is tinkering with the corresponding settings.  We had a recent thread on who did and didn't like to do that sort of thing.

[/quote]I deduce that one DCC throttle can control only up to 6 trains simutaneously operating and wonder if the up to 5 that are cut loose from the human operator at any one time are controlled by some sort of looping within the decoder or by simultaneous transmission of up to 6 streams of packets by the DCC control computer.[/quote]This is a place you could reference 'primitive' technology.  There is a limit to the addressable bandwidth in DCC, but to my knowledge little control storage apart from CVs.  Adding the necessary information plus overhead plus any guard band between addressed instructions will give you the required bitrate... which must fit within the modulation rate of the signal.

 

I infer the latter as it is hard for me to see how a decoder could preserve the signal internally.

A more modern decoder could be treated as a small special-purpose computer, running what might be a sophisticated running program interrogating other devices on the layout for context.  The DCC modulation could easily bootload such a program, run a firmware update utility, etc.  Ask your brother about some of the possibilities of AI/ES in such a context.

More importantly this offers the capability you indicate: running what may be complicated operations with either interrupt-based or as hoc acceptance of external commands... cutting down dramatically on the actual real-time modulation needs for the DCC control signals.  (I would also mention that data fusion from other modulations, including several 'flavors' of wireless, then become easy to do, and relatively easy to secure...)

That would also explain why power interruptions cause signal dropouts even with locomotive keep alive circuitry. The DC power can be stored onboard but not the DCC signal stream.

That would be correct.  Some of the 'issue' is concern over failing safe; if the locomotive keeps running on unexpected LOS (as some of the old command systems did) you can have disasters with little warning involving hundreds of dollars' worth of sensitive equipment.  Another concern is the absence of even rudimentary reverse-channel signaling from the 'connected environment' which would allow the kind of short-term request and retransmit of dropped-packet or LOS information that would preclude most digital quality-of-signal concerns.  As you note, any information 'stepped on' by noise or interruption is gone (and the device has to wait for the next valid command to 'know more') -- the deep voltage slew of the DCC modulation helps with the former but of course can't put back what disconnection eliminated...

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Wednesday, August 12, 2020 11:47 AM

 The only one that truly allows you to have live control of more than one train at a time is Digitrax, their full featured throttles have 2 knobs, so you cna have 2 trains actually under control (meaning not having to push any buttons or anything to switch between them). You can have many running at the same time, just good luck trying to switch back and forth between them. So I'm going to go with 2 as a practical maximum, because I can actually have them under control, just turn the respective speed knob, with my Digitrax throttle.

 The limit of 6 is very much system dependent. Right now I still have my original Digitrax Zephyr as a command station, it can handle up to 12 locos. The larger command stations can handle 100 or in the case of the DCS240, 400. Theroetically, you can start up all those locos from one throttle, the Digitrax system design does not limit the number of locos 'controlled' by a throttle to some limit based on the throttle's memory. In any case, only the loco who's address is shown on the display is actually under control of the throttle's knob and buttons.

 This is where you seem to misunderstnad something about DCC. The command station generates the DCC data. It is continually sending packets for every loco that is running. The throttle doesn't do that, althought he throttle does update the command station based on user inputs. Either by sending the updates in a peer to peer system like Digitrax, or by being polled like NCE. The decoder doesn't need to remember anything, it gets a steady stream of packets. There is usually a timeout setting in the decoder that tops it if it doesn't see packets for some period of time, likewise the command station may see a lack of updated information from the throttle as the loco no longer being controlled, and may also stop the loco. Packet timeout on the loco side is part of the NMRA standards, the throttle bus and command station communications are left up to the individual manufacturers to handle in their own way.

                                  --Randy


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Wednesday, August 12, 2020 11:53 AM

stevetx

I am confused about the subject of this thread.  I'm DC so most of this thread is over my 79 year old brain but I keep trying to learn.  I started out reading rrinker's starting post about LocoFi.

rrinker on 7/27/20 wrote:

". . . review the LocoFi system. . . . . . But they blew it - on their decoder, the function wires are positive and the common is negative - exactly opposite the DCC standard!" 

So, I went to the LocoFi web site and found in thier FAQ that the wiring was in fact not DCC compatible for 8 pin.  You do need to change the wiring in DCC ready locos to get LEDs to work.  Thus, rrinker is correct.

I thought this thread was all about LocoFi but then on page 5 rrinker brought up WifiTrax and said it was backwards.  

rrinker on 8/11/20 Wrote:

"That's a pretty lame excuse, that WiFiTraxx isn't making a decoder. No, they aren't. But they ARE making a device to be used IN PLACE OF a DCC decoder and even outfit it with the standard connection plug that is provided in many locos so you don;t have to solder wires in - AND THEN WIRE IT BACKWARDS." 

So, I went to the WifiTrax web site but I cannot find that a "used in place of a DCC decoder" product by WifiTrax.  They do make an 8 pin NMRA DCC standard compatible wireless interface to a DCC decoder that does process the incoming two track wires. I found this document for the inerface product "WDMI-35 Wi-Fi/DCC Loco Interface Module Getting Started" but it shows to be compatible to the 8 pin NMRA.  There is nothing that indicated that the common was backwards. 

What am I missing?  Yes, most of the commentary in this thread I do not understand but I am missing something in going to the LocoFi and WifiTrax web sites. 

 

Simplest explanation? It's why a gas filler and diesel filler nozzle are different sizes, so you can't put diesel in yoiur gas car, the nozzle won't fit (you can do the reverse, and make a hash out of your diesel engine, but let's ignore that for now). If they were both the same, you'd have (believe me, people ARE this dumb) people ruining their gas cars left and right with diesel.

WiFiTraxx has chosen to use the same style plug as a standard DCC decoder on their product, in an attempt to make it easy to install. Except they wired it backwards. So it fits - but does not work, without further modification. That's ALL I was saying when I started this thread. It's a mistake, a bad design, because you can plug it in expecting it to work right out of the package, so you can use direct WiFi control instead of DCC - except that the plug is wired backwards and it WON'T work. All this add-on DCC stuff in the thread comes from LastSpikeMike.

                                      --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Wednesday, August 12, 2020 12:02 PM

 Further on the DCC sisgnal - since the same packets are repeatedly sent by the command station, it's not actually an issue if a loco misses a packet or two addressed to it. Another will be along. That's why there is no need for a acknowledgement for packet reception. 

 The total available bandwidth allows for hundreds of locos to run at the same time. Most clubs don't come anywhere near this, but there is one well known layout, the owner has switched DCC systems more than once to get to the point where the wired and wireless throttles can reliably support the number of simultaneously running locos during his op sessions. If anyone stresses DCC's capabilities, it's him. Note the issues experienced over the years are entirely on the throttle/command station interface, NOT the DCC protocol. Even with so many trains moving at once (and this layout is big enough to handle them all, it's not a whole bunch of trains all chasing their tails), you press the horn button and the horn sounds. You slow down for a station stop, and the loco you are running slows down. 

 Not bad for a 'primitive' one-way communications protocol.

                                             --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    July 2006
  • 33 posts
Posted by stevetx on Wednesday, August 12, 2020 12:18 PM

Randy, Thanks for your quick reply. Your "Except they wired it backwards" still has me confused. All the wire instructions I see on the WifiTrax web site never gave me a hint of any 'backwards' situation like the LocoFi web site did.  The LocoFi product seems to be totally different product than the WifiTrax product but you say there the same.  my short coming.  I will go back and look again at WifiTrax to  find what I missed.

  • Member since
    April 2020
  • 526 posts
Posted by Ringo58 on Wednesday, August 12, 2020 12:23 PM

ATLANTIC CENTRAL

 

 
 

Why I am on this little rant?

Because your opening line, and to some degree, your whole post is condescending  to those who have decided not to use DCC.

You are saying this product is aimed at the "stupid" people who have not gone DCC.

How much are 10 DCC wireless throttles and 140 decoders? My system did not cost that much.

 

 

I think you took what he said out of context. He didn't call you out personally, he was referring to the people like me that don't want to spend the money on DCC. Never once did he attack DC or saying that it was behind in technology. All he said who the target audiance was and how the system is behind current DCC technology. Not how DCC is better than DC. 

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Wednesday, August 12, 2020 12:38 PM

 SO much back and forth in this thread now even I am getting confused. Two different prodocts - the WiFiTrax ones with the 9 pin JST connector appear to be made to actually have the same pinout as the 9 pin DCC plug found on many locos. So they DO just plug in. It's the LocoFi one that is backwards.

 Names are too similar. The whole "WiFi" name for wireless networking is yet another one of the idiotic marketing things ever present int he computer industry. A stupid acronym that conveys exactly zero information on what the technology is. Wireless Fidelity? Only a marketing major who spent 4 years in college partying could come up with that junk. They claim it never stood for that, but the brand consulting firm that came up with it (yes, a company who's sole purpose is to advise people on brand names for their products - what a world) says it was a play on HiFi audio, and that has always stood for High Fidelity.

 Then the engineers go the opposite way, and just tack different letter suffixes after the base standard designation. Which one is better, a, b, n, ac, ax? So now the current top dog, 802.11ax, is being braded as WiFi 6. 802.11ac was retoractively named WiFi 5 for branding. So I guess that makes sense now, 6 comes after 5 and is an improvement on it, but on the cellular side we have 3G, 4G, and 5G. Nothing in sync. Leave off the outright lying from AT&T on which tech they actually are using for cell phones.

 What does this have to do with trains? Not much, but the confusion of all these "*Fi" names gets the products all mixed up. WiFiTrax makes a controller for direct WiFi control of trains that is plug and play with the standard type of connector found in many DCC Ready locos. LocoFi makes a board with a common plug on it, but which does not directly plug in even though it's the same plug. It doesn't help that the LocoFi web site is wifimodelrailroad.com, further comfusing them with wifitrax.

(another example - I've seen people want to build modular layouts and use standard 120VAC outlets and plugs as the power link for track power between modules - about the worst idea ever, because what's to stop someone from plugging one into a real AC outlet, energizing the tracks - two strips of bare metal - with 120VAC house current. That's why you don't repurpose a standard connector for something else) 

                             --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Wednesday, August 12, 2020 12:40 PM

rrinker
Simplest explanation? It's why a gas filler and diesel filler nozzle are different sizes, so you can't put diesel in yoiur gas car, the nozzle won't fit ...

If I may, the fundamental issue can be illustrated in a slightly different part of the car.  Multiple DC circuits can be wired so either the 'source' or 'sink' of their voltage comes from one common point (to which a single common wire could be connected).  In a car this is conveniently the whole mass of what used to be metal chassis and other structure, so you only have to run one wire to a device to be powered (the other connection being where it screws to the 'common'.

Now' like it or not, the electrons in DC current flow from the 'negative' to the 'positive' (by early convention too pedantic to give here) and so with negative ground the whole chassis becomes the 'source' of current with only the voltage-dropped return having to negotiate individual wires.  Many British cars have positive ground, where the power goes out through the individual wires and only back to the battery through the common.

What Randy pointed out is that DCC assigned the 'utility' DC supply controlled by the decoder in a particular way, by standard -- and this facilitated the use of standard connection plugs that one could assume would reasonably follow the standard.  This is what LocoFi did not do that is the subject of this discussion.

In my opinion they could have avoided any concern by not using or touting a 'NMRA standard' plug if they wanted to wire that supply polarity backward.  Instead they crawfish and try to make it as feature not a bug'.  Randy pointed out early that the NMRA standard makes more objective sense electrically, so there was no performance reason they could assert.

  • Member since
    July 2009
  • From: lavale, md
  • 4,678 posts
Posted by gregc on Wednesday, August 12, 2020 2:29 PM

Lastspikemike
By 6 I meant the maximum size of the recallable stack of locos that have recently been or are being run by one throttle

very precise

an NEC PowerCab can handle 6 locos.   bear in mind that a single knob is not a good way to control multiple locos because you may have the knob set high for one loco and may not want it set that high when selecting another.  the PowerCab uses buttons to control speed.

greg - Philadelphia & Reading / Reading

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Wednesday, August 12, 2020 3:12 PM

 It is very specific to the system architecture - the stuff left up to the vendor, not controlled by the NMRA DCC specifications.

 NCE throttles are limited to no more than 6 loco addresses per throttle. That's in part because the cabs are really just dumb terminals, and all the information is held in the command station, and that's all the memory it allows for. Even though the PowerPro command station can handle a LOT more than 6 locos, you need more throttles to go over 6. Not that controlling 6 locos from one throttle is really possible for one person to do.

Digitrax's network works completely differently. There is no 'recall' because the information about which throttle is controlling which loco is stored in a completely different manner. If the command station is capable of 100 locos (DCS210 model) then I can theoretically use one throttle and start 100 different addresses running. 

(just because you CAN - doesn't mean you SHOULD< or that it's even practical)

Issue 2 is the type of knob. The NCE ProCab/PowerCab use buttions and an encoder wheel as speed control options. The Digitrax DT-series throttles also use encoder knobs (and there are buttons as well). With these, you can easily switch back and forth between multiple locos and none of the change speed unless you actually turn the knob. The NCE Cab-06p and the Digitrax UT4 use potentiometer knobs. So if you have one loco running at 1/4 throttle, switch to another one and accelerate it to 3/4 speed, then switch back to the first one, you have to guess and get the knob close to the right position, or else the loco will speed up from 1/4 to 3/4. And vice-versa. Just another reason I don't like potentiometer throttles. ESU solves this in the most typical German over-engineered way - their potentiometer knob is motorized, so when you switch to a different loco, the motor automatically dials the knob to whatever speed the loco is running.

 It's differentiation like this that makes a DCC market, where you can compare features between different brands, yet at the track level, they all work together, because they all output NMRA DCC packets. Same with decoders. 2, 4, 6, 8, 12 function wires. Sound/no sound. BEMF/no BEMF. Specific form factors for specific brands of locos, or generic wires, or one of several standardized plugs. Pick your desired options. Brand does not matter, they all respond to the same NMRA DCC packets coming in from the track pickups. 

                                         --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

Moderator
  • Member since
    June 2003
  • From: Northeast OH
  • 17,249 posts
Posted by tstage on Thursday, August 13, 2020 8:28 AM

That's why an encoder is preferred over a potentiometer when using recall stacks.  Moving the knob or encoder wheel increases or decreases the speed from where the throttle was last set for the recalled locomotive rather than from a fixed point on the potentiometer knob.

Tom

https://tstage9.wixsite.com/nyc-modeling

Time...It marches on...without ever turning around to see if anyone is even keeping in step.

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Thursday, August 13, 2020 8:49 AM

 There's another way to handle switching between locos running at different speeds, but it would be the least desireable way possible to do it. I have no experience with MRC DCC so I have no idea how they do it, but they were rather late to the party as they had several DCC systems prior to their existing lineup that had little expansion capability - if you wanted/needed more, you tossed the old one and bought a new one. The other big players from the beginning designed their systems to allow you to start small and add on without having to discard your original investment.

 That would be to just take the current throttle knob position as the speed. So you are running a loco at 3/4 speed. The knob is 3/4 of the way turned up from the stop position. You then recall a loco that you left running at 1/4 speed. The knob now assumes the curent position (physical, 3/4 of a rotation) is that 1/4 speed. Meaning you turn the knob back to the halfway (phycical) point and the loco stops, or you turn it to the 100% point (physically) and the loco now either runs at half speed, or it compresses the speed range of 1/4 to full into just 1/4 of the physical range of the knob. I really hope NO system does that, that's absolutely horrible. None of the ones I am familiar with do.

 But somewhere you have to take that into account - the the knod is at 3/4 and you select a loco running at 1/4, at some point the two have to be matched (remember, we are talking about potentiometer knobs here, where there is a fixed range of motion, usually 270-320 degrees). If you touch nothing, it's fine, the loco can continue to move at the set speed. But at some point, either the command station has to poll the throttle, or the throttle has to send the current state to the command station. Or you touch the knob and turn it just a little. Having the loco adopt the throttle position is how most systems do it, ESU uses the motorized knob to physically turn the knob to match the speed.

 This is all a non-issue if the knob is instead a rotary encoder, as the physical position in a 360 degree circle has no bearing on the actual speed. The speed setting is simply a value in the throttle that is incremented when the knob is turned clockwise, or decremented when the knob is turned counterclockwise. When you move from a loco running at step 100 to one running at step 20, the number stored in the throttle changes to the new loco's value, and neither the physical interface (knob) nor the loco have to change to perfectly match the speed of the loco you just switched to.

 Remember we are now talking about the user interface side, which is NOT part of the NMRA DCC standard, so not all systems behave the same way. NMRA DCC was only designed to guarantee compatibility at the track level. So talk about how many recalls there are and what they do is strictly a specific manufacturer issue.

 An example - the DCC system comparison charts you can find all over typically list a Recall feature, and Digitrax is listed at 0. OMG, what a junk system, XYZ has 6 recalls. Not so fast. With Digitrax, there actual IS a recall function in the throttle, but it is merely a shortcut memory to the last few selected addresses. Sort of like hitting the Undo menu option in a spreadsheet to change the value back to the previous one. It has no bearing on the actual operation of locos. Whereas NCE, and it seems MRC, the locos in the recall stack of the throttle are the ones that are actually running, and the recall stack is used to select between the running locos.

 People talk about 'storing' locos in their DCC system. That''s not how it works, nor how you should use the system. It was very common a number of years ago when Digitrax came out with the original Zephyr, and then when NCE came out with the PowerCab. Zephyr was listed as having a capacity of 10 trains. PowerCab is 6. The common question heard was : I have more than 6 (or 10) locos, how can I fit them all in, I guess I need a bigger system. No. You do not store your entire loco roster in the system, you call them up as you use them. The number of trains limit is how many can be moving AT THE SAME TIME.

                                   --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    February 2018
  • From: Flyover Country
  • 5,557 posts
Posted by York1 on Thursday, August 13, 2020 1:59 PM

I know next to nothing about the technical side of things, but ... 

My NEC powercab has both the thumb wheel and buttons to control speed.  If I have a locomotive running at 20, and switch to one running at 40, the thumbwheel automatically runs at that speed.  If you switch back, it's at 20 again.

I would think could somehow be accomplished with a knob.

If I understand, that's what Randy is talking about with a rotary encoder?

York1 John       

Moderator
  • Member since
    June 2003
  • From: Northeast OH
  • 17,249 posts
Posted by tstage on Thursday, August 13, 2020 2:26 PM

Yes, encoders do not have stops so you can spin or rotate them endlessly and they will only go to there pre-set max or min settings.

Tom

https://tstage9.wixsite.com/nyc-modeling

Time...It marches on...without ever turning around to see if anyone is even keeping in step.

  • Member since
    July 2009
  • From: lavale, md
  • 4,678 posts
Posted by gregc on Thursday, August 13, 2020 2:26 PM

the thumbwheel is an encoder

greg - Philadelphia & Reading / Reading

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Users Online

There are no community member online

Search the Community

ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT
Model Railroader Newsletter See all
Sign up for our FREE e-newsletter and get model railroad news in your inbox!