Auto-reversing for continuous operation of a point-to-point layout
Digital Railroader
the locomotive's direction, but we believe that it's for DC layouts only. Is
there a similar device for DCC layouts?
The Conductor
Digital Railroader LLC
dr-new...@wi.rr.com
Bill
We know of one device that automatically senses end-of-track and
reverses the locomotive's direction, but we believe that it's for DC
layouts only. Is there a similar device for DCC layouts?
Maybe the MRC #AD520 Auto Reverse Loop Module could be adapted to do
what you want. The additional switches and/or electronics could be
placed on/in the trackwork
Bill
Bill's Railroad Empire
N Scale Model Railroad:
http://www.billsrailroad.net
Brief History of N Scale:
http://www.billsrailroad.net/history/n-scale
Resources: Links to over 500 helpful sites:
http://www.billsrailroad.net/bills-favorite-links
Bookstore: http://www.billsrailroad.net/bookstore.html
Gregory Procter
Bill wrote:
> Digital Railroader wrote:
> We know of one device that automatically senses end-of-track and
> reverses the locomotive's direction, but we believe that it's for DC
> layouts only. Is there a similar device for DCC layouts?
> --------------------------------------------------
> Maybe the MRC #AD520 Auto Reverse Loop Module could be adapted to do
> what you want. The additional switches and/or electronics could be
> placed on/in the trackwork
That item is designed to reverse the track polarity so that a train can
negotiate a reversing loop without DPDT switches. DCC decoders don't see
polarity, but track joins do.
Regards,
Greg.P.
Bill
That item is designed to reverse the track polarity so that a train can
negotiate a reversing loop without DPDT switches. DCC decoders don't see
polarity, but track joins do.
Greg, thanks for pointing this out. My aging brain realized that just
after I posted the reply.
Robert Heller
In a message on Thu, 19 Jun 2003 00:57:10 -0400 (EDT), wrote :
B> Greg.P. replied:
B> That item is designed to reverse the track polarity so that a train can
B> negotiate a reversing loop without DPDT switches. DCC decoders don't see
B> polarity, but track joins do.
B> --------------------------------------------------
B>
B> Greg, thanks for pointing this out. My aging brain realized that just
B> after I posted the reply.
One *downside* of DCC is that it is not really good for 'dumb'
automation (dumb == fun with simple logic circuits driving relays).
For a simple store-display type layout, unattented operation, DC is
still useful.
It is possible to do this with DCC, but would require a *lot* more
electronics, since you need to form the DCC packets that implement the
reversing commands for the locomotive. It might be easier for more
Model RRs to just buy themselves a 'cheap' PC and write a VB (or Bash
Script, depending on O/S) script.
Beware of the axiom: "If all you have is a hammer, everything looks like
a nail." Sometimes you just need to use a screwdriver... DCC is no
more a universal tool than a hammer is a universal tool. Sometimes you
just need to use DC...
B>
B> Bill
B> Bill's Railroad Empire
B> N Scale Model Railroad:
B> http://www.billsrailroad.net
B> Brief History of N Scale:
B> http://www.billsrailroad.net/history/n-scale
B> Resources: Links to over 500 helpful sites:
B> http://www.billsrailroad.net/bills-favorite-links
B> Bookstore: http://www.billsrailroad.net/bookstore.html
B>
B>
Mike Tennent
>We know of one device that automatically senses end-of-track and reverses
>the locomotive's direction, but we believe that it's for DC layouts only. Is
>there a similar device for DCC layouts?
>
I don't think so. I make one for DC and have pursued the same question
unsuccessfully because I'd like to offer one to my customers. (I hope
I'm wrong and someone will point one out.)
The primary problem is that the technology is totally different. With
DC, you simply switch polarity of the track and the engine switches
direction. All you have to do is sense that the engine is in position
to change direction.
Switching track polarity on DCC doesn't effect the direction of the
engine. The Command Station has to tell it to stop and switch
directions, which is a whole different ballgame.
It can be done with a computer, position sensing, and the appropriate
software. Theoretically it could be done with a microchip of some sort
for a stand-alone system, but the variables involved would be
daunting.
Mike Tennent
"IronPenguin"
Operating Traffic Lights
Crossbucks
Special Effects Lighting
http://www.ironpeng.com/ipe
Steve Watson
> billsr...@webtv.net (Bill),
> In a message on Thu, 19 Jun 2003 00:57:10 -0400 (EDT), wrote :
>
> B> Greg.P. replied:
> B> That item is designed to reverse the track polarity so that a train can
> B> negotiate a reversing loop without DPDT switches. DCC decoders don't see
> B> polarity, but track joins do.
> B> --------------------------------------------------
> B>
> B> Greg, thanks for pointing this out. My aging brain realized that just
> B> after I posted the reply.
>
> One *downside* of DCC is that it is not really good for 'dumb'
> automation (dumb == fun with simple logic circuits driving relays).
> For a simple store-display type layout, unattented operation, DC is
> still useful.
>
> It is possible to do this with DCC, but would require a *lot* more
> electronics, since you need to form the DCC packets that implement the
> reversing commands for the locomotive. It might be easier for more
> Model RRs to just buy themselves a 'cheap' PC and write a VB (or Bash
> Script, depending on O/S) script.
??
Unless I badly misunderstand the way DCC works, sending reversing
commands to the loco isn't the right thing to do. "Forward" and
"backward" are determined strictly by which way the decoder is wired.
I gather there's a few ways of handling loops under DCC, but they all
basically work by detecting the polarity mismatch, and reversing the
track feed -- not too different in principle to how it's done in DC.
One obvious way, for the simple case of an end loop, is to feed loop
power through the loop turnout. When the train is completely in the
loop, throw the turnout. Loop polarity now matches the main at the
exit point.
-- Steve
Paul Newhouse
Is this:
"http://www.ttx-dcc.com/technews/powershield_icb.htm"
what you are looking for? In a reversing loop you don't want the
pick up wheels shorting out as it goes through the switch. Flipping
the "polarity" in the rails should not affect DCC operationof the
loco.
Paul
--
Working the Rockie Road of the G&PX
Stan Ames
Digital Railroader wrote:
> We know of one device that automatically senses end-of-track and reverses
> the locomotive's direction, but we believe that it's for DC layouts only. Is
> there a similar device for DCC layouts?
>
Auto reversing is different with DCC but it is can easily be done. I
have had a unit for years that does this.
In DC you reverse the polarity of the track.
In DCC you need to reverse the polarity of the locomotive.
What I have done is to put a detector at each end of the track and have
the detector control a relay. (same as DC)
The difference is that the relay controls the direction switch of a
handheld rather then the polarity of the track. You set the address of
the locomotive in the handheld and the speed and functions work normally
but the direction switch in effect reverses as the locomotive trips the
detector.
Any detector works, I have used both optical and current detection for
this purpose.
---------------------------
Stan Ames stan...@attbi.com
DCC Evangelist, DCC Book co-author, married to a DCC manufacturer rep
Home page with Disclosure Statement: http://www.dccsig.org/sra/
Paul Newhouse
Stan Ames <s...@mitre.org> writes:
>
>
> Digital Railroader wrote:
>> We know of one device that automatically senses end-of-track and reverses
>> the locomotive's direction, but we believe that it's for DC layouts only. Is
>> there a similar device for DCC layouts?
>>
>
> Auto reversing is different with DCC but it is can easily be done. I
> have had a unit for years that does this.
>
> In DC you reverse the polarity of the track.
>
> In DCC you need to reverse the polarity of the locomotive.
>
> What I have done is to put a detector at each end of the track and have
> the detector control a relay. (same as DC)
>
> The difference is that the relay controls the direction switch of a
> handheld rather then the polarity of the track. You set the address of
> the locomotive in the handheld and the speed and functions work normally
> but the direction switch in effect reverses as the locomotive trips the
> detector.
Way cool. Can you point me at a How-To, a url or something, for the
excruciating details?
Thanks,
Paul
> Any detector works, I have used both optical and current detection for
> this purpose.
>
> ---------------------------
> Stan Ames stan...@attbi.com
>
> DCC Evangelist, DCC Book co-author, married to a DCC manufacturer rep
> Home page with Disclosure Statement: http://www.dccsig.org/sra/
>
--
Robert Heller
In a message on Thu, 19 Jun 2003 15:14:19 -0400, wrote :
SA> Digital Railroader wrote:
SA> > We know of one device that automatically senses end-of-track and reverses
SA> > the locomotive's direction, but we believe that it's for DC layouts only. Is
SA> > there a similar device for DCC layouts?
SA> >
SA>
SA> Auto reversing is different with DCC but it is can easily be done. I
SA> have had a unit for years that does this.
SA>
SA> In DC you reverse the polarity of the track.
SA>
SA> In DCC you need to reverse the polarity of the locomotive.
SA>
SA> What I have done is to put a detector at each end of the track and have
SA> the detector control a relay. (same as DC)
SA>
SA> The difference is that the relay controls the direction switch of a
SA> handheld rather then the polarity of the track. You set the address of
SA> the locomotive in the handheld and the speed and functions work normally
SA> but the direction switch in effect reverses as the locomotive trips the
SA> detector.
Right. The handheld implements most of the 'smarts' (generating the DCC
commands), either by itself or with the help of its base station.
SA>
SA> Any detector works, I have used both optical and current detection for
SA> this purpose.
SA>
SA> ---------------------------
SA> Stan Ames stan...@attbi.com
SA>
SA> DCC Evangelist, DCC Book co-author, married to a DCC manufacturer rep
SA> Home page with Disclosure Statement: http://www.dccsig.org/sra/
SA>
SA>
Robert Heller
In a message on 19 Jun 2003 09:58:24 -0700, wrote :
SW> Robert Heller <hel...@deepsoft.com> wrote in message news:<d4862$3ef192eb$d0c7e1fd$14...@nf1.news-service.com>...
SW> > billsr...@webtv.net (Bill),
SW> > In a message on Thu, 19 Jun 2003 00:57:10 -0400 (EDT), wrote :
SW> >
SW> > B> Greg.P. replied:
SW> > B> That item is designed to reverse the track polarity so that a train can
SW> > B> negotiate a reversing loop without DPDT switches. DCC decoders don't see
SW> > B> polarity, but track joins do.
SW> > B> --------------------------------------------------
SW> > B>
SW> > B> Greg, thanks for pointing this out. My aging brain realized that just
SW> > B> after I posted the reply.
SW> >
SW> > One *downside* of DCC is that it is not really good for 'dumb'
SW> > automation (dumb == fun with simple logic circuits driving relays).
SW> > For a simple store-display type layout, unattented operation, DC is
SW> > still useful.
SW> >
SW> > It is possible to do this with DCC, but would require a *lot* more
SW> > electronics, since you need to form the DCC packets that implement the
SW> > reversing commands for the locomotive. It might be easier for more
SW> > Model RRs to just buy themselves a 'cheap' PC and write a VB (or Bash
SW> > Script, depending on O/S) script.
SW>
SW> ??
SW> Unless I badly misunderstand the way DCC works, sending reversing
SW> commands to the loco isn't the right thing to do. "Forward" and
SW> "backward" are determined strictly by which way the decoder is wired.
Right. A *lot* more electronics to implement the 'smarts'. It needs
to remember which *logical* direction the loco is going, not the
*physical* direction, along with the loco's address and current speed
setting. It is probably complex enough to worth doing with a 'postage
stamp' uProcessor (for those that know how to mess with VLSI chips) or
a PC with a serial-port-to-DCC-booster interface.
SW> I gather there's a few ways of handling loops under DCC, but they all
SW> basically work by detecting the polarity mismatch, and reversing the
SW> track feed -- not too different in principle to how it's done in DC.
SW>
SW> One obvious way, for the simple case of an end loop, is to feed loop
SW> power through the loop turnout. When the train is completely in the
SW> loop, throw the turnout. Loop polarity now matches the main at the
SW> exit point.
Basically this is all that is needed. Track polarity itself does not
matter to DCC, since direction is handled at a more abstract level, so
a reversing loop just needs to have things wired to avoid a short circuit.
The OP wanted a circuit for an *automated* back and forth layout (at
least that is what I assume). I guessing for sort of display (like in a
store window or some such).
SW>
SW> -- Steve
SW>
Dave Curtis
> automation (dumb == fun with simple logic circuits driving relays).
> For a simple store-display type layout, unattented operation, DC is
> still useful.
>
> It is possible to do this with DCC, but would require a *lot* more
> electronics, since you need to form the DCC packets that implement the
> reversing commands for the locomotive
*BZZZT* Thank you for playing.
This is not how reversing sections are done in DCC. The loco direction
is relative to the headlight, always, period. The DCC signal is square
wave AC, it has no polarity, instead it has phase. Two adjacent
blocks must be in phase as conducting wheels pass over the gaps.
There are multiple ways to handle this:
1) Set reversing section phase relative to the points on the turnout.
2) Use an autoreversing DCC booster or DCC power manager/distributer.
3) Use a reversing section controller such as Loys Toys ARSC.
To implement a "dumb" automated loop-to-loop layout with DCC, you only
need to solve the problem of setting turnout direction. This is just
about exactly the same complexity as a DC system.
In a properly wired DCC layout, you *never* have to think about how
the track is wired during normal operation. This, to me, is the great
advantage over DC. I don't give a rat's behind where the power blocks
are when I am operating, any more than I care about what kind of glue
is holding the ballast in place.
If the "complexity" of DCC is scaring you off of trying it, please
don't let it. IMO DC and DCC are, to a first approximation, about
equally complex for most layouts. Its just that the complexities are
different. If you understand how to wire a moderate size layout for
DC, DCC is not going to be hard, just different and a few new things
to learn.
-dave
David Harris
You could make one from a cheap DIY DCC controller -- like
http://www.minidcc.com, for instance. You would have to arrange for the
end of track sensor to 'push' the direction key, but that wouldn't be
too difficult.
David
Gregory Procter
Dave Curtis wrote:
The original question was: "how does one make an automatically reversing
shuttle train in DCC", or words to that effect.
Robert Heller
In a message on Fri, 20 Jun 2003 20:24:46 +1200, wrote :
GP> Dave Curtis wrote:
GP>
GP> > Robert Heller <hel...@deepsoft.com> wrote in message news:<d4862>
GP> > > One *downside* of DCC is that it is not really good for 'dumb'
GP> > > automation (dumb == fun with simple logic circuits driving relays).
GP> > > For a simple store-display type layout, unattented operation, DC is
GP> > > still useful.
GP> > >
GP> > > It is possible to do this with DCC, but would require a *lot* more
GP> > > electronics, since you need to form the DCC packets that implement the
GP> > > reversing commands for the locomotive
GP> >
GP> > *BZZZT* Thank you for playing.
GP> >
GP> > This is not how reversing sections are done in DCC. The loco direction
GP> > is relative to the headlight, always, period. The DCC signal is square
GP> > wave AC, it has no polarity, instead it has phase. Two adjacent
GP> > blocks must be in phase as conducting wheels pass over the gaps.
GP> > There are multiple ways to handle this:
GP> >
GP> > 1) Set reversing section phase relative to the points on the turnout.
GP> > 2) Use an autoreversing DCC booster or DCC power manager/distributer.
GP> > 3) Use a reversing section controller such as Loys Toys ARSC.
GP> >
GP> > To implement a "dumb" automated loop-to-loop layout with DCC, you only
GP> > need to solve the problem of setting turnout direction. This is just
GP> > about exactly the same complexity as a DC system.
GP> >
GP> > In a properly wired DCC layout, you *never* have to think about how
GP> > the track is wired during normal operation. This, to me, is the great
GP> > advantage over DC. I don't give a rat's behind where the power blocks
GP> > are when I am operating, any more than I care about what kind of glue
GP> > is holding the ballast in place.
GP> >
GP> > If the "complexity" of DCC is scaring you off of trying it, please
GP> > don't let it. IMO DC and DCC are, to a first approximation, about
GP> > equally complex for most layouts. Its just that the complexities are
GP> > different. If you understand how to wire a moderate size layout for
GP> > DC, DCC is not going to be hard, just different and a few new things
GP> > to learn.
GP> >
GP> > -dave
GP>
GP> The original question was: "how does one make an automatically reversing
GP> shuttle train in DCC", or words to that effect.
GP>
GP>
Right, NOT a reversing loop (which *I* know is trivial with DCC). The
OP wanted some sort of automated (unattended) back and forth setup,
probably with a single track with bumpers on the ends (*NOT* a reversing
loop).
Mike Tennent
>
>*BZZZT* Thank you for playing.
>
BZZZT, yourself.
Try answering the question the poster actually asked, like some of us
did. He was looking for a unit to do automated, back and forth
trolley action in DCC.
Mike Tennent
>The difference is that the relay controls the direction switch of a
>handheld rather then the polarity of the track. You set the address of
>the locomotive in the handheld and the speed and functions work normally
>but the direction switch in effect reverses as the locomotive trips the
>detector.
>
It would be nice if one of the manufacturers would come up with a
small standalone unit that does the same thing in a dedicated manner.
To simplify things, it could have a pre-set address - something basic
like 02 or 03 and only need to recognize a relay being tripped and/or
a signal from a detector. No keypad or display.
Gregory Procter
Robert Heller wrote:
> Mike Tennent <iron...@darientel.net>,
> In a message on Fri, 20 Jun 2003 08:25:17 -0400, wrote :
>
> MT> Stan Ames <s...@mitre.org> wrote:
> MT>
> MT> >The difference is that the relay controls the direction switch of a
> MT> >handheld rather then the polarity of the track. You set the address of
> MT> >the locomotive in the handheld and the speed and functions work normally
> MT> >but the direction switch in effect reverses as the locomotive trips the
> MT> >detector.
> MT> >
> MT>
> MT> It would be nice if one of the manufacturers would come up with a
> MT> small standalone unit that does the same thing in a dedicated manner.
> MT> To simplify things, it could have a pre-set address - something basic
> MT> like 02 or 03 and only need to recognize a relay being tripped and/or
> MT> a signal from a detector. No keypad or display.
>
> The thing is that there would not really be much of a real market.
> Yes, a handfull of DCC-is-everything people would get it, but for most
> people who would want to set up something like this, for such things as
> a store display or for an *isolated* trolley or mine track with *one*
> train or trolley car that only goes back-and-forth and has no (functional)
> switches and no connection with the rest of the layout, a *simple* DC
> type system is more cost effective and simplier to deal with. The loco
> or trolley would be in 'captive' service and would not need a DCC
> decoder.
>
> The whole *point* of DCC is running multiple trains in a prototypical
> fashion on shared trackage, possibly in opposite directions, with *each*
> *train* controled by its own control stand (eg hand held controller),
> without either changing control stands or messing with block switches.
> This functionally is wasted (unneeded) with a "captive" single-train
> (or trolley car) on isolated, *static* (no switches) trackage. Either a
> plain old circle or oval (possibly twisted into a 'pretzel') or a back
> and forth shuttle route (which may also be twisted into a 'pretzel').
The branch-line Doodlebug could be made to run from the main yard/station to the
"branch" which need be no more than a turnout, tunnel mouth and an extra foot of
track. There is nothing isolated about that situation and very little "operation"
for an extra operator, but it would involve the main throat of the station and
give a busier (more realistic/more interesting) appearance to a layout.
Regards,
Greg.P.
Donald Kinney
> MT> It would be nice if one of the manufacturers would come up with a
> MT> small standalone unit that does the same thing in a dedicated manner.
> MT> To simplify things, it could have a pre-set address - something basic
> MT> like 02 or 03 and only need to recognize a relay being tripped and/or
> MT> a signal from a detector. No keypad or display.
> The thing is that there would not really be much of a real market.
> Yes, a handfull of DCC-is-everything people would get it, but for most
> people who would want to set up something like this, for such things as
> a store display or for an *isolated* trolley or mine track with *one*
> train or trolley car that only goes back-and-forth and has no (functional)
> switches and no connection with the rest of the layout, a *simple* DC
> type system is more cost effective and simplier to deal with. The loco
> or trolley would be in 'captive' service and would not need a DCC
> decoder.
> The whole *point* of DCC is running multiple trains in a prototypical
> fashion on shared trackage, possibly in opposite directions, with *each*
> *train* controled by its own control stand (eg hand held controller),
> without either changing control stands or messing with block switches.
> This functionally is wasted (unneeded) with a "captive" single-train
> (or trolley car) on isolated, *static* (no switches) trackage. Either a
> plain old circle or oval (possibly twisted into a 'pretzel') or a back
> and forth shuttle route (which may also be twisted into a 'pretzel').
Okay, so I got this big loop that has a hidden siding. I want my Amtrak
train to go into the siding and stop. After a little while it comes back
out and then stops at the station on the main line. After the passengers get
off and on, I want it to start up and go back into hiding until the next
Amtrak stop at the station...
Also I want only the Amtrak train to be able to get on the siding. The
turnouts need to switch for Amtrak leaving and arriving at the siding. All
other trains go on the main line never on the siding....
Might as well throw in can have numerous control blocks on the main line and
if there is a train in a block just in front of the Amtrak then the Amtrak
comes to an emergency stop until block is cleared.
But just to keep it simple, the Amtrak train will use the same locomotive at
all times.
I bet a lot of home modelers would use a system that could do this:-)
Any ideas on how to make it work....What can I buy now, and what would I
have to do in order to make the parts work....
Donald
Rich Frieser
website reading the article index for the upcoming August 2003 issue.
In the DCC column they've list "Reversing Sections" as the topic for
August. This may shed some more light on the current discussion. The
issue is due at news stands on July 8.
Digital Railroader
news:cfbd6baf.03062...@posting.google.com...
Actually, their series is still very young, and it probably relates to
wiring reverse loops, wyes and turntables, not the direction reverser for a
point-to-point layout I was asking about.
Mike Tennent
y.
>
>The thing is that there would not really be much of a real market.
>Yes, a handfull of DCC-is-everything people would get it, but for most
>people who would want to set up something like this, for such things as
>a store display or for an *isolated* trolley or mine track with *one*
>train or trolley car that only goes back-and-forth and has no (functional)
>switches and no connection with the rest of the layout, a *simple* DC
>type system is more cost effective and simplier to deal with. The loco
>or trolley would be in 'captive' service and would not need a DCC
>decoder.
>
While I'm not sure there is a big market, I do have to question a
couple of your assumptions.
First, I think that most folks who have DCC prefer to have the entire
layout converted. So, I think there is probably more than a "handfull"
of DCC-everything folks out there. That's why some folks with DCC have
asked me about such a unit.
I agree that if the section is isolated, it's far easier to do it with
DC, but why be limited to that?
The trolley or back-and-forth section may not be isolated from the
rest of the layout. What if you want a switcher with a couple of cars
to go back and forth from point A to B, to add some action and a
little challenge for operating in that area?
I can see where a unit would add some flexibility and interest to a
DCC layout.
Frank A. Rosenbaum
--
From the computer of
Frank A. Rosenbaum
"Mike Tennent" <iron...@darientel.net> wrote in message
news:n4rdfvc03otekihdt...@4ax.com...
Has anyone suggested a buddy throttle with the reversing toggle on top? Take
the toggle out and put in a DPDT relay set up as a reversing switch and
using some type of sensor to activate it.
Gregory Procter
Robert Heller wrote:
> Mike Tennent <iron...@darientel.net>,
> In a message on Mon, 23 Jun 2003 08:19:47 -0400, wrote :
>
> MT> Robert Heller <hel...@deepsoft.com> wrote:
> MT> y.
> MT> >
> MT> >The thing is that there would not really be much of a real market.
> MT> >Yes, a handfull of DCC-is-everything people would get it, but for most
> MT> >people who would want to set up something like this, for such things as
> MT> >a store display or for an *isolated* trolley or mine track with *one*
> MT> >train or trolley car that only goes back-and-forth and has no (functional)
> MT> >switches and no connection with the rest of the layout, a *simple* DC
> MT> >type system is more cost effective and simplier to deal with. The loco
> MT> >or trolley would be in 'captive' service and would not need a DCC
> MT> >decoder.
> MT> >
> MT>
> MT> While I'm not sure there is a big market, I do have to question a
> MT> couple of your assumptions.
> MT>
> MT> First, I think that most folks who have DCC prefer to have the entire
> MT> layout converted. So, I think there is probably more than a "handfull"
> MT> of DCC-everything folks out there. That's why some folks with DCC have
> MT> asked me about such a unit.
> MT>
> MT> I agree that if the section is isolated, it's far easier to do it with
> MT> DC, but why be limited to that?
> MT>
> MT> The trolley or back-and-forth section may not be isolated from the
> MT> rest of the layout. What if you want a switcher with a couple of cars
> MT> to go back and forth from point A to B, to add some action and a
> MT> little challenge for operating in that area?
>
> The problem here is that a *prototypical* such 'regular switch job' would
> have a dispatcher or yard boss in charge of things. An autonomous back
> and forth switch job *mixing* with 'regular' (human controlled) traffic
> would require some smarts if sometimes there is other traffic that
> might interfere with it (or it interferes with, depending). That is,
> the control gadget would need to implement some sort of dispatcher or
> yard boss logic. Unless the control box had some serious AI in it (not
> likely) this means that a non-isolated automated section is a recipe
> for a rail disaster (Murphy's Law!). If one is going to do that, it
> makes more sense to just have a full fledged PC with a serial port =>
> DCC adapter, with all of the hand-held controls connected to the
> *computer* (ala LocoNet or the like) and the computer set up as a
> semi-intelligent dispatcher or yard boss, with some fail-safes coded in
> (the computer would *enforce* signals, etc.). This would require that
> the computer program have a knowledge base for the whole layout and
> everything on it, since if there is connectivity between the 'regular'
> layout and the switch job's route at one point there is connectivity
> between the switch job's route and the whole rest of the layout, since
> there is most likely connectivity between all of the parts of the
> layout at some point or another.
You're getting carried away with complex solutions to a simple situation!
Given that we're running the train from a shuttle bay of the station to a branch
with a turnout off the main line, all that is required to despatch the train is to
have the two turnouts correctly set - a Peco switch or microswitch on each point
motor - turnouts aligned = reverse module operates.
The turnouts are only going to be reset when there are no conflicting moves. It
would be possible to reset the turnouts in the middle of the run, but DCC certainly
allows that with individually controlled trains.
>
>
> For a plain dumb type back-and-forth gadget (two end-of-track sensors,
> a flip-flop, etc.) it really only makes sense for something like a
> store display or a truly isolated section, in which case DC makes more
> sense. DCC for this is just overkill (complexity without purpose) for
> these situations.
You're still determined to make this more complicated than the prototype does!
>
>
> To me, there are three main situations:
>
> 1) Some kind of simple 'store display' set up (this might also include
> museums or any of a number of other automated display layouts).
>
> 2) An automated *isolated* sub-layout within a regular layout, such as a
> trolley line or a narrow gauge logging or mine scene.
>
> 3) Automated *non-isolated* *traffic*.
>
> Case 1 and 2 make more sense as DC systems with a 'dumb' sensor pair
> driving a flip-flop and a DPDT relay. Case 3 [which is what you are
> talking about] seems (to me) require something much more 'intelligent'.
> I don't mean a positronic brain or anything, just something more than a
> two state state-machine -- something with a few more 'if statements' to
> cover some additional cases. Something that has a bunch more block
> occupancy detectors and with some sort of model of where everything is
> and with the ability to send a 'stop' (or 'slow') command to *any* loco,
> not just the automated traffic. I agree that adding 'automated traffic'
> is an interesting and useful idea and DCC is *definitely* the way to
> handle it, but with something more than a 'dumb' black-box. To me it
> simply make better sense to think in terms of automated traffic in this
> situation and not think in terms of a specific section of trackage, even
> if the automated traffic never leaves a certain specific section of
> trackage. If you are going to implement this for one switch job, you
> can re-cycle the 'code' for any regularly scheduled train anywhere on the
> layout -- perfect tool for when club members can't make the session for
> whatever reason.
>
It's easier with analogue block control.
add:
- a delay module. (available over the counter for the last 50 years)
- a braking position detector. (simple with any analogue momentum controller)
- turnout switch linked to momentum controller.
All done!
>
> One of the things *I* plan to work on is a generic train operating
> program that can operating a regularly scheduled train completely under
> control of the computer.
Regards,
Greg.P.
Ron Herfurth
iron...@darientel.net wrote on 6/23/03 8:19 AM:
Could you use 2 such reversers and 3 spring switches to make an automatic
wye? (of course you could do the same thing with a loop, one spring switch
and a polarity reverser but turning a short train or doodlebug on a wye
would be much more interesting)
--
Ron Herfurth
Charlottesville, VA
Gregory Procter
Ron Herfurth wrote:
With analogue control, that would be no real problem.
Joe Ellis
<top...@cstone.net> wrote:
>Could you use 2 such reversers and 3 spring switches to make an automatic
>wye? (of course you could do the same thing with a loop, one spring switch
>and a polarity reverser but turning a short train or doodlebug on a wye
>would be much more interesting)
>
Hmmm... I think you'd need only a single DCC-specific unit (Which, to the
best of my knowledge, doesn't exist yet... Anyone from BitSwitch
listening? Sounds like it's right up your alley...), and 4 detectors (two
on each wye) because the DCC unit would send messages to the decoder
rather than controlling a discrete piece of track. There's no reason that
the detector unit couldn't throw the turnouts as well as change the train
direction...
Another advantage of this is that you could use the decoder
acceleration/deceleration features to add momentum - something usually
lacking in a simple reverser.
Let's make the turn-around really interesting - make it a "twisted" wye,
using a wye turnout to split the main, then having the legs curve out,
then cross back over and go to the "crossbar" of the wye. You get more wye
in less real estate.
--
The Institute reserves the right to act first. Look, historically, governments
and people are reactive. They have to depend on popular consensus of some
sort. That means when a tyrant or a madman or a system looks evil, they have
to wait. It's been the same throughout history. There's this feeling that you
can't use applied or deadly force to head off disaster because you can't
absolutely prove there will be a disaster." - Nathanial Whaler
_The_Ecolitan_Enigma_, L. E. Modesitt Jr. http://home.mindspring.com/~filker/
Mike Tennent
>For a plain dumb type back-and-forth gadget (two end-of-track sensors,
>a flip-flop, etc.) it really only makes sense for something like a
>store display or a truly isolated section, in which case DC makes more
>sense. DCC for this is just overkill (complexity without purpose) for
>these situations.
>
I really can't disagree with you. Case 1 and 2 below are 99% of the
situations.
Case 3 would have to be quite simple to work.
In fact, if I was doing it to make operating interest, I'd be more
inclined to have the "back and forth" train (running in DC) simply
cross the DCC controlled mainline to provide an occasional conflict.
>To me, there are three main situations:
>
>1) Some kind of simple 'store display' set up (this might also include
>museums or any of a number of other automated display layouts).
>
>2) An automated *isolated* sub-layout within a regular layout, such as a
>trolley line or a narrow gauge logging or mine scene.
>
>3) Automated *non-isolated* *traffic*.
>
Mike Tennent
Digital Railroader
> DCC adapter, with all of the hand-held controls connected to the
> *computer* (ala LocoNet or the like) and the computer set up as a
> semi-intelligent dispatcher or yard boss, with some fail-safes coded in
> (the computer would *enforce* signals, etc.).
Well, we do have an extra PC we could use for such a thing, but I've seen
some of the software available from DigiTrax, and it is antiquated. Can you
recommend a manufacturer that offers a full package, with equipment and
decent Windows-based software?
KTØT
news:Vx_Ja.147920$jT4.2...@twister.rdc-kc.rr.com...
: "Robert Heller" <hel...@deepsoft.com> wrote in message
: CTI makes such a product and software. See it at
http://www.cti-electronics.com/
I'm using it on my model; haven't tried the DCC throttles yet but they
should work. My layout has two lines - one is DCC with the staging switches
and junction controlled by CTI and the other is DC with complete CTI
control - throttles and all.
--
73 de KTØT
Bob Schwartz
Modeling Waseca, MN in the 50s
Digital Railroader
> : CTI makes such a product and software. See it at
>
> http://www.cti-electronics.com/
>
> I'm using it on my model; haven't tried the DCC throttles yet but they
> should work. My layout has two lines - one is DCC with the staging
switches
> and junction controlled by CTI and the other is DC with complete CTI
> control - throttles and all.
I'm trying to get away from having to set up a DC layout just to accomplish
this. I could use one of the commercial auto-direction reversers in DC, but
all of my locos are converted to DCC. Too bad - looks like I'm back to
square one...
Gregory Procter
Digital Railroader wrote:
Use Stan Ames idea of allocating a modified hand controller to the specific
shuttle. Any competent electronics hobbyist should be able to modify it.
Regards,
Greg.P.
Charles Kimbrough
If you are using Digitrax you could modify a UT1 Throttle msrp $79.95.
Then you could select the loco address and still not have an arm and a
leg invested.
KTØT
: "KTØT" <kt...@mn.rr.com> wrote in message
Frank, I'm sorry I didn't make myself clear - CTI is supposed to be able to
operate as a DCC throttle via the serial interface to most DCC systems. Once
the system is interfaced you can write a program to do reversing, inertia,
and speed.
You can download the software free at
http://www.cti-electronics.com/download.htm
and check it out. The help files are excellent and cover the DCC operation.
Digital Railroader
news:3EF8ABC5...@ihug.co.nz...
> Use Stan Ames idea of allocating a modified hand controller to the
specific
> shuttle. Any competent electronics hobbyist should be able to modify it.
You're not talking about me! ;-) But seriously, I really need something I
don't have to put a lot of time into. I also don't really like the idea of
modifying a hand controller - I only have one, and it's radio.
--
Digital Railroader
> Frank, I'm sorry I didn't make myself clear - CTI is supposed to be able
to
> operate as a DCC throttle via the serial interface to most DCC systems.
Once
> the system is interfaced you can write a program to do reversing, inertia,
> and speed.
No problemo. I'll check out their website, and thanks for taking the time to
clarify what you said!
--
Gregory Procter
Digital Railroader wrote:
> "Gregory Procter" <pro...@ihug.co.nz> wrote in message
> news:3EF8ABC5...@ihug.co.nz...
> > Use Stan Ames idea of allocating a modified hand controller to the
> specific
> > shuttle. Any competent electronics hobbyist should be able to modify it.
>
> You're not talking about me! ;-)
I worded that fairly carefully to avoid being specific :-)
> But seriously, I really need something I
> don't have to put a lot of time into. I also don't really like the idea of
> modifying a hand controller - I only have one, and it's radio.
However you attempt to achieve some automated control, you're going to have to
purchase or manufacture some control equipment. A low cost basic hand-held is
probably going to be the simplest means. You're not actually going to hold it
while it operates automatically so the radio version would be overkill.
Greg.P.
Mike Tennent
>"Gregory Procter" <pro...@ihug.co.nz> wrote in message
>news:3EF8ABC5...@ihug.co.nz...
>> Use Stan Ames idea of allocating a modified hand controller to the
>specific
>> shuttle. Any competent electronics hobbyist should be able to modify it.
>
>You're not talking about me! ;-) But seriously, I really need something I
>don't have to put a lot of time into. I also don't really like the idea of
>modifying a hand controller - I only have one, and it's radio.
Hey, I do this sort of thing and I'm not inclined to go hacking apart
a hand-held controller or try to rig a Rube Goldburg contraption.
I did a quick search and the cheapest, no-frills, hand held throttles
are in the $50-$60 range. That's a bit pricey to simply play around
with and/or possibly devote to a dedicated operation like a trolley.
Which brings us back to your original inquiry. <g> There might be a
niche for such a product if it could be kept in the $30 range.
Stan Ames
Mike Tennent wrote:
>
>
> Hey, I do this sort of thing and I'm not inclined to go hacking apart
> a hand-held controller or try to rig a Rube Goldburg contraption.
>
> I did a quick search and the cheapest, no-frills, hand held throttles
> are in the $50-$60 range. That's a bit pricey to simply play around
> with and/or possibly devote to a dedicated operation like a trolley.
>
> Which brings us back to your original inquiry. <g> There might be a
> niche for such a product if it could be kept in the $30 range.
Gad
Such a simple approach is non considered a contraption.
Go to ebay and bid on any throttle that has an actualy switch as a
direction switch.
Replace this switch with a spdt center off switch.
Use a relay to that goes between the direction switch and the input to
the handheld. The relay is controlled by a pair of block occupancy
detectors.
It took about $40-50 to build mine. It has been in use for many years.
You of course could also simply use a computer throttle whose direction
switch was controlled in software by an arbitrary (set in advance) set
of block occupancy detectors. Cost in this case would only be the
detectors. (assuming you had the computer, its interface and other
hardware already)
It really is very simple in either case and has been in use by several
for years. For examply Mike Tylick uses it on his O scale display layout
when to have operation when operators are not around.
There are many cases where a one desires a train to go over a set route
back and forth. The computer approach is the generalized approach that I
first saw in use in 1993. The most sophisicated and best use of this
was in the ESU booth in Nuremburg where a series of locomotives were
operated to a set point, displayed their unique sounds and then returned
to a staging yard.
Wether a dedicated throttle or a piece of software, this is a trivial
operation.
---------------------------
Stan Ames stan...@attbi.com
DCC Evangelist, DCC Book co-author, married to a DCC manufacturer rep
Home page with Disclosure Statement: http://www.dccsig.org/sra/
Mike Tennent
>
>Such a simple approach is non considered a contraption.
>
>Go to ebay and bid on any throttle that has an actualy switch as a
>direction switch.
>
>Replace this switch with a spdt center off switch.
>
>Use a relay to that goes between the direction switch and the input to
>the handheld. The relay is controlled by a pair of block occupancy
>detectors.
<sigh>
Stan, it ain't trivial to most folks. Heck, I work with all the parts
you describe and it took me a while to figure out what you were
saying.
But frankly, I'm not sure I'd get past the first step:
1) Go to Ebay and find "any throttle that has an actual switch."
I did. Guess what? None of the (5) DCC throttles that show up appear
to have a manual switch.
I know the Digitrax doesn't. The Lenz listed doesn't look like it
does. Nor the MRC.
I looked at the Easy DCC site and their throttle doesn't look like it
has a manual switch.
I did find the Lenz LH90 on their site. It retails for $99. Ouch.
Which throttle did you use? Is it still available?
Stan Ames
>
> Stan, it ain't trivial to most folks. Heck, I work with all the parts
> you describe and it took me a while to figure out what you were
> saying.
>
> But frankly, I'm not sure I'd get past the first step:
Gee I gave you more credit :)
>
> 1) Go to Ebay and find "any throttle that has an actual switch."
>
> I did. Guess what? None of the (5) DCC throttles that show up appear
> to have a manual switch.
>
> I know the Digitrax doesn't. The Lenz listed doesn't look like it
> does. Nor the MRC.
Digitrax UT1, UT2 also relatively easy and inexpensive using a Zephyr
Jump port.
>
> I looked at the Easy DCC site and their throttle doesn't look like it
> has a manual switch.
I will have to check again but i believe the Easy DCC throttles all work
but take a little more work. The Lenz, NCE I believe use a membrane
which is a pain to do. The Easy DCC use a pushbutton which works fine.
I have never tried it with an MRC.
>
> I did find the Lenz LH90 on their site. It retails for $99. Ouch.
LH90, LH200 both work.
> Which throttle did you use? Is it still available?
I initialy used the circuit board from a Knobby and built a throttle.
It was a pain to wire not for the reversing switch but the handheld
itself has a lot of wires. If I were to do it now I would use a UT1 or
an LH90 (depending on system)
Please note the throttle still works as a throttle. I have a plug on
mine thich allows it to be used as a reversing unit. When the block
occupancy relay is not plugged in the reversing switch works as it
normaly does and thus the throttle works as it normaly does. Thus the
cost of the throttle is really not an issue if you need the throttle
anyways.
Paul Newhouse
Stan Ames <s...@mitre.org> writes:
> The Lenz, NCE I believe use a membrane which is a pain to do.
So if I'm already committed to NCE it's a pain or a computer solution?
Paul
--
Working the Rockie Road of the G&PX
Mike Tennent
>
>
>I initialy used the circuit board from a Knobby and built a throttle.
>It was a pain to wire not for the reversing switch but the handheld
>itself has a lot of wires. If I were to do it now I would use a UT1 or
>an LH90 (depending on system)
>
Stan,
What you're saying is that you made your one-off throttle and are
reasonably certain that it would be trivial to do so with other
throttles. I only disagree with it being trivial.
I think a lot of folks would get nervous opening up a $70 - $90
throttle (voiding any warranty,) hard wiring a switch in place of
another, then figuring out how to control it with a relay off of two
detectors. It sounds like a great article for one of the mags,
though. With step by step instructions, wiring diagrams, etc, a lot of
folks could make one.
Still, I think the discussion tends to support my feelings that a
simple, universal, dedicated unit at a moderate cost could find a
niche. One or two hard coded addresses, programmed to accelerate and
decelerate smoothly to 1/3 throttle (top speed handled by speed table
on decoder,) slows and reverses in response to input signal from
standard block detector.
Given the choice between buying a $30 unit for such a simple operation
or tearing into a $90 unit in a DIY project, I think most folks would
opt for the single unit.
Food for thought for manufacturers...
Regards,
Mike Tennent
"IronPenguin"
Digital Railroader
> simple, universal, dedicated unit at a moderate cost could find a
> niche. One or two hard coded addresses, programmed to accelerate and
> decelerate smoothly to 1/3 throttle (top speed handled by speed table
> on decoder,) slows and reverses in response to input signal from
> standard block detector.
>
> Given the choice between buying a $30 unit for such a simple operation
> or tearing into a $90 unit in a DIY project, I think most folks would
> opt for the single unit.
You've got my vote!
David P Harris
David
Mike Tennent wrote:
> Stan Ames <s...@mitre.org> wrote:
>
>
>>The difference is that the relay controls the direction switch of a
>>handheld rather then the polarity of the track. You set the address of
>>the locomotive in the handheld and the speed and functions work normally
>>but the direction switch in effect reverses as the locomotive trips the
>>detector.
>>
>
>
> It would be nice if one of the manufacturers would come up with a
> small standalone unit that does the same thing in a dedicated manner.
> To simplify things, it could have a pre-set address - something basic
> like 02 or 03 and only need to recognize a relay being tripped and/or
> a signal from a detector. No keypad or display.
Richard P. Kubeck
1. If you are using an isolated section of track.
Do not install the decoder but use the outputs for the motor hooked to a
regular reversing unit.
The decoder is the power pack of a dcc system.
2. Rather then modifying an expensive throttle use two MRC 2k's hooked
up the the standard reversing unit. One unit forward second unit
reverse.You will also gain a second advantage the the units when
switched start again with momentum.
UNCLE
YESTERDAY WAS
TODAY IS
TOMORROW WILL BE
Digital Railroader
news:3EFDA2E4...@telus.net...
> Well,I could build you one of those....
> David
>
> > small standalone unit that does the same thing in a dedicated manner.
> > To simplify things, it could have a pre-set address - something basic
> > like 02 or 03 and only need to recognize a relay being tripped and/or
> > a signal from a detector. No keypad or display.
Are you saying you could build the item Mike refers to above? If so, how
much would it cost? Would you need two of them - one for each end of a
point-to-point layout?
--
Digital Railroader
> 1. If you are using an isolated section of track.
> Do not install the decoder but use the outputs for the motor hooked to a
> regular reversing unit.
> The decoder is the power pack of a dcc system.
Except that in the locomotives I want to use on the point-to-point layout
already have decoders installed, and one of them is a SoundTraxx. Would hate
to be without that...
> 2. Rather then modifying an expensive throttle use two MRC 2k's hooked
> up the the standard reversing unit. One unit forward second unit
> reverse.You will also gain a second advantage the the units when
> switched start again with momentum.
I take it the MRC 2k's are power packs - this could easily cost more than
Mike's solution.
David P Harris
I would modify the MiniDCC project (see http://www.minidcc.com) - cost
would be <$100US, prob <$50. Same project but could delete: keyboard,
lcd. We would probably want some kind of interface to select enegine
number, pause timing etc, so maybe keeping the kb&lcd would be better.
I have one, so I could try out a mod of the software.
David
PS- is there any interest?
PPS- MiniDCC has a new intergrated controller/booster pcb. I have one
too, but it is larger, see: http://groups.yahoo.com/group/OmniPort/.
Digital Railroader
> I would modify the MiniDCC project (see http://www.minidcc.com) - cost
> would be <$100US, prob <$50. Same project but could delete: keyboard,
> lcd. We would probably want some kind of interface to select enegine
> number, pause timing etc, so maybe keeping the kb&lcd would be better.
> I have one, so I could try out a mod of the software.
> David
> PS- is there any interest?
I looked at their website, and frankly, I'm lost. Tell me more about how
this would work, and what the difference between $50 and $100 would get
me...
KTØT
: Except that in the locomotives I want to use on the point-to-point layout
: already have decoders installed, and one of them is a SoundTraxx. Would
hate
: to be without that...
It sounds like I'm pushing CTI doesn't it? I have no $ interest in CTI, I
just use their stuff and like it. Here's an application note for a passing
siding that uses most of the commands you would need for automating a
point-to-point layout, DC or DCC.
http://www.cti-electronics.com/app2.doc
The initial cost of a starter kit is $99 - includes four output relays,
four sensor inputs and software (windows type) plus an IR and a magnetic
sensor. You would have to add a computer and a MS-100 interface (assuming
you are using Digitrax) and, perhaps, another sensor and you should have
everything to control your DCC layout - and then some.
David P Harris
The MiniDCC is a PIC processor with a 16 key keyboard and a 2x16 LCD
display, and two addition push buttons for mode control and emergency
stop. It generates DCC.
There is a separate booster board which uses a single chip to generate
up to 3A DCC signal to drive the track. They have since combined the
two boards, so it is even simpler now.
At the moment the MiniDCC will control up to 4 locomotives (each row of
the keyboard becomes: faster, slower, change direction, and headlight
for one loco); program decoders, and control turnouts through DCC with
programmable routes.
This costs in the range of $50-100 depending on whether you use surplus
keyboard ($5), LCDs ($6), etc, and how much you put into the case, power
supply, etc. Dirt cheap, bare minimum, no box, own power supply = $40.
I was proposing to to modify the modify the MiniDCC so that the two
extra buttons would receive the end of track signals. When one is
received, then the train would be slowed to stop, would wait a certain
time, then reverse to the other end. It should be possible to program
in control over sound too.
Hey, this is nice little project! Main problem is changing the software.
My project (OmniPort) is a generalized controller. It is sufficiently
close to MiniDCC, that I can adapt the MiniDCC software to it and have
the same system The advantage is that I have more i/o, so I could add
features.
David
PS - I should say that the MiniDCC project got me interested in MRR
again, as I could see applying my interests in electronics and computers
with MRR.
Mike Tennent
>Hi-
>I would modify the MiniDCC project (see http://www.minidcc.com) - cost
>would be <$100US, prob <$50. Same project but could delete: keyboard,
>lcd. We would probably want some kind of interface to select enegine
>number, pause timing etc, so maybe keeping the kb&lcd would be better.
I think the main thing would be to keep the cost down, since it's
dedicated to such a simple task. That's why I suggested 2 hard coded
addresses and pre-programmed acceleration/deceleration.
Maybe a Pot to control the length of pause at each end...
If you're serious about this, email me and maybe we can do something
jointly.