A few more thoughts on being able to read back from decoders:

Being able to read back CV values is not necessary for running your trains.  If your record keeping is good, it is not necessary for programming your trains.  It is helpful but not absolutely necessary for trouble shooting your trains.  But it is absolutely essential for trouble shooting other people's trains.  The latter is based on the observation that people who need help trouble shooting often have no record of the programming changes they have made to their decoders and in some cases do not even know that they have made them.

DCC command stations with read back facilities built in are not the only way to read back CV values.  A faster and better way is to use a computer connected to your DCC system to do the job.  The cost of the hardware that connects the computer to the DCC system is often less than the difference in price between a DCC system that can read back and a similar system that cannot.  The required programs are available and in some cases are free.  The down side of this approach is (a) not all DCC systems can be interfaced to a computer and (b) there is a bunch of learning involved, usually at a time when the owner is also busy learning to operate the rest of his DCC system.  These systems are often plug-and-play only in the bad sense - you plug it in and then play around with it until you get it working.

Jim

Off hand, I see two return loops but one of them has multiple entrances.  One loop is around the roundhouse.  The second loop starts and ends at the uppermost turnout in your diagram.  Both tracks leaving the left side of that turnout need both rails isolated.  Then there is a third connection to that loop at the turnout in the lower left corner of your diagram.  This will require insulating both rails of the diverging route of the turnout.

I am assuming that the right hand route from the upper turnout is an elevated track because if it were not, you would need one very fancy turnout with a crossing built right through the middle of it.  While that is not impossible, it would involve at least five frogs and be a nightmare to wire.

As far as finding reverse loops goes, there is no need to keep track of the individual rails.  Just the fact that a train can end up going back over the same track but in the opposite direction also defines a reverse loop.  I too have problems remembering where I started from so I usually keep one index finger on the starting point while using the other index finger to follow the track.

Jim

added: sorry I missed the update on your original message.

A couple of thoughts:

Be sure the decoder is in 28/128 speed step mode.  Flashing lights that occur randomly are almost always erratic pickup.  Flashing lights that occur as you turn the throttle up or down are usually a speed step mismatch.

Be sure the decoder is set for DCC only.  If it is reverting to Motorola Trinary on loss of signal, it may be exaggerating the effects of minor power losses due to dust on the rails.

There has been no mention of what type of locomotive you installed this decoder in.  I assume you did a stall current test before the installation and found the stall current to be less than .7 amps.  If the stall current is close to or even exceeds .7 amps, you may be seeing the effects of different voltages on the rails.  I could find no specifications in the manual for your Roco Minimaus and am left wondering whether it regulates track voltage or not.  I suspect it does not.  Depending on what you use for input power, its track output voltage could be toward the lower end of the acceptable range for DCC.  The E-Z Command is known for being toward the high end of the range - some people moving up to a more comprehensive system complain that their trains no longer run like slot cars.  What I am thinking here is that the decoder may have marginal cooling and/or current limits for the particular application.  With the higher voltage from the E-Z Command, the decoder may be intermittently going into thermal or over current shutdown until the speed and current draw get high enough to reduce the track voltage to a more acceptable range.

If the locomotive in question was manufactured by Bachmann, make sure that the RFI capacitors have been clipped or removed.  With your Kuehn NO25 decoder, you can alternately set the motor drive frequency to 120 Hz instead of 16 kHz.  Bachmann locomotives with the capacitors still in place can have problems with high frequency "silent drive" decoders and when they do, the problems can be erratic, appearing and disappearing depending on the DCC system, the configuration  of the layout, and how much sugar your grandmother uses in her tea.

If the problems continue, take a peek under your friend's layout and see how he has done his track wiring.  Some configurations are more susceptible to "ringing" than others.  Running a DCC bus in a closed loop or eliminating bus wiring completely can be particularly troublesome once the size of a layout exceeds a couple of sheets of plywood.  One test you might consider doing is connecting his E-Z Command to your layout and seeing how it runs his locomotive there.  You might even want to make the testing more comparable to earlier tests by using the same number of his locomotives on your tracks during the test.

Bottom line, there are many variables involved.  If you can change just one thing at a time, you will have a much clearer understanding of what might be causing the problem.  And if you would be kind enough to let us know what you find, we will have a better understanding of this curious problem as well.

Jim    

Reading about your decoder's delayed start up concerns me.  Short of having the sound initially muted, then turning off the mute after 10 minutes, this should not happen.  The only other thing that I can think of that could cause that effect would be an electrolytic capacitor connected backwards.  When that happens, the capacitor leaks (electrically speaking) rather badly and heats up.  It may in time reform its insulating working layer, which could account for the sound system starting up after a delay, but the leaking and heating continues.  Left in this state, capacitors have a bad habit of exploding and spewing shredded aluminum foil in all directions.  Not exactly what you want thrown around inside your locomotive.

You might want to talk to the folks at Bachmann about this.  They may suggest you keep on running the locomotive as it is running okay now, but at least you will be on record if something untoward should happen.

Jim Banner
Electronics Technologist 

Donneldon brings up a valid point but there is an easy solution - used eased curves.  Eased curves start off at a wider radius then tighten up to the minimum radius.  Coming out of the curve, the radius again wides before going straight.  You can approximate this  by replacing the first and last pieces of 15" radius track in the  semicircular ends of your oval with pieces of 18" radius.  You know your rolling stock can go from straight to 18" radius.  Going from 18" radius to 15" radius should be equally easy for it.  Your table width will accommodate the slight extra width very easily.  I would suggest buying just one package (4 pieces) of 15" radius E-Z Track and testing this out before investing heavily in all the 15" radius track you need.

Jim

Point to point layouts are not for everybody.  Many of us enjoy showing our handiwork to friends, neighbours and relatives as well as running our trains back and forth.  But we usually do not enjoy both at the same time.  Visitors like to see trains running and they enjoy talking about them.  They are not so interested in looking silently at your back as you do some fancy switching maneuvers.  Thus many of us include loops, hidden or otherwise, in our track plans so that both we and our visitors can enjoy the best of both worlds.

Unless you are modelling in Z-scale, I doubt you can achieve continuous running on a 15" maximum width, U-shaped layout.  In H0 scale, it takes at least 3-1/2 feet of table width for a return loop or semicircle of track.  This is already a foot too wide to comfortably work on if the table is against a wall.  Increasing track radius to 22, 24 or 30 inch radius increases table width to 4, 4-1/2 or 5-1/2 feet, all of which require access from both sides or well placed access hatches, both of which may be beyond the space, skills or interests of the modeller.

Not everybody who starts a model railroad envisions a basement or garage full of trains.  More likely they are thinking of a smaller layout that fits comfortably in a corner of the family room or on the bottom level of Juniors bunk bed.  Something around 3-1/2 feet by 5 or 6 feet.  Often it is a major step to expand their idea to 4 by 8 feet even though that greatly increases their layout's potential.  If they do convince their spouses, parents or themselves that the increased size is justified, they are not about to reduce that potential by using larger radius curves - besides they already have the 18" radius curves that came with their set.  It is only later when they want to add more locomotives and rolling stock that they discover that not all of them will run on their choice of curves.  So there are many questions of the form "how small a curve can locomotive X and rolling stock Y run on?"  Even though we may see these questions as repetitious, they are still valid questions from people who need to know, and each and every one of them deserves an answer.

So now you know why.

Jim

Quote from: Jerrys HO on January 19, 2012, 03:59:42 PM
Most Bachmann DCC trains have dual decoders that will let them run both way's. If you are not certain or have any other type decoder I would hesitate running them on DC power as you could fry the decoder. If you are certain you have a dual mode decoder than yes it would be safe.
Jerry

Just to correct a couple of misconceptions, all Bachmann locomotives sold with factory installed non-sound decoders have dual mode decoders which are programmed to allow dc operation by default.  However, they will not operate on dc if someone has reprogrammed the locomotive to run only on DCC or if the dc is badly distorted by certain pulse type dc power packs.

You will NOT harm your locomotive by trying to run it on dc.  Even the decoders that are not dual mode will not be damaged by dc as per NMRA requirements.  This does not mean you should try to run your locomotive on excessively high dc voltages.  Voltages over 30 volts can burn out a decoder whether they are dc, DCC or any other waveform you can think of.  Then again, such voltages can burn out a motor even if the locomotive has no decoder at all.  Bottom line, no matter what decoder it has, you cannot damage your locomotive by trying to run it with a dc powerpack.

Jim

I have used the "conventional" construction Roger speaks of and find it extremely versatile for scenery both below and above track level.  My preference is to use 1" chicken wire tacked/stapled to the sides of roadbed and to its own risers, then cover it with a couple of layers of paper towels, or better, used Bounce sheets dipped in plaster of Paris.  The chicken wire makes it easy to visualize the general shape of the final scenery, rather like a net diagram.  Once the plaster of Paris has set, you have a strong base for the final layer of plaster.  This final layer can be plaster of Paris or Drywall Joint Cement, depending on how you are detailing it.  The only down side of this "Hard Shell" method of making scenery is its weight.  While lots of weight is unimportant  for a permanent layout, it is the last thing you want in a portable one.

Most portable and transportable layouts are limited to flat scenes or minor grades because of volume consideration during transport and storage.  So being able to cut only an inch or two below track level or build up only a few inches above track level is not a problem.  So foam is a great way to go for portables.  But if your dream is a home layout with high mountains and deep valleys, then conventional construction is a better way to go.  As far as strength goes, I find a 1 x 4 fame with 1 x 3 "joists" every 12 inches gives lots of strength for tables up to about 4 feet wide.  With the joists only a foot apart, no plywood is necessary under foam 1-1/2" or thicker, even in large scale.  The exception would be under deep lakes etc. that almost penetrate the foam.  Then pieces of plywood or scrap foam can be glue under the lake to provide extra support.

Access hatches are a snap with foam.  They can be easily cut with a knife, even after the scenery is down if you use soft scenery.  Just make sure you keep the knife angled a bit toward the center of the hatch when cutting down through the foam.  That way, the top of the hatch will be slightly larger than the bottom and will sit in the hole like a cork in a bottle - no extra support is needed.

Keep in mind that the foam we are talking about is Styrofoam or equivalent, usually coloured pink or blue.  White "bead board" that is made of Styrofoam balls pressed together does not have the necessary strength unless you use plywood or similar material to support it.  This added cost usually more than offsets any price savings you might have by buying bead board.  For gluing Styrofoam, I like LePage's PL Premium as it sticks to most model railroading materials including foam, wood and stone.  http://www.lepageproducts.com/ProductDetail.aspx?pid=84   I recently tried PL Premium Advanced but cannot recommend it unless you happen to own a high pressure air operated caulking gun.

Jim     

The one thing you need to do is read the manual.  The Dynamis requires quite a bit more learning than the E-Z Command.

Jim

Quote from: NarrowMinded on January 18, 2012, 12:03:15 AM
Careful calling Jim Mr. Banner he considers us all friends and his friends call him Jim.

Right Mr. Banner? eerr I mean Jim.

Thanks NM aka D.  It's definitely Jim.  When model railroaders get together, we are all model railroaders in spite of age, status, title, experience or you name it.  In Saskatoon, there are many modelers and railroad buffs who know one another through our local museum and we all call one another by our first names.  Even the three year olds who bring along their latest Thomas the Tank to show us, or the hundred and three year olds who stop by to tell us about railroading in the days of steam.  Some parents are aghast at their young kids using our first names, especially the first names of those modelers who are Dr. So-and-so or Bishop Such-and-such in more formal circles.  But we feel anybody at any age who is interested in trains deserves to be heard.  The result is often an outpouring of enthusiasm and knowledge from those we could easily dismiss as too young to have it or too old to have retained it.  I think the most positive feedback we get on the use of first names is when people who we first met as young tots come to the museum as adults to introduce their own children to us by their and our first names.

Jim     

Maximum radius on a 36" wide board would be about 17-1/2" radius, give or take a little depending on scale.  For odd sizes like this, you would have to use flex track or hand lay your rails.

If you can frame those pieces of mdf with 3/4" boards then you can increase that width to 37-1/2 inches.  With your 5/8 mdf, you will need a bit of extra support to keep it flat.  The usual material for this would be "1 by 4's" which are actually 3/4" by 3-1/2".  Nailing them to the edges of the mdf will give just as much support as putting them under it and will bring it up to a better width.  Thirty-seven and one half inches will allow you to use common 18" radius curves in H0 or something similar in other scales.

Jim

It is indeed an effective method of chlorinating the shallow end of the gene pool but can be pretty hard on the train crew.  One engineer I met was off work for 3 months and off trains for an additional three months after hitting a car and killing the driver.  She tried to beat the train but didn't quite make it.  If she had, she might have saved a minute or two.  What she lost was the rest of her life.  If you figure she might have lived another 40 years, she was risking 2 million minutes to save 2 minutes.  What a huge risk for such a small return.  What a loss to her husband and kids.  And all for two lousy minutes.

Jim

I live about 1/2 a kilometer from where a major city road crosses the CN's north-south main.  Not surprisingly, there are often idiots who manage to stop on the tracks while waiting for traffic lights a block away.  Most of the engineers who regularly work that line will blow a series of short toots when they have to stop for such idiots.  But there is one who blows his horn only once, holding the button down from the time he has to stop until the idiot clears the track.  Needless to say. other drivers are looking at and pointing at the idiot who learns by embarrassment not to stop of the tracks.  Even the guys looking and pointing are reminded to avoid stopping on the tracks.  Some people are annoyed by the noise.  But I figure that he saves just one life over the span of his career, the noise is worth it.

Jim 

Sorry for the delay in getting back to you.  DCC uses square waves, not sine waves on the rails.  If the square wave (one positive peak plus one negative peak = one wave) is short, it is detected as a binary one.  If it is long, it is detected as a binary zero.  The command station uses these ones and zeros to write messages to the decoders, much the way we write messages to one another by email using strings of ones and zeros.

Normally, the positive peaks are the same length as the negative peaks so the average dc voltage on the rails is zero.  But some DCC sets, including the E-Z Command but not the Dynamis, can make the positive peaks longer than the negative peaks (or vise versa) for the zero waves.  This is referred to as 'zero stretching' and is used to create an average dc voltage on the rails.  This dc voltage is what can operate a dc locomotive without a decoder on DCC tracks.

Rather than wade through 6000 words on the subject, go to the link below and have a look at half a dozen pictures of DCC waveforms.  These are photographs of an oscilloscope screen where a light beam moves across the screen from left to right while the instantaneous voltage moves the beam up or down depending on polarity, and at a vertical distance dependent on voltage.

http://members.shaw.ca/sask.rail/dcc/DCC-waveforms/DCC_waveforms.html

The waveforms a decoder uses to drive a locomotive motor can be found at the link Jerry gave above.  Note the difference between those waveforms and the ones discussed here.  The ones discussed here are either shorter OR longer than a certain length to make the digital ONES or ZEROS of a digital message.  The waveforms at Jerry's link can be any width and their average voltage is applied to the motor as an analogue signal.

If you are really interested in the specifics (and are a bit of a masochist to boot) follow this link to the NMRA website.

http://www.nmra.com/standards/DCC/index.html

Jim

If you really want smoke after all that has been said, it is possible to add a smoke generator to most models.  Often it involves cutting away metal from the locomotives' frames which reduces the weight of our little narrow gauge locomotives and reduces the number of cars that they can pull.  On a flat layout, this may not matter.  But if you like hills and tunnels and grades, you may not be happy with this reduction.

Seuthe is one well known manufacturer of smoke units.  Bachmann also manufactures smoke units.  Whoever makes them, you will usually find that you have to run the locomotive at close to full speed to get the smoke unit working on dc.  Running on DCC creates the opposite problem - the smoke unit will still be working at full blast even when the locomotive is stopped.  But there are solutions.

With dc, you can keep the speed down but still use high enough track voltage to work the smoke unit by adding resistors or pairs of back-to-back diodes in series with the motors.  With DCC, you can reduce the voltage to the smoke units by installing voltage regulators as is done in large scale or by adding resistors or back-to-back diode pairs in series with the smoke generator.  A more elegant solution (suggested by Mark Gurries) is to run the smoke generator off the motor output of its own decoder.  By setting the smoke generator's decoder to the same address as the motor's decoder, you can then use its speed table to control smoke output at different throttle settings.

I hope you don't mind, but after you have your smoke generators all set up and working, please do not invite me over to see it.  I like to see the smoke but am one of those who have problems breathing it.

Jim