The past two weeks has been busy in terms of model railroading, but I don’t have a huge amount of physical layout evidence to show for it. A lot of what I’ve done has been of the nature of experimentation and learning in preparation for the next layout steps. In addition to my
first attempt at ballasting track, here are the activities of the past fortnight:
1) Decoder installations - F59PHI.
I now have the large majority of engines running on the layout. One big task the past two weeks has involved the
Athearn F59PHI locomotive. I bought 10 of these about 3-4 years ago for the commuter fleet, and alas aren’t designed for a PNP (plug and play) decoder. Back in 2007 I installed
DZ143 decoders in two of them using a
procedure from the Digitrax web site. This worked, but not perfectly, and I fried one decoder in the learning process. I resurrected this project a couple weeks ago after my son, while cleaning the wheels of one of them, fried its decoder as well. This was not my son’s fault – my faulty installation left a motor and a power wire too close together, causing the short.
Fortunately, the benefit of all these mistakes is that eventually you will develop better skills in spite of yourself. I now have 5 of the locomotives running with decoders, and I’ve improved the installation process. When I do the next install I’ll take pictures and post the process.
Alas, Athearn just released an
updated, DCC ready version of this locomotive, so this process is already obsolete. This reinforces a lesson I’ve learned earlier – which is to delay buying a non-optimal version of a locomotive you want because an optimal version will probably be released in a few years. It also made me resolve to not buy any more locomotives without PNP decoder capability (but I still have 6 non-PNP locomotives in the queue waiting for decoder installs, plus the remaining 5 F59PHIs, so I’ll be doing this for a while longer).
Another learning point is to test/use decoders within a few months of purchase.
TCS has had a “no goof” warranty for several years now, which means that within a year they will replace the decoder for any reason, even if you clearly screwed up the installation yourself.
Digitrax now offers the same warranty for all their components. This is a really nice policy.
2) Decoder installations – F59PHI HO scale.
I
mentioned earlier that I was trying to get my son’s
HO Athearn F59PHI working with DCC. This locomotive does have PNP capability, but it didn’t work. After lots of email with the Athearn support people – who were very helpful and responsive – and getting two replacement boards from them I discovered that the problem was a short between the bottom of the motor and the metal frame. Of course, with DCC installs the first rule is “isolate the motor”, but a) this should already be done in a PNP locomotive, and b) my experience is that motor/frame shorts fry the decoder, but in this case all that happened was the decoder didn’t send power to the motor.
Even the Athearn people didn’t deduce this problem from the symptoms. Only after the second replacement board failed in the same way did I take the whole locomotive apart and find the issue. Once found the fix was easy -- some
Kapton tape between the frame and motor. Great learning experience. I could have bought a replacement locomotive for a fraction of the cost of the time I worked on this problem, even if you price my hours at minimum wage, but now I have a much, much better understanding of the guts of these things.
3) Motor problems – Kato F3.
Although the F3/F7 locomotive is way too obsolete for the modern time period I model, I have a bunch of them for my museum trains. 3 for the California Zephyr, 3 for the Empire Builder, 4 for the Super Chief, and 1 for an SP commuter train. Like my other Kato engines they all have worked flawlessly except one.
Turns out the motor would not turn unless the armature (the part that turns around) is nudged. I tried many things, consulted with friends who are experts in the area, and asked on-line groups. Learned a ton in the process and was told that the problem is probably a broken wire – something almost impossible to repair. Because a replacement motor is only $26 from
KatoUSA I eventually decided to take it completely apart. Learned even more, although I haven’t yet located the problem. I am showing the motor to a friend this week who probably will find the issue right away.
A learning point here is, similar to point (1) above, fully test locomotives shortly after purchase. Because I bought this one in 2006 the warranty has expired, although this had to be a manufacturer defect as the F3 in question had been unopened until I started the decoder install last month.
4) Switch management – control.
Thinking beyond my first ballast experiment I figured that before I ballast around any switches I will need to have them fully configured. Currently I have installed each switch with the manual spring still in place. There is a drill hole underneath to accommodate a switch machine wire, but nothing else.
This is a fairly involved task, and the first part was to figure out the control scheme. I have a
DS64 which I
mentioned last year, but although I purchased it and read the docs I never tried it out. Well, I set it up in a temporary configuration to control the two installed
Tortoises. This allowed me to throw the switch using the Digitrax throttle. I then added a push button to an input circuit, and confirmed that it will be easy to also allow the throwing of switches from a button on the fascia panel. I also worked out the wiring and power approach for the DS64, and sent out a message to a Digitrax discussion group to confirm that this will work. Everyone who responded said I was on the right track, and this in turn led to a few very useful email discussions on related topics.
I also have started the process of designing and building the fascia switch panels, including buying materials to test out the first panel. My intent is to have a single push button to toggle between “closed” and “thrown” positions on the switches, and to have lights indicating the current position of the switch. Long term, one thing I want to do is support “night” operations, which means “lights out” except for layout lights, and that means the switch boards will need illumination.
5) Switch management – Tortoises.
Worked on resolving two outstanding questions regarding the Tortoise switch machines. The first involved the crossover question. Briefly, in the case of a crossover between schematically parallel tracks the two switches are logically linked, in that it never makes sense for one to be closed (i.e. switched so that trains go in the straight, or “normal”, direction) and the other thrown (i.e. switched for the “diverging” direction). Either they both are closed or they both are thrown. This means that in theory you can get by with one switch control for both switch machine, or even one machine controlling two switches.
So, starting two years ago I struggled with trying to get one machine working with two switches. I actually got it working on the layout earlier this year, but it was not very "tuned". After working on a tuning strategy for a while I decided to see just how much money this would save me.
A pack of 12 Tortoises has a street price of $160 plus shipping. I have 19 crossovers, so the potential savings is 19 motors, or about $260. However, there are parts required for modifying a Tortoise to control two switches. My lowest estimate was a parts cost of about $6 per crossover, which is probably optimistic. So the net savings is around $140, or likely less. Given the extra hours required, PLUS the likely additional reliability issues, I finally dumped this idea. I will now buy one Tortoise per switch.
So, having decided that I now need to figure out how to tune each Tortoise to work best with my N scale switches. I think I have an answer, and I'll cover that in a future post when I've tested it out.
6) General Electronics
In many of these topics I've found that I'm getting blocked by holes in my knowledge of electronics. I bought a
book on the topic and am studying closely. So far it has helped in several cases.
