Labor Day Weekend Roundup

We just had a long weekend for Labor Day. Here’s what I squeezed in around family stuff.

Atlas 17,360 gallon Tank cars

Sat urday started out well with a couple of hours working on rolling stock. I like to replace the trucks on Atlas cars with MicroTrains trucks. Atlas N Scale couplers are less reliable than the MT couplers so they have to go. Quite frankly, I would be perfectly happy to buy Atlas cars without any couplers or trucks, especially for a few bucks less. I wish!!

I mainly go for truck mounted couplers. Body mount couplers are good when there’s be a lot of backing up with long cuts of cars. Since I don’t expect to be doing a whole lot of that with my cars, truck-mounted couplers are good. In fact, they provide two big advantages: 1) the coupler is always centered with respect to the track and that helps when coupling on curves, and, 2) there’s no messing around with getting the coupler height right.

My Richmond Belt Line plan calls for a 16 inch min radius and I’m interested to see how well body mounted couplers fair on those curves. For the most part, my locomotives have body mounted couplers. I’ll have to get round to testing couples on curves before committing.

Here is one of the GATX 17,360 gal cars with new trucks and low profile wheels. I’m not interested in the older MT “pizza cutter wheels. They look aweful and will have difficulty running on code 40 track.

This is the other one. Unfortunatley both have the same number.

If you look closely at the wheels closest to the couplers where the two tank cars couple together, you’ll see he Fox River Valley metal wheelsets I installed. My thought is to have at least one metal wheelset on each car, eventually with a resistor, to facilitate detection of cars and locomotives. We’ll see how things go with just the metal wheels in place. Two observations right off the bat. First, the metal wheels look good but are not as uniform as the MT wheels when it comes to being perfectly round so you need to be a little selective. Second, the noise is really noticable, at least at the benck. In a room full of operators it may not matter.

Atlas Santa Fe Extended Vision Caboose

This is a nice model that nicely replicates the Richmond Pacific extended vision caboose.

It comes with body mounted Atlas couplers. I removed them and ground away the mounting elements with a Dremmel tool hoping to make enough room to instal truck mounted MT wheelsets with couplers. Unfortunately, the coupler box on the MT trucks sits too high and clashes with underside of the end platforms. I’ll have to look to see if they have a lower one. In any case, I installed a couple of low profile body mounted MT couplers.

 I also wanted this car to be a lot heavier that other cars to provide some stretch when it rides at the back of a train and to reduce the tendency of lighter cars to bounce forward and aft a little on their coupler springs. I installed metal wheels on all four positions and added some lead strip where I could on the underframe.

The lead looks a little untidy here but is very hard to see from the side when the car is on the track and after a few passes with the airbrush, it will not be visible at all.

You can also see in the photo above, the MT kingpins holding the trucks in place. These fit the MT trucks better, the Atlas kingpins fit the frame nicely but allowed the MT trucks to move about 1/2 a mm side to side. This translates to a car body that wobbles which is not what I want.

Oh, one last item on the caboose. The stack sits a little higher than the cupola roof. This means when you turn the car upside down and set it on its roof to put in extra weight, the stack breaks off - three times actually. CA glue (super glue)seems to do the trick when repairing it though.

Operating Session on Don Marenzi’s Copper Pass and Western

I joined Jim Radke at an operating session on Saturday night at Don Marenzi’s layout. Great fun as usual. Don models a proto-freelance railroad based on the Alaska Railroad. Jim and I ran the North Main Local (NML) which had plenty to do and kept us busy for a couple of hours. As rookie conductor, I managed to collect a few extra cars along the way that weren’t supposed to be in our train but redeemed myself part way through the session by fixing a session ending problem with the DCC feed to the layout. We set out the extra cars once noticed. It will add to the fun of the next session!!

Don is a great host and helped us all have a good time. Thanks Don, for a great evening.

Local Railfanning

The long weekend usually dries up freight traffic in this area. After the usual Friday night departures, there’s some activity on Saturday, then Sunday and the holiday Monday are pretty dead, except for Amtrak which is not my main interest. On Saturday as Jim and I were heading out to Don’s we crossed over the MOARV at Buchannan and paced it a little up I-580. We pulled over for Jim to take a great afternoon light shot of the UP power leaning into the superelevated curves north of Buchannan only to find that Jim’s camera was still set up for delayed shooting! We had places to go and the MOARV would be spending at least the next 40 minutes or so, collecting the Richmond Pacific setouts at Central Ave. so no shots of that one.

After the operating session, Jim and I caught a UP autorack train at South Oakland. We took a slight detour to watch the action on the Embarcadero at Jack London Square. It was 12:30am by this stage. Jim tried a few shots to see if he could get one to work. I snapped this just to see how it would turn out.

At least it shows we were there!! We continued up to Albany where the same train had kept pace with us. We watched it again at Albany and called it a night. An appropriate way to end a decent day of roalroading!!.

Operating the M&N and DCC gremlins

I was on a roll so on Sunday, I cleared off the kitchen table and out came the Michieburg and Natalieville Railroad. I pulled out my Atlas Santa Fe blue and yellow B40-8 and set it to doing laps. Lo and behold, the unit was stuttering along with flickering headlights so I cleaned the track. I have noticed that the inside of the railhead can get gunky so I cleaned that as well. The unit was still running erratically and at times, stopping altogether. In the past, this unit has been very reliable so this was unexpected. I cleaned the unit’s wheels and cleaned off the phosphor bronze wiper that rubs on a pin on the sides of each truck. This helped a lot. I am also suspicious of the needle point contacts on the inside of the trucks but I wasn’t feeling like that kind of fussiness and the unit seemed to be improving so I left the disassembly of the trucks for another time.

However, despite the cleaning, the unit still stopped here and there. What I found was that both throttles on the DT400 Digitrax throttle hat the 7412 assigned to it. One from the last session and one from this session. I set the throttle I wasn’t using to 00 and the problem went away!! So once in a while, the throttle with the assignment from the last session must have been getting a 0 speed packet through to the locomotive decoder. Note to self: When finishing with a locomotive, assign the throttle to address 00.

Emboldened by my apparent high grasp of DCC, I fished out a project I started a year ago to put a Tsunami sound decoder under the layout. I’d gotten so far as half wiring the thing into an enclosure before putting it away till I had more time to work on it. I had been putting it off because it involved several aspects of DCC that I rarely use. And in the past, that has meant scouring the Digitrax manuals to make sense of the overly complicated methods and interfaces that pass as today’s ‘plug and play’ DCC.

Warning: DCC Diatribe…

[I think DCC is really great, but the implementation sucks!!! Documentation is pretty bad too. Even with the benefit of an electrical engineering degree and a computer science background it is not always clear to me what is going on going by the written descriptions. My Digitrax Super Chief manual for example, lists four possible programming modes, but only explains two of them. Why is that I wonder? The whole notion of CV’s is about the best way I can think of to make the system largely impenetrable, throw in hex numbers, controllers with more buttons than my worst TV remote control, tiny screens with too much information crammed in, and you have the perfect techno-storm to keep out all but the brave and the feeble-minded. Having programmed in assembly language early in my career, I am not phased by programming CV’s, but what must it be like for most everyone else who didn’t? And besides, the whole system is so non-intuitive that even for those who do understand how it works, it is all very unmemorable stuff. So it’s off to the manual for every little thing. It frequently seems to me that based on the DCC implementation and the lousy documentation that goes along with it that the designers and vendors of DCC didn’t really want it to catch on! Roll on DCC 2.0]

Okay, I’m back now!!

So feeling brave, I picked up the manual and after a short period that seemed like a lot longer than it actually was, I decided to go out to buy some electrical tools that I needed to finish the wiring job. At least that felt like progress! I need to be in a less feeble state of mind when dealing with DCC.

Monday - Sound decoder, 2nd try

After a good night’s sleep, I took on the Tsunami sound decoder. One smart move was to use aligator clip wires to temporarily connect everything up. I selected address 03 and as I advanced the throttle, the decoder sprang to life. There coming out of the littel speakers I had set up were the unmistakable sounds of an EMD 567 engine notching up!! Way, way cool. Hit button 1, a beautiful bell started ringing out, hit button 2 and the Nathan horn wafts through the room.

I then spent the next few hours making sense of how Digitrax handles, service mode addressing, programming on a programming track, reading back variables, and consisting. Along the way the sound decoder stopped working completely twice, even though the CV’s were reading back exactly as they should, the thing refused to make any sound at all.

In the end, and after a lot of looking through the 3 Tsunami manuals and the instruction booklet that all came with the decoder, I finally found the reset method. Often a DCC decoder is reset by writing to CV 8, the Tsunami DSD is reset by writing value of 02 to CV 30. CV 30 is called Error Information - as you might expect! The reset reference is included as an off-hand item in the section about EEROM errors! There is no index in the manuals and no reference to resetting the decoder to factory defaults. Fun, fun, fun!!!

Okay, so what did I learn? Here are the gems of my Labor Day labor…

1. You can’t rely on the Digitrax DT400 throttle display to know what to do during consisting when using the system infrequently, have the manual there and follow it to make sure
2. If you want to consist the sound decoder with a switcher, the switcher has to be added to the decoder with the decoder as the ‘top’ address. Otherwise, the function buttons don’t activate the bell, horn etc. on the decoder
3. You can ‘Dispatch’ the consist containing the sound decoder like any other consist
4. You can’t chage the Tsunami address using ops mode programming on the main, you must use the programming track and Pg mode on the controller. The manual should say you can’t do it successfully. If you attempt to program the address using ops mode, the decoder stops responding to commands. It’s not that you can’t do it, it’s that you shouldn’t do it!
5. When using the programming track, the decoder will not make any engine sounds. However, when varibles are written or read back, the decoder makes a chirping sound at the speakers. The throttle also indicates if the write or read was good or bad.
6. Reset the Tsunami on the programming track by writing CV 30 with value 02
7. Setting CV 29 on the Tsunami to enable the use of speed tables (which would facilitate matching the throttle up of the sound decoder with the speed of the locomotive) doesn’t appear to work as described. Setting CV 29, to 50 according to the manual, caused the decoder to not respond. Has extended addressing been messed up? It doesn’t appear to be by reading back variables in CV 17, 18 and 29. More messing round needed to figure this one out.
8. Volume controls on the decoder CV 50 do not appear to be working as described. A value of 0 did not mute the sound, values between 1 and 15 did not change the volume. More messing round neededto figure this one out too.
9. Leaving the EMD idle sound on with family members in the same room reduces the length of acceptable “train time” for Coxy by a factor of 10, your mileage may vary

Fortunately, I was able to get the sound decoder working again, twice. Next time i have the layout out, I’ll dig a little deeper into the programming as I would like to have the sound match the speed of the locomotive. Bottom line is you have to press ahead even when it involves DCC programming. Despite all the unnecessary difficulty,  the end results of DCC are in fact worth it. But oh, what a wasted opportunity - what DCC could have been.

All in all, a great collection of railroading for the weekend and tomorrow night is my railroading night out!!! Can’t wait!

Coxy

Comments

[Denny]

Hey Coxy,
looks like you got a lot done in your holidays weekend. I use mainly body mounted MT couplers (except where they don't just fit, like in tank or cover hoppers cars where I use MT truck mounted couplers) and they run excellently on my sharpest curve (15").

It looks like on cars with body mounted couplers the pulling force is distributed on the frame, rather than on trucks.

As for DCC 2.0 you better get rid of your Digistuff and get yourself an NCE. I was able to program a decoder with a PowerCab without even pulling the manual off the box :)

[Duncan McRee]

Hi Cory,

I have enjoyed following your blog - I have the RSS feed on my browser toolbar. I grew up in the Bay Area and enjoy hearing about train happenings around there as weel as watching your layout grow.

Your diatribe on DCC programming is right on - what you need is JMRI Decoder Pro. It makes programming decoders fun and obvious and saves all the locomotives in a database. I have used it for quite a while and find it fantastic. You need a DCC computer interface but there is a good explanation of how to set that up in the documentation: http://jmri.sourceforge.net/help/en/html/hardware/index.shtml

Cheers,
Duncan
San Diego

[Martina]

These are all nice picture and these all train is very stagger.