DD54 IMPERIAL DIESEL LOCOMOTIVE
KATO 7010-3

N-SCALE LOCOMOTIVE

(This locomotive ended up being a dummy locomotive less than a month later. Read about it here.)

12/12/15- Yikes! This one screams to be repainted a darker shade of orange, with the grays done up in white and decorated with longhorn decals. Of course, I'd have to figure out a way to shoehorn in a sound decoder and speaker, just so I could upload a custom horn sound, "The Eyes of Texas". Yee haw.

Orange isn't my favorite color-- looks great on a LeMans racer, but for a train... not so much, IMO. I think I got this because it was another early JNR Imperial locomotive, and a diesel one at that. It comes with the crossed Japanese flags herald for front mounting and ID number boards for DD 54-1 and DD 54-3. A pair of Kato knuckle couplers are included as alternatives to the Rapido couplers.

At this time, I don't have any plans to get another sound decoder and speaker for a custom sound car. I don't know what it would be pulling, besides being consisted with the Imperial train. It would be neat to have both electric and diesel sound sets in the Imperial train's sound car, but I currently don't have any solid ideas about how I'd do that. Otherwise, the preferred sound set could be uploaded into the sound decoder, a 10 - 15 minute long process.

The Kato locomotive isn't really suitable for a sound decoder and a decent speaker. The majority of the interior is the metal chassis, and that's to improve its drive performance. Generally, I wouldn't compromise that for sound, especially when a sound car works pretty well.

This article is mainly about installing a DCC motor/light decoder. Normally, that wouldn't be a particularly interesting or challenging task, but this one was different. This isn't one of Kato's "DCC-Friendly" trains, and is pretty close to being "DCC-Hostile". Fun!

How it Works: In DC mode, the front and rear trucks (two traction wheels, one on each truck) pull the DC power into two long brass contact strips that are anchored in the white metal chassis halves. The chassis halves are separated and held in position by upper and lower plastic frames and the plastic motor cage. At the top, a lighting circuit board is powered by sliding in and anchoring within the two frame halves; a pair of contacts are clipped on to the sides of the circuitboard and extend downwards to provide power to the motor.

What makes this more "DCC hostile" than usual is that there's very little free space inside and the metal frame halves act as wiring in DC operation. With DCC decoders, you want to the decoder to supply power to the motor and LEDs, and that's done surgically through wiring. In theory, there's nothing wrong with having the whole thing surrounded by track power... except, if some things short against it, it could mean bad trouble.

Deconstructing the DD54: The first step in planning the decoder installation is disassembling the locomotive to see how it works: How it picks up power, how the motor gets power, how the LEDs work. Disassembling this sucker wasn't easy without a parts diagram to understand how it was assembled. Actually, the first step is searching the Internet to see if anyone else can show you how to do it. I found a few pictures at some Japanese websites that helped a little bit... but info is scarce, which is the reason for this article.

Actually, I saw pics of a decoder that looked like a relatively "drop-in" replacement for the lightboard, but I don't know who makes it or its model number, or whether it needs to be modified to fit. I didn't pursue that since it's more fun and interesting to fit a generic decoder. Personally, I'm invested in the ESU line of decoders and happened to have an unwired LokPilot pulled from Hobbytrain's WR200-V20 "Doppellok".

Getting started isn't too hard: The body shell removes easily by pulling upwards; it's a snug fit. Once inside, you can see that the light circuitboard has to be removed before the black upper frame can be removed. Remove the small black plastic retaining clip on the top. Under that, unclip the brass contacts clipped onto the sides of the circuitboard (shown clipped to the circuitboard after removal). Slide the circuitboard back, and lift out. Splay the edges of the black plastic upper frame and lift upwards. The side brush contact clips should come up with it, housed in the rectangular retainers.

The bottom section is a bit more tricky.

First though, take care of that peculiar single wheel truck in the center before any parts get lost. When I was trying to remove the lower frame, it fell off. It was much simpler than I'd imagined, since there are only two parts: The truck assembly with wheelset and the light spring between it and the body. It's lightly held in place by some clipping and the spring, and slides side-to-side within a channel in the lower frame. It doesn't conduct power, so it really doesn't do anything except replicate the look of the prototype. In practical terms, it just makes it slightly harder to put the train on the track since there are two more wheels to align in the track.

For the last and most crucial step- separating and removing the frame halves- there's no easy way to describe it as a sequence since it all happens in a single step: Releasing the locking clips and prying the lower plastic frame partially lets you pry the metal frame apart slightly at the bottom, which lets the trucks fall out so you can remove the lower frame, which frees the frame halves. Before figuring this out, I spent a lot of time trying to figure out how to remove the trucks as a separate first step. It's all got to happen at one time. Reassembling it is relatively easy, once you know how it's disassembled.

Planning the Decoder Installation: The brass contact strips are a good sign; they can be soldered, so track power can be conveyed by wiring upwards throught the frame to the topside to the decoder. Likewise, the motor brushes can be soldered and wired for decoder control. The part that made me nervous was the way the brass contact strips are mounted: They're wedged into the metal frame, which means that the metal frame is electrified. It would be better if it weren't. Minimally, you want to be absolutely sure that the two halves don't touch since that would short out the track.

I spent a lot of time trying to make the frame not conduct track power, insulating some parts with Kapton tape and Pliobond rubber contact adhesive. Ultimately, I think this was a waste of time. I tested conductivity with an Ohmmeter at several times during reassembly and it seemed like it was working, but after assembly, power from one rail was getting to one side of the metal chassis. I think there are just too many places for the insulation to leak.

The plan was to bring the track power and motor leads topside, where they could be soldered to the decoder. With the lighting circuitboard installed, the decoder wouldn't fit. The circuitboard is useful for positioning the LEDs for the light tubes, so ideally that function would be preserved; however, it would have to be cut up to give room for the decoder. I marked the metal frame so I'd know where the LEDs were located, just in case.

The reused decoder had to be prepped for this installation: The pins were straightened and the short-trimmed blue common wire had to be replaced with a longer one. When bending decoder pins, be very careful-- I learned the hard way with my other Doppellok decoder. The first time you bend them is easy. Bending them straight caused several pins to break off, so the plug housing had to be cut off and wires soldered directly to the circuitboard. This time, I carefully straightened them by "crushing" the bend in the pins gradually with tweezers.

Although there isn't much free space to be found, the little space above the brushes was large enough to store a small amount of excess wiring during assembly. Although the wires need to be trimmed to near the exact length, a little excess is very helpful when soldering them to the decoder. It also avoids the frustration of redoing wiring if you discover that you've trimmed a wire just a hair too short.

Test-fitting the decoder showed that it was too thick to fit in the space-- it might have fit, but it was awfully close, and that's something you don't want to find out after everything's been soldered, when you finally fit the body shell on. I milled the frame rail slightly with a Dremel to recess the decoder just a little bit (about 1.5 mm).

At this point, it's near test-ready for DCC functionality (after tack-soldering the wires to the decoder). Test continuity frequently, to make sure that there aren't any shorts or opens. The motor is tested to see if it works with a DC power source; once the trucks are installed, verify that they're properly plugged into the flywheels. The incremental testing revealed that one truck wasn't conducting power on one rail because of the way that the contact strip had seated. It's better to find out stuff like this before the body shell's on, when it's relatively easy to fix.

With the trucks on, the driving test can confirm the driving direction for connecting the headlight correctly-- you want the headlights to match the direction of travel! You can probably plan this by observing wire color-coding conventions, but I never remember the "proper" wire color orientation for the motor (the rail color coding doesn't matter since either end of the locomotive can be the front). I find it easier to deal with the lighting as the last step.

The Headlights: The lighting circuitboard is simple and easy to trace. The only component I removed was a small SMD noise-reduction capacitor; I believe they interfere when reading the decoder. I carefully removed the SMD resistor (approximately 500K) before cutting out the middle section of the circuitboard. The resistor was relocated to the forward circuitboard that was wired to the positive common (blue wire) of the decoder. After the resistor, a long blue wire was run to the rear circuitboard. This recreates the circuit path of the circuitboard before the middle section was removed.

I thought I'd been very clever and efficient in my wiring of the split LED circuitboards. I tested the LEDs, checked for continuity and shorting during every step; The final test before fitting the body shell and calling it "done" was with the LED circuitboards slipped into position. Once I did and fired it up, I got a sick feeling... I'd forgotten that the frame halves weren't insulated, and I'd left the circuitboards largely unchanged-- circuitboards that were designed to draw power from the electrified chassis. The result: Two blown LEDs that added another hour to the project. My SMD LEDs weren't the same size and had to be adapted; they jumped into the carpet and defied being found. To avoid the possibility of this happening again, I simplified (grinded) one of the circuitboards and covered the other one with Kapton tape since that one had too many opportunities for shorting.

Eventually though, the repairs were tested and the body shell was fitted-- thankfully, it went on easily, and everything worked properly. The collection of removed and unused parts is small: The small black retaining clip, the two small brush contact clips, the brush contact retainers, and the center portion of the light circuitboard.

Taillights: (12/20/15) I got this idea from a Japanese website... Before that, I didn't even realize that the body shell had taillights! Lo and behold, there are red light pipes on the inside at both ends. All that was missing were the LEDs to light 'em up.

The frame was milled at the four corners to recess the LEDs; otherwise the body shell wouldn't fit. As with the Flying Hamburger project, the taillight LEDs were wired in parallel with the headlight LED at the opposite end. I used the same type of LED for the headlights and taillights, so I didn't need to add any additional resistors. I was concerned that they might be too bright, but as it turns out, the taillight light tubes are fairly opaque which reduced the brightness to the right level. The biggest problem was light leakage along the bottom of the body shell.

;

 

Video Clips: DD54 hauling the Imperial train cars. (Apologies to any Japanese railheads who cringe at the ignorant mish-mash of Japanese-esque stuff I've thrown together! No offense intended.)

In the first video, I turned off the electric locomotive drive sound because I hadn't installed a diesel sound set in the sound car... so this shows what the train sounds like without sounds, other than an occasional horn honk.

The second video demonstrates the sound car with a diesel drive sound (I think it's from 54878-LSNV4.0-Diesel-BR643-Talent) in slot 1. Otherwise, it's identical to the sound set that I used above (cloned from the EF58-61 sound set). Note that both EF58-61 and DD54 are hauling the coaches, all decoders set to the same ID; the speed curves in the LokPilots of both are identical and it was fairly easy to get a close match of the motor speed curves.

Dual Drive Sounds? After Googling, searching the manuals and turning up nothing, I didn't think this was possible. So I tried it: I saved the Diesel drive sound to the Library, opened my Electric sound set, and copied the Diesel drive sound to sound slot 2. I mapped sound slot 2 to a function key, uploaded the sound set to the sound car and... it worked! Function key 1 starts up the Electric locomotive drive sound and function key 5 starts up the Diesel locomotive drive sound. Either or both can be turned on and running. Wow. So easy! That opens up a lot of flexibility for using sound cars set up for a variety of locomotive types and multiple locomotives. The limitations are the number of additional locomotive-specific sounds (like coal-shoveling) you can fit into 24 sound slots, the number of simultaneous sounds, and the amount of memory the samples take up. If you're fussy, you may not like the Electric locomotive's brake sound doing double duty for the Diesel locomotive. For me, it's a small sacrifice for the flexibility it gives, and a lot faster than loading a sound set to suit a single locomotive. Hmmm... triple drive sounds?

PREV   NEXT

N-SCALE INDEX