Locomotive Intermittent Power: Kato DD54
12/32/15- Symptoms: Kato DD54 locomotive ran perfectly during numerous tests after the decoder was installed; about two weeks later, headlights intermittently exhibited brief glitches in specific areas of the track, at switches and curves. The condition worsened to the point where it would intermittently stop in those areas, which then included specific areas of straight track. At first, touching the locomotive would get it moving again; at worst, it took rocking from side-to-side, or twisting in the track, or bearing down with pressure. There appeared to be no known event that led to this problem or made it worsen. None of the other locomotives had this problem.

Suspect#1- Digitrax Command Station: Whenever locomotives start exhibiting wonky drive behavior, I reset the Command Station because it's relatively easy to do. This clears problems of multiple throttle assignments and fixes odd things that only digital technology can screw up behind the scenes. It didn't fix this problem.

Suspect#2- Dirty Track: Since the locomotive seemed to stop at specific areas of the track, dirty track was suspected. Cleaning the track in those areas was the first remedy tried. It seemed to help a bit at first but didn't solve the problem. The track was slightly dirty, but it wasn't localized to those areas. Even though no other locomotives had problems with these areas of track, I thought that the train's shorter span between trucks might have played a role.

Suspect#3- Dirty Wheels: This wouldn't explain why the locomotive would intermittently stop at specific areas of the track, but wheels are easy to clean, and it's always a good practice. Also, drive problems may have a combination of causes. Cleaning the wheels didn't fix the problem.

Suspect#4- Locomotive Dirty or Sprung Contacts: I began to suspect this when the problem was at its worst. Kato locomotives get track power from contacts that slide against the wheels; those contacts slide across brass strips as the trucks swivel. Good conductivity relies on the spring pressure exerted of one contact against the other. If the contact points are dirty or aren't firmly touching because of insufficient spring pressure, conductivity will suffer. DeOxit contact cleaner was applied, but didn't solve the problem. I began to suspect sprung, or improperly positioned contact strips-- plausible, because they might not have been positioned correctly during reassembly after decoder installation. A truck contact can fall off the edge of an improperly positioned contact strip. Although that didn't appear to be the case, disassembly would let me examine the situation more clearly and inspect/bend the contact strips to increase their spring pressure.

Tests during disassembly/partial reassembly didn't show any improvement after cleaning the contacts, bending the contact strips, or resoldering decoder power leads. To remove spring contact as a cause, I soldered wires from the trucks directly to the contact strips. This didn't solve the problem.

Suspect#5- Wheel Spacing: I suspected that the wheels weren't making proper contact with the truck contacts, so I narrowed the wheel spacing slightly (geared truck contacts are on the inner surface of the wheels, unlike ungeared passenger coach trucks). This didn't solve the problem.

The wheel spacing was increased slightly (less than a millimeter). This solved the problem!

Apparently, some of the wheel set spacing in the trucks was slightly off, which caused intermittency in certain areas of the track-- probably because the rail spacing in those areas was off by a fraction of a millimeter as well. Fixing the track would be impractical because there's so much more track to nudge into a state of perfection... especially since you can't easily see where it's slightly out of tolerance.

This explains the symptoms of intermittency, and why nudging, rocking, rotating the train on the tracks and applying downward pressure would improve contact. The spacing of all the wheels was checked with a micrometer to ensure consistency, and after reassembly, the locomotive ran as well as it had originally.

Conclusion: Wheel spacing should be one of the first things to check when troubleshooting this kind of problem. I'd been misled because the locomotive had worked well at first. I wasn't aware of any event that might have affected or changed the wheel spacing. It therefore wasn't on my radar in the list of suspected causes of the intermittency. Apparently though, it can happen, possibly from a change in temperature? (It got colder.) The difference is not something that you would necessarily notice with your naked eye. A rail gauge or micrometer are useful tools for diagnosing this kind of problem. (Note to self: Get a rail gauge!)

Always Wait a Few Days Before Proclaiming Victory...

It seemed like adjusting the wheel spacing had fixed the problem, but when the problem returned a few days later, I realized that the problem was deeper than that. Although I'd cleaned the truck contacts, I hadn't completely disassembled them to see exactly how they were constructed. It's a Kato-- hey, I trusted the design!

Trust no one.

I was sooooooo disappointed when I saw how the brass contacts were drawing power from the wheels. The stiff brass contacts have semi-circular openings along the bottom edge, and the short bare metal section of the axles press into the openings.

(Ignore the strip of brass soldered onto one of the wheel contact brushes. Explanation follows...)

I believe that this is one of the worst pickup designs I've ever seen. It's even worse than any spring-tapped axle pickup I've ever run across. It doesn't take a genius to know what happens when an axle spins within a slot in soft metal: It wears the slot, making the opening bigger! The brass contacts are thick and stiff, so there's no spring loading to press the brass contact against the wheel or the axle. After a few weeks of running, the axles were sloppily knocking around within the contact slot, the only contact pressure being the weight of the locomotive. This explains why the locomotive worked well at first when the contact slots tightly held the axle. This explains why it later started balking at start up and slow speeds, and why it worked better at high speeds. This explains why changing the wheel spacing seemed to help for a while.

I tried to fix this, first by bending the contacts slightly to give better contact with the wheels. That helped for a short while. I soldered springy brass strips to the contacts; they caused the wheels to bind.

Unfortunately, there's no easy fix for this; the trucks would need to be heavily modified (redesigned) or replaced by others with a better design. Factory replacement trucks, even if available, wouldn't be a solution since the design is broken. The wheels are thin and flat-ended so wouldn't be an easy swap for a needle-point wheel and cup contact bearing. Such a wheel would need to have an installed gear of the same size. (Frankly, I'm a little apprehensive about disassembling any of my other Kato locomotive trucks to see their pickup system design.)

If I were really determined to make this run, I'd do a shared power modification with another car that had reliable power pickup. However, the DD54 is a spare locomotive, envisioned for multi-locomotive duty in a consist. It wasn't the primary dedicated locomotive in any particular set of coaches.

The sad fact is, a locomotive with intermittent power pickup is unusable in a multi-locomotive consist. The other locomotive may help it get started, but if the power cuts in and out, it will likely lurch, causing the other locomotive's trucks to skip rails.

My solution was one that folks who want to run multiple locomotives in a consist dream about: Having a locomotive that can be guiltlessly gutted and turned into a dummy locomotive! No need to worry about speed matching or changing the consist address. As a bonus, now I've got a spare decoder, motor and pre-wired SMD LEDs! Yep, life is sweet.