Well, this is a first! (Loco Meltdown)

[tehachapifan]

Was out at the layout tonight tinkering and, as I often do, had one of my custom N scale SW1500 kitbashes crawling around the branchline at speed step 4 (approx. 25 scale MPH) with 3 cars in tow. I took notice that the loco, built on one of those old Con-Cor/Kato switcher drives, was running extra smooth and silent tonight. Then, as I was about to go in for the night, I brought the loco to a stop and noticed that the sides of the short long hood directly over the motor had completely melted! Any ideas on what may have happened?? I should note that, for this kitbash, the thin bit of frame metal that originally covered the motor on this drive was milled away to fit in the narrow Atlas MP15DC long hood. I've never noticed these drives getting particularly hot before, especially at such slow speeds.

[mmagliaro]

Does it still have the old Rivarossi can motor in it?  If so, "I'll take Rivarossi can motors for $100, Alex"

[tehachapifan]

No, these are a later version with a more typical N scale pole motor with exposed armatures.

[peteski]

So the loco with that melted shell was fully functional before the meltdown, and after the meltdown is still works perfectly?!

How about posting some photos of the shell and frame laying on the side, with the shell lined up over the frame (to see how it aligns with the motor)?

[tehachapifan]

Yep! Ran its best ever just before and was still running nice when I discovered the damage. It was making a slight growl-type noise (not uncommon with these drives) at the end and I
now think it was from the motor armatures striking the now-melted shell. I may try to post pictures later (busy today), but the melted sides are directly over the motor. The only thing that recently changed is I added a Run-N-Smooth module to the loco and installed it in the fuel tank. However, the fuel tank and module were not hot to the touch upon initial inspection, but the motor certainly was. I will add that, because it was running so nice and whisper quiet tonight, I probably let it run longer than I usually run one of these (I made 5 with a couple more in the works).

[peteski]

The Kato USA's first American loco (GP38-2) had a thin aluminum shield placed over the open sides of the motor for protecting the adjacent shell walls  from heat, and that was on a DC-only loco with DCC not really being utilized much in N scale, especially in narrow hood models.

So, you might be correct that the motor's armature was the culprit. As you said, it is quite clear where the damage occurred.  The inside of a model is relatively air tight so the heat generated by the armature is not readily vented out, accumulating in the motor cavity.

Actually the slow running might have contributed to the problem.  Since the decoder uses PWM for running the motor, the pulses sent to the motor are always at full 12V, even when the average voltage is lower.  Those full voltage pulses likely heat the coils on the armature at slow speeds almost as much as if it was running at much higher speeds.

[tehachapifan]

That's not encouraging news, especially since these are my favorite locos and now I'm afraid to run them!

These were very involved kitbashes that I almost consider like children and I don't want to lose any more.

Is there a common DCC adjustment to reduce the voltage that can cause excessive heating at slow speeds?

These are equipped with Zimo MX621 decoders.

Also, I wonder if some kind of foil on the inside of the shell over the motor would help reduce the risk or would it make it worse?

[nscaler711]

That's not encouraging news, especially since these are my favorite locos and now I'm afraid to run them!

These were very involved kitbashes that I almost consider like children and I don't want to lose any more.

Is there a common DCC adjustment to reduce the voltage that can cause excessive heating at slow speeds?

These are equipped with Zimo MX621 decoders.

Also, I wonder if some kind of foil on the inside of the shell over the motor would help reduce the risk or would it make it worse?

Ah Now I see the need for MP15s. That sucks, I'm usually wary of DCCing old equipment for that reason myself so as of late, I've been looking at newer sound units to throw old shells on but they dont compare to new molds anymore.
Also did you manage to grab some pictures? I am ^morbidly curious to see how bad it melted...

[tehachapifan]

Yep, this is the reason I need the replacement shell (sniff, sniff ).

I haven't taken the pictures quite yet. Maybe tomorrow. Getting pics taken and then uploaded seems to be an involved process on my end these days.

[C855B]

Also, I wonder if some kind of foil on the inside of the shell over the motor would help reduce the risk or would it make it worse?

Worse. A "merely obvious" is that the armature cools itself by moving air as it spins. Combine what Pete mentioned about the PWM (versus linear voltage) with slow rotation and you have a recipe for cooking things, remove from oven when golden brown. Shielding the armature cavity would worsen the effect by cutting off air circulation.

If it were me, I'd be shopping around for suitable screen material to model the vent details. Or not run at ultra-low speeds for long periods of time.

[nscaler711]

May I suggest a really cool product?
https://www.digikey.com/en/products/detail/sunon-fans/UF385-100/1739059

Ok, even if you could find one, it would still take up too much space and still wouldn't work well in a Switcher or even a hood unit.

[tehachapifan]

CAUTION: THE FOLLOWING IMAGES ARE GRAPHIC. VIEWER DISCRETION ADVISED!!





As you can see, the motor armatures are visible directly in line with the melted area.

OK, take the pics away! I can't bear to look at them!

[nkalanaga]

If you plan on running slowly all the time, could you set the maximum voltage in the decoder?  That might force it to send pure(r) DC at full throttle, and minimize the heating.

(I know NOTHING about operating with DCC, except that most modern decoders don't do well on my DC layout.)

[nscaler711]

Oh ouch, there really is no fixing that.
However I will say outside of the Melty loco, it looked pretty good.

[Doug G.]

This is why it's more important to monitor motor current draw than voltage.

Doug