They are a current limiter

They don't really prevent the shorts from occurring... but they limit the current which flows in case of a short. NCE even makes a circuit board which holds six 1-amp auto lamps.

https://ncedcc.zendesk.com/hc/en-us/article_attachments/200585885/CP6.pdf

What actually makes a turnout "DCC Friendly"?

Dear Kirk,

File this one under "know the basics, then extrapolate to any compound situation".

The definition of a "DCC Friendly" turnout is as follows:

- _ALL_ rails actively powered
("dead frogs" are a cheat false-economy, not an all-permutations-covered solution)

- All Frogs isolated from all other rails
(allows correct "polarity" powering under any given set of routing conditions)

- Each Switchrail hard-powered/permanently-connected to the same polarity as its adjacent stockrail
(IE eliminate reliance on switch<> stockrail pressure-contact for power transfer,
and make "stockrail> wheel tread> wheel backside> switchrail" short-circuit conditions _impossible_).

Nothing in there about _how_ the frogs are powered,
(micro-switch, toggle, ground-throw contacts, Tortoise contacts, etc etc)
  
and given that a "protection bulb" is just a current limiter, not a "2-in / 1-out" polarity switch,

the inclusion of "an auto lightbulb" does not forfill the definition of "DCC Friendly".

NB that unlike a simple turnout, a diamond-crossing has no mechanical moving-elements from which simple microswitches or similar can obtain and action "routing polarity" info. Thus, the "simplest" solution is to use an active-frog-switching unit, like a TVD Frog Juicer.

"Glancing/intermittent shorts" with a loco navigating one of the "diverging routings" of this dbl-crossover _will_ be traced down to one-or-more of the above not-being-addressed...

EDIT: it has been noted that the Shinohara Code 100 dbl-crossover may exhibit "shorts at the central diamond" 
IF all 4x turnouts are NOT set simultaneously.

IE - ALL turnouts set "straight" = OK
- ALL turnouts set "diverging" = OK
(although this also creates the mechanical situation where 2 trains may try to cross the diamond simultaneously... oops)
- ONE route/pair-of-turnouts set "crossover", while the other turnouts are still set "Straight" = Possible Short

You can easily test this by consciously setting _all_ 4x turnouts in-sync,
and seeing if this clears the "fault" condition,

READ AND HEED :  http://www.wiringfordcc.com/DoubleCrossover_Galyon.pdf

and, again. I should note that if the 4x turnouts + 1x diamond was truly wired "DCC Friendly",
then this would not be an issue...

Happy Modelling,
Aim to Improve,
Prof Klyzlr

Making Shinohara DCC Compatible

Interesting Prof Klyzir, I'll have to keep that all in mind when I install my Code 100 Shinohara in my two mainlines here
 

Seems to me that years ago when I was putting 2 dbl crossovers on my Central Midland layout (only DC), I found a way to power all four of the turnouts with only one motor, and it worked find for that.

I just don't wanna start cutting it...

This is on a friends layout and his space is limited. I hate the darned switch, but I know he spent BIG $$$ on it and the last thing I want to do is suggest the removal. Especially considering how delicate it is and I would hate to ruin it.

The VERY first thing I considered was a short through the crossing when all 4 switches were not thrown. WRONG !!! it still shorts which is discouraging. The only thing I have not done is try several different locos on it to ensure it is not the loco. I can only work on it every so often and therefore, do not have the luxury of testing some new idea to make it work when it is fresh in my mind.

I suspect the easiest way to make this work is to cut the point rails, a total of 4 different places, just before the frog and then power it using a frog juicer or something like that. I don't want to do anything more complex than is necessary, especially since it is not mine.

Who has the solution?