Cheap and easy short protection for your DCC layout

[ednadolski]

I'm not talking about shorts where the current path is strictly through the metal wheel (which is a robust piece of metal which can handle it no problem) but many times the current path of a short can be through the loco itself. Either between the front and rear truck or even between the same-side truck wheels.  In those cases the full current goes through delicate or thin conductors which will heat up and melt the surrounding plastic if the short persists for longer than a second or two.  Yes, the full booster current (5A, 8A, whatever) will be limited to only 2A, but I still think that 2A can do some damage if not immediately stopped.

Here is an example of such damage (upper right metal strip is melted into the sideframe).  I have seen this type of damage more than once. That is the kind of damage which can occur if a high current flows through fairly delicate parts of the model. 

OK so the cases you outline the short circuit current is flowing thru the pickups, but not the decoder itself.   So it's a matter of which "fairly delicate parts" you mean     The sideframe/pickup strips can probably tolerate a few amps without melting, but yes I can see they would get hot enough to damage the adjacent plastic.

Shorts thru the wheels can be a problem too, if there is any arcing that could cause pitting damage on the wheel tread plating.

The modern problem is hard-wired decoders typically use #30 or #32 wire. #30 is rated for 0.8A, #32 is 0.5A. If, heaven forbid, you would experience a hard short between the wheels and decoder, with that bulb passing 2+ amps you are all but certain to melt insulation, further reinforcing any short, and probably generating enough heat in the closed space to deform the shell before the wire(s) eventually melt apart.

That sounds like the kind of short one might get from mis-wiring the decoder, or a or a loose/exposed part contacting the metal loco frame, etc., rather than an op-time, across-the-rails type short.   But there are no limits on odd cases: my P:48 GP9 has all-brass trucks which by their normal design conduct power as part of the pickup system, and so the wheels on one side of the truck have to be on insulated bushings.  When I narrowed the truck to scale width, it brought the brass brake shoes into close proximity with the opposite-polarity wheels, and I would get occasional wheel-to-frame shorts until I covered each brake shoe with a small piece of Kapton tape.   

Ed

[C855B]

... That sounds like the kind of short one might get from mis-wiring the decoder, or a or a loose/exposed part contacting the metal loco frame, etc., rather than an op-time, across-the-rails type short. ...

Exactly. As frustrating as it can be sometimes reassembling a locomotive with all these wires everywhere, the possibility of something getting pinched and exposing wire(s) is pretty good. Heck, I'm sitting here cringing about the brass GP30B on my bench needing a decoder - this is a classic example of this sort of potential problem.

[peteski]

OK so the cases you outline the short circuit current is flowing thru the pickups, but not the decoder itself.   So it's a matter of which "fairly delicate parts" you mean     The sideframe/pickup strips can probably tolerate a few amps without melting, but yes I can see they would get hot enough to damage the adjacent plastic.

Shorts thru the wheels can be a problem too, if there is any arcing that could cause pitting damage on the wheel tread plating.

The other example of delicate parts is the contact point between the rounded brass nub coming up from the truck sideframe touching the flat brass strip in the chassis. This applies to both locos and also to illuminated Kato passenger cars (since they use similar electric pickup method).  If you manage to have the front and rear truck sit across different voltage potential (like running through a power routing turnout which is against you) then the front and rear trucks become the path of a short. The very small contact area of the truck's nub and the flat pickup strip will heat up when 2A of current are going through that area, heating up the metal parts and the plastic around them. I have seen passenger cars with melted floors due to that kind of short.

[RonsTrainsNThings]

No one seems to remember that behind these bulbs you srill have the breaker in the DCC system. Decades ago DC power packs didn't offer the same.

[RonsTrainsNThings]

And I'd like to see an example of damage that occured to a locomotive through a buld protecting it with a modern DCC system.

[peteski]

No one seems to remember that behind these bulbs you srill have the breaker in the DCC system. Decades ago DC power packs didn't offer the same.

But you are missing the fact that your light bulb circuit protector only allows enough current to pass through the short to possibly heat up and damage some delicate parts of the model while not conducting enough current to ever trip the DCC system breaker.

Actually the better older DC power packs did have thermal circuit breakers which would totally cut the current when a short lasted more than few seconds, but just like a light bulb short protection, it wasn't perfect either.

[RonsTrainsNThings]

"Possibly" but again show me one case where a locomotive was damaged by a short through an 1156 bulb on a modern DCC system in HO or N scale. I'm not saying it has never happened, but I have never seen it. Practically I think the likelihood is tiny.

[bobthebear]

Come across a problem with light bulbs - specifically the NCE lightboard. A friend uses NCE, and it can't cope with, for example, the Scale Trains GTEL with twin Loksound decoders. Bulbs glow brightly and train won't move.

[Bendtracker1]

p.s. ..... To me the cost saving does not add up in the risk to my trains.

This pretty much sums it up for me.
Not that the bulbs are a bad idea, but I don't want to take the risk.  I just installed several of the PSX breakers and I can now pretty much run things worry free.

[Greg Elmassian]

Not trying to pile on, but WHY would I add a bunch of bulbs when I have a correctly functioning solid state breaker? Unnecessary, in fact the bulbs most likely would hamper the short circuit detection ability by hampering the "magnitude" and "quickness" of the current flow.

Old circuit breakers only tripped at max amps, modern stuff basically can analyze what is happening, in fact my Zimo system can be programmed for duration and magnitude of the short circuit to optimize detection. The very good DCC specialities PSX series can be adjusted for trip current and there is the "weak system boost" option, besides the microprocessor that analyzes the current to distinguish between "decoder capacitor inrush" and a real short.

If you have been doing this for 15 years, perhaps you have been thinking that solid state breakers are still back in the stone age with fuses?

No one seems to remember that behind these bulbs you srill have the breaker in the DCC system. Decades ago DC power packs didn't offer the same.

[Mark W]

Decades ago DC power packs didn't offer the same.

Which makes this a viable option.. decades ago. 

I respect the notion that "this is just one option of many", but at some point we have to recognize that there are now such better and safer methods that presenting such a risky out-dated option is more harmful than helpful.

[peteski]

Ron,
The solution you presented here is nothing new and does work for limiting the current during shorts, but not for protecting today's delicate N scale model trains.  It would however protect the DCC booster or a DC throttle from overloading and tripping its own "real" circuit breaker.

You say that it has worked for you for 15 years. But we don't know the details. Maybe the current in the type of shorts you experienced was not going through the delicate parts of the model. Maybe you were always able to remove the shorting loco fast enough to prevent any damage. Maybe your booster or throttle does not provide enough current to do any damage to the shorting model. I don't know. But whatever it is, you are lucky.

When I first saw your post I simply wanted to present my own (also experience-based) opinion as to why it is not the optimal solution, and that it might actually cause more harm than good.   And before you bring it up, I also realize that a DCC manufacturer (NCE) sells a similar circuit, but theirs limits the current to only 1A. Still, I would not use that NCE circuit on my layout.  Just because a product is commercially available it doesn't automatically make it usable. There are plenty of commercially-made duds out there.

If you look at the responses in this thread, they are pretty much unanimous.

If you do some Google searches for this circuit, you'll find plenty of info out there. Here are links to info related to the light bulb over-current protectors.  If you actually bother to read completely through this info, you'll find warnings there, very similar to what we are telling you here.

https://sites.google.com/site/markgurries/home/technical-discussions/dcc-circuit-breakers/light-bulbs
http://cs.trains.com/mrr/f/744/t/215408.aspx
http://model-railroad-hobbyist.com/magazine/mrh-2014-11-nov/di_short-protection
http://www.modelrailforum.com/forums/index.php?showtopic=10173
http://www.wiringfordcc.com/track.htm
http://www.rr-cirkits.com/Notebook/short.html

But hey, if it works for you - keep on using it. But I also felt that it was my duty to present my warning.  I'm done here -there is nothing more I can say about this subject.

[Dave V]

Might be worth remembering the audience here too.

[Kentuckian]

Hey, I resemble that remark! 

[C855B]

"Possibly" but again show me one case where a locomotive was damaged by a short through an 1156 bulb on a modern DCC system in HO or N scale. I'm not saying it has never happened, but I have never seen it. Practically I think the likelihood is tiny.

There are a number of us on this board with decades of MRR experience and even some with formal education in electronics engineering. Just because this inexpensive solution has worked for you and a couple of other folks does not mean it's an especially good idea once examined from a learned electronics design approach. It was a cobbled-together make-do 50-60 years ago when it was the rage, when train models were simple and relatively crude electromechanical widgets. When transistor throttles came along (I built my first SCR throttle in 1968), taillight bulbs as load balancers went out the window as insufficient.

That you have not seen "it" happen also doesn't mean we're not about to. Statistically, there are so few modern modelers especially in N employing the taillight bulb scheme that the likelihood of our hearing of a failure is equally slim. If you are successful in proselytizing this method to more than a few, I have a pretty strong hunch the issues will bubble to the surface. @bobthebear 's experience, f'r instance.

Personally, I don't like the bulb solution as general advice because it creates yet one more unexpected variable when our expertise is solicited to solve somebody's operating problem. So we'll have to ask the "protection" question in the diagnostic process especially tracing anomalous behaviors. So I guess for that, thanks for the warning, now we know what to look for.