Atlas Switch Machines Melting

[StewRRFan]

I am relatively new to N Scale.  We purchased a small layout which was set up for DC.  The DC transformer controlled the power sections and the switches.  Upgraded to Digitrax DCC and it works great.  I moved the switches to a separate transformer with the switch power wired to the AC.

The problem is the switch machines begin to smoke and melt down on a somewhat regular basis.  The melt point is directly on the top and it continues until you can see the copper coils.

Any help or suggestions are greatly appreciated.

Ross

[ljudice]

You have momentary contacts on the switches controlling the switches?

Otherwise they will melt for sure....

[StewRRFan]

Just the standard Atlas controllers.

[Zox]

Can you put a meter on the output of your transformer and see what kind of voltage it's supplying? (Remember to set the meter for AC voltage, not DC.)

If you're using the Atlas controllers, they should be working as momentary contacts, so the next likely option is that you're simply hitting them with too much power every time you "flip" them. You shouldn't need more than about 12 volts, and probably not that much, to make the switch machines work reliably.

[StewRRFan]

It is reading 19.1.  The transformer is a Railpower 1300.

Thanks

[wm3798]

It's God's way of telling you to upgrade the turnouts to at least 1970's technology.

Atlas Controllers are NOT the same as the blue Atlas switch throws.

Controllers are designed to route continuous current from one circuit to another.  The switch throws (the ones with the blue button) are Momentary contact switches.  Twin coils work on a quick burst of relatively high current.  ZAP and it throws.  Apply constant current, and you will quickly have a Three Mile Island style melt down.

But seriously, get rid of those old coffins.  They look horrible, work less than reliably, and will burn down your house if not wired and operated properly.

Lee

[Zox]

It is reading 19.1.  The transformer is a Railpower 1300.

Thanks

Yeah, that's a bit much. It doesn't take very long to melt a coil that's over-volted. (Think about how long it takes an incandescent light bulb to get white-hot after you flip the switch.)

I'd suggest stealing a 9-volt wall wart off of some obsolete electronic device, cutting off the plug and stripping the leads, and using it instead of the transformer's AC output.

[StewRRFan]

Lee -

Thanks from another long suffering Orioles fan.

The layout does have the Atlas Blue Button controllers.  Are you suggesting scrapping these and what would be your suggestion?

The intent of this layout is to use it as a test platform so I can learn as my son and I build a layout from scratch.

Ross

[hpwrick]

First off let's get the voltage down to 12 volts AC.    Next check the old Atlas blue push and slide switches to make sure they aren't making contact excepting when you push them.    If it turns out they are broken and stuck on. your multimeter can varify this, then it's time to discard them.    May I suggest a momentary on, DPDT toggle center off and wire these in to replace the blue sliders.

[peteski]

If you used a DVM (Digital Volt Meter) then the 19 VAC is probably more like 13.4 RMS VAC (which without getting too technical is the voltage that actually does the work). If you used a DVM [Digital Volt Meter] (which without getting too technical is the voltage that actually does the work). (see my next post few down from this one) That is fine - no need to get another power supply. After all you are using a model train power pack which has the accessory output designed for things such as switch machines.

As others said, if you don't use a momentary button (or hold the momentary button for more than about half a second), the switch machine coils will start to get toasty.

[Zox]

If you used a DVM (Digital Volt Meter) then the 19 VAC is probably more like 13.4 RMS VAC (which without getting too technical is the voltage that actually does the work).

Not to hijack the thread, but--why would a meter, assuming the correct range is set, produce a grossly incorrect result? If I set my multimeter to "AC volts," I expect it to tell me AC volts. And since volts RMS is the correct way to measure AC voltage, I would expect a meter that claims to measure AC volts to use volts RMS.

I ask because if what you say is true, I may have committed heinous electrical sins against various devices... 

[peteski]

I guess the only way to know for sure is to look at the specs for your meter (and for StewRRFan to check his).
The other possibility (if the meter does read RMS AC voltage) is that the voltage coming out of the transformer will drop when a load is placed on it (voltmeter is pretty much seen as an open circuit).  I guess one could hook up a light bulb (or some sort of a load) then measure it again.  Come to think of it, I think you're right. Standard pedestrian digital meters most likely do measure RMS AC voltage.

As far as your heinous sins, I don't think you have to worry about it. If what I said was correct, you would have only underpowered them. 

[alhoop]

peteski, your first post about rms reading on a DVM is possibly misleading. Most DVM's will read true rms on a true sine wave, maybe something different on  a square wave, triangle wave etc. Your second post about the no load versus load output of the transformer  is closer to being correct.

Al

[peteski]

Thank you for your confirmation alhoop - I did correct myself in my 2nd post.   

[Atlas Paul]

Atlas switch machines need 16-18 V AC to operate properly.  The MRC 1300 probably only has enough amperage to throw one machine at a time, maybe 2 if you are lucky.

If only one machine is melting, then the problem could be with the button.  If more than one machine it melting, there is something with the wiring where there circuit is being constantly powered.