@gunrunnerjohn and other electronics gurus, I need your help!

I would like to build a circuit that "conditions" track power to feed a wireless receiver / decoder that was originally designed for battery RC operation.

Let's assume that I'm putting the typical 18 or 19 volts, 60 Hz unregulated AC into the track, with a current capacity of ~8 amps.  For various reasons, there may be places on the layout where the voltage at the rail head drops as low as 16 volts.  Also, the voltage on the rails may drop slightly due to load when multiple locomotives, lighted cars, etc., are in operation.  My locomotives are powered by DC can motors not drawing more than a few amps (and probably closer to 1 amp most of the time.)

The decoder / receiver i want to use was originally designed for G-scale locomotives, to be fed from a 14.8V Li-ion battery.  

I can imagine that I would need a high-current bridge rectifier to change the track AC to DC of the correct polarity.  Also, a filter capacitor to smooth out the "ripple" from the sine wave or modified sine wave.

I was warned that this decoder is very fussy.  To maintain consistent speeds, it supposedly needs to be fed a tightly regulated DC voltage of perhaps 15V or more.  I proposed using a zener diode to limit the excess voltage to this value, but the developer stated that a zener wouldn't work.  He didn't offer details as to why it wouldn't work.  I don't have any experience with zener diodes, except I've heard that they can be used to "clamp" voltage to a max value.

The developer also stated that a high-current voltage regulator circuit such as I'm proposing might be too large, and that it would get hot.  He re-emphasized the point that voltage must not vary with load.

I've never built a circuit like this.  I know it would be a good practice to heat-sink the rectifier to a metal surface.  But beyond that, I'm not sure what's so complicated.  I mean, firstly, how would the decoder "know" whether the DC is coming from a battery, or rectified DC from the track?  Also, can you think of any reason why operation would be compromised if the DC voltage fluctuates slightly in the range of 15 to 17 volts?

Thanks in advance for sharing your opinions about this project!

Creep, coast, and pull.  We're not talking about cold fusion here.

Original Post

You can likely do this with a very inexpensive buck regulator module from eBay.  eBay: 323405451143, $2.68 shipped free.

The Zener diode is not at all suited for this application, but this module should do the trick.


Photos (1)

Look on eBay, lots of choices. 

Truthfully, I suspect the incidence of the motors drawing 2 amps is pretty small.  Also, the supply has overload protection, so it'll just fold back if you stall the motors.  In order to get a can RS385 can motor to draw more than 1A on the bench, I have to get a pretty firm grip on the flywheel and really slow it down.

My personal opinion is this would be fine, it has a 3A peak output.

Output current range: up to 2.2A (the regulator chip can withstand a maximum current of 3A, can work at 3A output current for a short time.

From the TI datasheet...

  • Thermal Shutdown and Current-Limit Protection


Well for my purposes you are right it is probably fine.  I've just read these can motors can at stall or power up have a high transient peak ... when i've poked around looking for these voltage converters i've failed to find any that support, oh I dunno more than a few amps.  Then again, what I should do is simply get one, use it and if there's a problem -- do something.   I will add unexpertedly that the few motors I've played with do seem to run a tad better, especially on the low rpm side if the motor controller is high current.


I should just make one.  just buy better parts from digikey.  little problem, not exactly sure how but have a just enough to be dangerous inkling.  I see a linear regulator in the pic (I think), 2 capacitors, some connectors and a tiny screw dial.  then there's the thing with a 470 on it.  I have no idea that is -- and another vertical part on the left side flush against the edge and perpendicular to the far capacitor in the pic. I see I think on the right hand side an led, then in front it a resistor.  10 parts!

You do NOT see a linear regulator, that is a switching regulator.  A linear regulator handling the power you're talking about would be a hot potato!  The ferrite coil you see is the inductor for the switcher, the LM2596HV chip.  It will run MUCH cooler than a linear supply and also handle way more current than a linear supply that's anywhere close to this size.

Truthfully, do you really think you can do better for $3?

no i would expect to spend < $100 and possibly learn something though.  It's been a vexing problem for me since I don't quite know what to buy or look for, quite what to match up against motors + small cpu (such as an arduino) and motor control boards -- and understand it to be a well understood and easy-ish problem for those in the know.

I've been using batteries but they are cumbersome.  My 18v ryobi tool battery appears to work pretty darn good but it's the wrong shape.  It's ok otherwise and I already have the chargers.

I married that to a pololu motor controller(s) sized to a wide voltage input and fairly large -- i had a rasberry pi on there but since have gone to the arduino. 

But none of that from track power.  anyway it's my problem not yours.  i was just curious about the limitations.

Severn posted:

It's been a vexing problem for me since I don't quite know what to buy or look for, quite what to match up against motors + small cpu (such as an arduino) and motor control boards -- and understand it to be a well understood and easy-ish problem for those in the know.

I guess I'm just confused as to why with this statement you'd suddenly decide to start designing your own power modules.  I may simply be missing something in the translation.

With the new single chip designs, switchers are not that hard to assemble, and for any appreciable power, are usually a better choice.  For the LM2596HV used in the module I recommended, all it takes is four outside components and for AC input, a bridge rectifier.  Not a whole lot more complicated than a linear design.



Photos (1)

well I found this nifty seeming kit for learners but of course it's a runs hot type...

eBay 322392708398

and here's another non kit by the same seller. it certainly looks a quite a bit bigger than the first. funny I don't remember all those heat sinks in my engines. to be fair I never really looked that hard.



So all your locos have DC can motors ?   Why not convert to DC track power if you want to use G scale boards designed for a stable DC supply.  Seems you are making this more complicated than need be.  If you want AC on the track why not use ERR TMCC boards ?          j

If you can find a ready made board that does what you need, they are almost always the best choice from a cost and size perspective.  Most of the time I can't find the parts for twice the cost of such boards.  As for learning how to build basic circuits like buck/boost converters, linear regulators and deciding when to use each, there are countless resources online, often as simple as typing your question into youTube's search bar.  I'll recommend Jeremy Blum's introduction to Arduino for a basic overview of micro-controllers and simple electronic components found here:  https://www.youtube.com/watch?...t=PLA567CE235D39FA84  

For more advanced things I very hightly recommend Dave Jones EEVBlog and associated youTube channel.  The content varies but there are numerious older videos going into detail on various things from a professional engineer from how components work to PCB design to dumpster diving and repairing electronics.  https://www.youtube.com/user/EEVblog

There are many other sources, and any $50 electronics or Arduino started kit will get you through the basics.  All that said, for the case of a couple amp ac to dc regulated power supply, it's very hard to beat just buying one unless you need hundreds of them that fit a specific physical design.  

As a side note, you may not actually need that highly regulated of a power source to drive the motors.  If your electronics use a typical H-bridge chip it may be possible to power the radio off of a well regulated supply which only needs to provide a fairly low current and power the H-bridge directly off of filtered DC before a regulator.  

As a last note, if you're really interested in learning some electronics, I wouldn't even bother trying to adapt whichever product to work, but rather try starting from scratch, and build your own drivers and transceivers.  Between Arduinos, ESP8266, and nRF24l01+ modules there is little that can not be done at really surprising prices.  

$ This is John Galt speaking.  $

“I swear by my life and my love of it that I will never live for the sake of another man, nor ask another man to live for mine.” 



Add Reply

OGR Publishing, Inc., 1310 Eastside Centre Ct, Suite 6, Mountain Home, AR 72653