* golf clap *
Disclaimer: This post may contain humor, sarcasm, and/or flatulence.
Michael Mornard
Bringing the North Woods to South Dakota!
After diligent research, the LION has determined that, yes, you CAN fry a DC circuit board.
Please pass the Mustard.
ROAR
The Route of the Broadway Lion The Largest Subway Layout in North Dakota.
Here there be cats. LIONS with CAMERAS
And to add to the data overload - when you use the typical 1K resistor, you end up with less than 10ma, so not only is is safe for the LED (less than half the typical white LED's limit, it's also definitely safe for the decoder, at less than 1/4 of what the function can support.
Also, decoders tend to have 2 ratings for functions - a maximum for each individual function wire, and a maximim total function current that in the case of decoders with 4 or more functions is often NOT the sum of the individual limits. This is because the blue wire positive common comes from after the bridge rectifier on the decoder and there is an absolute current limit it can support.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Those Proto 2000 PA locos that I recently purchased have those 3-wire incandescent bulbs to simulate Mars lights with alternating filaments.
I am installing D13SRP decoders and the function outputs are 40 mA. So, I am replacing the incandescents with LEDs.
Sure don't want to fry these decoders.
That's what led me to ask the question about the possibility of frying DC boards.
Rich
Alton Junction
If you follow the trace from one of the function wires, you will see it leads to a tiny speck with 3 connections - that's the transistor that switches that function. Again a little engineering magic is applied, the function outputs act as switches to ground, called a current sink - also a very efficient way for the transistor to operate in either full on or full off. For dimming and flashing effects, PWM is again used. This is why the function color wires are negative, each function 'output' is just a switch to ground. This is also why you can connect multiple function wires to the same lamp or LED. Cheap 'OR' logic. What difference does it make if there are 1 or 5 connections to ground? The problem is, if you short the function output by directly connecting the blue wire to the function wire, or allowing a function wire to touch the track power, you are again dumping full voltage (if the blue wire, after being rectified) across that tiny speck of a transistor. Through a resistor, such as used with an LED, maybe it won't blow, since the current is limited. But short before the resistor - no chance of survival. Even with incandescent bulbs - some decoders recommend a small value resistor even if the bulb is rated for 14V or so - there is an inrush current. Same concept as using 1156 bulbs for power protection to the track - those bulbs allow around 2.5 amps until they finally light up and the current then drops to under 1 amp. A 30ma bulb - that's 30ma in the lit state. The inrush before it glows can be twice that. Momentarily switching 60+ma through a transistor rated for 50ma may not make it fail immediately but it is being over stressed and will fail sooner rather than later.
Thanks, Randy.
Actually, when I think about it, the decoders that I have fried in the past didn't wreck the entire decoder but, rather, one of the lighting function outputs. It seeems that the function outputs are where you have to be especially careful.\
It's not that they are really that fragile - in fact most decoders today have protection against shorts on the motor output. Short the motor wires and you most likely won;t smoke the decoder. Also consider the driver bits that actually control power to the motor are just one tiny little spec on the decoder, yet they can handle in many cases as much as 2 amps - at 15V tot he rails that's 30 watts. It all works because of the use of PWM - transistors really only get hot when they are partially open - all excess current gets dissipated as heat. At full on, and full off, they are quite efficient. So PWN works by flipping them on and off at high speed, always full on or full off. So a device that might be able to stand only a few milliamps at 25% on can actually handle much more current because it is only ever all the way on or off.
The killer is when the track input gets connected to a motor output. Full track voltage, at up to the limit of any circuit protection, is applied across the entire decoder circuit. The only things remotely heavy duty are the rectifier diodes att he power input, but the direct track voltage is also applied to the microcontroller since it has to decode the DCC signal. Any inputs and outputs on those chips ARE only rated for a few milliamps, and hitting it with a 5 amp surge even for a microsecond until the breaker trips (if it even does) will toast it.
Stevert, thanks for that info. Very informative, and very helpful.
In addition to what was already stated about smaller traces and components on a decoder vs a dc light board (required to fit everything a decoder needs into the space available), there is also a difference in the type of components.
A DC board ganerally has what are considered "passive" components such as resistors, capacitors, diodes**, and chokes, whereas a decoder also has "active" components such as transistors and microprocessors. Active components are generally more susceptible to over- or reverse-current, ESD, etc.
**Diodes can be active or passive depending on the characteristics of the circuit they're in. But generally speaking, on a DC light board application they would be considered passive. However, even then, in the case of over- or reverse-current or ESD damage, any diodes in the path of that over- or reverse-current or ESD event would most likely be the first to go.
[quote user="richhotrain
So, what is it about a DCC decoder that makes it so fragile and subject to frying if not careful?
[/quote]
Rocket Science..101, size of components.
Just saying!
Frank
Thanks for the replies, guys.
Can't manufacturers design them with better protection?
richg1998 Bachmann locos have two 4.7 uh chokes in series with each motor lead. No idea on how much current they can handle
I had a Bachmann Consolodation back in my DC days and something smoked in the tender on the PC board. The locomotive never derailed (shorted) and there was no problem with the motor wiring. I seem to recall Bachmann having a spate of problems with this particular Spectrum run. A letter to Philadelphia explaining the situation and I had a whole new tender a week or two later.
So, yes, it sure can happen. Ed
If the motor shorts, maybe. Would depend on the voltage applied and how much current the power pack can supply. The control would have to be a full power. If the wires from the pickups to the PC board are #30, they might burn up.
Bachmann locos have two 4.7 uh chokes in series with each motor lead. No idea on how much current they can handle.
I think this is more a hypothetical question.
If you ever fall over in public, pick yourself up and say “sorry it’s been a while since I inhabited a body.” And just walk away.
I suppose if the motor leads shorted together, and you had one of the more powerful DC power packs, you could fry stuff - if the loco has directional constant lighting on DC then that board has some diodes in the motor circuit, and they can certainly fry with too much current. If say one of the wires to the Mars light in those PAs shorted to the track, or to one of the other wires, it may fry the flasher circuit.
Anything's possible, if you try hard enough.
I just converted two DCC Ready locos to DCC, being careful not to fry the decoder in either case. I left the original DC circuit boards in place, plugged the decoders into the 8-pin connectors on the boards, and successfully completed the installations.
But, it left me wondering, on a DC loco, operated on a DC powered layout, is it possible to fry a DC circuit board like you can, if not careful, with a decoder on a DCC installation?