DCC decoders have a current rating. A decoder should be selected that has a current rating that exceeds at least the slip current of the locomotive (i.e. when the wheels are turning but the locomotive is preventive from moving) and possibly the stall current (when the wheels are prevented from turning). Mark Gurries (NCE) provides some explanation suggesting many decoders may be able to sustain excessive current over a short period of time without damage.
I'm curious about what may happen when the decoder current rating is exceeded. Some possibilities:
greg - Philadelphia & Reading / Reading
The ultimate control is with the motor drive H bridge. Most of the time, this is an integrated driver, not individual transistors (some of the really high current large scale decoders have individual power transistors or darlingtons). It all comes down tot he rating of this component, and then the quality. Just because one chip came off the line with hundreds of thousands of others, they are not all equal. Sometimes you get lucky and have a better one, sometimes you have one that just barely meets the specs. Obviously the better one can withstand more abuse before it finally fails.
Second you have the operating mode. Since motor drives are almost alwyas PWM, the transistor is either full on or full off. It shouldn't be operating at any in between state. So the heat generated is related to the full on resistence. Then you have the speed. At slow speeds, short pulse duration, there is more time when the transistor is off for it to dissipate heat. This suggests that a larger momentary overload can be sustained at slow speed than when operating at high speed.
Overload too long, and the drivers will smoke. Slight smoke, if things get shut down fast enough, and the part may still work, although generally at reduced capacity and highly prone to failure. If it goes all the way, it's pretty obvious. In extreme cases the heat generated is enough to blow a piece out of the epoxy chip package.
BEMF only measures voltage feedback from the motor. It's possible to calculate load based on the difference between what the motor is being told to do and what it actually is doing - I'm sure you could set some sort of threshold and say any greater difference means the motor is overloaded because it can't turn fast enough to match what it's being told to do.
Throttles in DCC don;t do anythign except tellt he command station to generate DCC packets for whatever it is you are trying to do. The booster is not going to shut down fast enough to prevent an overload - it's designed to supply 5 amps or more, so one loco drawing 3 amps in overload is not enough to trip anything. Even external circuit breakers would have this same issue, set them to trip at lower than the decoder's limit means even a decoder operatring fully within tolerance would trip the breaker. Kind of defeats the purpose of being able to run multiple trains on the same track with independent control. But theoretically if the system only supplies 1.7 amps before tripping and the decoder can stand 2 amps, you should never blow a decoder due to overload. Linking a motor output to the power input though, and it doesn't matter, the driver is going to smoke.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
as an engineer I can imagine a lot of different solutions to various problems and know that the final solution is often the one that is least expensive or the one customers are most willing to buy.
My NCE PowerCab can report the current draw of the decoder. I assume this is done in the decoder, especially one capable of BEMF. If this is correct, couldn't the decoder protect itself from excess current damage by quickly shutting down for a few seconds like the PowerCab?
are customers willing to pay a little extra for a feature that protects a decoder?
The PowerCab current readout is done at the command staiton (inside the handheld). It's the total being drawn on the rails, not the draw of any one given decoder - put 2 or three running locos on the track at the same time and you'll see.
Some decoders have short protection, so that if there is a dead short on the motor (miswired, failed windings) it will do something like blink the lights. An overload is more difficult to detect. Look at circuit breakers and boosters - you can set devices to cut powe rif the draw exceeds say 3 amps, but you can have a 2.9 amp "short" and it will stay powered all day long because it hasn't exceeded the limit. 2.9 amps at 15 volts is a 43 watt heating element. I'm not too keen on touching a running 40 watt light bulb - it's hot! Hot enough to melt plastic and do other things. Lower the threshold, and now you can't run the desired number of locos in the same area. If you have all the money in the world to throw at this, you could break the layout into less than train length sections, each with its own circuit breaker set to maybe 1 or 1.5 amps, so theoretically it would trip before the 2 amp peak on many HO decoders is hit
The Ring radio receivers have current sense and temperature monitoring built in. But they are much bigger than typical decoders and even the non-sound one is $40 or so. Are people willing to pay extra for a decoder that has these features so you can abuse it and not damage it? I don't think anyone's tried to market such a device, but based on many posts here, everyone is looking for the cheapest decoder they can find. Doesn't help sell the expensive one when several manufactures now offer no questions asked warranty replacement, so even if it's your own fault you fried it, you get a free replacement, the first time, anyway. DCC's been around long enough that it seems pretty well proven that as long as you match the decoder to the motor it's driving, you won't have any problems. Try to run a 5 amp big O scale open frame Pittman motor with a 1 amp N scale decoder and it's fairly obvious that it won't work, even if you don't really understand what an amp is.