DCC-AC-DC Rectifier

[claycts]

Im a little dense on this one.
1. DCC is AC voltage in the rails
2. HO power is DC from the rails 0-12v about
3. There is a AC/DC rectifier IN the decoder to allow the DC motor to run on AC power?
I would like to get this into my itty,bitty brain before I start doing the ConEdison things on my MESSterpece.
Thank you
George P.
"no train work till AFTER the car show"[xx(]

[cacole]

Go to Tony's Train Exchange's web site and read his "DCC For Beginners" document.

http://www.tonystrains.com

DCC is high-frequency square wave AC digital command signals. The decoder in the locomotive deciphers these signals, rectifies the motor current into DC, and applies it to the motor in accordance with the percentage of power and polarity called for by the DCC command signal.

Tony's explanation is probably the simplest and easiest to understand on the Internet.

[Anonymous]

I guess we are all a little lost when it comes to understanding this "concept". I was under the impression that DCC overlayed square sine wave signals over DC. And I am totally lost understanding how DCC Cab 00 can operate a DC non decoder equiped locomotive. May be time to revisit Joe Fugates "Twelve years experience DCC" clinic, or perhaps Randy Riker can shed some light on this concept for those of us lost and confused.
Will

[cacole]

Yes, the DCC signal is superimposed over DC track power -- but I was trying to keep the explanation as simple and non-technical as possible.

If you really want to know the technicalities of DCC, go to the NMRA web site and read their DCC Standard and Recommended Practices, which is what the manufacturers are going by. That should get you totally confused unless you have a thorough understanding of digital communications.

When you tell a DCC controller to run a non-decoder locomotive on address 00, the decoder takes the positive or negative half of the square wave DCC signal and stretches it out into a positive or negative DC pulse, thereby causing the motor to run in one direction or the other. A normal non-stretched DCC square wave signal is too high a frequency for a motor to respond without this "bit stretching" technique.

Most non-DCC motors will buzz and can overheat if left sitting on a DCC powered track for an extended period of time.

A friend of mine swears up and down that a DCC system demagnetized the motor in one of his locomotives, but I think the motor was on the verge of failure to begin with.

[selector]

I've tried to get my mind around this, and here is what I have decided:

The DCC signal to the track IS AC, according to Tony's information. The decoder converts the bottom part of the signal to pure DC energy, pulsed positive or negative to provide reversing to the DC motor, which the DC motors can only respond to. The top part of the signal is directive information to the decoder to perform function A with parameters x and y....

Address 00 is a filtered AC wave that has been filtered to all positve or all negative to provide the non-decodered DC motor impetus. These pulses are what causes the motor to "sing" because it would normally get a pure DC signal that is essentially pulseless. It is this freqency that the Soundtraxx DSD 100LC decoders are infamous for...they make the loco motor growl and groan by directing what turns out to be largely untunable signals. I can set this pulse freqency in my Heritage 0-6-0, but have yet to find a feqency that the DC motor likes enough to go quiet.

I don't know how realistic this scenario is, but it seems to fit with my "world-view" of DCC.

[Anonymous]

Overdurff makes reference to "square sine wave signals". A sine wave is not a square wave and a square wave is not a sine wave. My point is there is no such thing as a "square sine wave signal". Regular housecurrent is a sine wave. A square wave is just what the name implies, a SQUARE wave. When you're taking about DCC don't forget those ever present PACKETS. Just know that it works, and works rather well if you take your time and wire everything carefully and keep your polarities straight.

[Anonymous]

Claycts, It appears that us old duffers don't need to "understand" it, because if we connect it properly it works. Sort of like the dial tone, touch tones and the ringing sound on the telephone, it gets to who we want to talk to, who cares if it goes by way of Nome, or bounced off a satelite. To answer your question DCC is both DC and AC with a whole lot of information that the decoder works magic with to make lights, sound and motion. We should not be staying up nights wondering how the gerbiles got proded to run in the cage in the right direction.
Hope the rest of the discussion blew some of the mist that has the subject disguised. I thank those who particpated in the topic, I think it is a little more "understandable" to me.
Thank you.
Will

[rrinker]

Yes there is a rectifier in the decoder, but it does NOT run the motor. It only provides the power to the decoder electronics and the function outputs. If the rectifier had to supply the motor current as well, a couple things would appen. One, the rectifier itself would have to be much larger. Two, the motor drive circuitry would get MUCH hotter, especially at lower speeds. If a nice low speed for a motor was say 4 volts, and the track voltage was 12 volts, 8 volts would have to be 'wasted' by DC drive electronics to reduce the motor voltage to 4 volts. That's 8 votls times the current draw of the motor watss of heat to be dissipated - say a 1 amp motor, that's 8 watts! Seems small, but consider small components with small heat radiating surfaces, and also figure 8 watts of heat in a small locomotive shell. (note, before the 'rivet counters' (electron counters?) jump all over me - I know this isn't strictly accurate, but for someone who is not an engineer it's accurate enough)

Instead, decoders use pulse-width-modulation. A typical driver is known as an H-bridge, because schematically it looksliek an H, with the motor as the crossbar of the H. There are 4 transistors, two on each of the motor terminals. One of each pair is turned on with the positive half of the square wave, and the other is turned on with the negative half of the square wave. One + and one - on opposite sides of the 'H' puts a potential across the motor in one direction. Flip-flop which two transistors turn on, and you get the voltage across the motor in the opposite direction, for reverse. Couple of advantages. For one, when the transistor is off, it's off. No current flows, there is no heat dissipation. And when it's on, it's on. There is little heat dissipation because the only voltage in the transistor is that of the semiconductor junction, about .6 volts. Times 1 amp motor and it's only .6 watt, less than 1/10 the DC votage waster method.

How does this vary speed? By turning the transistors on and off at a high freqency (the 'pulse' part) and leaving them on for longer or shorter times (the 'width modulation' part. The longer the power is applied, the faster the motor goes. This is similar to the address 00 'zero stretching' except the opposite polairity component is missing, plus, at least in 'silent running' or 'supersonic' decoders, it occurs far faster than the DCC track frequency so the motor does not buzz.

--Randy

[Texas Zepher]

QUOTE: Originally posted by ebriley

Overdurff makes reference to "square sine wave signals". A sine wave is not a square wave and a square wave is not a sine wave. My point is there is no such thing as a "square sine wave signal". Regular housecurrent is a sine wave. A square wave is just what the name implies, a SQUARE wave. When you're taking about DCC don't forget those ever present PACKETS. Just know that it works, and works rather well if you take your time and wire everything carefully and keep your polarities straight.

[#ditto]
While technically DCC is an AC current because the polarity reverses, calling it so just confuses people because everyone associates AC with the sine wave electricity, as is used to run Lionel type trains. There would be so many fewer questions like this had no one ever associated the two.

[claycts]

Thank you all, it does make sense to me. Both Randy's and Texas Zepher. The "00" address made the reason for the question. I am going to setup an either or route for the older engines. Toggel switch operation of regular DC or DCC depending On what I was running. With full isolation if the systems so there is no way both can be on at the same time. We have a person on the local club that is a Digitrax Rep and is going to do seminars for us on DCC after the 1st of the year. I will take notes and pass them on. I would hope that since he will be looking at the members track plans the information will be of more use than a general Q&A on the subject.
He will be looking at 4 different set ups. From SMALL, 4x6, 2 trains no reverse loops to Large 18 ft x 22 ft, up to 10 engines, 7 operators, 3 reverse loops and a turntable. This should be a good head session on DCC.
As soon as the cars shows are over I am going to try and have something running other than back and forth on a 6 ft test track.
Thank you for the information and time to answer. I now understand the basics for DCC with the electronic control of current via transistors.
Take Care
George P.
"Car show Aiken SC. 11/12/2005"

[samgolden]

Randy is right on the, "Pulse Width Modulation". I use to work in Electrical Maintenence for a large company and had to repair machines that used this method to control the movement of (mostly) milling machines. The machines were called "Numerical Controled" Milling Machines. A computer program told the machine where to position the tables, but the motors that ran the tables, were P.W.M. exactly like the "H" transistor arangement Randy explaned, except I am talking about 2 to 10 Horse Power Motors. The first machines with this control were operated on the normal 60 Hz frequency (60 cycle), and when the motors were runing real slow, you could hear them Whine, but the newer machines, they used a frequency higher that the normal audio range so you didn't hear the whine anymore.
Retired in '93, so don't know what they have come up with now.

Sam

[MisterBeasley]

Yesterday, I hooked up a diode bridge to the tracks (with alligator clips) to measure the DC voltage I would be getting out. I need DC for LEDs in a subway car which will not have a decoder. It looks like I'll be happy with a 1K resistor in series with each LED. Is there anything wrong with doing this?

[rrinker]

I use 1K resistors for my LEDs. With about 12.5 volts on the track from my Zephyr, that ends up being around 10-12ma for a Golden White LED and is PLENTY. If you are going to connect an LED/resistor directly to the track, instead of to a decoder, you should install a standard diode across the leads of the LED but the the opposite orientation - exposing an LED to too much voltage in the reverse direction can damage it, even with the current limited. The plain diode will conduct for the reverse direction, limited by the same current limiting resistor, and the LED will see only the 0.6 volts across the diode in the 'wrong' direction, perfectly safe. Standard diodes have a much higher inverse voltage limit (the PIV - for a standard silicon diode line a 1N4000 series, depending on the exact diode, this is 50, 100, 200 or even 400 volts.)

--Randy

[Anonymous]

QUOTE: Originally posted by claycts

Im a little dense on this one.
1. DCC is AC voltage in the rails
2. HO power is DC from the rails 0-12v about
3. There is a AC/DC rectifier IN the decoder to allow the DC motor to run on AC power?
I would like to get this into my itty,bitty brain before I start doing the ConEdison things on my MESSterpece.
Thank you
George P.
"no train work till AFTER the car show"[xx(]

Well, time for an electornics engineer to see red, not at your post George, but at some of the horrific misinformation given in repsonse to it.

1.. DCC is AC power on the rails. The DC component of a DCC signal is, or should be very close to, ZERO. But it is not AC like you get from Con Edision, it is a high freqnewcy square wave AC which is modified by the DATA which is being sent ot the decoders.

2.. DC power (what you refer to as HO power) is somewhere between 0 and about 15 or maybe even 18 volts depending on your power pack.

3.. Yes, their is a rectifier on the decoder board. And YES, if does supply the power for the motor! But not directly. Here is how a decoder works:

The AC signal from the track is passed through a rectifier, a small amount of signal is 'sniffed' off the big signal and is sent to the decoder to be decoded. The bulk of this power is then filtered and becomes pretty good clean DC.

Another part of the decoder is called the 'H' bridge. You dont need to know why, but it is because when the circuti is drawn out it looks like an 'H'! This 'H' bridge is used to control the power flow to the motor to control its speed. But varying a DC voltgae, like you do in a DC power pack, is horribly inefficient and produces a lot of waste heat. SOmething we cannot deal with in the loco. So the itty-bitty little microprocessor on the decoder produces what is known as a PWM, Pulse Width Modulated, signal which turns the 'H' bridge on an off to control the AVERAGE output to the motor. This is very efficient, because the control circuitry is ON or OFF - both states dissipatihng virtually no power and resulting in pretty close to zero heat output.

The motor, by virtue of its mechanical inertia effectively integrates this signal and thus it has the same effect as DC.

But this is also the 'growl' or 'whine' you hear in some older and cheaper DCC decoders. If the PWM signal is within the audible frequency range the motor windings can act like a loudspeaker and you can 'hear' the PWM.

ERRORS SPOTTED ELSEWHERE:

The rectifier diodes on the average decoder board that I have seen are quire capable of handling between 1 and 3 Amps. When it is kmoving the average N or HO loco only consumes a few hundred milliamps. But the average decoder is capable of handling anywhere from 1.5 to 2 Amps to handle the loco when it stalls. The limitation on the current that is avilable to the motor is a fiunction of the 'H' bridge more than the rectifier.

Whether we like it or not, the signal on the rails in DCC is AC. It is not sine wave, and in fact it is not square wave either!

If you want to power lamps from DCC powered tracks, you dont need the rectifier, just use a resistor to limit the voltage to something reasonable.

If you need to use LED's then use the bridge. You can either use 4 individual 1N4001 or similar diodes, or you can buy small bridges in an integrated package that would be more conventient due to the reduced amount of soldering. Ask your friendly local supplier for something rated 50PIV or greater at about 1 AMP.

I hope this helps clarify some things.
John