When a decoder has specs that say it requires a minimum of 7.5 volts, does that mean the DCC signal must (at minimum) have its highest peak at +7.5 volts and lowest peak at -7.5 volts (for a total of 15 volts)? Or would it only need to have it's highest peak at +3.75 volts and lowest point ant -3.75 volts (for a total of 7.5 volts)
Well that's an interesting question, but I'm goign to go with the +/- 7.5V option, for the simple fact that the peak voltage (and RMS voltage) of a square wave that goes both _ and - with respect to a common is the same as the peak - not peak to peak, but peak. When calculating area under the curve, there is a squared termin in there, so the - half of the wave is inverted to fill in the space between the two positive sections. When a DCC system says it puts 12, or 15, or 18 volts on the rails, it is talking about peak to peak, because that's what the decoder sees when it rectifies the track voltage, since there is no reference to the common in the booster (and not all booster swing +/- around a common, some have the common set to the lowest part of the waveform and the peak is the actual peak to peak voltage - totally unimportant where the decoder is concerned since it has no reference witht he booster common. So the decoder see +/- 7.5V for 15 volts total(less 2 diode drops) when rectified. The minium voltage for the decoder is in this case 7.5 volts peak to peak, or +/- 3.75 volts relative to common. This is the minimum voltage that, after rectification (2 diode drops) is still high enough to operate the microcontroller reliably in the decoder. Of course the loco will not go very fast on that low of voltage, since the nominally 12V motor will only see a maximum of about 6 volts, or slightly less - the diode drops in the rectifier PLUS another set of diode drops in the H bridge driver.
Digitrax is one that swings +/- around the common, which is the terminal they label ground. So you can measure DC volts from Rail A to Ground, and Rail B to ground, and add them to get the track voltage.
There's a whole section on the DCC Wiki on measuring track voltage that goes into some crazy stuff about multiple sine waves (maybe because the opposite is true - you can use multiple square waves plus a smoothing filter to generate sine waves - a DDS (direct digital synthesis) function generator does this). It's true you can represent a square wave as an infinite series of sine waves and harmonics, but if you integrate that all together to find area under the curve, it's the same as the peak voltage, or so close to it as to not matter. This is why a cheap non-RMS meter actually reads pretty close to the actual DCC voltage. A basic RMS meter will not, as it will assume all incoming AC signals are sine waves and display the RMS approriately - which is the wrong value for DCC. A True RMS meter also looks at the wave form, sine, square, or triangle wave, and calculates the approriate RMS, assuming it's fast enough (ideally it has to sample at least 2x the DCC frequency). There is no need for specialized equipment to measure DCC track voltage. AC current is a different matter.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.