i believe the following schematic for an auto-reverser is representative of most auto-reverser circuits derived from the MERG design (see ARC). I'm trying to understand the purpose of having two mosfets in each path which provide a circuit breaker function more than auto-reverse (see relay).
If you notice, the pair of mosfet (e.g. Q2 and Q1) are wired with polarity reversed of each other and each has a resistor tieing the gate to the source lead. (i couldn't find answers to my questions on the web).
Am i correct in saying that because of the alternating DCC voltage two mosfets are needed so that one of the gate resistors (R3 and R4) is at ground, shutting off the mosfet when Q1 blocks the gate voltage?
this is easy to see for Q2 because it's connected directly to the DCC supply, but Q1 (track side) would only see ground if there were a short on the track?
would the DCC voltage, when positive on Q2 actually turn on Q thru R3 and the circuit relies on the other path (Q9 and Q10) to block current through the track?
greg - Philadelphia & Reading / Reading
Most auto-reverser circuits are also designed to be circuit breakers, so your own explanation may be the correct one.
It takes an iron man to play with a toy iron horse.
Yes, you need the two to block both halves of the DCC waveform. Having all 4 mosfets in there is to cut off both rail outputs - there are some breakers that cut off only one side. However, in an autoreverse, if only one side is cut off, it could be the other side is the one bridges past the gap with a metal wheel or through loco or lighted car wiring to the opposite phase on the main, meaning the short would continue. So for an AR to work fully as a breaker, both sides have to be opened. Compasre a PSZ and a PSX-AR, the PSX has just 2 mosfets in one of the rail lines, the other is effectively passthrough except to tap off for power. The PSX-AR has 8 mosfets, two in each side for the breaker functionality and 4 more wired as a reversing switch portion.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
two MOSFETs are needed in each path because the current can only be controlled in one direction. However, unlike a BJT transistor, a MOSFET, as the symbol indicates, conducts current when the polarity across its source-drain is reversed. In other words, the gate controls the impedance of the juntion when the drain is at a higher voltage than the source. But current flows freely when the source is at a higher voltage that the drain. Only one MOSFET is ON at a time, but the other conducts through the "intrinsic" diode.
resistors pull the gate to the source voltage, shutting the MOSFET off when there is no control voltage applied through the diodes.
The gate voltage is supplied from the positive DCC voltage. On the path providing the negative DCC voltage, the gate voltage is significantly greater that the source voltage (-DCC). On the path supplying the positive DCC voltage, the source voltage must be slightly (~1V) less than the gate voltage which is already slightly less than the +DCC voltage. Therefore, the DCC voltage is reduced somewhat through each path, but more so on the +DCC path.
having 4 MOSFETS is only required to provide a circuit breaker function where the short can be on either/both paths. An auto-reverser requires quickly and temporarily breaking the DCC path until a relay or H-bridge reversers the polarity to the track and therefore only two MOSFETS on one path are required.
For me, building circuits is part of the joy of the hobby. Understanding how such circuits work creates inexpensive options. (if you build your own turnouts, why not build your own circuits).
rrinkerSo for an AR to work fully as a breaker, both sides have to be opened. Compasre a PSZ and a PSX-AR, the PSX has just 2 mosfets in one of the rail lines, the other is effectively passthrough except to tap off for power.
Don - Specializing in layout DC->DCC conversions
Modeling C&O transition era and steel industries There's Nothing Like Big Steam!
Most of the ones with relays. The PM42 breaks both sides, it uses DPDT relays, wired as a reverse switch. Fo using a secotion as an autoreverser, you connect both legs to each rail, so the rail A output is either rail A input or rail B input depending on which way the relay flips. For breaker use, you connect just the one leg of the relay to the rails so when it trips, it's an open circuit to the rails.
I think the original PowerShield broke both legs, have to dig up a picture of one.