Yup, and depending on the mounting, there may be little reason to adjust the throw - certainly not for each and every unit, once you figure out a value that works.
There is a commercial product that also uses the double relay system. They offer it as an option - you can get their unit with no relay (use a microswitch, or a Frog Juicer), one relay (most DCC friendly turnouts truly only need a single relay, as the point rails at least are electrically tied to the adjacent stock rail and so never have opposite polarity from the stock rail they are touching), or two relays, working as mine do, where power is completely removed before movement starts, then the polarity is set after movement stops, and finally power is restored. The older non-DCC friendly Shinohara turnouts are the ones that most need this, because both point rails are the same polarity, and the same polarity as the frog, so if the frog polarity changes before the point rails movbe, there's a short, or if the point rails finish mocing before the polarity changes, there's a short.
I had the extra outputs, so i added the feature. Removing it does ntoo free up enough IO pins to make each board control 3 servos, so I figured why not. If you don't need it, you can leave off the relay, drive transistor, and resistor.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Interesting. So am I correct in assuming by using the two relays no microswitches are required? Would that mean the only required adjustment would be servo throw and servo throw adjustment is in the software?
Lee
Been a while, but I finally had time to work on it, and I now have a breadboard prototyype fully functional, with completed software.
A little different than my original design, this one now has 2 relays per turnout, one that turns off the frog power, and one that sets the polarity. Soo even touchy turnouts like old Walthers/Shinohara can work with no special mechanical alignment of the relay. When an input is received, first the frog power is cut off, then the servo moves to position, then the frog polarity is set, the LED indicators are set, and the frog power is applied. No chance of a short.
There are also remote lock and remote move inputs, these will come from my sort-of CMRI node boards when I get those designed.
Position is saved by writing to the microcontroller's EEPROM. Writes are spread out so that by the time any position hits the datasheet recommendation of 100K writes, withg normal usage it will be more than 50 years from now. So I'm not worried.
You can leave off parts you don't need. If you don;t need frog polarity, leave off the relays. if you just need polarity but not the power control, you can leave out the one relay. If you don;t need remote control (I plan to use the same circuit for my yard) you can leave off the connections and protection resistors for the remote inputs. Or no buttons, you can leave off the components associated with the buttons.
I will be demonstrating the prototype this Friday night and Saturday at the Reading Modelers Meet. While the focus of the show is on prortpye modeling, the guy running it said to bring my stuff. I doubt too many will be interested, but I will be there anyway. He said something about a clinic next year - perhaps I will have the PCB for this circuit plus at least a protoype working for my 'sort of' CMRI nodes that will drive the turnouts, signals, and read block detectors.