I read the article title "Improved hands-free uncoupler control" in the Nov'2015 Model Railroader and was greatly intriqued by it. I plan on making one of these for my N scale layout and many thanks to Jean Piquette with sharing the design.
Now, while it has been over 40 years since I had my circuits classes in college, I was able to understand how it worked up to the use of diode D2.
Can anyone explain why D2 is needed? I don't doubt that it is needed and I will build it as designed, curiousity is driving my question.
D2 is to shunt the BEMF from the Kadee uncoupler coil. It's maybe not the cleanest schematic I've seen, but if you use where it connects, it's across the coil leads. With any sort of electromagnet, when the power is killed and the magnetic field collapses, it generates a potentially large surge. You'll see the same thing where there is a relay controlled by a transistor.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
another way to think of it is to compare a coil to capacitor. A capacitor maintains a voltage while a coil maintains current.
When a circuit containing a coil is abruptly broken, a diode provides a path for the current in the coil to circulate through the coil and diode until depleted by resistance.
In a capacitor, the current and change in voltage (dv/dt) are related. In a coil the voltage across the coil (back emf) and change in current (di/dt) are related. Breaking the circuit and abruptly changing the current (di/dt) results in a very high voltage.
this same principle allows charge-pumps, switching power supplies and car ignition systems to generate higher voltage than their input voltage.
greg - Philadelphia & Reading / Reading
Thanks to both Randy and Greg for great responses.
It is spooky how these things come back to you after all these years, but every word of their responses made sense from what little I remember from college.
I've got my parts on order, can't wait to get started assembling the thing.
I am putting this circuit together and it appears that there is a mistake in the parts list. The parts list calls for a Radio Shack 275-1555 momentary push button. The text also calls it a momementary button. However, my continuiity tester tells me that the 275-1555 is not momentary. When the button is pushed it switches to the other position. Anyone else addressed this problem.
275-0609 is a similar size momentary pushbutton. Not sure why the circuit designer would have used a SPDT toggle pushbutton when the idea is to only energize the uncoupler magnet for a set amount of time. I don;t think Kadee even recommends leaving it powered continuously as would happen with the specified part. The trick of the circuit is that after pressing the MOMENTARY button the magnet stays energized for a few more seconds so you don't have to hold the button with one hand, drive the train with the other, and juggle your paperwork with the third...
Circuit only 1/2 works, why? I built the circuit largely with patch cords to test it before putting it on the layout. It 1/2 works. What I mean by that is the magnet is not forcing the coupler parts to separate enough to achieve clean decoupling. When I measured the voltage at the magnet leads it is only 12 volts and less than 1.5 amps. The voltage coming into the circuit is 18-19 V DC. Why would this drop in voltage happening? Would I lose that much voltage because of poor connections? Is there some other reason? I will appreciate any help you can give me.
Sounds about right. You have one diode drop for the reverse voltage protection (D1), that.s .7 volt, and the Vce for the transistor is 3 volts at saturation, so you are dropping almost 4 volts through the circuit.
Not a huge fan of this circuit, if I were doing this I'm probably take one of Rob Paisley's simple 555 time circuits for the time delay, and use the output to drive a power MOSFET to switch the Kadee coil - or even a relay. The Kadee coil would have its own power supply independent or what drives the circuit. There is probably the exact circuit on Rob's site, except it probably lights up an LED for x seconds. The time is adjustable by tweakign resistors on the 555, and the LED could be repalced by a relay that turns on the Kadee uncoupler. Two diodes would be needed, one across the relay coil to protect the 555 and one across the Kadee coil to keep the relay contacts from arcing. No need to hold the button or anythign with a circuit liek this, you push the button and the coil stays energized for some number of seconds and then turns off, quick press or long press of the operating button would not matter.
Randy -- many thanks. My electronics experience is limited to 2 courses at the local tech collge and working through the Make electronics workbook -- so I can hook up a 555 timer -- but not really technology savvy as you are so appreaciate your help. It sounds like an excellent idea. When I look at Paisley's site -- are you referring to the #20 Time Delay circuits? Many thanks.
Yes, what I think would work for this is #2, the negative recovery circuit. When you push the button, the output line goes high for as long as you hold the button PLUS the time delay factor - when you let go of the button, the output stays high for an additional time calculated by the values of the resistor and capacitor (the 1 Meg and 1uF caps in the circuit - the 10K one is a pullup for the button.)
You would connect a relay between the output and ground, with a diode reversed across it to protect the 555. You need a relay with contacts capbale of switching however much current the Kadee uncoupler needs, and the relay coil voltage would need to be about what the power supply voltage is (for the timer circuit - this can and should be an independent power source from the uncoupler electromagnet). If you also go down further, #30 the power off delay is a little more complex but exactly what you are looking for - use a momentary button for the switch. The difference is that #30 also kills power to the 555 when it shuts off.
After fiddling with the original circuit for quite a while it is now working reliably -- for now. If it starts giving me trouble I will switch to the 555 circuit you recommend. Again, thanks for the time.