Kato turnout controller using capacitors

[Maletrain]

I have some Kato turnouts that I want to control from the fascia, instead of using the bulky blue Kato manual devices.

But, I also don't want anybody to lean on a momentary switch and burn out my turnout coils.

I have looked at what Stillwell developed, and am thinking about using a similar, but not quite the same approach.  I am thinking the circuit would use a capacitor that is either connected across the power supply to charge-up or connected across the turnout coil when throwing the turnout.  A SPDT momentary contact pushbutton switch would use the normally closed contact to charge and maintain charge in the capacitor, and pushing that switch would disconnect one lead from the power supply and connect the capacitor through the turnout coil with the normally open switch contact. 

A DPDT switch in the circuit loop to the turnout coil would determine which way the turnout points move.  I am happy with the DPDT switch position indicating where the points should be.  I know that the turnout manual slider switches could be used to put the points in the position not indicated by the DPDT switch, but a simple push of the control button could  assure correct turnout positioning without needing to actually look at the turnout, if there was any doubt.

I like this design better than Stillwell's approach, because it does not leave the Kato turnout solenoid connected to the power source all of the time.

I am wondering about the voltage for the power supply (DC of course) and the values of the capacitor(s) to use.  Stillwell seems to settle on 12 VDC and 1000 mfd.  My experience with the Kato mechanical controllers using various input voltages is that I need more voltage for throwing multiple turnouts, such as 2 for crossovers, and particularly the 4 in the Kato double crossover part.  My thinking is that I should be able to pick a voltage that will reliably throw a single turnout with a particular capacitance, and then simply add more capacitors in parallel for more coils to be controlled by a specific push button on the fascia.

So, is 12 volts DC and 1000 mfd what I need to reliably throw one Kato turnout?  I have read that some others were doing well with as little as 500 mfd, but did not tell me exactly what the power source was for doing that.  I know I can throw the whole 4 coil double crossover reliably with the Kato manual device and an old Radio Shack wall wart labeled 14.5 VDC 800mA, but it really puts out more voltage to an open circuit.  Lower voltage wall warts that I tried first had failed to throw all of the points in the Kato double crossover. 

So, what can others with real experience here tell me about using capacitors to throw Kato turnouts?

[peteski]

So you want a DC power supply to charge a capacitor through a resistor, so when it is called for the energy to be released into the switch machine coil, the jolt of energy stored in the cap is what the coil will see?  Just like capacities discharge used for standard twin-coil machines?

If that's the case then the current rating of the power supply is not all that important.  You could have a supply rated for 50mA, and it will charge the capacitor just as well as a 2A rated supply.  The series connected resistor is what limits the current supplied by the power supply if the coil is left connected for extended period of time.

You should design the circuit for the voltage specified for the Kato turnouts.  Probably 12V.  It is the size of the capacitor which you need to figure out (you can just experiment).  If the cap is too small, it will discharge too fast.  I don't use Kato turnouts but 1000µF sounds like it will likely work for throwing a single switch.

[Steveruger45]

Here is something using capacitors similar to what you are talking about.
https://www.ebay.com/itm/124111747142.

I’m wondering why be so complex when could do this by DPDT momentary switches like Mike Fifer.
https://www.fiferhobby.com/how-to-make-kato-turnout-control-switches/

[reinhardtjh]

I’m wondering why be so complex when could do this by DPDT momentary switches like Mike Fifer.
https://www.fiferhobby.com/how-to-make-kato-turnout-control-switches/

Because the circuit allows you to use regular non-momentary contact switches.  It also prevents you from burning out the switch coil if you "forget" to let go of the switch and allow the momentary part to work.  Or if the switch sticks on and doesn't go back to off.

[Maletrain]

Yes, my intent is to limit the amount of energy that can be sent to a Kato turnout solenoid with a single activation, whether it be by holding a toggle in the "on" position for too long or by unintentionally leaning against a push button on the fascia while reaching into the layout for some unrelated purpose (e.g., clearing a short on the track, rerailing a car, etc.)

The Stillwell design does that, too, but it also leaves the turnout solenoid continuously in the circuit, with the capacitor in series blocking DC current but not AC ripple.

Peteski brings up the business about having a resistor to limit the recharge rate of the capacitor by the power supply.  I guess I need to calculate that to keep the charging current below the power supply continuous rating.  For the 14.5 VDC 800 mA wall wart I am now using (with mechanical Kato switches), I get 14.5 V / 0.8 A = 18.125 Ohms and 14.4 V x 0.8 A = 11.6 watts.  But, the watts are not continuous - only the initial charging current, which should quickly go to zero.  That looks like not much of a resistor.  I am wondering if I really need it. Stillwell did not show one in his circuits, but he was running the current through the turnout solenoid, which he said was 22 ohms.  So, I guess I should put a 20 ohm resistor in series with the capacitor on the power supply side of the SPDT momentary switch.

[Steveruger45]

I personally wouldn’t hold down a momentary switch for more than one or two seconds but I do understand the risk and magic smoke possibilities if some one would.   Anyhow this peaked my own interest and I stumbled upon this that might be suitable.

https://www.rr-cirkits.com/Notebook/Kato-Switch-Machine.html

[reinhardtjh]

... the risk and magic smoke possibilities if some one would.

It's always the "someone" that's the risk...

Anyway, the circuit is to help keep it from happening.

[Maletrain]

It is really not that unusual for somebody to try to throw a turnout with the wrong panel switch.  So, while looking at the turnout they expect to move and not seeing it move, they may hold the switch in the "on" position while the turnout they are not looking at smokes its coil.  That is the main reason I don't want to rely on a knowledgeable operator instead of a limiting capacitor.

 I had seen the RRCircuits link, before.  I was wondering why it shows the capacitance as "2200/35V" instead of 2200/15V.  Stillwell used 2200/22ohms to get 1000 mfd.  Then, I looked at the picture and I see that it means the capacitors are 2200 mfd rated at 35 volts.  That is a lot of capacitor for a single turnout solenoid!  I suspect it is intended to throw 4 coils in a double crossover. 

I would not want to put 4 x 2200 mfd on a double crossover.  So, I am still wondering if anybody here has messed with this enough to know a good capacitor value per solenoid.  I am not an electronics guy with a drawer full of spare parts to experiment with, so I will have to buy whatever I need to do any experiments.  At this point, I am thinking to start at 500mfd and see if 1, 2, or 3 are needed per solenoid for the wall wart I am using now.

[peteski]

Do you have a link to the Stillwell diagram? Just to be clear Steve, my idea is not to have a large capacitor (with a bleed resistor in parallel with the cap), connecting in series with the switch machine energizing the  coil while the cap is charging.

Mine idea has the DC power charging a large cap through a series connected resistor, and discharging it through the coil.  Sort of reverse from the Stillwell circuit.  The resistor's value can be high enough so if the coil is left connected, only small amount of current will pass through the circuit. Of course if the resistor value is high, the cap will require several second to recharge.

Also, when we discuss electronic circuits and components, I feel that it is important to use proper nomenclature to prevent any possible confusion (and maybe even damaged components).

You keep mentioning "mfd". In my eyes that is not right on couple of levels:  First of all lower case "m" denotes "milli" not micro.  Second: Farads are represented by upper case "F", not "f".  A 1000mF cap would be 1000,000µF.  It is also accepted that the "µ" is not readily available on a keyboard, so "u"  (as un "uF") is generally accepted as a reasonable alternative.  It might be just me, but I don't like the ambiguity of "mfd".  I realize that someone with electronic knowledge will know the impliedvalues just from the context of the discussion, but someone with no electronic packround could get confused.

The 35V rating for the caps in that RRCircuits diagram is just a safety margin.  It is not like it will magically get charged up to 35V. I don't believe electrolytic caps are made rated for 15V. I usually see 16, 20, 25, 35V ratings.  Plus, you don't want to operate electrolytic (or any type of cap for that matter) at its rated voltage.  You need some headway for safety.

As for the AC ripple current damaging the coil I wouldn't worry too much.  Or to be safe use a filtered DC wall-wart (or even better filtered and regulated power supply).   Those are often available in surplus electronic stores and look like laptop power supplies.  Some cable  boxes came with 12V 2A filtered and regulated power supplies which are often available as surplus. Or like mentioned in another thread Dell laptop supplies produce higher voltages (like 18 or 20V).

[Maletrain]

Stillwell uses 4700 µF for a double crossover, based on the measured coil resistance of 5 ohms for the 4 coils wired in parallel, giving him 2200/5 = 4400 µF and going to the next higher available capacitor value.

So, Stillwell seems to be saying ~ 1000 µF per coil, maybe a little more.  But, he is also using a 12 VDC power supply, not a 15 VDC supply like the RR-cirkits link or my 14.5 VDC wall wart.

[Note that I found the ASCII character for "micro" so I have stopped using "mfd" for microFarads and started using "µF".  I can get the "µ" character by holding down the LEFT Alt key while I enter "230" WITH THE NUMBER PAD.] 

I had the Stillwell link this morning and printed it, but now I can't find it where I thought I saved it, nor with a quick Google search.  I'll find it again and post it here when I do.

[Maletrain]

OK, I found Stillwells file on my hard drive and attached the pdf here. [ Guests cannot view attachments ]

[Maletrain]

Here is a sketch of the circuit I am trying to develop parts specs for:
[ Guests cannot view attachments ]

[mmagliaro]

The limiting resistor on the recharge side is a good idea.  True, the massive current is very brief.  But you are using a momentary contact pushbutton, so every time you fire a switch and let go of the button, that massive instantaneous current zaps through the NC switch contacts for a few microseconds.  (And that current can be something like 50-100 A, believe it or not, even if only for a few microseconds.).  Just putting a 10 ohm resistor in there would limit it to 14.5/10, or about 1.5 A.  The only thing you give up is that the cap won't recharge "instantly".  But it will still recharge mighty fast, and your switch contacts will thank you.  1.5A x 15v = 22.5 watts for an instant.  So make that resistor something like a 10W power type.

[Maletrain]

Max, If I add more caps to power more coils with the same switch, I guess I should be, in effect, adding resistors in series to keep the peak current to about the same level, right?  So, for a double crossover with 4 coils, the capacitor would be something like 4000 µF and the resistor would need to be more like 40 ohms and a higher watt rating, too, right?  I am thinking just size the resistor for the 4 capacitors, whatever they turn out to be, and use that for any lesser number of capacitors, too.  I am thinking that the capacitor(s) would still recharge plenty fast enough, especially if they are a bit oversized for the coil current requirement.

Additional thoughts?

[Point353]

I don't believe electrolytic caps are made rated for 15V. I usually see 16, 20, 25, 35V ratings.

Vishay still has a few legacy ex-Sprague parts among their product offerings with 15V ratings.
For example:
https://www.vishay.com/docs/42036/39d.pdf
https://www.vishay.com/docs/42054/672d.pdf