Little help with a board.

[Martijn Meerts]

While we're building the big layout, my father wants to build a smaller one using my Tomix track/turnouts, and his Selectrix system. He's got a LOT of turnout decoders (enough for about 50 of them :)), but they're meant for the European style 3-wire turnouts (Minitrix, Fleischmann, Roco etc.). To get them to work with the 2-wire Tomix turnouts, it needs a couple of extra components.

 

I found the following circuit which is supposed to do the trick: http://encyclopedie.beneluxspoor.net/index.php/Bestand:E10.04-10.jpg

Diodes are 1N4001, resisters are 100k, transistors are bd679.

 

 

It does sort of work, but the turnouts barely make the switch. Initially with a 15 volt, 3 amp power supply (a Lenz one meant for their turnout decoders) it wouldn't switch. Trying the one we used on the old layout, a 23 volt, 3 amp power supply, it did work but just barely. I just know I'll get problems with turnouts refusing to switch if I leave it as is.

 

Once, a long time ago, I studied car mechanics, which included electronics like this. Unfortunately, I've pretty much forgotten it all. So, can someone with more knowledge have a glance, and maybe come up with ideas to improve it? I guess the output needs some more oomph...

[inobu]

Martijn,

 

I think the Kato and Tomix turnouts require a DC pulse. The momentary pulse energized the coil which creates a magnetic field, that magnetic field pulls the switch as the switch is connected to a permanent magnet.

 

Apply voltage, create a magnetic field, magnetic field attracts the magnetic connected to the switch armature.

 

The question is does the BD679 create a momentary pulse at the correct voltage?

 

I think the switch can be thrown with a 9v battery that bd679 may be doing something else.

 

Inobu

 

 

   

[CaptOblivious]

Funny, I've just been working on this kind of thing.

 

My Tomix points respond to 12V, ~75ms square pulses, about 1-2A (not sure how much, but I can't do two at a time on my 2A power supply).

 

FWIW. I'll look at more detail later. I'm using an L298N H-Bridge to generate the pulses, though.

[KenS]

I can't quite tell what's going on with that circuit (the diodes seem backwards, which suggests that I'm confused about something).  But basically it looks like it's using the red and green upper inputs as controls to reverse the current one the lower two (assert red, get positive on one of the outputs, assert green and get positive on the other).

 

Inobu is correct about the function of the switch machines: they need a pulse to operate.  Or rather, they need a current to operate, but too long a current will damage them.  Ergo, a pulse is the best choice, as long as it contains enough power to throw the solenoid.

 

But it looks like this circuit provides a continuous output, suitable for something like a Tortoise switch machine which is always-on.  The Kato (and Tomix) solenoids take a momentary current pulse of 12V (not a huge one in terms of current like some turnouts, but not small either).  Apply a continuous current of more than a second or so and you're going to burn out the solenoid.  Unless you're sure this circuit was intended for Kato/Tomix use, or you're sure the input in momentary, I wouldn't recommend using it.

 

Kato's blue switches are basically a metal spring that momentarily applies current in either polarity, then springs back to cut it off.  There's enough current from their power pack to provide the necessary pulse in that time.

 

The BD679 basically acts as an amplifier, with a very small current on "B" causing a larger current from "E" to "C". It's not inherently self-terminating.

[The_Ghan]

Martijn,

 

I think you've got the wrong end of the stick here!  How about modifying the point motor?  I haven't looked into it myself yet, but I know the motor has two coils.  I'd take a look at modifying one if I were you.  I'm guessing, but turning those two coils into separate circuits can't be that hard.

 

Cheers

 

The_Ghan

[Martijn Meerts]

Inobu, the circuit does throw the switches, so obviously it's doing something right, it's just not throwing them with enough force which means there's too much of a chance for the switch to not throw at all during regular operation.

 

 

Cap, I can't really measure anything, since all I have is a cheap multimeter which is too slow to catch the pulses sent out by the circuit ;)

 

 

Ken, the pulse is generated by the turnout decoder which connects to the 3 inputs at the top. The turnouts decoders are designed for 3-wire turnouts (1 wire for straight, 1 wire for curved, 1 wire for power). The circuit itself is just to go from the 3 wire system to what the Tomix turnout needs.

 

 

TheGhan, I hadn't considered that actually. I could try modifying one, the question is whether 1 coil generates enough of a magnetic field to switch the turnout.

 

 

 

For what it's worth, the original circuit from Lenz for use with 1 of their turnout decoders. They also sell pre-made version of this thing, but they're around 10 euro each... Anyway, you'd expect Lenz knows what they're doing :)

[inobu]

Inobu, the circuit does throw the switches, so obviously it's doing something right, it's just not throwing them with enough force which means there's too much of a chance for the switch to not throw at all during regular operation.

 

 

Martijn,

 

I think "right" maybe be the operative word. The setup is doing something but its not doing it "right" (as far as the switches are concerned).

 

From an operational stand point the Tomix and Kato switch needs a magnetic field to pull the magnetic attached to the points armature from one side to another. It is the correct voltage/amperage that will create the electromagnetic field designed to draw the armature open or closed.

 

You can actuate a Kato switch with a 9V battery. A 6 volt battery will not work.

 

I don't think that unit is producing the correct voltage to energize the winding wrapped around the coil in the switch.

 

Based on what you initially describe the 15 and 23 volts @3A cycle/test.......most likely that fused the windings on the core and impaired/destroyed the solenoid. The wire turn density, wire gauge and voltage all plays a role. I got a feeling that switch took a beating.     

 

With an electromagnetic field there is a transition point where it stops being a electromagnetic field and becomes a heating element.

 

Doubling the current passing through a wire will increase the heat generation 4 times.

 

Take it for what its worth but try a 9v battery and see how another switch responds. It will snap.

 

Inobu

[Martijn Meerts]

The circuit didn't destroy the turnout motor, the coils are just fine. In fact, you can hold your fingers on the solenoid while operating the it, and it won't heat up at all. I'm also 100% certain the turnout motor still works because it does indeed switch with a 9v battery.

 

The turnout switches completely, even 2 turnout motors hooked up to the output of the circuit switch. In about 50 attempts throwing the turnout with the circuit, it switched 50 times. It just doesn't snap from 1 position to the other, so I believe there's a chance it at some point won't switch completely, and cause havoc.

 

What it looks like to me, is that the circuit does do what it's supposed to do, but it's not outputting enough power, or rather, it's outputting JUST enough power to throw a switch (with the more powerful supply)

[The_Ghan]

Hey Martijn,

 

Are you going with the circuit?  I've got a spare motor at home and can look into it over the weekend if you like.  I can wire it to a Digitrax DS-64 as a 3-wire (+ + -) and see how it goes. 

 

Let me know if you'd like me to have a go.

 

Cheers

 

The_Ghan

[Martijn Meerts]

Ghan, had a quick look at modifying the turnout motor for 3 wire, but it's not that easy.. Not all that much space, and there's just no telling if 1 coil is enough to reliably throw the switch. I also might want to use them for an analog layout at some point using standard Tomix switches for the turnouts.. (I want to at some point do a completely analog layout with Tomix controllers and turnout switches etc. :))

[The_Ghan]

OK Martijn,

 

I'll stick to watching the cricket then ... 

 

Cheers

 

The_Ghan

[Martijn Meerts]

It seems getting it working was as simple as replacing the resistors with smaller ones ..

 

I did some more searching, and noticed there were several drawings, some showing 100k ohm resistors, others 100 ohm, and yet other 10k ohm.. I figured I'd just give it a go, and replaced the 100k ohm ones with 47k ohm ones. Now the turnouts switch just fine ;)