Connecting two loops with back to back 022's

[Paulct]

Please forgive my 3-rail novice skills! I've done a lot of searching so far, but can't seem to get this right. I co-run our annual Great Trains show here in Wilton, CT and this year we have built a new O layout. Its all 027 track. 8x12 layout. It consists of basically two loops (more detailed inner loop, but for electrical reasons, just consider it two loops). We connected the two loops with back to back 022's so we could switch trains between loops if needed. We have multiple transformers we can use. For now, we used a large ZW (I think?). Had the "U" posts on the outside rails, the A on the inner loop and the D on the outer loop. We insulted the CENTER pins on the switches to isolate the two loops. Just can't get this to work. The switches sit there and hum, almost like it is trying to do the auto switch thing! What's the trick? We realize that there will be some challenges with "timing" in getting a train out of the inner loop when there is already one running on the outer. But can someone just tell me the basics for wiring this up? After two hours of reading links, I need to ask the pro's. I'm an HO DDC guy, but am trying to learn this 3-rail stuff. Gotta admit, its pretty cool! Thanks, Paul

[ADCX Rob]

If you are indeed crossing over with a pair of 022 switches(or 1122/1122E and their descendants), you will need 3 fiber/insulating pins at the juncture... one for the center rail, and one for each of the two non-derailing trigger rails - ALL THREE.

[Michael6268]

Yep. What he said. All 3...

[Paulct]

Thanks Rob! Just got back from testing it, and it worked. Paul

[lionelsoni]

I would advise you to arrange your transformer connections so that you can power both loops from the same transformer output turminal when crossing between loops.  You can do this with a single-pole-double-throw switch to connect the center rail of each loop to either A or D.  (Or even leave one loop's center rail permanently connected to, say, A, and use just one SPDT switch to connect the other loop to A or D.)

The reason for this is that, if you neglect to set the A and D controls nearly exactly to the same voltage, a large fault current will flow as the locomotive (or lighted-car) pickups connect the two loops together when traveling through your crossover.  This current does not flow through the circuit breaker, so it will not trip to protect your transformer.

[phillyreading]

Are you using true 022 switches or are they the 027 switches? 027 switches will hum if you have a car parked on them. The post war 022 O gauge switches have built-in protection to keep from humming or harming the switch. As for insulating pins from what you describe you will need at least 4, one on each outside rail that controls the switch and one at each end of the loop for the center rail.

[Paulct]

The are true 022's. It seems to be working fine now with all three pins isolated in the middle of the two switches. The anti derail feature is also working. I was able to switch a train from loop to loop by matching the transformer settings. I have noted the warning about current spike from the other poster. Thanks.

[lionelsoni]

I count five, two per turnout for the control rails and one to separate the center rails of the two loops.

[BigAl 956]

lionelsoni

The reason for this is that, if you neglect to set the A and D controls nearly exactly to the same voltage, a large fault current will flow as the locomotive (or lighted-car) pickups connect the two loops together when traveling through your crossover.  This current does not flow through the circuit breaker, so it will not trip to protect your transformer.

Absolutely not true! If one loop is at 10 volts and the other is set for 20 the voltage will drop to 10 volts. That's all that happens. When the two loops are bridged by a car or locomotive it's as if you reached into the transformer and yanked out the extra coil of wire. In essence you are duplicating the power terminal.There is no 'fault current'. Current flow is from the center rail to the outside rail regardless of the difference between the controls.

Don't believe me? OK I'm going to set this up and video my tests. However Lionel sets have been run this way for over 75 years.

[lionelsoni]

"When the two loops are bridged by a car or locomotive it's as if you reached into the transformer and yanked out the extra coil of wire."  Al, you seem to be saying that, if I connect together two points on a transformer secondary, no current flows between those points.  It's as if the secondary winding turns between them had disappeared.  That should be true for whatever two points I choose, right?  So, what if I connect two points on the center-tapped 240-volt secondary of the transformer on the pole behind my house?

The end of the winding that has the black wire goes through a circuit breaker at my service entrance, just like the U terminal of a ZW; I'll leave that alone.  The white center-tap neutral doesn't go through a circuit breaker, just like the A terminal of a  ZW;  I'll leave that alone.  The other end of the winding that has the red wire also goes through a circuit breaker; I'll put a shunt around that breaker so that it is like the D terminal of a ZW.

Now I've got a higher-voltage version of a two-block layout:  The black wire is the outside rails, the white wire is the 120-volt center rail of one loop; and the red wire is the 240-volt center rail of the other loop.  (The trains are running faster on the red loop.)  Next I'll simulate a train passing from one loop to the other and connecting the center rails together as it does, by connecting the red wire to the white wire.  Do you believe that no fault current will flow?  Do you doubt that the red and white wires, with no circuit breaker in either of them, will fuse?

If anyone should try this experiment, leave the circuit breakers in place.  When the fault current in the red wire trips the breaker, that will prove to you that there was a fault current, without burning down the house.

[servoguy]

Bob, your advice on all this is absolutely correct.  As an electrical engineer with 51 years of experience, I agree with everything you are saying.

[BigAl 956]

Just to be clear I am saying on a transformer with two controls like a KW with an A and B for two separate tracks, if you short A and B together and crank one up all the way and the other half way nothing will happen.

The voltage of the higher control will drop to the voltage of the lower control. You can leave it like that all day. Nothing will happen. Try it I dare you! 

TQt BLKnPNQ

[lionelsoni]

If I had a KW, I would have done that already.  I do have some Zs; and I have done it with them and measured high fault currents.  Al, are you saying that you get no fault current whatsoever?  Have you tried to measure it?  Just connect an ammeter between A and B.

Of course the voltage drops, with a dead short across a part of the secondary winding.

Actually, the worst case is not with one control in the middle, but with one at the minimum non-zero voltage, which is 6 volts for the KW, and the other all the way up (20 volts).

Have you noticed the fine print in the Lionel service manual for the KW (which actually applies to many other Lionel transformers too):  "Note that the circuit breaker does not protect binding post combinations A-B, B-D and C-U"?

[webenda]

BigAl 956

...Lionel sets have been run this way for over 75 years.

True, our transformers seem to take momentary shorts in stride. However, park a train in neutral (different voltages to track sections still on) with rollers bridging the gap long enough and something happens. You can't see it, but you can smell it.

And don't touch the rollers unless you like roller branding your fingers. (OK, exaggeration, my fingers healed without branding marks.)

Edit 1: Added "in neutral (different voltages to track sections still on)" 10/22/2014 web

[BigAl 956]

I think the 'engineers' on this forum need a refresher course on Ohms Laws. Think I=E/R (where E=0)

This is really old school stuff. Rather than continue to hijack this thread with a transformer current discussion I'm going to create a new video and thread to explain why you can bridge multiple track outputs on transformers with a common return and not experience a short or fault.

[webenda]

BigAl 956

I think the 'engineers' on this forum need a refresher course on Ohms Laws. Think I=E/R (where E=0)

You are correct of course for what I posted. Train stopped would mean zero volts, were it not for the neutral position in our engine controllers. Failure to communicate on my part. I will go back and edit my post.

[lionelsoni]

Al, I did notice in your video that you seemed to think that, because you measured no voltage between the ends of the wire that you had connected between A and B, there must be no current flowing in it.  Ohm's law, solved for voltage, says

E = I * R

There is no violation of Ohm's law when the current is finite and the voltage and the resistance are both zero, as in a perfect conductor. (There are perfect conductors, with truly zero resistance.)  And, if R is small, so is E.  Your conductor is not perfect; but it is good enough that you simply didn't detect the voltage caused by the current that was flowing in it.  That's why the best way to detect whether there is a current in a pretty good conductor is with an ammeter, not a voltmeter.  I assume that you have not yet tried that.

If you look at my posts, I think you will find that I have not claimed that I am right by virtue of my being an electrical engineer.  I resent your implication that I have made any argument from authority.  Please let your arguments stand on their own merit and avoid ad-hominem attacks.

[servoguy]

I just ran the test that Big Al suggested using terminal A and terminal D of a ZW.  The results were what Bob Nelson and I predicted.  With A set to 10 on the dial and D set to 20 on the dial, the voltages read 11.3 and 20.25 respectively without A & D connected together.  I put an Amprobe clamp on ammeter on the wire between A & D, and then shorted A & D together.  The current went to 28 amps.  This is enough to cook the secondary of the transformer, so obviously I didn't leave A & D connected very long.  During the time A & D were shorted, terminal A voltage went to 12.5 volts.  I didn't measure the voltage of terminal D.  It would have been higher than terminal A due to the resistance of the connecting wire and the internal resistance of the transformer.  

Big Al, if you want to run this test and leave the terminals connected for some time, I suggest you do it outdoors so you don't set your house on fire.

[ADCX Rob]

Thank you, Bruce.

[webenda]

BigAl956,

I appreciate your video. I love experimental proof of a concept or theory. Good job!

You said in the video, amongst other things, these two interesting statements,

STATEMENT 1: " At this point, if you look on the top, I have set the B throttle almost all the way up and I am going to put my meter from U to A and you will see it measures 20.4 volts. Also, I'm going to look at my A throttle, which is turned down right now, and since the A and B are shorted together, it too measures 20.4 volts, even though A is turned off."

At the time you said, " Also, I'm going to look at my A throttle, which is turned down right now…", the video showed you measuring from B to U.

STATEMENT 2: "If you have one control on one block of track and another control on another block of track, they are set at different voltages, and a car crosses between them--creating a momentary short--what will happen is the voltage will drop on both controls to the lower setting."

Those are two contradictory statements. According to statement 2, the voltage at both A and B should read the lower value. You said, in statement 1, A was turned down right now. Therefore A is the lower value. The lowest value, with respect to U, is 6 volts. For statement 2 to be true, the voltage measured in statement 1 would have to be 6 volts.

I offer this simplified schematic of a Lionel KW transformer for you to study.

[webenda]

BigAl956,

Based on results of the experiment you so kindly performed, videotaped and posted on YouTube for us, you have a faulty circuit in your transformer between terminal A and the transformer windings. Possible points to trouble shoot are; connection to back of A terminal, Reverse switch, Horn and Bell switch,the roller and wire connections to all of these.

[webenda]

Paulct,

I am very happy Rob (and Michael) solved your problem right off the bat, before this discussion on high currents (when a train crosses from one loop or block to another if there is a voltage difference) started. Good work Rob (and Michael.)

[servoguy]

Here are Lionel's comments re connecting A & B together on a KW:

http://pictures.olsenstoy.com/searchcd31.htm?itm=657

Look on page 2 for the wiring diagram for a KW and for Lionel's comments