Stan Trzoniec's article includes the suggestion of using one ZW output for level track and another for a grade up a hill on the same loop. Some of you could have predicted that I would object to that. The problem is that every time the train makes the transition from one block to the other, its pickups create a short circuit across a part of the ZW's secondary winding, unprotected by the circuit breaker.
I have long advised against this practice for level track, where there is at least a chance of setting the two blocks to the same voltage. But in this case, the voltages are deliberately set different. I don't know what his settings, track length, and feeder arrangements are; but here is an example of how things can go wrong:
Suppose that the voltage difference is 5 volts and that the total resistance from the ZW's A terminal to its D terminal, through the train's pickup, is 100 milliohms. This could be about 25 feet of center rail or 30 feet of 14 AWG feeder, or a combination. Now 14 AWG would ordinarily be safe for wiring a ZW, but not in this case. The fault current when the pickup crosses the gap is 5 volts divided by 100 milliohms, or 50 amperes.
The ZW may well be able to supply this, since the low voltage of 5 volts means that the power drawn is only 250 watts, not a great overload for the primary winding. But the power dissipated in any 14 AWG feeders will be about 10 times what the wire can stand: Wire is rated according to the current that will not heat it to the point of melting the insulation. The current here is 3 1/3 times the wire's rating; so the temperature rise is over 10 times what the insulation can tolerate, assuming of course that the layout was wired with 14 AWG.
All of this is guesswork, since we don't know the particulars of the actual wiring. But the numbers I used are plausible, and the situation could be dangerous if the train ever stalled across the gap.
The other tip that my "1/2" refers to is number 7, wiring a signal to a turnout. This is a good idea; but it should be noted that it does not work for Lionel O27 turnouts, which also have 3 terminals, albeit in a different order. The problem is that it depends on the the internal contacts that shut off the solenoids of 022 and similar Lionel turnouts. (Some brands of O27 turnouts also have these contacts and can be used in this way).
Bob Nelson
I have heard this before. While I understand the logic behind this fear I can tell you that this never happens. Most model railroad hobbyists at one time or another park trains on sidings and across different blocks where there is a difference in voltage and there is no damage whatsoever. Lionel enthusiasts have been doing this for centuries now and there is just no credible evidence of this practice being a problem. Recently on another forum we had a few operators trial parking various cars and engines across different blocks set at different voltage levels. In no case was any damage or overheating reported. The current flow was negligible. Maybe there is an EE out there who can explain why reverse current flow does not occurs .
A low-enough voltage difference, a high enough track or wiring resistance, and a short enough duration all can make the likelihood of damage arbitrarily low. Many modern "transformers" will not have the problem at all. You will have to give the specifics of the setup for anyone to explain a particular result. As I pointed out, these details were not given in the article; so I made up some plausible ones. The danger is that someone will take the "tip" to be unconditionally safe and try it on his own layout, with bad results.
Lionel once effectively admitted that the circuit-breaker arrangement that their postwar transformers shared was flawed, in the fine print of the service manual for the KW: http://pictures.olsenstoy.com/cd/transfmr/pskw2.pdf
I have an interesting exhibit. It is the actual transformer out of a type-Z that I bought for parts. That transformer lacks a whistle control but is otherwise very similar electrically to the ZW shown in the article. There is a section of the exposed secondary winding that is badly burned. It is in the middle of the winding, with undamaged turns beyond it at either end. I think it is pretty clear that the fault current that must have burned it had to flow through two of the rollers stationed at either end of the burned turns. Otherwise the same current would have flowed throughout the secondary winding, and it would all be burned. So at some time there was a short circuit between two of the output terminals; and a current flowed, long enough and heavy enough, to severely damage the transformer; but the circuit breaker did not trip. I don't know how to post the pictures here; but I will happily e-mail them to anyone who would be willing to do that for me.
(It's not a "reverse" current, but a fault current.)
(My job does not require me to be licensed, but I hold BSEE and MSEE degrees and have 47 years of experience.)
Bob,
I follow your advice religiously. I am on the other side of young with no academic electrical knowledge. Some argue that your advice is somewhat theoretical and, therefore, not applicable in the "real world." I favor your approach. The practical aspect is, I think, undeniable and I have not seen an answer to the following basic question. Please excuse me you have previously offered such advice. I have searched and have not been able to find it.
Is there a "convenient" way to safely move a locomotive, on track, from one power source to another as desired? Of course "convenient" is a relative term. I suppose I'm looking for a solution to a recurring issue that the unknowing can apply :-)
Best,
John Mucci
John Mucci, Meredith, N.H.
You do this by isolating sections of track into blocks, then using switches you chose which power source is supply power to each block.
https://brentsandsusanspicutures.shutterfly.com/
I, also, have a BS in math and an MSEE with 50 years of experience. Bob Nelson and I are in perfect agreement about the damage fault currents can do. I suggest everyone follow his advice and avoid smoking your layouts.
I have heard of passenger cars with two rollers burning the wiring because one roller was powered by one transformer output and the other roller was powered by another transformer output.
I have owned Lionel trains since 1950 and have many years of experience with them.
While this is good information to have, in all my years of running 3 rail trains I have never had a problem running trains from one power output to another. As long as they are in phase of course.
"IT's GOOD TO BE THE KING",by Mel Brooks
Charter Member- Tardis Train Crew (TTC) - Detroit3railers- Detroit Historical society Glancy Modular trains- Charter member BTTS
Is there any way to get the benefit of applying different voltages to the level and inclined track without the risk of excessively high fault current? Perhaps a resistor in series with the feed to the level track block?
Thanks,
George
That's how to do it, George. But, even better than a resistor, whose voltage drop will depend on the current that the train draws and therefore may require readjustment for different trains, you can use anti-parallel connected diodes, which will have an almost-constant voltage drop.
The basic unit is two rectifier diodes with rated forward currents at least about 5 amperes. Connect them in parallel with the diodes pointing in opposite directions. Put this in series with the center rail of the level track to get about 1/2-volt drop. Put more of these pairs in series to get a greater drop.
You can make the equivalent of two of these pairs, with a total drop of about a volt, from a bridge-rectifier module: Connect the + and - terminals together and use the other two (~) terminals to wire the module in series with the track. As before, add more modules in series to get a greater drop. If you need a finer adjustment, use the +- junction of one of the bridges as a 1/2-volt tap. (Note that the bridge-rectifier module is not being used here as a rectifier, just as a convenient way to get 4 diodes in one package.)
Wouldn't a time delay fuse or circuit breaker on one (or preferably both) of the throttles protect against this?
J White
It would make the setup safe, assuming that the track feeders and any internal wiring connecting pickups together in the locomotive or lighted cars is heavy enough to carry the rated current of the added circuit breakers. It's the way Lionel should have designed their transformers, rather than skimping by putting a single circuit breaker in the common return wire.
But the overcurrent would still be there; and so would any arcing or welding caused by it. An intermittent short circuit is also a good way to generate high-voltage spikes, of the sort that often do in modern electronics-heavy locomotives.
lion888roar,
I can understand two different power sources for two different isolated blocks.
It's my understanding that lionelsoni is referring specifically to the issue of straddling blocks where pickups will, at least momentarily, reside in two different isolated blocks.
Perhaps I do not understand this, but it sounds to me the only way to avoid the issue, referred to by lionelsoni, is to create a scenario where power would be regulated to on/off almost immediately. A locomotive or powered car entering block A from block B would need special handling as it crossed between blocks A and B.
When the locomotive enters block A (the lead pickup crosses the line), power to A would need to be off until the trailing pickup enters block A. At the moment, and not before, the trailing pickup enters block A, the power to block A would need to be on. I could see this happening manually be coasting a locomotive over the dividing line and only providing power to block A when all locomotive pickups are in block A. This would be impossible to manage manually if lighted passenger, or other powered, cars were part of a consist.
I "think" I understand lionelsoni's objections, but am not positive. I am sure I don't know how to implement a solution to the problem I "think" he is describing. If there were a viable solution, it seems to me that a recurring discussion might be put to rest?
John
This is a problem for prototype electric railroads too. They have to go between sections powered from different sources. They avoid connecting those sources together by coasting through an unpowered stretch of track.
George posted a very practical solution above. That is to drop the voltage on the level track with a series resistor. I further suggested replacing the resistor with diodes, to keep the voltage drop constant. In either case, the important thing is that connecting the two center rails together between blocks does no more than place a short-circuit around the passive voltage-dropping element, whether resistor or diode network. This is harmless.
The more common situation, which is not the one here, is moving from block to block between two supplies set to the same voltage, or intended to be set to the same voltage. It is better to avoid this situation entirely, by doing what lion88roar suggested, switching the blocks to be powered always by the same source. With two sources, this is easily done with a single-pole-double-throw-center-off switch per block.
John,There is no 'safe' way to move between two tracks powered by separate transformers (as far as I know) on the fly.My layout has two mainlines on the lower level connected by 4 switches in a crossover. Two switches on the outer loop and two switches on the inner loop, these switches are at the end of one block on each track. There is also a reversing loop inside the inner loop that is contained within the same block as the crossover switches with a shutdown toggle, and a switch leading to a grade to the second level. Each lower level main line also has a separate block.When I want to transition from the upper level to the lower level I have to:1. Park the train on the inner mainline inside the non-reversing loop block - disengage power to this block2. Change the source for the reversing loop to the source for the second level3. Change the source for the transition grade block to the source for the second level4. Engage power for the grade and reversing loop5. Throw the switches to the transition grade and reversing loop6. When the train is on the lower level 1. Park the train in the reversing loop 2. Disengage power to the grade block 3. Disengage power to the reversing loop 4. Throw the grade block switches to normal 5. Change the source for the reversing loop to that of the inner mainline 6. Engage power to the reversing loop 7. Engage power to the secondary inner mainline block 8. Move both trains on the inner mainline
You have to pay attention to what you are doing, but having the parked trains' blocks powered down ensures you will not 'cross' transformers.
My setup is somewhat simpler. I have a yard and two loops with a double crossover. The yard is a block; and each loop is divided into two blocks. So I have 5 toggle switches, one for each block. Whichever of my two transformers a train starts with, it stays with that transformer as it moves about the layout, as I throw the toggle switches to assign the blocks ahead to that transformer.
lionelsoniI have an interesting exhibit. It is the actual transformer out of a type-Z that I bought for parts. That transformer lacks a whistle control but is otherwise very similar electrically to the ZW shown in the article. There is a section of the exposed secondary winding that is badly burned. It is in the middle of the winding, with undamaged turns beyond it at either end. I think it is pretty clear that the fault current that must have burned it had to flow through two of the rollers stationed at either end of the burned turns. Otherwise the same current would have flowed throughout the secondary winding, and it would all be burned. So at some time there was a short circuit between two of the output terminals; and a current flowed, long enough and heavy enough, to severely damage the transformer; but the circuit breaker did not trip.
I have an interesting exhibit. It is the actual transformer out of a type-Z that I bought for parts. That transformer lacks a whistle control but is otherwise very similar electrically to the ZW shown in the article. There is a section of the exposed secondary winding that is badly burned. It is in the middle of the winding, with undamaged turns beyond it at either end. I think it is pretty clear that the fault current that must have burned it had to flow through two of the rollers stationed at either end of the burned turns. Otherwise the same current would have flowed throughout the secondary winding, and it would all be burned. So at some time there was a short circuit between two of the output terminals; and a current flowed, long enough and heavy enough, to severely damage the transformer; but the circuit breaker did not trip.
..........Wayne..........
Many thanks to Wayne for posting those.
The upper picture is one of the sides where the rollers go, as you can see by the two circular tracks. The opposite side, for the other two rollers, is similar. The lower picture shows the (once) insulated edge of the winding that is almost vertical at the bottom of the upper picture. As you can see, the insulation is completely incinerated in the middle section where the fault current flowed. I am confident that the current was well over the 15-ampere rating of the circuit breaker, but was not interrupted by it since it did not flow through the circuit breaker.
All,
Thank you very much for your patience and efforts to help me clearly understand some of the issues surrounding multiple power sources, multiple blocks and the effects upon my trains. Your recent posts have confirmed to me that I do understand, at least to a degree, the difficulties imposed by electric conventional control.
My most current set-up, which I am intending to expand, indeed allows me to take a locomotive through multiple blocks followed by the same transformer. Essentially, it seems to me, I must wire my layout to allow complete layout access, modified by block on/off switches, for each power source. I'm speculating that this might manifest itself with a series of hierarchical block switches for each power source.
Thanks again,
I'm glad this topic is helping, John.
I'm not sure what you mean by hierarchical block switches. How many transformer outputs are you planning to use? A single toggle switch (SPDT-CO) per block works well with two; but the concept can be expanded to more sources. You can connect to 4 with 1 SPDT-CO and 1 DPDT, 6 with 1 SPDT-CO and 2 DPDT, and 8 with 1 SPDT-CO and 2 3PDT per block. Some folks prefer a rotary switch to a cluster of multiple toggle switches per block; but, unless the rotary switch is augmented with an overall SPST on-off switch, you can get in trouble switching an occupied block from one source to another past active sources. This is why a toggle-switch cluster should always include a center-off switch at the center-rail end of the switch tree.
You may be thinking of on-off control of sidings and yard tracks. This can be done with just an SPST connecting the track to the associated main line or the yard lead. Instead, I have gone to the trouble of modifying turnouts to route power in these situations, in the American Flyer fashion; but control-panel switches work too.
BigAl 956 I have heard this before. While I understand the logic behind this fear I can tell you that this never happens. Most model railroad hobbyists at one time or another park trains on sidings and across different blocks where there is a difference in voltage and there is no damage whatsoever. Lionel enthusiasts have been doing this for centuries now and there is just no credible evidence of this practice being a problem. Recently on another forum we had a few operators trial parking various cars and engines across different blocks set at different voltage levels. In no case was any damage or overheating reported. The current flow was negligible. Maybe there is an EE out there who can explain why reverse current flow does not occurs .
David,
I love your control panel design, especially the old style meters.
If one ammeter is showing current before the train leaves that ammeter's block and the other ammeter shows zero current, do both ammeters indicate current as the train transitions from one block to the other and then the first ammeter show zero current after the train leaves that block? Are there any speed control resistors in the circuit at the time of transistion from one block to the other?
Yes, the thread has been helpful to me.
I am intending to use three, maybe four, power sources for running trains.
"Hierarchical block switches" was intended to describe a series of switches that would control an increasing block size. In other words, one might provide a switch (SPST) to control power to each of two+ yard or accessory tracks and a switch to control power to all as you describe in your comments regarding expanding power sources. Your comments pertain exactly to what I intend with at least two main lines and one yard area.
I will control these blocks from a control panel, at least initially. Routing power via turnouts does sound intriguing however, but it may be over my head at this point. First, I've got to settle on a track plan.
"Recently on another forum we had a few operators trial parking various cars and engines across different blocks set at different voltage levels. In no case was any damage or overheating reported. The current flow was negligible. Maybe there is an EE out there who can explain why reverse current flow does not occurs." -- BigAl 956
I guess we need to know more about their set up. If they turned the power off to park a train across the gap, of course there would be no problem. A car without pick up rollers would not be a problem. A car with one pick up roller would only be a problem if you could get the roller to contact the track across the gap. Some insulated pins prevent that by having the gap area a larger diameter to keep rollers out of the gap. My Alco only has pick up rollers on one truck, so centering the engine across the gap would not cause high current flow.
Dave has the ideal set up to repeat the test you report Al. If he stops the train on the gap between blocks with rollers bridging the gap by holding the train stopped with his hands, both his meters should peg just before a circuit breaker pops. If he speeds across the gap the circuit breakers will not open because of thermal lag time, but the ammeters should peg momentarily.
I have done some tests using a type-Z transformer to get some actual fault current numbers. In the real world, the transformer is not an ideal voltage source, but has some finite impedance effectively in series with its outputs. There are three distinct forms that these impedances can take, according to how they vary with the output-control setting:
1. Primary-winding impedance. This is the primary-winding resistance plus the stray primary inductance resulting from incomplete linking of primary and secondary magnetic flux. At the secondary, it is proportional to the square of the output-control setting. At higher settings, it tends to decrease fault current.
2. Secondary-winding impedance. This is a portion of the secondary-winding resistance plus the stray secondary inductance resulting from incomplete linking of primary and secondary magnetic flux. It appears directly proportional to the output-control setting. So its effect on fault current tends not to depend on the control setting.
3. Secondary fixed impedance. This is the constant resistance of the roller (if any) and internal wiring. At lower settings, it tends to decrease fault current.
I first tested a single output, with the circuit returned to the U-terminal common. These tests were very brief, since the circuit breaker tripped quickly. The open-circuit voltage range available was 8 to 26 volts.
Open-circuit Fault current intovoltage, volts a short circuit, amperes
8 7012 8017 6026 45
Impedance type 3 seems to be having some effect at the lowest voltage; and type 1 seems to have taken over at the high end.
I next tested a pair of outputs, with the circuit between the A and B terminals and no connection to the U terminal. These tests were also brief, to avoid damaging the transformer. However, as predicted, the circuit breaker did not trip when the current exceeded 15 amperes.
Open-circuit Open-circuit voltage Fault current intovoltages, volts difference, volts a short circuit, amperes
8, 26 18 30 8, 13 5 1412, 17 5 1421, 26 5 14
The first test is the worst-case open-circuit voltage. The other three gave the same result, just under the transformer's rating, which is to be expected, since all three impedance effects are unchanged from test to test.
We can expect the layout wiring and track resistance to lower these numbers on an actual layout. However, to the extent that good wiring practices cause the layout to approach the ideal of low voltage drop between the transformer and the train, the unprotected fault currents will remain high.
These tests apply directly only to powering adjacent blocks from different outputs from the same transformer, which was the case in the CTT article. However, we can get some insight into David's case from them. When the blocks are powered from different transformers, any fault current goes through both transformers' circuit breakers; so there is no problem with safety if the layout wiring can carry the rated current at which the circuit breakers will trip.
However, the fault current does still flow. But the source impedance can be much higher. Instead of the small impedance corresponding to points a few volts apart on a single secondary winding, each source has in series the much larger impedance corresponding to the full track voltage of its block. Furthermore, the impedances of the two transformers are in series with each other as they drive the fault current. The result is that the fault current is probably substantially reduced. Depending on the transformer types and the layout wiring, it might not even be noticed.
Outstanding experiment Bob! We learn from it why the practice of using different transformer voltages across block gaps appears to work without apparent problems. The fault currents are high, but not high enough to immediately fry the transformer(s). Still, I imagine, if Dave examines the center rail at the gap he will find pitting of the tin plating from arcing that occurs during the time both rollers are producing a fault current.
We need to keep in mind that David's situation is not the one that the CTT article recommended and which my original post warned against. David has the advantage of using two separate transformers (LWs I think), which puts two circuit breakers in series with the fault current and furthermore greatly increases the series impedance. This is the case that Wayne's first diagram shows. He still gets a fault current, but his setup is not unsafe.
The situation in Wayne's later diagram is the dangerous one, with the circuit breaker out of the circuit and the fault current limited only by the layout-wiring resistance and the much lower transformer impedance.
Thanks for the excellent pictures, Wayne.
I have 4 Post War ZW's set at same voltage. Now 1 is outer mainline, 2 is inner mainline, 3 is sidings and 4 is switching yard. Run from one to the other through switches, blocks and back and forth. Run remote control modern command control MTH, Lionel and Atlas and Post War. Never had a problem.
God bless TCA 05-58541 Benefactor Member of the NRA, Member of the American Legion, Retired Boss Hog of Roseyville , KC&D Qualified
o Running between blocks powered by two different transformers is not unsafe if either has a circuit breaker adequate to protect the transformer and wiring.
o Running between blocks powered by two outputs from the same transformer with the usual postwar Lionel design is unsafe since it requires that both transformers are without fail set to the same voltage.
o Absence of evidence of a problem does not generally constitute evidence of absence.
I sent an email to Mr. Trzoniec to let him know we were discussing his Tip #8. I really appreciate Mr. Trzoniec for taking the time to respond to us.
-----Original Message-----
From: Wayne To: Stan Trzoniec Sent: Mon, Aug 12, 2013 2:00 pm Subject: A dozen easy electrical tips
Hi Stan,
I enjoyed and appreciated your article in the September Classic Toy Trains Magazine, A dozen easy electrical tips. Lots of great tips.
Tip #8, Up-and-down operation, is being discussed on the CTT Forum. Some of us do not think tip #8 is so great. I know applying higher or lower voltage to grades is common practice, but to an electrical engineer it is bad practice. As the train engine crosses the insulating gap it produces high fault currents.
Reference: http://cs.trains.com/ctt/f/95/t/219551.aspx?sort=ASC&pi350=1
Sincerely,
Wayne Benda Metrology Engineer Retired Tucson, AZ
-----Response Message-----
From: Stan Trzoniec To: wayne Cc: Cswanson@kalmbach.com Sent: Tuesday, August 13, 2013 6:25:44 AM Subject: Re: A dozen easy electrical tips
Hi Wayne,
Your email is interesting and appreciate it very much. While I'm sure it has good merit, in all the years I've been running up and down grades I've never had any problems maybe because the run is so long and the trains never stalled over the gaps. This is not a new idea, some of my other friends do the same thing and I have passed this on to them.
Again, I appreciate the interest and the words of caution. Thank you....
Stan
Stan Trzoniec OutdoorPhotoGraphics.com Publishers of Premium Nature and Railroad Books 562 South Street Shrewsbury, MA 01545
Well, it's two issues later; and the November CTT on page 28 is again recommending powering consecutive blocks from ZW outputs deliberately set to different voltages. Ironically, the last item in the article warns against running wires through foam insulation boards, to avoid fires due to overloaded wiring! That's a good idea; but I would combine it with avoiding a practice that might overheat the wires in the first place.
Our community is FREE to join. To participate you must either login or register for an account.
Get the Classic Toy Trains newsletter delivered to your inbox twice a month