someone told me a story that they hitched a ride with a trucker who sped up as he crossed a bridge. He sped up because his truck exceeded the weight limit on the bridge.
it's common practice to make a reverse section longer than the longest train. My understanding of this is to avoid the possibility of metal wheels simultaneously staddling the gaps at both ends of the reverse.
but I think the probability of a train having some pair of wheels exactly the distance of the reverse section is low.
So a train longer than the section would cause the auto-reverse to switch polarities many times as one set of wheels or another straddles the gaps at either end. Is there anything wrong with this?
But what if the train does have some pair of wheels at exactly the distance of the reverse section ... and is moving?
Would this really be a problem if the auto-reverser is relatively slow to change polarity because it uses a relay, even if it ends up toggling the polarity several times until wheels cross the gaps?
of course a short of any duration is undesirable. It causes excessive heating of electronic components which over time may cause them to fail. Does this mean that there is a maximum number of shorts that can occur before an autodetector or the power supply fail?
So if a long train causes an auto-reverser to switch polarity a dozen times instead of just once undesirable because it can cause an early failure of the auto-reverser or power supply?
is it the frequency of the polarity switches or the time between them that are bigger factors to consider?
greg - Philadelphia & Reading / Reading
I've got a short reversing section on my layout. It's a "crossover," most easily pictured as a diagonal track from one side of an oval to the other. Trains are frequently longer than the auto-reverser controlled section. I use a PS-REV, the older model from Tony's Trains / DCC Specialties that has been replaced by the PSX-AR. My PS-REV units are solid state, not relay-driven.
I've upgraded my fleet to all metal wheels to improve rolling resistance and keep the track cleaner.
Most of the time, trains roll through the loop just fine. If I watch the LEDs on the PS-REV, I see them flashing a lot if the train is long enough to be on both ends of the controlled section at once.
The flaw in your logic, and the times I have problems, lie with illuminated passenger cars. Even if there's only one pickup truck on each side, that still means two connected metal wheels separated by an inch or so, which is plenty of time to cause a short and an auto-reverser polarity flip, or a dead short if both ends get this at the same time. Yes, running the train faster may let you coast over these situations.
I don't think there's a problem when this happens. I've been running mine for years with no issues.
If you do see cutouts where the engine stops, though, I'd avoid running fast. You're not going to get a smooth slow-to-stop with controlled deceleration, but rather a sharp, jerky stop. This can lead to derailments and even coupler damage as a long train suddenly compresses when the engine stops dead.
I'm thinking of reconfiguring the gaps for my crossover section, to include not just the crossover itself but also the turnout and part of the main. That would lengthen the crossover section.
It takes an iron man to play with a toy iron horse.
I have a loop I use to turn trains in Durango that's longer than most trains thru it. It generally works fine, but every once in awhile, things line up just wrong and there's a issue. Not enough to worry about, I just give a nudge to it when it happens.
One thing that can help is to stagger the gaps, instead of making them directly opposite. IIRC, this recommendation is even in the PSX-AR manual. Less chance of things hanging, but may not totally solve the problem.
Despite lots of reversing and shorting action, I've never had a PSX circuit breaker or auto-reverser fail in action in nearly a decade of DCC action.
Mike Lehman
Urbana, IL
The Digital Specialties manual for the PSX-AR also says you can actually double up the gaps at either end, provided the space between these close-set gaps exceeds the length of the longest truck with metal tires/pickups.
You can have MU'd locomotives on one consist with one of them outside the reversed section. The decoders don't care what they're getting under their feet, and don't care if it's being switched in phase. It stands to reason...if they don't mind while inside the gaps in a reversed section, the loco outside won't mind if it gets it rail feed switched under it at the same time.
What you are worried about, or should be, is if a single metal tire bridges any one gap such that the phases are in conflict across that gap. But wait...that's what the reverser is meant to detect. As stated higher by the OP, the likelihood of two metal tires simultaneously bridging gaps is somewhat low, but it depends on how precisely they are aligned transversely across the axis of the track. So, to minimize the trouble, as someone else pointed out, Digital Specialties says stagger each gap across from each other by about two or three mm. If you double the gaps to handle a long truck, stagger the second set of gaps by a couple of mm as well. That should reduce your risk of a hard short down to less than 0.001.
don't understand how staggering the gaps helps. can someone explain this more, possibly with a picture?
If you cross just one gap, the reverser knows what to do. It's either A>B or B>A. If you have two gaps exactly opposite each other, it tries to do both at once -- and that's why things usually stall. It goes into protect mode.
Same thing if you cross the gaps at one end of the reversing section. If it's done on one side, it may still hang, but the train momentum is isually enough to remove the fault. But if both sides, then it further confuses things enough the reverser can't sort through quick enough and it then self-protects.
A very non-technical explanation and not a picture, but maybe helps sort it out.
let me get this straight (... you put the lime in the coconut ...)
there are two gaps at each end of the reversing section. I've assumed that the gaps are exactly across of one another and that both wheels on an axle span both gaps at exactly the same time. But this is unlikely.
more likely is that that the gaps are just a bit staggered (< 1/16") and each wheel on an axle will span each gap at a slightly different time. There are 4 wheels. So there isn't just one instant in time when there could be a short but at least a couple.
but by staggering the gaps at each end by ~ half a truck length (?) allows the reverser to handle one short without having to deal with another one just after it completes switching.
am i saying this right?
Right on.
Basically, so long as the reversing action has enough time (microseconds) to flip in the reverser, the gap difference need only be quite minimal. Best to add a few fractions of an inch, but the total difference should be less than the minimum wheelbase of any trucks on metal wheeled rolling stock operating through it.
Note that this isn't absolutely a cure, but it is 99.8% of the time, because some odd combination could still potentially overcome these efforts. If using it on hidden track, then you would want to have some way to monitor things -- which you likely would anyway. If you used trains that always used the same rolling stock in the same order, like passenger trains, then once you've tested thoroughly going through it, then it should work reliably after that so the issue of watching things becomes less.
In the manual for the PSX-AR, here is what is said about staggered gaps:
When setting up gaps for reverse sections, we recommended that the gaps be staggered about 1/8 “. Perfectly aligned gaps may reduce the current needed for PSX-AR to reverse properly.
Rich
Alton Junction
since it appears that a reverse section does not need to be longer than the longest train and may become a problem if longer trains are run, what is the smallest feasible reverse section?
the length of the longest locomotive or consist?
this may be a problem for passenger cars that may see opposite track polarity on different trucks which would provide the same polarity (+ or -) on the pick-ups, hence no power. This would be a problem for a locomotive or consist if it were in the middle of a train.
Greg, the section of the PSX instructions we are referring to is meant to cover trains longer than the reversed section....hence the twin sets of gaps at each end.
On my last layout, the one with the helix, I had a reversing side of a scissored turning wye arrangement. My Y6b always tows a train of about 18-22 cars plus a caboose. That was at least six cars longer than the reversed bit. Worked without a hiccup every time.
selectorhence the twin sets of gaps at each end.
so there are more that 4 gaps? 8?
----------|--|----------
I just thought the gaps at each end were offset.
----------|-------------
-------------|----------
if there are more than 4 gaps and therefore more that one section of track that is separately powered, how is it connected to the auto-reverser?
Staggering gaps won't change a thing. The short will occur between the rails that are out of phase. It only takes one of the rails. When a metal wheel crosses the gap it can be in contact with the cut ends of both rails at the same time. If those two rail sections are out of phase you get a short, the AR swaps phase and the short is gone.
Problem occurs when the AR sees a short but it is actually ywo shorts occuring at the same time. If those two shorts are directly across from each other, then when phase is swapped, both shorts are corrected and the train keeps moving.
Now in the event of a train crossing the gaps at both ends of a reversing section at the same time, those two shorts will not be fixed. That's because when the AR swaps phase on the reverse section, it will always be out of phase to the adjacent rails at one end or the other.
So front wheel of engine crosses gap at end of reverse section and a short occurs. AR swaps phase on the reverse section and now the AR section matches the exit section. It no longer matches the approach section. Since some of the train's wheeels are still in the approach section the next wheel that crosses the gap will cause a short. AR sees it and swaps phase. Now the reverse section matches the approach section but no longer matches the exit section where the train is also still crossing gaps. Short accurs again. As long as only one wheel crosses a gap at a time the AR may be able to keep up. If a wheel at each end makes contact at the same time, AR will keep swapping phase to correct but the train will stop because the resulting short will not go away quick enough to keep the engine moving.
A lot has to happen at the same time, but since it is possible, just be aware that it could happen and the train will simply stop.
Martin Myers
I have four reversing sections on my layout, each controlled by a PSX-AR. I have never consciously staggered gaps. I simply gap each rail, once, at each end of each reversing section. If the gaps are staggered, it is unintentional. I have never experienced a problem.
i'm confused by the inconsistencies in the discussion, including not understanding the rational for the staggered gaps described in the PSX-AR manual
the following seems valid (consistent with what several people have said that they've done nothing special and have no problems)
all that's required for the auto-reverser to trigger is for just one gap separating rails of opposite polarity to be bridged with a metal wheel
gregcI'm confused by the inconsistencies in the discussion, including not understanding the rational for the staggered gaps described in the PSX-AR manual
I'm even more confused than Greg.
If someone could draw it all out and explain it, axle by axle, track by track, that would be great. Say the offset between gaps is only 1/8" ......I can see that Axle #1 left and right wheels might only trigger one reversal. But doesn't Axle #2 reverse it again? Then there is the second truck with two more axles.
Henry
COB Potomac & Northern
Shenandoah Valley
In a reverse section situation you have three nadjacent isolated districts. The one in the middle is the reverse section. By nature it will only be in phase with one of the sections adjacent to its ends at a time. If a train longer then the reverse section causes a short by having metal wheels cross both ends simultaneously, there will be a short at one end. The auto reverser will correct the short. this causes an immediate short at the other end. The problem flip flops infinitely because AR's work so quickly. Now by nature a short that continues long enough will trip the booster's circuit breaker. Now to get the booster back on, you will need to phisically shove the train over the gap so that track power can be restored.
The above is possible with a train that is longer than the reverse section. It may be a rare occurence but it can happen. The exact length of the reversev section and the exact spacing of the wheels need be "just right". Basically, you are jusr testing Murphy's Law. BTW, if you run plastic wheels, there should be no problem.
my original thought was what happens if different trucks on different cars are crossing the gaps at either end of the reverse section. Does a reverse section have to be longer that the longest train? (no)
a train enters a reverse section, causing a polarity reversal. It travels thru the reverse section without causing any polarity reversals until it reaches the opposite end of the section.
a polarity reversal occurs when the wheels hit the gaps exiting the section. But then a polarity reversal occurs when wheels near the end of the train enters the section. At this point, there is a polarity reversal for each wheel set exiting the section at one end and entering the section at the other end.
There are many more polarity reversals in this case than the typical two that occur when the reverse section is longer than the train. But is there anything wrong with this? (no)
The other potential problem are gaps at both ends being bridged simultaneously. In this case, there will always be a short. But since the train is moving, the wheels have crossed the gap before the reverser has sensed a second short immediately after a polarity reversal (especially with a relay). Is this anything different than any other short detected by the auto-reverser? (i think no)
can staggered or twin gaps help this situation in any way? (I thought I understood, but now I see no need).
of course there is a problem if the train is stopped with wheels bridging gaps at both ends of the section. This is a short and the train requires a nudge, so don't stop.
The most likely time for that simultaneous short to occur and cause the booster to trip is if multiple wheels on the same trucks are electrically connected. As with Passenger car trucks with track power pick up (and engines). This would allow the bridging of the rails to last a longer period of time.
Like I said:it is unlikely but not impossible. The AR also gets one hell of a workout.
gregc i'm confused by the inconsistencies in the discussion, including not understanding the rational for the staggered gaps described in the PSX-AR manual the following seems valid (consistent with what several people have said that they've done nothing special and have no problems) a reversing section can be shorter than the length of the train the section must be longer than the longest consist all that's required for the auto-reverser to trigger is for just one gap separating rails of opposite polarity to be bridged with a metal wheel
Staggered gaps separated by more than 1/8" are not necessary and are not any more effective than 1/8" staggered gaps. The most important thing to understand about a reversing section of track is that, once the auto-reverser reacts to reverse polarity by flipping the polarities inside the reversing reversing section to match the polarities of the two adjacent sections of track, the polarities remain matched until the first set of wheels exit the reversing section.
Using "best practices" as a guideline, the reversing section should be longer than the longest train. That best practice will prevent the auto-reverser from reacting prematurely. Unless the layout is too small to accommodate a reversing section of the proper length, this best practice should be followed without exception.
One exception to this best practice is a train where the rolling stock has plastic wheels. In that case, the reversing section only needs to be as long as the longest locomotive consist.
Another exception is a train where the rolling stock has metal wheels, but unpowered metal wheels. In that case, you may get away with a reversing section longer than the longest consist but shorter than the longer train. But it is not a best practice. Depending upon the design of the rolling stock, the auto-reverser may still sense the presence of reverse polarity and try repeatedly to correct the situation.
BigDaddy gregc I'm confused by the inconsistencies in the discussion, including not understanding the rational for the staggered gaps described in the PSX-AR manual I'm even more confused than Greg. If someone could draw it all out and explain it, axle by axle, track by track, that would be great. Say the offset between gaps is only 1/8" ......I can see that Axle #1 left and right wheels might only trigger one reversal. But doesn't Axle #2 reverse it again? Then there is the second truck with two more axles.
gregc I'm confused by the inconsistencies in the discussion, including not understanding the rational for the staggered gaps described in the PSX-AR manual
So, on a 6-axle locomotive, once the first wheelset trips the auto-reverser, the remaining 5 wheelsets cross the gaps and enter the reversing section without causing the auto-reverser to trip again.
I followed the instructions and staggered my gaps. I am not qualified to judge their recommendation, but I believe what they are saying is that 2 shorts at once will reduce the current below what they can detect.
One side effect I've noticed is that because only one of the 4 wheels on a truck passes over the gap at a time, the rigid truck causes the wheel to ride over the gap rather than dropping in and creating a short. Mine doesn't reverse when a steel wheeled car rolls over the gap - only a locomotive triggers it.
This recommendation from the manual is the one that really boggles my mind:
If your train is longer than your reverse block and has metal wheels, you may need to cut additional gaps into the reverse section. Simply cut another set of gaps at both ends of the reverse sections inside of the original gaps. The distance between these gaps and the original gaps should be longer than the wheelbase of any metal truck.
I have the right to remain silent. By posting here I have given up that right and accept that anything I say can and will be used as evidence to critique me.
The likelihood of bridging the gap at both ends at the same time can depend on the kind of cars being used. On my old layout, normal freight and passenger trains fit in the reverse section (using a DCC auto reverser), but ore trains were too long so the last 8-12 cars and caboose were entering the section as the engine(s) were exiting. Since the ore cars are so short, with so many metal wheels so close together, it was very common to get a short. Fortunately at that time I still had a fair no. of ore cars with plastic wheels so always made sure to run them at the end of the train.
carl425 - from PSX-AR manualIf your train is longer than your reverse block and has metal wheels, you may need to cut additional gaps into the reverse section. Simply cut another set of gaps at both ends of the reverse sections inside of the original gaps.
does anyone know how to wire these sections between the additional set of gaps?
If I understand this correctly, by cutting these extra sets of gaps there are three sections: the main reverse section which is several feet long and a very short section (~1/4") at each end.
if these short sections are wired to either the reverse or mainline section, what's the point in cutting the extra gaps?
Thanks Rich
gregcdoes anyone know how to wire these sections between the additional set of gaps?
I sent an email to DCCSpecialties. Their response is
The short isolated sections of track are not powered. They're simple isolating dead rails.
gregcThe short isolated sections of track are not powered. They're simple isolating dead rails.
That way only something long enough to straddle the short dead section can cause a short to trigger the reverse.
gregc carl425 - from PSX-AR manual If your train is longer than your reverse block and has metal wheels, you may need to cut additional gaps into the reverse section. Simply cut another set of gaps at both ends of the reverse sections inside of the original gaps. If I understand this correctly, by cutting these extra sets of gaps there are three sections: the main reverse section which is several feet long and a very short section (~1/4") at each end. if these short sections are wired to either the reverse or mainline section, what's the point in cutting the extra gaps?
carl425 - from PSX-AR manual If your train is longer than your reverse block and has metal wheels, you may need to cut additional gaps into the reverse section. Simply cut another set of gaps at both ends of the reverse sections inside of the original gaps.
What I find perplexing about that last sentence is that it states that the distance between these gaps and the original gaps should be longer than the wheelbase of any metal truck. Since wheelbase is defined as the distance between the front and rear axles, that distance on a truck is inevitably going to be greater than 1/4". Also, that sentence refers to a "metal truck". What about a plastic truck?
Quite frankly, that whole paragraph leaves me confused. I just don't have such problems on my layout and I have four PSX-AR units, each controlling a different reversing section with a single pair of gaps on each end of each reversing section. But, each of my reversing sections is longer than the longest train including all locomotives in a consist plus the rolling stock (all metal wheels).
Second gaps and staggered gaps are different animals. If you do the second set of gaps and the distance between the two pairs of gaps is less than the wheelebase of a metal truck, you run the risk of fixing nothing, in that the lead axle could be crossing the original gaps while the rear axle is crossing the new second set of gaps. While at the same time, another powered unit or some such is crossing the entrance gaps of the loop. Avoided, if the reversing section is longer than your longest train - none of this should be necessary if that is the case, just an isolated section powered by the AR. It's when you start bending this rule that problem soccur. Note is is usually not an issue with a wye, either the tail track is as long as your longest train, or it becomes so because only a train that fits completely on the track and clears the turnout can be reversed on said wye. Simple balloon loops too are self regulating - you can't run a train longer than the loop through it, or else the engine will strike the cars that haven't yet cleared the turnout. It's the more complex and tricky almost hidden loops that can have issues with train lengths.
Now, the whole idea of leaving dead sections as long as the longest metal truck wheelbase is asking for stalls, if you ask me. I don;t run passenger cars, so the longest conducting truck would be the ones on my Trainmaster (which are longer than most any passenger car anyway). So you would be effectively making a dead section that puts one whole truck on unpowered track. Such would be the case if you made the gaps spaced to handle a 6 axle passenger car but also run ordinary 4 axle diesels - not just tiny things like critters (those have zero chance of making it past such an arrangement without keep alives), but regular size things like standard F units and Geeps.
Best plan: just avoid such situations altogether. Relocated gaps if necessary, so there is no way a train can cross to sets of gaps that require the reverser controlled section to be opposite polarities. If the track design allows a 'sneak path' through part of the reverser controlled are but that part does not require any actual reversing to keep the track all consistent, there's no problemIf polarties already match at both pair of gaps, it doesn't matter if they are crossed at exactly the same time or not. You cna take advantage of that to extend an overly short reversing section so that the part that actually does need polarity reversal is long enough to hold your longest train. Don;t worry about the old, too close together gaps - just connect the feeders for the rails between the old, incorrect gaps and the new, proper gaps to the same bus feed as the track that always was between the gaps. No need to re-lay the track or otherwise fill in the existing gap.
==1==//====2====//====3====//====4===
If the old gaps were between 2 and 3, and the new gaps are between 3 and 4, you just wire 2 and 3 together - the new reversing section is 2 + 3, instead of just 2.
(yes, that should be pretty obvious....)
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
One thing worth mentioning is that rolling stock (passenger and freight cars) with metal wheels that are unpowered may not actually trigger the auto-reverser, depending upon the trip level of current. So, in some situations, it may be possible to get away with running a train that is longer than the reversing section of track, while still using only one pair of gaps at either end of the reversing section.