Rick, any chance that you can post a photo of the NE DXO?
And, can you tell us how that DXO is put together?
Is it three Atlas Snap track turnouts and a W/S curved turnout, plus a crossing?
Rich
Alton Junction
richhotrain Heck, you don't even need a loco. Just use a voltmeter. Turn the Z-1000 all the way up, that should feed 18 volts to the rails. With a voltmeter probe on each rail and the voltmeter set to AC voltage, see if you get a reading between 12 and 18 volts. Measure on both the NW and NE end of that DXO on the outer loop. If you don't get a good reading, add the jumpers and measure again to see if you get sufficient voltage. Rich
Heck, you don't even need a loco.
Just use a voltmeter. Turn the Z-1000 all the way up, that should feed 18 volts to the rails. With a voltmeter probe on each rail and the voltmeter set to AC voltage, see if you get a reading between 12 and 18 volts. Measure on both the NW and NE end of that DXO on the outer loop. If you don't get a good reading, add the jumpers and measure again to see if you get sufficient voltage.
DCS runs a constant 16-18v at all times. I tested with the VM and can confirm that all rails at the NE DXO in all directions are live & showing appropriate voltage.
I was thinking it was just a matter of the spacing of the insulated frogs on the lower two turnouts in combination with the XO; that the spacing of the trucks just matched the distance between insul frog portions, and that the unit was losing power. But given that the NE DXO is a mirror of the NW DXO, and the engines did not stop cold when running in either direction there, that can't be the cause.
So my guess is the problem with stalling going from the figure 8 to the outside loop is some combination of the insul frogs and the AR unit/reversing section. Odd.
rallison richhotrain Rick, what about putting jumpers on the NW and NE sides of that NE DXO on the outer loop? Once again, I repeat, the additional power feeds should eliminate the stalls on that DXO and solve problem #2. Rich Rich -- Yes; I expect once I get the engines repaired I'll replicate the 'stall' to confirm it's a lack of power at particular parts of the NE DXO. Then add temporary juice as needed to correct, as you suggest, & then make permanent whatever power addition fixes it. I just didn't mention it because I can't do much about it 'til I have engines to run over the DXO...
richhotrain Rick, what about putting jumpers on the NW and NE sides of that NE DXO on the outer loop? Once again, I repeat, the additional power feeds should eliminate the stalls on that DXO and solve problem #2. Rich
Rick, what about putting jumpers on the NW and NE sides of that NE DXO on the outer loop?
Once again, I repeat, the additional power feeds should eliminate the stalls on that DXO and solve problem #2.
Rich -- Yes; I expect once I get the engines repaired I'll replicate the 'stall' to confirm it's a lack of power at particular parts of the NE DXO. Then add temporary juice as needed to correct, as you suggest, & then make permanent whatever power addition fixes it. I just didn't mention it because I can't do much about it 'til I have engines to run over the DXO...
Rich -- thank you (and the other contributors) once again for all the time and thought put into my troubleshooting request. While I'm getting the engines repaired I will do additional testing for short-circuits on and around the 3WT, place current limiters/breakers to ensure no more fried boards; and (as you and others have suggested) likely pull up the 3WT to simplify the testing of that portion of the figure 8 route and of the 3WT itself.
I'll update the thread once I have the engines back and/or when I find anything conclusive about the cause(s) of problem #1 or #2.
Again, my thanks to all for the assistance.
Another new day, still thinking about those two problems.
Regarding problem #2, locos stalling as they cross the diamond on the NE DXO. It still seems to me that there is a lack of power feeds on the NW and the NE legs of that DXO. If you don't want to permanently add feeders, then temporarily add jumpers from the bus to both sides of the DXO. If stalls, not shorts, are the only issue, then problem #2 is an easy fix.
Regarding problem #1, locos are shorting and boards are frying. Given the additional details that Rick has provided, it is sounding less and less like the 3-way turnout is causing the problem. If the 3-way turnout is not the problem, what's left?
I am beginning to wonder if the MTH loco boards are the problem. I am no expert on MTH DCS, but I can say that a DCC locomotive would not result in a fried decoder under such circumstances. The derailments don't seem to lead to the shorts because those shorts occur with and without derailments.
If it were me at this point, I would pull out the 3-way turnout and temporarily install a straight piece of track. Then, I would disable the Bachmann AR unit and wire the reversing section to match the polarity of the non-reversing section. At that point, there is nothing to keep the locos from shorting through the reversing section. Verify that, and we can proceed to trouble shoot the 3-way turnout and the Bachmann AR unit.
richhotrain [...] ...why are those cars derailing and where? On the 3-way turnout?
[...] ...why are those cars derailing and where? On the 3-way turnout?
The cars that derailed on engine failures 2 and 3 were not on the 3WT.
The car that derailed for event 2 was at the next turnout up the line to the NW, and I'm not sure (it's been the better part of a month now) but I think I'd left it in diverging position, as if I were backing up that top spur. Cars passed correctly down the line toward the 3WT, but as soon as I started backing up (because the fig. 8 was not complete) one wheelset of a car near the middle of the train went up the spur, I think. Doh! User error. But it should not have killed the engine...
Event 3's derailment was the car right behind the engine. The spot where it derailed (on the curve between the 3WT and the figure 8 spur turnout just above the yard) is a tight (18") curve with a height transition, too. The springs in the remote operating Proto-couplers on the MTH HO engines are a bit stiff, and they sometimes deflect the following car's coupler all the way to the side on a tight curve--the shank hits the side of the gear box. Normally not a problem, but in that position the slightest lifting force (i.e., when there's a change in grade) lifts the lead truck off the rails. In this particular case, I had a heavy duty t-pin stuck in the foamboard a quarter-inch away from the ties; when the derailed car hit that, it halted the engine and poof! That was it. Again, can' see why it killed the engine to be halted. The wheels and axles on the derailed car were plastic, so I know they did not short out the track section.
So, nothing mysterious about either derailment; they just happened to halt the engine with one truck on either side of that oddly treacherous gap.
richhotrain Do the shorts only occur with derailments?
Do the shorts only occur with derailments?
The first engine failure was solo with no derailment, so--no.
richhotrain Do any shorts occur without derailments?
Do any shorts occur without derailments?
As above, the first engine failure was solo with no derailment, so--yes.
Quite the puzzle, innit?
rallison richhotrain Regarding issue #2, now that the issue has been restated, it seems that the AR unit is always operating properly, but that the loco is stalling (not shorting) as it crosses the diamond on the NE DXO headed NW or NE. If it is stalling, not shorting, there is a lack of power. This goes back to my earlier point of adding power feeds to every end of every turnout and placing gaps further downstream. So, is the Bachmann 44912 AR working flawlessly and consistently? Rich -Rick
richhotrain Regarding issue #2, now that the issue has been restated, it seems that the AR unit is always operating properly, but that the loco is stalling (not shorting) as it crosses the diamond on the NE DXO headed NW or NE. If it is stalling, not shorting, there is a lack of power. This goes back to my earlier point of adding power feeds to every end of every turnout and placing gaps further downstream. So, is the Bachmann 44912 AR working flawlessly and consistently? Rich
Regarding issue #2, now that the issue has been restated, it seems that the AR unit is always operating properly, but that the loco is stalling (not shorting) as it crosses the diamond on the NE DXO headed NW or NE. If it is stalling, not shorting, there is a lack of power. This goes back to my earlier point of adding power feeds to every end of every turnout and placing gaps further downstream. So, is the Bachmann 44912 AR working flawlessly and consistently?
-Rick
You mentioned earlier that.......................
"The two gaps at center bottom are built in to the Wathers/Shinohara DXO, and the two odd ones at the top mirror those gaps on the opposite rail at the top. Granted, it's an odd arrangement; but, along with the built-in ones in the DXO at bottom, it prevents each of the outer and inner tracks from forming a complete uninterrupted electrical loop because I've been told that DCS engines do not like loops".
I am not real familiar with DCS, but I would add a pair of feeders on the west side of that outer loop beyond the NE DXO as well.
You need to adequately power that NE DXO.
Rick, sorry, I wasn't directing that information overload comment at you. The thread in general is long and complicated, necessarily so because of the dual problems. It is just a lot to absorb, and it is hard to get one's arms around these two complex issues.
richhotrain INFORMATION OVERLOAD !
INFORMATION OVERLOAD !
Agreed! I am a wordy dude, sometimes. Okay, often.
richhotrain Back to issue #1. Without the AR unit wired in place or the reversing section completely connected, there are damaging dead shorts occuring when the loco wheels come to rest on those gaps on the tail end of the 3-way turnout. It could have something to do with the 3-way turnout. It bothers me that the DSC commander circuit breaker doesn't always react. So, I suppose that the DSC commander is suspect as well. As far as the Z-1000, its circuit breaker is 6 amps, so no wonder.
Back to issue #1. Without the AR unit wired in place or the reversing section completely connected, there are damaging dead shorts occuring when the loco wheels come to rest on those gaps on the tail end of the 3-way turnout. It could have something to do with the 3-way turnout. It bothers me that the DSC commander circuit breaker doesn't always react. So, I suppose that the DSC commander is suspect as well. As far as the Z-1000, its circuit breaker is 6 amps, so no wonder.
Bothers me, too. That's why adding ~2A breakers to each of the power feeds makes sense to me.
As far as I can tell Problem #2 is not a short--it does not fry the engine board, and does not trip any breakers; the engine just stops cold when exiting the figure 8 at the NE DXO. Once I get problem #1 solved I'll spend more time determining exactly where the wheels are when the engine stops cold, and determine if it also happens with the F3 (which is shorter). I don't think it's a matter of the AR not flipping polarity correctly or promptly, because it does so in both directions at all the other points where the reversing section meets the rest of the layout.
So I think I would answer your question "As far as I am able to tell, yes, it is working correctly."
My answer is 'qualified' only because I haven't determined the actual cause of either problem #1 or #2 yet.
Thank you, Frank -- trust me, if I pull it I will test it ten ways from Sunday... !
I keep on thinking about those derailments. Very few of us have bullet proof track work, but why are those cars derailing and where? On the 3-way turnout?
Just wondering.
rallison If there's no definitive result, I will yank up the 3WT (alas, as it took some fiddling to get it and the switch motors positioned just right, oh well...) and temporarily replace it with a flex-track section, pending repair of the engines. I will check and triple check the 3WT's under-tie connectors to ensure there are no flaws.
If there's no definitive result, I will yank up the 3WT (alas, as it took some fiddling to get it and the switch motors positioned just right, oh well...) and temporarily replace it with a flex-track section, pending repair of the engines. I will check and triple check the 3WT's under-tie connectors to ensure there are no flaws.
LOL I used to coach girl's softball when my daughter was a youngster, and one father told me that the instruction was more than his daughter could absorb. That seems to be what is happening here - - - too much informaation to absorb.
In any event................
Rick,
Just for a sanity check. If You do decide to remove the three-way turnout...test it at Your work bench with a 9volt battery, with alligator jumper wires to the point side of the turnout and test each frog in the straight and diverging routes. Should not get any reading on any of the frogs. If you do, its a possiblity that the frog is touching one of the jumpers on either the straight or diverging route. It has happened before and tough to figure out without a lot of testing. A continuity test will not always catch it.
Just a thought.
Good Luck!
Frank
You have the relevant gaps right. (There are also gaps in each rail of the outer and inner loops, but at opposite sides per rail, in order to prevent electrical loops which--I'm told--DCS does not handle well. Both loops are isolated from and therefore do not in any way affect the figure 8 sections.)
For event #1, power was transferred to the partial/future reversing section via wired alligator clips connected (per FIG 1 from my original post) to the end of the middle track of the small switching yard at bottom, to a point well up the partial curve heading NNE on the incomplete portion of the East/right lobe of the figure 8. Polarity was therefore hard-wired to match the lap section. Engines repeatedly passed through that area successfully before the first failure.
For Event #2, power was transferred with the wired clips from the same source, but connected to the left end of the spur that hangs West from the turnout at the bottom of the partial curve/incomplete East lobe of the figure 8. That's not shown on FIG 1 because (electrically) it is theoretically the same, and again I am quite sure it matched the polarity of the lap section. And Engines crossed the problem area repeatedly before the second failure occured.
For Event #3, the figure 8 was complete, gapped as described, and the AR until was installed. Power is fed to the AR unit from the power distribution block, and then from the AR unit to the rails of the ascending, completed curve on the East/right lobe of the figure 8, as shown in FIG 2 link from my original post. FYI, I made sure the 'default' polarity of the AR unit matched that of the lap switch section by marking the input and output wires before hooking them up at either end and plugging them into the AR unit. The reversing worked properly (and as far as my ear could tell, instantly) in the directions indicated by green arrows in FIG 2.
In all cases all the polarities were tested with the voltmeter before I put the engines on powered track.
I am pretty well persuaded that something on either the reversing section or, most likely, on the 3WT/Lap Switch itself, is causing an intermittent short. The design itself seems relatively sound, and all the voltmeter testing I've done bears that out; plus the AR unit did work correctly except when traversing the NE DXO from the figure 8 to the outer loop, where (per the red arrows in FIG 2) I believe the combination of AR gaps and dead frogs plus the 'wheelbase' of the GP-35 leads to stalls.
Alright, back to work for me 'til tonight.
rallison In at least two of the failures (#1 & #3), I am 100% sure the affected engine was well clear of the frogs. I'm fairly sure it was clear in event #2, as well.
In at least two of the failures (#1 & #3), I am 100% sure the affected engine was well clear of the frogs. I'm fairly sure it was clear in event #2, as well.
Clear of the frogs is fine ... but sounds like you were shorting because of bad jumpers.
Lemme make sure I've got all the gaps and everything right.
1. North West crossover, south of the frogs (2 pair)
2. North East crossover, south of the frogs (2 pair)
3. Fig 8 West Side (south of the turnout) (1 pair)
4. Fig 8 East Side (Southeast of the lap switch) (1 pair)
5. Storage tracks (lap turnout center route) (1 pair each storage track)
Where was power being fed into the reversing section via jumpers, just at the lap switch, or from a different point (e.g. the crossovers, or the western switch that leads into the fig-8)?
-Dan
Builder of Bowser steam! Railimages Site
To recap:
#1. The GP-35 is running solo heading SE from the 3WT to the temporarily hard-wired & incomplete reversing section, as shown in FIG 1 linked in my topic post. I hit 'DIR' to halt the engine after clearing the 3WT. DCS momentum halts it with the rear-most wheelset of the rear truck straddling the insulated gap. A BIG, visible spark occurs at that wheelset, and the engine dies. Neither the Z-1000 breaker nor the DCS Commander breaker trip. The front of the engine is fully on the temporary section, so the entire engine is well clear of any frogs or points. The Bachmann AR unit was not yet installed.
#2. The F-3 is running a 6- or 7-car consist, plus lighted caboose, heading SE through the 3WT toward the temporary section/insulated gap. I brought the engine to a halt with all of its wheels fully on the (future) reversing section. I reversed direction ('DIR') and began to back the train towards the upper-left/NW DXO. A car derailed and halted the train. I did not see any spark at the engine; there may have been a short further back on the train at the derailed car, as it has metal wheels/axles--but I'm not sure. Regardless, the engine dies; The Z-1000's breaker was not tripped; but this time the DCS Commander breaker WAS tripped. The Bachmann AR unit was still not installed, so power to the temporary section was still hard-wired with jumpers, & 100% in the correct polarity. NOTE: I'm not sure exactly where the engine was in the 3WT/gap/temporary section when it died, because right away I moved it by hand further along the track to clear the derailment before restoring power to the layout. But my recollection is that the front truck was on the temporary section, and the rear truck on the 3WT--but probably SE (therefore clear) of any of the 3WT frogs.
#3. The (now repaired) GP-35 is running a 6- or 7-car consist heading NW onto the 3WT from the reversing section, crossing the insulated gap. A car derails and halts the train with the front truck on the 3WT, perhaps as far as the first set of points (set to through route) but no further, and well short of the first frog; the rear truck is still on the other side of the gap, on the reversing section. The engine dies; the Z-1000 breaker was not tripped; but the DCS Commander breaker WAS tripped again. The Bachmann AR unit WAS installed at this point, and had been functioning with no noticiable audible delay or engine hesitation when flipping polarity.
In #1 and #2 I am 100% sure the polarity was matched as designed and tested--i.e., if there's no intermittent short somewhere in a track section, the 3WT matched the temporarily hard-wired/future reversing section.
In #3, the reversing section polarity would initially have been reversed to match the outer loop when the engine crossed from the OL to the reversing section at the upper-right/NE DXO, heading clockwise around the right/East lobe of the figure 8. It would have flipped (assuming it was functioning correctly) to match the polarity of the 3WT as soon as the front wheelset of the front truck of the engine crossed the insulated gap. I do not recall whether or not I heard the AR relay trip when the engine entered the 3WT.
So...
Per a suggestion from an e-mail conversation I am going to test every section from the reversing feeders all the way to the section with the yard/spur feeders by pressing down with varying force on a lighted caboose to see if it causes an intermittent short. If there's no definitive result, I will yank up the 3WT (alas, as it took some fiddling to get it and the switch motors positioned just right, oh well...) and temporarily replace it with a flex-track section, pending repair of the engines. I will check and triple check the 3WT's under-tie connectors to ensure there are no flaws. And I will review all this with MTH if they'll spare me the time, as I don't want to put a DCS engine back on the figure 8 until I can get this 100% nailed down. Plus I'll be adding some circuit protection at ~ 2A trigger point.
'Til this happened, I was really enjoying the layout coming together and the operation of the engines under DCS, though. So I hope I can get it sorted out.
To that end I continue to be very grateful for everyone's thoughts/guidance on this issue.
And, remember, those shorts occurred even before he installed the AR unit, and all of his feeders were in phase - - - no mismatched polarities.
y'know -- that makes total sense! From his picture, it looks like the points end of the lap switch is where the gaps for the reversing section are ... if that's the case, he might be shorting things out fast enough by traversing the switch that the AR can't keep up.
I know my B'Mann one is a little slow on the uptake (needs about 1/2 - 1 second between flips*, else nothing happens -- not that I've timed/tested it super-extensively though).
*i.e. if the tracks are +/-, and then flipped to -/+, it needs a "cooldown" before it'll switch them back to +/-
NeO6874 I mean, from your drawings, it LOOKS like it should work (barring your lap (3-way) switch misbehaving -- I've heard that sometimes you can short from a stock to a point rail, due to tight clearances). Actually, having read the other posts after your response to me, it sounds more and more like that lap switch is the culprit.
I mean, from your drawings, it LOOKS like it should work (barring your lap (3-way) switch misbehaving -- I've heard that sometimes you can short from a stock to a point rail, due to tight clearances).
The OP mentions that a short occurs when the loco stops with one of the wheelsets resting ON the gaps. But I wonder if the real problem is where the other wheelset is resting.
I have three of the Walthers Shinohara Code 83 3-way turnouts lined up end-to-end on my layout, and I initially experienced shorts where the wheels touched a rail of opposite polarity right after the frog (see my photo). Some guys use nail polish as an insulator. I used a small thin strip of masking tape which you can see in the photo.
That could well be the problem. With a moving loco, the short is intermittent, not a dead short. But leave a loco sitting on that short and who knows what might happen if the breaker doesn't trip.
Yep, I agree that isolating the entire figure 8 would work.
Or, maybe I should say that it "should" work. Still not sure what is wrong with his layout.
rallison I'd originally expected to make the whole figure 8 a reversing section, but was advised that only a portion long enough to handle a train either direction from the upper-right/North-East DXO would be needed. I did do what you've suggested, but using a paint program instead of a highlighter on line art of my layout. If you look closely at FIG 2 linked in my original post, you'll see that, rather than red and black as with the two outer loops, yard, and spurs/engine house section, the reversing section is shown in purple and blue. The gaps at that NE DXO, and at the 3WT, and at the East lobe of the figure 8 where it meets the yard turnout are what isolates the reversing section. I can move the gaps 'inward' at either end of the reversing section, but not too far on the East end as it would require much shorter trains...
I'd originally expected to make the whole figure 8 a reversing section, but was advised that only a portion long enough to handle a train either direction from the upper-right/North-East DXO would be needed. I did do what you've suggested, but using a paint program instead of a highlighter on line art of my layout. If you look closely at FIG 2 linked in my original post, you'll see that, rather than red and black as with the two outer loops, yard, and spurs/engine house section, the reversing section is shown in purple and blue. The gaps at that NE DXO, and at the 3WT, and at the East lobe of the figure 8 where it meets the yard turnout are what isolates the reversing section. I can move the gaps 'inward' at either end of the reversing section, but not too far on the East end as it would require much shorter trains...
maxman I have always been told that turnouts should be fed from the point end, and gaps should be at the frog end. In this case, if I am understanding the diagram correctly, there is no power feed to the points. So the points are being backfed by the pair of feeders beyond the 3-way. It was my understanding that feeding from the point end was always good practice, no matter dc or dcc.
I have always been told that turnouts should be fed from the point end, and gaps should be at the frog end. In this case, if I am understanding the diagram correctly, there is no power feed to the points. So the points are being backfed by the pair of feeders beyond the 3-way.
It was my understanding that feeding from the point end was always good practice, no matter dc or dcc.
So, in the present case, I would add feeders on the three pairs of frog rails on that 3-way turnout and then place the gaps a little further east of the tail end of the 3-way turnout. That way, power feeds can also be added on the tail end of the 3-way turnout. Why? Because a turnout with an isolated frog raises the possibility of disruption in the electrical continuity of the turnout if any of the turnout's jumpers fail.
That said, the lack of power feeds and/or the presence of gaps should in no way cause a short on a turnout with an isolated frog.
Thanks again -- off to catch some zzzzzz... and to think on all the suggestions.
Maxman -- apologies if it seemed like your reply was ignored. As a newb, my posts were sitting in moderation queue all weekend. But if you look above you'll see I replied to you on Saturday, September 06, 2014 6:20 PM. As noted above, the 3WT is an unmodified 'DCC-Friendly" Walthers Shinohara code 83. I gather I'm no longer filtered through the mod queue.
rallison The short on the unused turnout was complete, left rail to right rail on the straight route. No apparent cause--it looked normal underneath, etc., but there was zero resistance when tested with the voltmeter, one probe on each rail...
The short on the unused turnout was complete, left rail to right rail on the straight route. No apparent cause--it looked normal underneath, etc., but there was zero resistance when tested with the voltmeter, one probe on each rail...
maxman How would that make a difference? (I wish I could have quoted entire post, but somehow the quote thing has suddenly changed format) I have always been told that turnouts should be fed from the point end, and gaps should be at the frog end. In this case, if I am understanding the diagram correctly, there is no power feed to the points. So the points are being backfed by the pair of feeders beyond the 3-way. I asked previously if that 3-way was a code 100 or a dcc friendly code 83. That question was ignored. If the 3-way is not dcc friendly, who knows what effect that might have at the point end. It was my understanding that feeding from the point end was always good practice, no matter dc or dcc.
How would that make a difference? (I wish I could have quoted entire post, but somehow the quote thing has suddenly changed format)
I have always been told that turnouts should be fed from the point end, and gaps should be at the frog end. In this case, if I am understanding the diagram correctly, there is no power feed to the points. So the points are being backfed by the pair of feeders beyond the 3-way. I asked previously if that 3-way was a code 100 or a dcc friendly code 83. That question was ignored. If the 3-way is not dcc friendly, who knows what effect that might have at the point end.
The 3-way turnout is a DCC Friendly Code 83. Read back through the thread and you will see his replies.