The Perplexing Case Of The Melty Lounge Cars

[inobu]

"the layout chucks out a current that is rather unchecked when an issue does arise".

 

"A transformer with no safety break."

 

These two comments points to a discussion of circuit breakers. Adding a safety break will just place a limit on the maximum amperage that will flow through the spring. In either case the break must be establish below the rating of the spring which will not resolve the issue.

 

Lets look at the issue from another angle. 

 

Lets say a car battery is 12 volts @ 75 amps. The little marker light is a 12v 4.9 watts. Being that the bulb is rates at 4.9 watts it will draw .35 amps. Regardless the bulb will only draw .35 amps. Placing a breaker in the circuit will do nothing unless the breaker rating is below the .35 amps.

 

Getting closer to your problem

 

If the bulb is using is 22 gauge wire (thickness) then everything will be ok. Because a 22 gauge wire can handle up to 7 amps. It does not matter how many amps is at the battery, it is what the bulb draws that counts.

 

Here is your problem.

 

If you were to run a 33 gauge wire to the 4.9 watt bulb then a problem will arise. A 33 gauge wire will start to get hot this is because the load of the bulb (.35 amps) will strain the 33 gauge wire. A 32 gauge can handle up to .43 amps. The wire will get hot is the bulb is drawing .35 amps. 

 

If your spring is getting hot, it is because a higher current draw is running through your circuit. In this case your curcuit is

1. your transformer
2. the track feeder
3. the track connector
4. the bogie wheels
5. the bogie pickup
6. the spring
7. the light pick up
8. bulb.

 

The spring seems to be revealing the current flow.

The spring is getting hot because of current or voltage drop across the circuit so either the bulb is rates too high and the connection is poor somewhere along the path causing the current draw to climb. 

 

I believe that kvp stated that poor connectivity as a possible cause which I think it may be.

 

 

Inobu

Edited May 19, 2016 by inobu

[cteno4]

Inobu,

 

well said. 

 

the thing that has me scratching my head is the springs getting fused with only 1A power supplies. i guess a sustained 1A draw could get it hot enough and i would expect the springs could draw more than 1A so would trip the breaker if it were in a situation where it was getting full power. 

 

the spring fusing ive seen has never been in a lighted car, so im guessing all were motor cars and the motors were stuck at some point to allow the high current draw. should not be a problem with tomix distributed power pickups as well as the high current could get drawn thru many springs and thus no single one would get really high.

 

we now have solid state 1A breakers wired in and not seen any of the fusing or melting situations occur now with straight dc on 1A supplies, so our  on and off issues may have been solved by doing this.

 

jeff

[mrp]

Ok, well here's what I don't understand…

  

The very first post in the thread said "…the only coaches that have light circuits operating are just the standard end coaches, of which this isn't one…" - which I read as saying that the first bogie that melted was on a coach with *no* lighting.

  

If that's the case, then where's the load coming from?

  

Maybe I just read it wrong.

[Ochanomizu]

 

Ok, well here's what I don't understand…

  

The very first post in the thread said "…the only coaches that have light circuits operating are just the standard end coaches, of which this isn't one…" - which I read as saying that the first bogie that melted was on a coach with *no* lighting.

  

If that's the case, then where's the load coming from?

  

Maybe I just read it wrong.

 

 

Hello mrp,

 

I agree with you.  This is why I suggested dismantling the car in an earlier post to check the metal strips that run beneath the plastic seating piece.

[inobu]

Internet trouble shooting is difficult as we can only use the information presented. That is kinda why I tend to present the basic principle in order for the problem to be understood and to invoke thought. As the OP is the only one who can see the variables.

 

Just as mrp reiterated "the coach has no lighting" comment bring up an anomaly that needs to be verified but only seems to validate the problem is an connectivity issue. 

 

Here is my guess........

 

Looking at the layout...... it is an advance level build but seems to have an issue. It appears that the track has been painted and that the paint is creating a poor contact surface between the wheels and bogie. I could be wrong but there is no reflection on the rails. 

 

Inobu

[Davo Dentetsu]

But with the last week or so, the coaches got multimeter tested with no shorts or odd things of the sort, so at the moment I'm stumped.

[mrp]

Here's another thought:-

 

You said that you'd recently "assembled sets in a yard that has a lot of isolators and power feed switches ".

 

Say one specific bogie (or one coach if the coach has an internal lighting buses connecting the bogies) is straddling an isolation gap, and the power to one side of the gap is switched to reverse.  That could also cause a ton of current to flow up one wheel and down another - and you wouldn't necessarily notice if the locomotive wasn't crossing the gap at the same time.

 

Something like this:

 

 

[kvp]

Yes, that was my initial idea a few pages back. As you have pointed out, it could also happen at the border between two on-off (power/brake) type PWM controllers, one having power-ground, while the other having ground-ground on its outputs. (constantly or due to jitter) This would only damage one side of the bridging bogie. Having a lower current limit or a short detection on the outputs would mean a similar situation would trip one or both controllers/blocks.

Edited May 20, 2016 by kvp

[mrp]

Yes, that was my initial idea a few pages back.

 

Right.  I guess what hadn't really registered for me before was how easy it is to melt things just trying to do simple DC shunting with two 'blocks'.  Like this picture.  Looking at it now, I can see that there's just no way for it to work.  I guess the rule has to be that if any part of a train straddles two blocks in any way - then those blocks have to have exactly the same DC power and/or PWM pulse.  It's good to know that before I accidentally start melting my own rolling stock and fusing pickup springs.

  

[Davo Dentetsu]

Oh, I see now.  Yes, starting to make sense now.  That is definitely possible on this layout if done wrong.  Sorry to all for the confusion.

[Ochanomizu]

 

Here's another thought:-

 

You said that you'd recently "assembled sets in a yard that has a lot of isolators and power feed switches ".

 

Say one specific bogie (or one coach if the coach has an internal lighting buses connecting the bogies) is straddling an isolation gap, and the power to one side of the gap is switched to reverse.  That could also cause a ton of current to flow up one wheel and down another - and you wouldn't necessarily notice if the locomotive wasn't crossing the gap at the same time.

 

Something like this:

 

IMG_0745.JPG

 

 

Hello,

 

This is why I asked about the reversing loop in my first reply.

[cteno4]

Ok here is a possible culprit for these kinds of meltdowns one of our club members just discovered on his new rapit (we were playing trains yesterday and talking trains)! He was finding it having odd stall outs (and higher current draw) in a semi regular pattern that was slowly moving around the layout, and was not due to dirty track or wheels. So started a chain of tests to zero in where the shorting was coming from.

 

Then ran just the motor car and ran fine.

 

Then added an end car and ran fine with lights proper

 

Then with the other end car on the problem came back!

 

So first thought some issue with the lighting circuit. Pulled off the shell and lighting circuit and problem still there with just chasis and wheel pickups

 

So pulled off trucks first to disassemble chassis bottom but first popped the truck on the rail with the motor car and problem still there!

 

So pulled all the wheel sets off and tried them one at a time and one set would give the problem if you rolled it around the track while current on and motor car running.

 

The wheel sets were the kind with the all metal wheels that have a part axle sticking out and these two half axles are joined by the hunk of black plastic tubing. Gauging seemed right, so he went ahead and pulled them apart and just took off a fraction of a mm off the ends of the metal half axles and re assembled. Gauging still right and tried it and problem gone!

 

So either there was a little fleck or burr of metal in there or a half axle just a tad too long that with the proper amount of wiggling shorted then perhaps the heat from the short pushed things apart just enough to stop it before it could cool and then repeat.

 

I would have guessed they would leave a large gap in there to prevent this ever from happening, but somehow here it did! I'm going to pull apart a few of this style of wheel sets I have to see what those gaps are. But it certainly is something to look at as this one was not shorted all the time, you had to roll it around a bit to get it to short some. But with a good short short it could do the spring fuse w.o tripping a breaker or with a serious, longer short and slow breaker melt a bogie.

 

Was one of the odder situations I've heard of! Luckily Matthew is a software engineer so very good about systematic testing to diagnose the cause of a bug like this!

 

Cheers,

 

Jeff

[beakaboy]

Good news Jeff. My faulty axle may have same issue. I will try the same trick to see what effect it has. thanks!

[Davo Dentetsu]

Mmm, great thoughts indeed.  Someone else had suffered a total melting of one axle so there's even more interest in having the issues behind this fixed.  I've resumed running for the meantime.

[mrp]

Mmm, great thoughts indeed.  Someone else had suffered a total melting of one axle so there's even more interest in having the issues behind this fixed.  I've resumed running for the meantime.

 

Did they install the overcurrent protection that you mentioned was missing before?  I think that would reduce the risk of melting more wheels.  Specially if it was suitable for N-scale - like about 1A (or less).

 

[Davo Dentetsu]

The wheels of bureaucracy turn slow.  That and a big fat exhibition this week, so we're having a fortnight off running from today anyway.

[Atomsk]

I've had a similar issue with some Kato trucks.  Kato uses plactic axles, and I've seen both the axles and sideframes melted on three different trucks.  Always the inside axle (nearest the car center).  The only truck that didn't show signs of damage was one that I had adjusted the metal cups that hold the axle ends inward, to give better contact.

 

Both of these cars were equipped with my home-made interior lights, but these only draw about 10 to 20 mA.

 

Now I'm looking for replacement trucks, (part #5133-1D) but noone seems to have them.

 

I have other Kato cars, with the same set-up, and have had no problems with them.  The offending cars were both SuHa 43 series.  The only cars of that type on my roster.

[cteno4]

Atomsk,

 

It may be this issue of some of the wheel sets shorting out periodically die to the inner axle bits being just a tad too long and squeezed in a little too far. Periodic shorting then causes heat that can melt the plastic truck frame. Did it have any jerky running? Was it using a high amp power supply?

 

No derailments not found for a while,mcorrect? We have had a few of those that have shorted out on a point and caused meltdowns.

 

Doubt friction with these weights could cause enough heat to melt the frames.

 

Jeff

[Atomsk]

Atomsk,

 

It may be this issue of some of the wheel sets shorting out periodically die to the inner axle bits being just a tad too long and squeezed in a little too far. Periodic shorting then causes heat that can melt the plastic truck frame. Did it have any jerky running? Was it using a high amp power supply?

 

No derailments not found for a while,mcorrect? We have had a few of those that have shorted out on a point and caused meltdowns.

 

Doubt friction with these weights could cause enough heat to melt the frames.

 

Jeff

Sounds about right.  One of the cars had flaky contact, and the interior lights kept flickering.  I took the copper contact strips out of one truck, and bent the ends in a bit, so they would engage the points on the wheel centers more positively.  That was the only truck that didn't melt down.  That seems to point to loose contact between the axle and the contact strip as the source of the problem.

[cteno4]

The metal half axles on each wheel are held together by the black plastic tube piece. On the one our club member found the two half axles actually made contact in the center and filing one back a mm solved the problem. Why they don't have a huge gap and not have the hole go all the way thru (I just assumed that's how it was designed, never pulled one apart) I don't know! Maybe it's some have a small metal burr in there. But you might slap a multimeter on some across the wheels and wiggle the wheels to see if they short now and then.

 

Jeff

[Davo Dentetsu]

Happened again...

 

More info: https://m.facebook.com/story.php?story_fbid=1853547294864291&id=1830525997166421

[kvp]

I would seriously consider not running anything on that layout until proper current limiters and short circuit shutdown is added. Imho it costs too much to replace damaged parts and sooner or later something will get damaged that doesn't have spares available.

 

ps: At least replace every wheel with a plastic one as a stopgap measure...

[Davo Dentetsu]

Well, a few things did happen since the initial first post.  The layout binned it's self-titled "arc welder", got itself a new power supply and had a temporary setup of bulbs that took away killer current replaced with a fabbo 3A per track auto-cutout.  It has been in for a month and extensively tested with no issues.

But if you read the FB post, this still did not prevent an accidental collection of 6A.  Our whizz electrical chap, who had a look, is totally perplexed now.

So thoughts are now we need to redo that part of the layout to simplify and add safety to it somehow.  But I shall look at that Friday, not much I can do until then.  Oh, changes should be a lot easier since I became the head honcho of this layout a couple of months ago, so I'm a bit more hopeful something can be done now.

Edited November 13, 2016 by Azumanga Davo

[kvp]

This could happen if the current limiters are not the trip type. This means after a brief full power they should cut out or limit to a few mA until power is disconnected.

 

Adding polarity indicators between power districts could also help. This could be a signal/sound/light or cutout relay that is activated when a crossover is thrown and activates in case there is a polarity mismatch.

[inobu]

Insanity is doing the same thing over and over expecting a different results.

 

You have two problems.

 

The first problem is something is drawing too much current through the bogie. This high current drawing is heating up the conductors turning them into heating elements.

 

The second problem is the power management system is not detecting this current draw.

 

If you are completely DC (Non-DCC) and using the PM42 in a DC configuration then it should be wired as a negative return path. Digitrax does not specify what they mean and which ground wire to use. most likely it the 

reference ground in regards to the power supply but you should verify. It may be that's the issue as the track is completing a path outside of the PM42 hence no trip protection. 

 

In any case you should start eliminating components.

 

If it were me. i would disconnect the PM42 and connect a blue Kato transformer to your blocks and watch where it trips. 1.5 amps on the Kato box is low enough to catch the problem. If nothing trips and everything runs cool then you know its the PM42 back. 

 

Inobu 

Edited November 13, 2016 by inobu