No, I can't refer you to any documentation. Was something I read somewhere some time ago. But, I was at a small RR museum some 15 years ago and smelled something hot/burning. I was the only visitor, and there was only one staffer. I called him and we searched, finding a phone wall wart very hot. Note this goes beyond the hobby, since homes are full of them.
Before buying something that comes with a power supply, who checks for a UL symbol on the power supply?
@Arthur P. Bloom posted:How is a leakage of 120 Volts through 11,000,000 ohms creating a voltage reading of 87 Vac from line hot to a (supposed) isolated track system load? I await the explanation with bated breath.
Many electronic meters have a very high impedance input and would most likely indicate a voltage like that through 10-11 megohms. I just measured my 18V bench transformer through a 10 megohm resistor and I got 9 volts with two different meters, so I assume they have a 10 megohm input impedance.
@gunrunnerjohn posted:Many electronic meters have a very high impedance input and would most likely indicate a voltage like that through 10-11 megohms. I just measured my 18V bench transformer through a 10 megohm resistor and I got 9 volts with two different meters, so I assume they have a 10 megohm input impedance.
Well, let's see-
If we have two wall warts @ 11M each to ground that is 5.5M.
Using the voltage divider equation, assuming a 10M meter, we get a division of 10M/(10M + 5.5M) = .65
.65 x 120 volts = 78 volts - Well withing the ballpark, given the uncertainties.
I forgot about the second parallel wallwart, makes perfect sense. 
I went back this morning, moved the two warts to the workbench, and took readings. On one, readings between either receptacle prong and either output prong are all 10.13 megohms and varying. On the other, the reading is a fairly steady 10.8 meg. No question: these should be trashed.
I installed two good wallwarts to power the TIUs and fired up the layout. Suprise, I got an 86 volt reading between a hot receptacle terminal and the outer rail, and 43 volts between a common receptacle terminal and the outer rail, I disconnected all transformers and this didn't change. I disconnected the two new wall warts and no change.
T unplugged the powerstrip from the wall outlet, and got infinite resistance between the outside rail and any receptacle terminal on the power strip. There are no wires departing the layout other than the power strip.
I pose a question for any electrical engineer out there. I have a common ground wiring system, with everything tied into a ground. We have 60 Hz AC. If I put one lead of a test meter on the hot of a 120-volt outlet, and the other onto a connection to hundreds of feet of track many items with motors, lamps, and coils (loco and accessory motors and transformer secondaries), would there be a flow of electrons to charge and uncharge--capacitive & inductive action if you will--the layout?
Another factor to consider is that the layout ground is always connected to the hot low voltage wiring, through lamps and other devices, which also adds to the "load," in effect doubling the amount of wire that is charged negative and positive 60 times a second.
I have not tried to light a lamp, due to the complications of trying to connect one side only to hot and the other to an outside rail, and mytest meter is high impedence so it doesn't take much current flow to operate.
@RJR posted:I went back this morning, moved the two warts to the workbench, and took readings. On one, readings between either receptacle prong and either output prong are all 10.13 megohms and varying. On the other, the reading is a fairly steady 10.8 meg. No question: these should be trashed.
I installed two good wallwarts to power the TIUs and fired up the layout. Suprise, I got an 86 volt reading between a hot receptacle terminal and the outer rail, and 43 volts between a common receptacle terminal and the outer rail, I disconnected all transformers and this didn't change. I disconnected the two new wall warts and no change.
T unplugged the powerstrip from the wall outlet, and got infinite resistance between the outside rail and any receptacle terminal on the power strip. There are no wires departing the layout other than the power strip.
I pose a question for any electrical engineer out there. I have a common ground wiring system, with everything tied into a ground. We have 60 Hz AC. If I put one lead of a test meter on the hot of a 120-volt outlet, and the other onto a connection to hundreds of feet of track many items with motors, lamps, and coils (loco and accessory motors and transformer secondaries), would there be a flow of electrons to charge and uncharge--capacitive & inductive action if you will--the layout?
Another factor to consider is that the layout ground is always connected to the hot low voltage wiring, through lamps and other devices, which also adds to the "load," in effect doubling the amount of wire that is charged negative and positive 60 times a second.
I have not tried to light a lamp, due to the complications of trying to connect one side only to hot and the other to an outside rail, and mytest meter is high impedence so it doesn't take much current flow to operate.
Any object has capacitance, and with a sensitive enough meter you can indeed read the charging current. That is one reason why metal electrical equipment is generally grounded, to prevent it from acquiring a voltage, even though there is no direct connection to a source of energy. Try your test with a metal tool box or office chair or similar large object and you will probably see a similar result.
When you measure 43 volts to the powerline neutral, you are seeing another manifestation of the same effect, but this time the current is coming from capacitance to the hot side of the power line.
To put this in perspective, a 60 Hz current of about 0.25 milliampere (.00025 amperes) causes a noticeable but not painful "tingle." Assuming you have a 10 megohm meter, it will read 87 volts at a current of about 8.7 microamperes (.0000087 amperes,) about 1/30 that amount.
This effect is the reason experienced industrial electricians often use a Wiggy or similar low-impedance measuring device when doing troubleshooting. In a factory environment, where there are many wires in a conduit or panel, even completely disconnected wires can acquire a voltage (ghost voltage.) The low impedance meter needs much more current than stray capacitance can generally provide.
@RJR posted:Arthur: I would like to hear from other forumites what results they show from such a test.
Just did your tests on my small toy train layout, basically a Fastrack version of the Lionel 1957 catalog layout. That layout has exactly one electrical connection, a single plug that feeds an older under-monitor computer power strip. The power strip feeds a 1063 transformer, an MTH Z-1000 brick, a PH-180 brick, a Base 1L and the power supply for a couple ASC2 modules. Nothing else.
Meter was a Fluke 87, 10M input resistance
Just sayin'
@Mellow Hudson Mike posted:This is a seriously important statement, from a safety perspective, with potentially huge implications for us hobbyists since most of us use several of them.
Those that are UL listed, or other international equivalent, should never have this kind of problem.
How broadly does it occur? Do you have a technical reference that supports this claim? Or is it simply internet hearsay and/or urban legend? (I'm not trying to badger you. I'm just trying to find the facts before such a fire shows up on my layout.)
Mike
Let me chime in and tell a very short story. I was away at work and when I arrived home my wife told me she could not start the sewing machine. My reply was I will look later. Next day went upstairs for something and smelled what I will call Electrical burning odor (EB0) at top of steps. went to sewing room - no odor. Front bedroom - no odor. North bedroom - no odor. Hall to TRAIN ROOM - Slight odor getting stronger as I approach train room. PANIC has now set in. Since this room is set on three circuits it should be easy to check. Lights work OK - circuit one OK. Half the layout works OK - circuit 2 Okay. No power to other half layout - circuit 3 is the problem. I immediately unplug the closest surge protector strip (high dollar one). Upon checking breakers in basement I discover a breaker tripped for half the outlets in the room. Upon going back upstairs I decided to unplug everything on that circuit and reset breaker. Second outlet has an Alexa Smart Plug (for turning on my Menards Morton Salt and Pepsi buildings). It also has a UL Approved wal wart for a digital clock. Upon trying to remove the wal wart I discover that the out face cover, the wal wart and the outlet have fused together. Bottom line I changed the receptacle and ditched the wal wart. THE WAL WART WAS THE PROBLEM! I was very fortunate I did not burn the house down only created a BLACK SOOTY V from the receptacle to the bottom of the layout. SO UL MEANS NOTHING TO ME ON WAL WARTS! Beware. You have been told!
Curtis
PS I had an electrician check every thing behind me and he agreed. WAL WART!
Curtis,
Was the digital clock imported? Could the UL listing on its the wall wart be fake?
Wall warts are very easy to test. Plug one in, short the output, stand back and see what happens.
If there's no fuse or fusible link inside them then they must be "thermally self-limiting", i.e. no fire if a dead short circuit.
This is the test UL runs on the design. In order to get the UL certification it must pass.
It's an example of a 'type certification', meaning the design is certified, but not every one is tested, because it's a test to failure. If they tested every one they manufactured there would be none left to sell.
There could be several issues here but we also can't ignore the possibility that your wall wart might not be authentic, but a cheap imported copy labeled as though it were authentic.
Mike
All the 20a wall plugs to the train room run thru a 20a switch on the wall. Shut the switches off and all the power to the plugs will be dead. Do you see any issues with a wal wart that's gone bad in this scenario when the power is off.
Tom
@Mellow Hudson Mike posted:Curtis,
Was the digital clock imported? Could the UL listing on its the wall wart be fake?
Wall warts are very easy to test. Plug one in, short the output, stand back and see what happens.
If there's no fuse or fusible link inside them then they must be "thermally self-limiting", i.e. no fire if a dead short circuit.
This is the test UL runs on the design. In order to get the UL certification it must pass.
It's an example of a 'type certification', meaning the design is certified, but not every one is tested, because it's a test to failure. If they tested every one they manufactured there would be none left to sell.
There could be several issues here but we also can't ignore the possibility that your wall wart might not be authentic, but a cheap imported copy labeled as though it were authentic.
Mike
Made in the USA Clock. Than does not mean that the wal wart was. Seems a lot of items now claim MADE IN USA but are really only ASSEMBLED HERE! From all that appears it is an official UL approved wal wart!
Here is a question for you. DO you check every UL label to make sure it is legit? Chances are you don't just like everyone else you assume it is. The actual % that is not is extremely low. BUT they are out there!
Curiosity compelled me to open the wall warts. I found a resistor connected between the wide (neutral) blade of the plug and the wire to the output and the transformer secondary. Pictures attached. To me, that is crazy. I can see no necessary electrical function that it performs. The wide blade is also connected to one terminal of the transformer primary
I'm sure the "intent" is static bleed, but I can't imagine that meets UL guidelines.
Access to this requires an OGR Forum Supporting Membership
