=== Disobey Kirchhoff's Laws ===
Splitting a line into two branches should divide up the current, according to [[w:Kirchhoff's_circuit_laws#Kirchhoff's_current_law|Kirchhoff's Laws]], and maintain the same direction (ie, polarity) of current for both branches, yes?
[[File:KCL - Kirchhoff's circuit laws.svg|thumb|The current entering any junction is equal to the current leaving that junction. {{math|1=''i''2 + ''i''3 = ''i''1 + ''i''4}}]]
This law, also called '''Kirchhoff's first law''', or '''Kirchhoff's junction rule''', states that, for any node (junction) in an [[w:Electrical_circuit|electrical circuit]], the sum of [[w:Current_(electricity)|current]]s flowing into that node is equal to the sum of currents flowing out of that node; or equivalently: ''The algebraic sum of currents in a network of conductors meeting at a point is zero.''
But what if Kirchhoff isn't always right? What if, sometimes, those laws can be broken?
Take [https://www.youtube.com/watch?v=XInN3jk1Hy0 MrPreva's example] on YouTube translated by [https://www.youtube.com/watch?v=GFqJ5D6mkO0 MrJohnK1] and explained by [https://www.youtube.com/watch?v=0XsXe9DJiXk Chris Sykes] ...
He has split a flow of current into two branches by shorting out both sides of a transformer and reducing the inductance on one side of his transformer (by reducing the number of turns/windings) relative to the other side. This other side possesses a greater inductance and more turns/windings making it into a step-up transformation. Oddly enough, this causes A/C to reverse its polarity of current on the larger coil and add this negative current to the other (smaller) coil which graphically heats, and lights, up this smaller coil in his YouTube video!
It is possible to simulate this, under ideal conditions, using a mirror of [http://falstad.com/circuit/ Paul Falstad's simulator] demonstrating that the A/C voltage source shifts into the realm of negative watts ''all the time'' rather than alternately every half-cycle!