So you’ve been troubleshooting some device or fooling around with a mains circuit and pop, goes the breaker. Well OK, let’s discuss a few ways of dealing with this situation.

Here are some common methods:

1. Fuses
2. Ballasts
3. Secondary, lower rated breaker
4. Dedicated line

The methods explained:

Fuses

The first and probably simplest method is to use fuses, however it soon becomes apparent that it’s also an expensive idea. Fuses should be, in my opinion, one of the last defences.

Another problem comes from an improperly wired plug (or non polarized as is the case in some countries) then you could have an open fuse by the netural side, but your device could still be at live potential! — This could be lethal.

So even if you’ve got an open fuse, don’t assume it’s safe to get your paws in there just yet, unplug it first. Likewise with power switches, if they aren’t bipolar then don’t assume, be smart.

Ballasts

As simple to implement as a fuse, but it won’t cut-out if you short the line, it will just provide whatever current the ballast lets through.

A ballast can be inductive or resistive, typically you’d want to use a resistive ballast, examples are room heaters, incandescent light bulbs, etc.

The down-side here is that if you were to use a ballast big or small enough, you can actually cause more damage with it than without it.

Explanation:

A “big” ballast (low value, allows for a lot of current) will most likely damage your device in case of a dead short, because it will continue to use your device as the current path, whereas a breaker would most likely trip open very quickly. Now if your device failed short, something already went wrong, but if you keep powering this device, chain reactions can occur and more damage can be caused. Because of this, sizing the ballasts properly is very important.

A “small” ballast (high value, allows for very little current) will display a noticeable resistive divider behavior between itself and the load, this means that the load (device) will see a lower voltage and this can cause trouble in certain cases, for example poorly (or cheaply) designed switch mode power supplies can have it’s switching / rectifier sections fail due to over-current (lower voltage forces the controller to use a higher duty cycle to compensate for this, thus longer “ON” times are seen). This can actually lead to yet another chain-reaction event.

Remember, as the smps starts up it needs to charge the input filter capacitors and this is a relatively heavy load, at this stage the voltage on the power supply may be way below it’s minimum specified rating thanks to your ballast.

Secondary breaker

You can wire up a box with an inlet and an outlet with a small breaker in between, typically breakers have a higher trip current than rated to allow for high inrush currents as seen on inductive loads such as motors, etc. So definetly do take this into consideration.

One issue however is that lower rated breakers can either be harder to find or more expensive, so keep that in mind as well.

Dedicated line

Having a dedicated line just for testing devices is ideal, if you add a secondary breaker and a properly sized ballast where applicable, you’ll never have to worry about blacking out or damaging something ever again.

You can add your own ground fault (switchable in and out) as well as residual current devices, etc. I would even suggest implementing phase indicators like you’d find on most properly installed residential panels.

It should be clear that a combination of all methods can only be ideal if it suits your needs, so there is no definitive answer here. Investigate, experiment and figure out what’s best for you.

Cheers.

It is said that you can disrupt the normal heart rhythm by passing no more than 10mA across the heart, this should be enough to cause fibrillation in most cases. What this means is that eventually you’ll die of cardiac arrest if you don’t get defibrilated in time (assuming of course you are in a shockable rhythm or can be brought to one to begin with). Heart failure is just one of the many risks involved.

Because of this you’d want to take precautions when working with mains at home, whether it’s 110VAC or 220VAC it doesn’t matter; both can be lethal if the right (or wrong) conditions are met.

A big part of safety comes from your behavior. As a general rule of thumb never grab things with both hands, specially if they have a metallic chasis and they happen to be energized — This is why most people will tell you to place your left hand behind your back or in your pocket, the reason for this is just so you don’t create that return path through your ticker with your hand by touching something that happens to be on the neutral side of the circuit while on the other hand you may be mistakenly touching the live side and vice versa.

Now, given the right conditions you could be grounded and thus your feet, or legs (point of contact) would become the return path, if you are touching live with your left hand chances are you may not make it.

So you can’t assume any safety measures are in place or functional, for instance the earth / ground wire or an RCD may be damaged or non existent. Even if they are tested and functional you still have to conduct proper behavior to avoid accidents.

Remember: Behavior. Don’t assume, don’t fool around — Be smart or be with Darwin.

We’ll discuss how to safely work on mains related projects at home on the next entry.

I see this one a lot,

People wondering why their current modes “aren’t working at all” on their multimeters. Well, chances are the fuse has been blown. Why? — Perhaps you tried to perform a parallel measurement in current mode, which would’ve caused a short.

However the point of this entry is not to point fingers, instead I’d like to remind you of the industry standard test for open fuses on multi-meters:

1. Put your DMM in Continuity mode, if it doesn’t have one just use the lowest Ohm range (In case it’s an auto-ranging DMM, just select Ohms)
2. Connect your red probe to the suspect banana jack. In this case the uA/mA input
3. If you get a low ohms reading, the fuse is fine. If you get an open circuit, it’s blown
4. Repeat step 2 for the “A” (Amps)  input and follow up with step 3.

Why it works

This circuit demonstrates how a reverse biased silicon diode can be used as a makeshift temperature sensor.

Note: If you can't find a 1 MOhm variable resistor or potentiometer simply try different fixed resistors, ranging from 1MOhm to 470K, 220K, etc.

The diode, when reverse biased, will present a certain leakage current that happens to be proportional to the junction temperature. This little current needs to be amplified, so we cascade three transistors in darlington configuration for maximum gain. The variable resistor allows us to set the overall “sensitivity” of the circuit. Without the transistors this small amount of current would be useless for us.

The small 10nF capacitor helps us obtain a certain hysteresis in the circuit, it may be removed if it’s effect is undesired.

A simple variation of this circuit would allow an LED bargraph display to indicate the temperature level. To achieve this we must create a voltage drop for each output LED, we can do this with regular silicon diodes. Then each LED would be driven by an NPN BJT with 1K base resistors.

Now, due to the high input impedance of this circuit you musn’t touch the sensor diode, likewise if you want to do remote sensing you must use a shielded cable. Failure to do so will introduce all sorts of noise into the base of the transistor rendering the circuit useless.

An even simpler version!

This version of the circuit involves only a 1n4148 diode, a high brightness LED and a resistor.

A few uAs of leakage will turn most high brightness LEDs on.

When you heat up the diode with a lighter, the leakage current is enough to turn on most modern LEDs producing visible light! — Of course, the temperatures required will most likely damage the diode in a very short period of time, however it’s one neat little experiment proving the concept at a bare minimum cost.

Hopefully you’ll be able to experiment with this circuit and come up with some useful variations!

Cheers.

What’s new:

Several packages were revised and new parts added, specifically two of the msp430 microcontrollers I use the most from the G series. And several ground related symbols. (supply symbols yet to come!)

You may visit the original post for a list of parts and download links.

Minor modifications:

• Several SMD footprints had their silkscreen redesigned with rounded corners and more accurate outlines.
• I added the package type and size at the bottom of each DIP IC footprint, for now it seems to be a good idea — we’ll see.

New parts: