This is trace to the ground pin on my shield. Right at the top there. There was a short circuit and the poor little strip of copper couldn't handle the heat. That should have been a fuse.
But I know almost nothing about fuses other than what they do. How to choose a size? Where to put them? What type to use? What things to consider?
So I wanted to ask if one of you EE types would put us together a primer on fuses. Maybe some examples of types and a brief discussion of considerations when choosing a fuse.
Why do you think so? Without knowing the current and the voltage you cannot select a fuse. The trace made a pretty nice fuse. Add a jumper wire and continue.
You also need to consider where you will place a fuse. Do you want to put one in each signal wire?
Because I saw it happen. It was just a quick short as a piece of wire touched Vin and Ground at the same time. Stupid of me to be moving things while it was powered. It happened so fast.
See, that's why I was asking for someone to do a write-up on the basics. I knew that a fuse would have protected my circuit, but I know next to nothing about how to actually do that.
I'm not so much looking for help with this particular board. It going to be fixed with a little piece of wire in a few minutes. But the fact that I know nothing about how to even start with adding a fuse to a circuit bothered me. I'm sure other electrical noobs like me could benefit from such a discussion.
I think I want something that goes in-line on the wire from the battery and not on the board itself. I just want to protect against shorting the main power and burning something up. I'm not trying to bullet-proof a circuit, but I don't know even what level of protection is worthwhile.
Again I don't know how to answer. I don't understand what that would accomplish.
To start, after 50+ years of experience in EE, the conclusion is fuses are protected by the circuit
In general, fuses are useful for protecting against catastrophic shorts.
i.e. a car 12v battery (250 amp) is being used to power an Arduino 10 feet away.
Place a 1 amp fuse at the battery not at the Arduino.
Of importance is fuse coordination; 12 above such circuits requires one 12A fuse at the battery feeding twelve 1A fuses each connected its own Arduino cable.
Probably not, in that case, unless it was very carefully chosen.
For experimentation, a current limiting lab power supply is invaluable. Set the current limit to a bit above the maximum that you would expect the project to draw, and it will respond instantaneously to overcurrent. Latching power off is an option on some, as is short circuit protection.
In addition, don't be misled by the voltage rating of a fuse. That ONLY tells you that the ARC when the element melts will self extinguish at the voltage or lower. Of equal importance is the fuse holder, if one is used. Many of the older circuit boards use a thru-hole fuse as one of the components on the board.
A fuse holder must maintain a very firm electrical contact with the fuse connections and be rated for the fuseing value.
No.
The time it takes to blow is many orders of magnitude greater that the time it takes you circuits to blow.
An IC is a device that protects a fuse from blowing.
I once had a neutral wire drop off the spur my house was connected to. This caused it to be supplied by two phases of the three phase voltage. Result was that the AC in my house effectively nearly doubled. I was using an audio amplifier as a baby monitor at the time and the amplifier had a mechanism to protect the right channel. It was called the left channel.
@Grumpy_Mike - I was hoping to hear from you. Again, this is nothing about any specific circuit, more of a general question because I realized that there was a gaping hole in my knowledge.
What about basic protection. For instance I am using a LiPo battery. It would be very bad for that to get shorted and stay that way. Even if the circuit fries, it would be nice to not have a full on house-fire.
One question that vexes me is choosing a size. Let say I have a device with some servos and motors that normally pulls a little under 2A at full draw. I obviously don't want to just put a 2A fuse there because it will just pop any time I stall a servo. And a real short is going to be a LOT more amps than 2. So how much head room should I leave myself?
Fuses protect the wires and the supply, not the circuit being fed. They interrupt currents large enough to start fires, as others have noted electronic components are perfect for protecting fuses, not the other way round. Choose a fuse just above the highest current the circuit will draw in normal operation. Some fuses are filled with silica (sand), which quenches the arc that forms when it blows. Consider how much current the power source can supply into a short circuit, for example a car battery can probably supply a couple of thousand amps for a brief period, the same with lithium batteries. HRC, high rupture capacity (with the sand), are appropriate for breaking those kinds of current, especially with DC.
So, consider how much current the supply can feed into a short circuit and how much damage that might do. Use a fuse that is rated to interrupt that much current and use the lowest value fuse for the load being fed. Place the fuse close to the source of power. In a multy cell battery place a fuse between the cells to guarantee they are protected.
Don't assume that. I have seen 13A mains fuses last 5 minutes with 30A before they blew. Not advisable but a good demonstration of the gap between the fuse rating and what it takes to blow them. Try a 2A fuse, if it doesn't blow then it's ok. If it blows try 3A.
Not all fuses will blow in the same time. There are both specialised slow blow fuses as well as fast blow fuses. Then you have resettable fuses, either the type that act like a relay with a mechanical action like the residue ground current you get in mains, or the over current solid state ones like you get in the path between a USB connector and a device.
This latter type I used to use when designing access control systems. The time taken to blow depends on the size of the excess current, sometimes several minutes. So you have to carefully assess each situation. You can also consider the excess current monitoring circuits used in power supplies. The so called crow bar circuits. With this sort of thing you can design your own characteristics to be as complex as you need.
So for your example above I would look at the sort of slow blow fuses you can get and see if they will do.
No. I have seen equipment fuses with silver elements that will handle 100 percent of the rated current without heating, but will open with perhaps 10 percent or less overload. My plant had three 200 amp, three-phase services that were fused to blow at just a bit over 200 amps. Turning on a big transformer would sometimes pop a fuse and they were $75 each, so we just left the transformer on 24/7/365.
Ok, so we've determined that a lot of things won't benefit from a fuse.
What about places where you absolutely must have one? (or a breaker). I'm particularly interested in the sorts of DC applications we run into here. Should I have some sort of fuse on my LiPo to protect it from ever going dead short? How about automotive applications? Cars have lots of fuses.
I would say so. If it can handle inrush current, then it is just a slow blow fuse. Sure it will blow with sustained over current.
I am reminded of a line in the satirical news show called "On the hour". It was and spoof advert for a food processors:-
"We let the fruit mature on the tree slowly, and then we harvest it quickly. "
Here are some data from Littelfuse:
3AG Fast-acting fuse 312/318 Series
0.062A --- 10A: @ 100% load will open in 4 hours Min
135% 1 hour Max
200% 5 seconds Max
3AG Time Lag fuse 373/375 Series
12A, 15A: @ 100% load will open in 4 hours Min
135% 1 hour Max
200% 5 seconds Min
60 seconds Max
3AB Very Fast-Acting fuse 322/332 Series
1A --- 10A: @ 100% load will open in 4 hours Min
250% 0.2 seconds Max
Are you saying that because you can imagine a failure mode that can't be prevented then we shouldn't bother to try to prevent any of them?
I'm sorry Larry, I don't have another way to say this. That sounds US Government level stupid.
By your logic there shouldn't be any fuses anywhere in the world.
I'm sorry I asked. I really wanted to try to learn something useful. But it seems like I've run into grouchy old engineer wants to tell me none of it works.
Can someone please talk about the actual question, how to choose and size and place these things instead of just imagining all the potential places they might not help?
Some kind of battery management must be an integral part of these batteries.
Both in charging and discharging; this would be the responsibility of the manufacturer.
For cost saving, I am sure many manufacturers cut safety in their designs.
In our region, there have been several instances of spontaneous fires with LiPos.
One can imagine this can be due to: manufacturing, abuse physical and electrically, deterioration, and environmental.
Should we add our own protection circuitry to prevent fire ?
One needs to ask if the technology is such that there is an inherent/potential problem with LiPos. Shouldn’t we first question if we should even use them, or why should it be up to the end user to provide solutions to battery design flaws.
If I had to use LiPos, I would want two levels of overload current discharge protection.
A fast acting fuse, possibly rated at 3/4 the recommended discharge rate and an electronic disconnect of some sort.
As for LiPo batteries that spontaneously catch fire the only solution is don’t use them in the first place.
Maybe the newer technologies will have a break thru, such as solid state, sodium . . .
