a car battery has an extremely low internal resistance.
@arduinoaleman
I said impedance not resistance. Have you any idea about the difference between the two?
Please stop this very naive view of what goes on inside a car electrical system.
you got it. i am not a native english spreaker - mixing up impedance and resistance. many people in europe will call a capacitor a condensator. i know, this is not correct in english. however, forgive non native speakers that they are not in perfect control of YOUR language.
when it comes to naive views ... (these were your words)
i built my first frequency counter back in 1986 with TTL chips. 3 amperes at 5 volts. this project required lots of capacitors. a 100nf for each chip and several 4700µF capacitors on the power rails.
so when it comes to capacitors - i know what i am doing and i know what i am recommending to other people.
i may be a newbie in the arduino forum compared to your thousands of posts.
however, i do have a university degree in electronic engineering and and other one in computer science.
and what is more important, i do have more than 35 years of experience in electronics design.
and believe me, it does make a difference between believing if something works or having had the experience if it works or not.
or in other words: just try it.
or in other words:
do you think the arduinio designers are idiots? - they adjusted their power regulator
to the maximum requirements of their board considering external fluctuations. and any external regulator WILL give you a cleaner voltage. no matter if you use capacitors or NOT in a DC environment (AC is different).
and please do not tell me that you could compensate for a large voltage drop in a car with a 1000µF or even a 4700µF capacitor.
well, if this large capacitor would be between the voltage regulator like the one i recommended and the arduino, the arduino might still work.
however, after such a voltage drop you would have a big, a very big problem with your car.
and you would not care anymore if your arduino is still working.
if you did not hate germans before, you will from now on.
however, in that case i would not share your general despise of other nations. why do you think my english is halfway acceptable?
we simply have different opinions - so that is OK.
i do not despise my neighbor just because he is is voting for an other party then me.
when it comes to batteries you should just distinguish between the small batteries we are using in our househoulds and larger stuff.
telcom companies are feeding most of their equipment with battery arrays that supply 48 volts. and that means 100s of amperes per computer rack. what size of capacitor would you recommend in such an environment. in a high current DC environment capacitors just dont make sense. computer companies like dell/hp/ibm are offering 48v DC/DC power supplies for their blade enclosures.
let us not be enemies - i am not working with hidden weapons (just using things i know)
arduinoaleman
ps:
you are a real expert - one of very few in the forum. and i do appreciate your posts.
REALLY.
so lets work together.
however, no expert knows everything.
arduinoaleman, just something I wanted to say-it is condenser not condensator. 
And a big cap will always help. For example: I use a LM2576 5.0 to power servos. It refuses to work unless I put a 470uF+ capacitor across the input rails.
FYI, I do not hate Germans. As a matter of fact, I was celebrating when they won the World Cup.
But a car is a very noisy environment. The spark plugs can use 20,000 volts. See my attached picture.
thanks for your friendly answer
condenser, condensator or capacitor. depending on the country you grew up you will mix up all these words.
i do have the tendancy to use more capacitors than required. capacitors are required to compensate power delivery if the power source cannot deliver enough current.
however, in a high current environment they do not make sense if you just want to attach an arduino that requires 200mA.
car batteries do not have a problem delivering up to 100 amperes for a short time. in such a scenario no capacitor would make any sense.
if you are using smaller batteries you DO NEED capacitors (100-4700µF) to compensate for current fluctuations.
and yes i do understand how these little things work:
capacitor: my body´s ability to store fat
resistor: my wive
transistor: ( amplifier /... / no words - political correctness)
arduino: frustration and fun (mostly)
in other words:
when the sun is shining you do not need a raincoat.
if you got one, you are fine, but you might not need it.
the same is true for capacitors.
however, if you are not sure, you better use them.
arduinoaleman
hi isaac96
a spark plug is like a transformer.
you have 12 volts on one side and maybe 20kv on the other side.
if you would see more than 20 volts on the side of your cars power system
at least half of your car electronics would be dead.
however, this will not happen.
the car battery has such a low impdance that it will compensate for
voltage spikes and current changes.
A spark plug is not a transformer, it's just a plug, with the spark gap at the end designed to be reliable The high voltage is generated outside the spark plug That's why the wires that connect to the spark plugs have such thick insulation.
hi DrAzzy
you are absolutely right. This is why spark plugs need these very well isolated cables.
I'd have gone for Tipp-Ex over the numbers on the meter and put new ones on with a Biro.... 8)
But on a more serious note, presumably the chargers that come with things like Garmins and so on, are able to withstand the vagaries of car power? (Or, do they rely on the 12V to the power outlet (cigarette lighter as it used to be called) being "good" in the first place, which may not be the case in a '79 Alfa, assuming it has a lighter socket.)
mixing up impedance and resistance. many people in europe will call a capacitor a condensator.
The difference being that capacitor and condenser mean the same thing and impedance and resistance do not.
You seem to be saying that making a power supply in a car is simple, just chuck in a linear regulator and all is fine. If life was as simple as this then people like ST would not be doing things like this:-
http://www.st.com/web/catalog/tools/FM116/SC1079/PF261390?ecmp=pf261390_enews_automotivenews_jun2015&sp_rid=NjkwODIzMjYzMTAS1&sp_mid=11948392
A car's environment is much more hostile that you seem to be acknowledging, that is the problem I have.
arduinoaleman needs less ale. Obviously too much has been consumed to give advice:
a spark plug is like a transformer.
I suspect that was the case prior to classes at uni too.
you have 12 volts on one side and maybe 20kv on the other side.
Please do not follow the advice of arduinoaleman, it is dangerous and unfounded. You will not find 12V on a spark plug at anypoint in the ignition cycle, other than during the instantaneous rise to discharge. 12Kv to 40Kv is what you will find on a spark plug when it is energized to fire. It is a spark gap designed to arc many times before failure.
Many kinds of instantaneous transitions occur in automotive electrical systems. The circuits used to power the electronics are highly filtered and use protective devices that are meant to keep the power clean to the critical electronics, even during electrical failure. Capacitors (condensors) are only a small (but very necessary) part of the protection built in.
Linear regulators are not sufficient in themselves to protect against the transients that occur in normal operation in a vehicle. They must have additional help, even some capacitors to make sure they too are protected. A simple fuse is not fast enough to protect a semiconductor in any circuit, they might protect against fire if properly rated.
@arduinoaleman
Please be careful when you are posting. You are saying things that can get people hurt, and cause property damage. You may think you know what you are talking about, but you simply do not communicate it well, and it is not a language, nationality or race issue. Post responsibly.
-fab
As Azzy said, a spark plug is a pair of electrodes. 20k volts will pass through the gap. If that escapes at any time, something is going to blow. Again, a car is full of voltage spikes. USE CAPACITORS!
I have just attached an oscilloscope to my BMW and had a look at the spikes. At 1000 RPM I saw less then 200mV. At 4000 RPM (equivalent to going 200kmh) i saw 250mV of voltage ripple.
so attaching an electronic device with its own voltage regulator (like Arduino) requires absolutely nothing. the input side of the device only has to accept input voltages up to 15V (more is better).
i just want to run the power regulator of my Arduino a little bit cooler when it draws more than 100mA. that is the only reason why i would use a 7809 or 7808 in this case.
so the only big voltage drop you will see is when you start your car.
i have lately disassembled an old car radio and could not find any large capacitors on the input side.
How old is said BMW? The OP's car is a - if I remember right- a '79 Alfa. 36 years old! :o
My car is a BMW with 200.000 miles (320.000 km).
It is a has a diesel engine with a different ignition system.
So you might get other readings when attaching an oscilloscope.
arduinoaleman:
It is a has a diesel engine with a different ignition system.
In the case of a Diesel engine, you don't have an ignition system at all.
arduinoaleman:
I have just attached an oscilloscope to my BMW and had a look at the spikes. At 1000 RPM I saw less then 200mV. At 4000 RPM (equivalent to going 200kmh) i saw 250mV of voltage ripple.
......
so the only big voltage drop you will see is when you start your car.
I assume what you did here was attach the scope directly to the battery terminals. This will probably be the point in the car system with the least amount of noise. However, just for fun go measure at the cigarette lighter terminals.
Besides when starting, I should also expect transients in the system when the air conditioner kicks in and out, and when the alternator field kicks in and out, and whenever any other inductive loads like power brakes, etc, are switched. And they will be different at different points in the car.
"On the right, zooming in shows that while the car is idling, the accessory output has 1/2 volt spikes every 28 milliseconds, due to the ignition firing. Note the voltage on the left is much noisier with the car running than on battery - the line on the left is thin, and the line on the right is thick."
"The transients on the automobile power supply range form the severe, high energy, transients generated by the alternator/regulator system to the low-level “noise” generated by the ignition system and various accessories."
ian332isport:
In the case of a Diesel engine, you don't have an ignition system at all.
A mere detail....
oric_dan:
http://www.littelfuse.com/data/en/application_notes/an9312.pdf
I like the look of those. Is it just a case of putting one across the incoming power rails coming from the car ?
ian332isport:
I like the look of those. Is it just a case of putting one across the incoming power rails coming from the car ?
What I would do is read a bunch of articles and try a few schemes that look good. There are a bazillion articles online.
It might take several suppression devices to be effective. Find what works. You want the protection devices near to the board being protected, in order to minimize inductive pickup - ie, short leads between protection device and protected device.
Thanks Dan,
I've made a custom board that's been running in the car for months, but was looking to add extra protection when I next re-spin the board. I'm already using automotive grade components wherever they're available, but I would like to give it the best chance of survival in the long term.
I'll do some more reading.
Ian.