Car automotive "produces several hundred volts"

I don't believe for a second an alternator (with a regulator onboard) and a 12v battery would ever allow transistient voltages to rise much past 18vdc let alone 200vdc

The battery would absorb most of the spikes.

And lastly... if this was true why is it i've yet to see a car stereo or infact any electronics i've used in the car over the last 30 years die or explode, not one linear regulator ever...

What am i missing? That others seem to know about who keep telling people this?

Maybe they mean on the HT side, but of course that's in the 10s of thousands, not hundreds.

What's the biggest current drawn by a starter?- maybe they mean amps?

Some things are like fish stories. The hands keep getting farther apart with each telling of the big one that got away.

The short answer is that it can't without some help, '200' A represents a very low resistance. However a 10 A load switched on at the end of 4 to 6 meters of 1.5mm wire connected to that 200 A current source can create short pulses of that magnitude easily. When you realize that in a car the wire, with the adjacent steel of the automobile forms the core and 'coil' of an inductor. The inductance of a plain wire in a loom can be as much as several millihenries. Mutual coupling between wires at those power levels is also a problem as the other wires in the loom are windings on a transformer.

Doc

Have you ever observed this?

Most of the thoughts expressed were mine when dealing with an issue I had about 25 years ago with a micro that intermittently reset itself.. A couple of beads fixed most of the issue, a pair of 100 pF caps did the rest. Since the wires leaving my device were well protected initially ( I Thought ).. I spent some time to track down the source of the pulses. There is another condition that can cause the same problem and that is dirty battery connections at the battery and initially I thought that was the issue and wrote it off with a note to the project file. When it happened a second time I found that it was an intermittent issue affecting only one of the field test vehicles but I decided to look into it and found it was the electric fuel pump that was causing the pulses and a loose ground to the car's frame that was the intermittent. Yes I've seen the voltage spike and saw it spike on several different wires as well at various levels depending on the condition of the ground.
The condition with dirty battery terminals or worse a full load drop can easily cause the alternator voltage to go to several hundred volts but not often or for long.. If it is dirty terminals the transient can occur, A load drop is just that, when the battery is disconnected from a working alternator dropping it's load.

Doc

What am i missing? That others seem to know about who keep telling people this?

Who are the others? Perhaps the "others" heard it from a psychic talking dog.

No one that knows what they're talking about. Only those losers over at JEDEC, for instance... XD

Docedison:
Most of the thoughts expressed were mine when dealing with an issue I had about 25 years ago with a micro that intermittently reset itself.. A couple of beads fixed most of the issue, a pair of 100 pF caps did the rest. Since the wires leaving my device were well protected initially ( I Thought ).. I spent some time to track down the source of the pulses. There is another condition that can cause the same problem and that is dirty battery connections at the battery and initially I thought that was the issue and wrote it off with a note to the project file. When it happened a second time I found that it was an intermittent issue affecting only one of the field test vehicles but I decided to look into it and found it was the electric fuel pump that was causing the pulses and a loose ground to the car's frame that was the intermittent. Yes I've seen the voltage spike and saw it spike on several different wires as well at various levels depending on the condition of the ground.
The condition with dirty battery terminals or worse a full load drop can easily cause the alternator voltage to go to several hundred volts but not often or for long.. If it is dirty terminals the transient can occur, A load drop is just that, when the battery is disconnected from a working alternator dropping it's load.

Doc

Well, if the battery is not properly connected, then of course you're going to get voltage spikes as the car cranks over at different rpm's the onboard regulator on the alternator should keep it steadyish voltage but yeah I could easily understand 20vdc+ spikes unless you have a heavy load.

But feel free to check for voltage spikes please make sure the terminals are tightly connected before you do...

And just what kind of fluctuations were you witnessing? how many volts peak?

The condition with dirty battery terminals or worse a full load drop can easily cause the alternator voltage to go to several hundred volts but not often or for long.. If it is dirty terminals the transient can occur, A load drop is just that, when the battery is disconnected from a working alternator dropping it's load.

It has been my experience that with dirty battery terminals, the engine will barely turn over and the lighting gets really dim. If the alternator has any output, it is going to the starter and other electronics, lights, etc.. When Chrysler replaced generators with alternators in their cars, they did a demonstration of starting a car, removing its battery, then driving it (as best as I remember) from New York to Chicago.

Relays switching on and off will also cause localized spikes. I've seen people measure (DSO 1MOhm inputs) upwards of 200-300V transients (on the order of uS or shorter in duration) from these. This is the cause of amplifier "thump" in badly designed aftermarket audio setups.These spikes though are very localized, by the time they get outside of the very immediate area, they drop rapidly as the charge dissipates away. Additionally, these spikes aren't much risk for low impedence devices as the amount of charge is very low. For high impedence devices like microcontrollers, they can be fatal. Otherwise, short spikes in the 20-30V range are very common but full system spikes in the 100-200V range sounds suspect to me.

That is not the cause of thump in cheap audio amps. That is the noise of the output transistors settling down to their final bias point.

A poorly chosen ground can result in some strange things appearing on a 'scope screen.

Docedison:
Most of the thoughts expressed were mine when dealing with an issue I had about 25 years ago with a micro that intermittently reset itself.. A couple of beads fixed most of the issue, a pair of 100 pF caps did the rest. Since the wires leaving my device were well protected initially ( I Thought ).. I spent some time to track down the source of the pulses. There is another condition that can cause the same problem and that is dirty battery connections at the battery and initially I thought that was the issue and wrote it off with a note to the project file. When it happened a second time I found that it was an intermittent issue affecting only one of the field test vehicles but I decided to look into it and found it was the electric fuel pump that was causing the pulses and a loose ground to the car's frame that was the intermittent. Yes I've seen the voltage spike and saw it spike on several different wires as well at various levels depending on the condition of the ground.
The condition with dirty battery terminals or worse a full load drop can easily cause the alternator voltage to go to several hundred volts but not often or for long.. If it is dirty terminals the transient can occur, A load drop is just that, when the battery is disconnected from a working alternator dropping it's load.

Doc

Poor battery connections can also cause static build-up. I had a lot of trouble with static on my car until someone suggested tightening the battery connections. No trouble since then.

...wow, just managed to delete my own post on my phone - how annoying.

The transient experienced by disconnection of a charging alternator is known as a load dump, not a load drop. Load dump on a 12V system often lasts for around 200-400ms at around 100V, but is very much dependant on alternator speed at the time and load current. The full formula for calculating internal resistance of an alternator can be found in an Infineon app note somewhere...

Transients around several hundreds volts however are much shorter in duration, but much more common (i.e. occur in normal operation, whereas load dump is really a fault) and are caused by switching through the harness transients. They are however becoming less common through better harness design and local suppression. It must be noted of course however that transients due to switching injectors for example shouldn't be clamped, clamping an injector with a flyback diode for example can increase its closing time by a factor of around 5, effectively making large injectors unable to pulse small idle quantities of fuel, and therefore - one must just protect against these.

The reason your radio or any other device in your car hasn't blown up, is simply because they're protected against these transients. Take a look at ISO6737-2 which outlines the common transients and their sources - don't believe for a second auto transients are a wives tale, it's well known the automotive power supply is one of the harshest operating environments.

Note also, to 'view' these transients, the positive terminal isn't the place to hook up a scope - you need to probe a 12v distribution line within the harness itself, or the 12v supply to other devices such as an airbag controller or radio for example. However, depending on your vehicle - these may all be supplied by a conditioned source as opposed to a direct battery connection.

EDIT - Another challenge of designing for an automotive supply is the cranking voltage, trying to keep MOSFETs that are passing not insignificant current fully switched on with the hugely reduced voltage can be a significant challenge.

I had my tectronix (744a) scope on the fuel injector looking at the idle pulse width; and there was a significant shutdown spike, well over 60V, everytime the injector fired. IIRC the injector pulse width was about 3.2ms, and the shutdown spike was ~0.3ms(+/-).

Considering that my Arduino Mega doesn't function well withing 3 feet of a florescent light, keeping an eye our for electrical noise is not a trivial task

morrisonandboyd:
I had my tectronix (744a) scope on the fuel injector looking at the idle pulse width; and there was a significant shutdown spike, well over 60V, everytime the injector fired. IIRC the injector pulse width was about 3.2ms, and the shutdown spike was ~0.3ms(+/-).

Considering that my Arduino Mega doesn't function well withing 3 feet of a florescent light, keeping an eye our for electrical noise is not a trivial task

Thanks - this illustrates my point quite well. At the particular load side you measured, 3.2ms of fuel was required - if you then tried to suppress that spike with a flyback diode across each injector - you'll add approximately 1.5ms of fuel - well outside of any closed-loop fuelling strategy, at those sorts of extra fuel levels a fuelling error will simply be reported (not to mention idle fuel isn't often controlled closed-loop).

60V however is likely a clamped value, internally clamped by the ECU - injector spikes will often exceed 60V unclamped. The fuel tables will however be populated for that particular voltage clamp.

Thanks! Nice to get informed observations.

60 volts was an estimate because the spike was outside my focus of study and going off screen. I was mostly interested in looking at the injector drive pulse.

Hadn't really given much thought to inductor kickback in something as simple as an idling fuel injector, but there it was big as life.

Funny what you find when looking at other stuff.

We do auto ignition systems at work, interesting to see how hard a secondary looks for ground.... :smiley:

I see all manor of electronics working in close proximity to HV, ....well...trying to work.

Tomorrows work involves an oscilloscope, oven mitts, and high voltage. ]:smiley:
I admire my ability to get paid for this job.