EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: lordvader88 on June 23, 2019, 05:13:26 pm
-
I have a 52Vrms transformer, so around 73V rectified. And I notice it sparks a little, connecting/un-connecting it Usually I'm at LV so I never notice this. So far I've never shocked myself on a breadboard by touching two live points with my finger. But I'm hardly ever over 12V.
When do u start getting enough voltage to get shocks that way ? I've touched the +35V on live BJT cases, but I will not be testing to see if that's high enough to overcome skin resistance if I bridged it.
-
It's not all that clear cut. It depends on the moisture content of your skin as well as the location of the connections on your body.
My friend touches a 9V battery with her fingers, wet from saliva, to test it. She feels a slight tingle.
The general consensus is that 60 V is the threshold above which it starts to get dangerous. Many of us have connected with the power line or even a still-charged television tube and have lived to tell the story.
A friend of mine got zapped by about 700 V and it threw him across the room; he had serious damage to his hand but it eventually healed.
Best not to be paranoid, simple use common sense. The rule is to keep one hand in your pocket so that any shock you get won't pass through your chest with potential stimulation of the heart.
-
It's best to follow the safety standards, not necessarily exactly, because they vary from place to place and time to time, but for example, 50V for AC and 100V for DC are good general limits. (AC here means anything from about 10Hz to about 10kHz. High frequency AC is a different beast.)
Electroboom on Youtube has some videos where he quantifies the amount of pain caused by different voltages and frequencies using his own pain index numbers.
Sparks have surprisingly little to do with voltage alone. Even very low voltage sources can spark quite a bit, especially DC, given a lot of current capability, some inductance from wires, and bad contacts.
-
Sparks have surprisingly little to do with voltage alone. Even very low voltage sources can spark quite a bit, especially DC, given a lot of current capability, some inductance from wires, and bad contacts.
Perhaps the best demonstration of that is to look at the commutator of a small brushed DC motor running on only a few volts.
Arc welders also operate at relatively low voltage but very high current.
-
Sparks have surprisingly little to do with voltage alone. Even very low voltage sources can spark quite a bit, especially DC, given a lot of current capability, some inductance from wires, and bad contacts.
Perhaps the best demonstration of that is to look at the commutator of a small brushed DC motor running on only a few volts.
Arc welders also operate at relatively low voltage but very high current.
Yes. You can think about this more deeply, though: there are high voltages actually involved!
Just because you have a power supply that outputs, say, 10 volts, doesn't mean all voltages supplied by it are limited to 10 volts. The arc welder, for example, has a massive inductor inside by design. Now, when the low voltage supply (say 30V) is shorted to the workpiece, the current starts increasing. When the connection lifts away for a tiny fraction of second, the inductor tries to keep the current flowing, producing a massive voltage peak - look up the schematic of a boost converter, it's exactly the same.
The same happens in any situation, to a lesser extent, thanks to the inductance always present in the wires. The inductance is smaller, but OTOH that means the current can rise faster as well. The larger the current (supplied by capacitors, for example), the more energy stored in the inductance, and you get sparks. This way, even if you have a wimpy 12V, 1A wallwart, it's well possible that you see currents way over 20-30A (supplied by capacitors), and voltages way over 50V, when you short it. An intermittent connection to a dirty, rough surface gives more sparks, as there very frequent "pumping" of energy. Like a boost converter operating continuously!
With a tiny bit of wire inductance, these overvoltages won't last for a long time, but it's enough to help starting the spark. Once the spark is running, much lower voltage can sustain it.
-
If you think only inductance matters, tap both poles of a car battery with a low inductance, short buss bar. I guarantee you'll see sparks. :-DD
-
It's not all that clear cut. It depends on the moisture content of your skin as well as the location of the connections on your body.
My friend touches a 9V battery with her fingers, wet from saliva, to test it. She feels a slight tingle.
I thought many people (myself included) tested 9V batteries by putting them on your tongue. It's unmistakable the difference between a good and a bad 9V battery... :D
-
If you think only inductance matters, tap both poles of a car battery with a low inductance, short buss bar. I guarantee you'll see sparks. :-DD
The battery also has quite a lot of inductance (think about the path the current needs to take), and the shorter and "lower inductance" the busbar is, the lower resistance it is as well, so the L-to-R ratio stays - meaning more current, which again means more stored energy per L, so even if the L is now smaller, the energy storage is still there - enough to help setting up the spark.
Assuming 100nH (from an online wire inductance calculator, for d=100mm, len=300mm, approximating the current path in the car battery), for example, and 1000A limited by resistances, the stored energy is 1/2 LI^2 = 0.05 J, which would be 50W average over 1 ms, or 500W average over 100 µs, or 5000W average over 10 µs, in a very small space (way-less-than-a-millimeter gap). I'm sure there are better ways to model this, but as a rough calculation using the most basic definitions. Released only once, it's not much, but it's capable of sparking over an airgap of, say, 0.1mm to 0.5mm, which 12V cannot simply do. Then you have some ionized air there locally, and the spark can go on at 12V.
And now indeed, if you short-circuit, or release the short, in a seriously planned, steady but quick motion (been there, done that), you won't get sparks that big. This is because the number of connections and disconnections is low. The number of energy pumping events is limited. But if your contact points bounce, you first have a short circuit for, say, some tens to hundred µs, then it breaks and you release the 0.05J of stored energy, then you again have a contact, and this goes on. The timescale is so fast that tens or hundreds such events may be seen in just a quick ZAP.
It's very difficult to emulate not having the inductance in the real world, because it's always there.
Hope this helps.
-
https://www.youtube.com/watch?v=-5R-KBa18ME (https://www.youtube.com/watch?v=-5R-KBa18ME)
-
I have a 52Vrms transformer, so around 73V rectified. And I notice it sparks a little, connecting/un-connecting it
When opening a contact mechanically you initially won´t have air available as an insulator. If enough energy to ionize air is present then it will arc over and ionized air is a good conductor, so you see a spark even at low voltages. The faster you get distance between contacts, the more voltage is required to arc over.
Inductors as switched loads are more likely to arc, because the sudden change in current leads to back-emf, supplying more energy for arcing over.
It would require a lot more voltage for arcing over a fixed distance, which makes these cases different from what you describe.
-
First time in comms class..we were playing with CB radios (12V)..had the bare center coax lead touch on my arm (sitting on a steel topped table) & keyed the mike.
Ow, it hurt!..small RF burn.
Another time, had my fingertip on a big glass PIN diode (35W Xmitter) & keyed it..another bad plan.. got another very hot, extremely fast, RF burn..
Will I ever learn? ..probably not! ;-)
-
AC generally considered to be more dangerous than DC. If you have dry skin, you'll probably be able to touch the 72VDC rectified output of the transformer, without feeling more than a slight tingle and you'll experience the strongest sensation when the current is initiated and interrupted. On the other hand touching the 50VAC from the transformer will probably deliver a fairly decent shock.
Note: if you do the above, check neither side of the transformer, rectifier or the circuit it's powering is connected to earth at any point and only touch the two conductors with one hand only, that way the current will not flow through your heart.
-
This is my personal practice.
When voltage exceed 100V, either DC or A/C peak, I start taking extra precautions. Check and double check. Verify where my left hand is, verify where my pinky on right hand is, make sure my legs aren't touching grounded anything, check connections, before anything happens. (I'm right handed) I have much more rigorous process for voltage exceeding 500V. I will not touch anything over 1000V.
I have personally touched 500V and I have a coworker who did the same with 3000V. We both lived but with nasty and deep burn. It was a sheer luck my colleague lived. It entered his right hand and left his elbow. Should some of it go through his chest area, he'd be dead. Power supplies in both cases were capable of around 1A.
Or, when anytime current exceeds 50A or higher, AND there is no current limiting at or around 50A. Even at 2 to 3 volts, 200A can cause serious damage. If you happen to short that much current with a wrench or something, it will turn red-hot until source explodes, or load melts down. I have seen this happen and battery explodes. Very nasty and dangerous situation.
Also, RF has different issues. RF burns are often deep. But I'm going to stop here, because you weren't asking about RF.
-
Generally speaking, 50V is the highest that is considered "low voltage" and safe to handle from the standpoint of getting shocked. As others have said, high current has other hazards even at much lower voltages. 1V with enough current can spot weld or start fires.
-
It pays to be careful, people can and have been killed by 120V mains, the amount of sweat and salt on your skin and where you happen to touch, where the current finds a path to ground can all make a big difference.
-
The battery also has quite a lot of inductance (think about the path the current needs to take), and the shorter and "lower inductance" the busbar is, the lower resistance it is as well, so the L-to-R ratio stays - meaning more current, which again means more stored energy per L, so even if the L is now smaller, the energy storage is still there - enough to help setting up the spark.
Assuming 100nH (from an online wire inductance calculator, for d=100mm, len=300mm, approximating the current path in the car battery), for example, and 1000A limited by resistances, the stored energy is 1/2 LI^2 = 0.05 J, which would be 50W average over 1 ms, or 500W average over 100 µs, or 5000W average over 10 µs, in a very small space (way-less-than-a-millimeter gap). I'm sure there are better ways to model this, but as a rough calculation using the most basic definitions. Released only once, it's not much, but it's capable of sparking over an airgap of, say, 0.1mm to 0.5mm, which 12V cannot simply do. Then you have some ionized air there locally, and the spark can go on at 12V.
Some interesting number crunching - but I feel you are missing something rather fundamental...
Is the "spark" a visible phenomenon of an air gap breakdown - or is it the incandescent glow of bits of metal being vaporised? The former needs some notable voltage, the latter just needs enough energy.
-
The battery also has quite a lot of inductance (think about the path the current needs to take), and the shorter and "lower inductance" the busbar is, the lower resistance it is as well, so the L-to-R ratio stays - meaning more current, which again means more stored energy per L, so even if the L is now smaller, the energy storage is still there - enough to help setting up the spark.
Assuming 100nH (from an online wire inductance calculator, for d=100mm, len=300mm, approximating the current path in the car battery), for example, and 1000A limited by resistances, the stored energy is 1/2 LI^2 = 0.05 J, which would be 50W average over 1 ms, or 500W average over 100 µs, or 5000W average over 10 µs, in a very small space (way-less-than-a-millimeter gap). I'm sure there are better ways to model this, but as a rough calculation using the most basic definitions. Released only once, it's not much, but it's capable of sparking over an airgap of, say, 0.1mm to 0.5mm, which 12V cannot simply do. Then you have some ionized air there locally, and the spark can go on at 12V.
Some interesting number crunching - but I feel you are missing something rather fundamental...
Is the "spark" a visible phenomenon of an air gap breakdown - or is it the incandescent glow of bits of metal being vaporised? The former needs some notable voltage, the latter just needs enough energy.
It's both, to varying ratios.
-
AC generally considered to be more dangerous than DC.
Depends, although AC has a lower threshold to when it feels and can couple through shoes for example, DC is much more dangerous IF you get shocked to a high voltage line, because the current does not go through zero, thus preventing a chance to disconnect yourself by your own muscle force. So AC increases the risks of getting zapped, but DC decreases the survival after you get shocked.
That's not true at all. AC is better at disrupting the heart's rhythm and the zero crossing time is far too short to make any difference.
And what do you mean by coupling through the shoes? The capacitance between your feet and ground is too small to make any difference, at mains frequencies.
I don't know why the myth that DC is more dangerous, than AC still proliferates so widely today, when it's been know that the reverse is true for well over a hundred years.
https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19730001338.pdf (https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19730001338.pdf)
http://www.wright.edu/~guy.vandegrift/wikifiles/Electric%20shock%20voltages%20Dalziel.pdf (http://www.wright.edu/~guy.vandegrift/wikifiles/Electric%20shock%20voltages%20Dalziel.pdf)
-
In my experience DC systems (125-250VDC) are seldom ever able to deliver a shock. This is because they are ungrounded DC supplies. I can touch either the + or - voltage and feel nothing. Because they are ungrounded & the topography of the systems I work with makes it unlikely that anybody can come into accidental contact with both + & - simultaneously. Not impossible, but unlikely.
What do we know about AC mains voltage ? It is grounded. You are in frequent contact with that earth reference, be it the ground or a metal chassis. It becomes much easier & likely that you can accidentally come into contact with an ungrounded conductor & experience a shock. I have many times in the distant past been shocked by a single point of contact, of course my feet were always the second point.
Safe voltages to handle with bare hands: Up to 48VDC & 25VAC. The numbers may be more conservative than the ones posted so far. This is my employers upper limit, & they are more stringent than the industry standards. Above this point gloves are required.
Any non-conductive glove will reduce the potential for shock. I like Nitrile gloves because they offer a good degree of protection & you can still feel through them. They are cheep too, one can get 50 pairs of either 7 or 9 mils thick for $12. I have personally tested the 9 mil to 250VAC between gloves. (120VAC to earth). At work voltage rated gloves with leather protectors are required.
-
The battery also has quite a lot of inductance (think about the path the current needs to take), and the shorter and "lower inductance" the busbar is, the lower resistance it is as well, so the L-to-R ratio stays - meaning more current, which again means more stored energy per L, so even if the L is now smaller, the energy storage is still there - enough to help setting up the spark.
Assuming 100nH (from an online wire inductance calculator, for d=100mm, len=300mm, approximating the current path in the car battery), for example, and 1000A limited by resistances, the stored energy is 1/2 LI^2 = 0.05 J, which would be 50W average over 1 ms, or 500W average over 100 µs, or 5000W average over 10 µs, in a very small space (way-less-than-a-millimeter gap). I'm sure there are better ways to model this, but as a rough calculation using the most basic definitions. Released only once, it's not much, but it's capable of sparking over an airgap of, say, 0.1mm to 0.5mm, which 12V cannot simply do. Then you have some ionized air there locally, and the spark can go on at 12V.
Some interesting number crunching - but I feel you are missing something rather fundamental...
Is the "spark" a visible phenomenon of an air gap breakdown - or is it the incandescent glow of bits of metal being vaporised? The former needs some notable voltage, the latter just needs enough energy.
You're forgetting something though, a spark often results when a circuit is broken, this means the air gap starts at zero and then grows progressively larger. The breakdown of a few molecules of air gap is pretty low, then once it ionizes the arc can be stretched over a larger distance.
Welders and carbon arc lamps can run at 20-30 volts. With both you have to strike the arc by briefly touching the electrodes.
-
You're forgetting something though, a spark often results when a circuit is broken, this means the air gap starts at zero and then grows progressively larger. The breakdown of a few molecules of air gap is pretty low
This - but do note that 12V as such isn't able to break down even the minuscule air gap (I forgot the exact value for a breakdown to happen at all, but AFAIK there really is a threshold, and it's higher than 12V). Hence, the inductance is needed.
-
You're forgetting something though, a spark often results when a circuit is broken, this means the air gap starts at zero and then grows progressively larger. The breakdown of a few molecules of air gap is pretty low
This - but do note that 12V as such isn't able to break down even the minuscule air gap (I forgot the exact value for a breakdown to happen at all, but AFAIK there really is a threshold, and it's higher than 12V). Hence, the inductance is needed.
Inductance is always present though.
Yes, 12V will never cause a breakdown at ambient air pressures, unless something is ionising the air such as x-rays. Look up Paschen's law.
https://en.wikipedia.org/wiki/Paschen%27s_law
-
There may be another factor at work here, metal vapor resulting from a very small contact patch vaporizing. I have a small hot cathode mercury UV lamp that maintains a mercury discharge at just 10V.
-
Growing up in a rural area you at some point contact the 'hot wire' that surrounds cows, pigs or chickens, those are around 2000 volts at 200ma but they switch on/off about 300 times a second, the switching
means you can let go or an animal can back away and also the higher frequency creates heat or a burning sensation on the skin which is more a deterrent than just the unpleasant shocking sensation.
Grounding is really important with an install because without the ground connection the wire doesn't work and it really helps you understand at a young age how getting shocked and grounding are related.
-
I thought many people (myself included) tested 9V batteries by putting them on your tongue. It's unmistakable the difference between a good and a bad 9V battery... :D
I'm glad I'm not the only one.
-
Growing up in a rural area you at some point contact the 'hot wire' that surrounds cows, pigs or chickens, those are around 2000 volts at 200ma but they switch on/off about 300 times a second, the switching
means you can let go or an animal can back away and also the higher frequency creates heat or a burning sensation on the skin which is more a deterrent than just the unpleasant shocking sensation.
Grounding is really important with an install because without the ground connection the wire doesn't work and it really helps you understand at a young age how getting shocked and grounding are related.
I thought electric fences worked at a much lower frequency than 300Hz, which is too fast for the zero crossing point to allow someone to let go. Even mains frequencies are too high for that.
I thought they generally uses short pulses at around 1Hz, so a low duty cycle.
-
The house I grew up in had an electric fence across the back of the property which belonged to the neighbor. It definitely pulsed at about 1Hz, I remember daring friends to grab onto it when we were kids, and also you could touch it with a damp stick and get a small spark to your finger without getting a noticeable shock.
-
With fencers you don't want to kill the cow/horse, just give them a jolt to move away. So the pulse is tailored for pain receptors, not muscles.
I remember OSHA lowered the safe voltage limit last year, based on people getting electrocuted on 24-30V. Does anyone know about it?
-
Electric fences here in NZ use 2 standards depending on whether they are mains or battery powered.
Mains units can only operate at 1 Hz whereas battery units might have 3 speed settings, slow for battery saving and already well trained animals, normal 1 Hz operation or fast for training livestock.
10KV is permitted for battery operation as the energy (Joules) is relatively low.
Mains units on the other hand might emit some 40+ Joules and that's more than enough to kill if the duty cycle wasn't low, in the order of low single digit ms.
With this amount of energy mains units are generally 7KV or less as carbonizing insulators should the pulse track over them is an ongoing fence maintenance issue. Mains units are generally matched to the farmers needs by a measure of km of electric fence line so to not create problems from applying to much energy.
Once units were dumb and just dropping the entire charged caps across the primary of the step up transformer but in order to increase their pulse grade caps lifetime they now have some fancier control to emit bigger pulses after the first hit to ensure better animal control.
Trust me you don't want to mess with a modern mains electric fence ! :scared:
They'll sit you on your arse in an instant.
When you see a 1000lb bullock bounce 4" completely vertical after getting a belt you know to respect them.
-
There may be another factor at work here, metal vapor resulting from a very small contact patch vaporizing. I have a small hot cathode mercury UV lamp that maintains a mercury discharge at just 10V.
Metal vapour sounds plausible.
Yes, when the electrode is heated, the breakdown voltage reduces, your UV lamp probably runs at a much lower pressure than barometric and I believe mercury vapour also has a lower breakdown voltage, than air, hence the vastly lower operating voltage.
-
Metal vapour sounds plausible.
Yes, when the electrode is heated, the breakdown voltage reduces, your UV lamp probably runs at a much lower pressure than barometric and I believe mercury vapour also has a lower breakdown voltage, than air, hence the vastly lower operating voltage.
I'm quite sure it does, but the discharge also spans a distance of around 10mm once it reaches operating temperature. When the gap is a few microns as the contacts of a switch first begin to open it would not surprise me if a discharge could start at atmospheric pressure.
-
Electric fences here in NZ .... When you see a 1000lb bullock bounce 4" completely vertical after getting a belt you know to respect them
I learned about electric fences in NZ when I visited there back in '75. I got a job fixing electrical stuff on a huge cow / bull farm ....
Came across this single flimsy wire surrounding the compound and laughed at how the fook it was supposed to keep the animals in ... I found out :-)
They had to run the fences full belt at 3Hz !! Rules were a bit relaxed back then !
The 2nd time, I was driving the tractor, with 2 guys on the back dropping off hay bales and feed .. we had to drive right along the E-fence line ... I misjudged one time and clipped the fence, which snapped and sprung back, wrapping itself around the tractor :-) Ever see 3 guys dancing at 3Hz on a tractor ? Lost that job :-)
-
I thought electric fences worked at a much lower frequency than 300Hz, which is too fast for the zero crossing point to allow someone to let go. Even mains frequencies are too high for that.
I thought they generally uses short pulses at around 1Hz, so a low duty cycle.
I should have been clearer, the frequency is not 300Hz, the pulse width is 1/300th of a second.
-
With fencers you don't want to kill the cow/horse, just give them a jolt to move away. So the pulse is tailored for pain receptors, not muscles.
I remember OSHA lowered the safe voltage limit last year, based on people getting electrocuted on 24-30V. Does anyone know about it?
I see low voltage work jobs still quoting 48V in requirements ,so I think OSHA is still at 50V
-
It's also printed in some of the manuals from multimeters the high voltage point or signal eg fluke 17b+ . It also notice that current will flow as much you skin allow it.... as in a pratical bad and tragical example:
https://www.youtube.com/watch?v=6Dd6_TghcE0 (https://www.youtube.com/watch?v=6Dd6_TghcE0)
The continuation:
https://www.youtube.com/watch?v=GlM6PE2kKVY (https://www.youtube.com/watch?v=GlM6PE2kKVY)
-
I see low voltage work jobs still quoting 48V in requirements ,so I think OSHA is still at 50V
Aside from OSHA there is also the NEC in the US and similar regulatory bodies in other nations. As I recall a low voltage electrician doesn't need to be licensed, anyone can do it. For line voltage stuff you need to be licensed unless you're the homeowner in which case you can do your own work but you're still supposed to get a permit and have it inspected. I don't think low voltage work needs either.
-
A good rule of thumb is this:
Voltage hurts.
Amps kill.
Tasers operate in the hundreds of thousands of volts, but nano-amps, so they hurt like the dickens, but that's about it.
Mains is only 120v, but it's more than capable of putting 50mA across your heart, killing you instantly.
-
A good rule of thumb is this:
Voltage hurts.
Amps kill.
No, this is total bullshit, makes absolutely zero sense, and confuse the shit out of beginners. This has been debunked countless of times by countless of people. Mehdi for example has some good commentary on this.
-
A good rule of thumb is this:
Voltage hurts.
Amps kill.
No, this is total bullshit, makes absolutely zero sense, and confuse the shit out of beginners. This has been debunked countless of times by countless of people. Mehdi for example has some good commentary on this.
just check this post above:
https://www.eevblog.com/forum/beginners/sparky-sparky-when-does-voltage-become-more-dangerous/msg2529075/#msg2529075 (https://www.eevblog.com/forum/beginners/sparky-sparky-when-does-voltage-become-more-dangerous/msg2529075/#msg2529075)
and more:
https://www.youtube.com/watch?v=XDf2nhfxVzg (https://www.youtube.com/watch?v=XDf2nhfxVzg)
And do not try this anywhere!!!
-
A good rule of thumb is this:
Voltage hurts.
Amps kill.
Tasers operate in the hundreds of thousands of volts, but nano-amps, so they hurt like the dickens, but that's about it.
Mains is only 120v, but it's more than capable of putting 50mA across your heart, killing you instantly.
If that were true, they wouldn't design modern welders to be high current low voltage machines.
-
It is the amps that kill, but you need sufficiently high voltage to push the amps through your skin. Same deal with the taser, you have an unpredictable contact resistance and skin resistance depending on where it hits so it produces a sufficiently high voltage for that to not matter.
-
Keithley SMU's have a safety interlock which limits the output voltage to 50V. It's obvious the commonly accepted limit is around that number somewhere.
-
It is the joules (voltage x current) that kills.
Energy has to be transfered & absorbed.
-
IEC / IET identify three main voltage levels. Extra low voltage (<50 Vac rms / <120 Vdc), low voltage ( 50<U<1000 Vac RMS / 120 < U 1000 Vdc ) and high voltage (>1000 Vac / >1200 Vdc). Yes, so called mains voltage is considered as low voltage, which doesn't mean it is safe in any respect.
It would be nice if people in electronics would actually use this terminology correctly.
-
It may depend on where you are. Here in the US, "low voltage" means under 50V, at least to an electrician. A "low voltage electrician" is a real job, they install things like network cables, alarm systems and industrial control stuff.
-
FYI for clarity and educational purposes.
"Joules" is not the unit to quantify what will kill you. It only represents the amount of energy delivered.
A Coulomb a unit of electric charge (Current per second) is.
-
FYI for clarity and educational purposes.
"Joules" is not the unit to quantify what will kill you. It only represents the amount of energy delivered.
A Coulomb a unit of electric charge (Current per second) is.
A coulomb is an ampere (unit of current) times a second, not per second.
An ampere is a coulomb (6.24 x 1018 electrons) per second.
-
You're forgetting something though, a spark often results when a circuit is broken, this means the air gap starts at zero and then grows progressively larger. The breakdown of a few molecules of air gap is pretty low
This - but do note that 12V as such isn't able to break down even the minuscule air gap (I forgot the exact value for a breakdown to happen at all, but AFAIK there really is a threshold, and it's higher than 12V). Hence, the inductance is needed.
From Paschen's law, breakdown voltage reaches a minimum and starts going up below some minimum distance. Many people are familiar with how breakdown voltage of gases decreases with decreasing pressure, up until it starts increasing again as the gas becomes to rarefied to ionize. The x-axis of the Paschen curve is actually the pressure-distance product, so the same effect applies as the distance is decreased. For air at atmospheric pressure the minimum is 327 V. Attached plots are from RCA's excellent "Electron Tube Design" from 1962, pages 796-797.
Spark formation on breaking a current-carrying connection is maybe a bit more involved, as current density and heating will go up as the contact area and force are reduced, possibly evaporating some of the contact metal before a gap has time to form. Whether this metal vapor is thermally ionized or not will determine if the arc obeys Paschen's law as far as I can imagine, but I'm not an expert on this.