An electric spark can be formed when there is high potential energy between two conductors, right? My question is can a spark be formed with high current and low voltage or only vice-versa?

  • 4
    \$\begingroup\$ Not sure why this has attracted such a mess of answers .. \$\endgroup\$
    – pjc50
    Commented May 22, 2018 at 9:06
  • 13
    \$\begingroup\$ @pjc50 Hot Network Questions... It's always the fault of the Hot Network Questions list. \$\endgroup\$
    – dim
    Commented May 22, 2018 at 10:56
  • 6
    \$\begingroup\$ high potential energy means high voltage \$\endgroup\$
    – jsotola
    Commented May 22, 2018 at 15:42
  • 1
    \$\begingroup\$ Immediately before the spark forms the current is zero. The rest should follow naturally. \$\endgroup\$
    – J...
    Commented May 23, 2018 at 12:18
  • \$\begingroup\$ Depends on the spark. Is 9V high voltage? electronics.stackexchange.com/questions/20699/… \$\endgroup\$ Commented May 24, 2018 at 3:08

4 Answers 4


You need a high voltage to produce a spark through air.

There are two ways to get a high voltage. One is to make a high voltage intentionally.

The other is that you can get a high voltage unintentionally by breaking a large current in an inductive circuit. As all conductors are inductive to some extent, a sufficiently high current going through an opening switch will create a spark as the contacts open and try to stop the current. Supplying a headlight bulb from a 12v battery via jump leads, and then pulling one away will usually make a spark as the connection opens.

  • 8
    \$\begingroup\$ And then St. Otto said: "First shalt thou connect the battery to the coil. Then shalt thou attach the coil to the distributor. With one big thick lead shalt thou connect the coil to the distributor! Then shalt thou connect the distributor to the spark plugs, and with big-arsed thick leads as well. And with much insulation shalt thou surround the leads, so that thy current shall not leak. Then shalt thou crank the engine, and it shall go bang at the right times!". And lo, it was so, and there was much rejoicing. \$\endgroup\$ Commented May 23, 2018 at 11:49
  • 1
    \$\begingroup\$ "You need a high voltage to produce a spark through air" - Presumably if you have the ability to reduce the electrical resistance of the air between your electrodes, that would also work. \$\endgroup\$
    – aroth
    Commented May 23, 2018 at 13:16
  • \$\begingroup\$ @aroth true; high humidity or fancy tricks like forming a plasma with lasers can reduce the required strike voltage. \$\endgroup\$
    – pjc50
    Commented May 23, 2018 at 15:51
  • \$\begingroup\$ I get high voltage unintentionally by stroking a dry cat. Aroth's comment (reducing the resistance of the air) would drain the high voltage and prevent a spark. I have never tried to pet a humidified cat. \$\endgroup\$ Commented May 27, 2018 at 20:46
  • \$\begingroup\$ Does the amps play a role in a spark? If so which one? \$\endgroup\$
    – JinSnow
    Commented Sep 26, 2019 at 18:37

Before the spark, there is no current at all, only a voltage (potential difference) between two points.

Arc discharge happens when the voltage is high enough to overcome the gap, and then continues when conductors are drawn apart until the plasma dissipates. This depends on how large the gap is; you can easily draw visible sparks from a 12V power supply by rubbing some conductors together. Tiny arcs form across the few microns of gap between surfaces that aren't perfectly flat.

Once an arc is struck it's a fairly good conductor, so the voltage across it will drop and current increase until it's limited by the rest of the system.

Van der Graff generators and similar "static electricity" systems are effectively capacitors charged to huge voltages that produce a fairly high current for an extremely short duration. This enables them to produce long, brief sparks.

Conversely arc welders operate with comparatively low voltages, maybe as low as 20V, but extremely high currents (hundreds or thousands of amps). This requires a very short distance - you have to touch the material being welded with the electrode.

  • \$\begingroup\$ Depending on the welder type, the electrode does not actually touch the weld material; for example in a TIG welder the electrode may be a quarter inch away (depending on weld power). \$\endgroup\$
    – skyler
    Commented May 22, 2018 at 17:22
  • \$\begingroup\$ @Skyler, I think what pjc50 is referring to is "striking an arc", this is what you have to do in arc wielding to start the arc. You briefly touch the work surface to start the arc, then pull the electrode away and keep it at the right distance to maintain the arc. \$\endgroup\$
    – Glen Yates
    Commented May 23, 2018 at 19:12
  • \$\begingroup\$ @GlenYates not all arc welders use touch to strike the arc. TIG welders use a brief high-voltage pulse to strike the arc and then switch over to the low-voltage, high-current circuit. \$\endgroup\$
    – skyler
    Commented May 23, 2018 at 19:49
  • \$\begingroup\$ "Before the spark, there is no current at all, only a voltage". Well, consider the dual situation: you have a closed circuit where a current is flowing and no appreciable potential between point A ad A' 1um away from two touching conductors. Then you separate the conductors and--- spark. \$\endgroup\$ Commented May 24, 2018 at 3:15

It all depends on how you define a spark. If burning metal particles count as a spark, you can create one with very low voltages. Shorting a 1.5V AA battery creates such sparks which can be easily seen. What you need here is sufficient current to melt the metal, typically currents of at least 1..5 A are needed for sparks to be observable in daylight.

If we're talking electric arcs between fixed electrodes, you need to fulfil the conditions of Paschen's law which relates voltage, pressure and distance between electrodes. In air at atmospheric pressure, you need at least 327V to create a sustained arc over 7.5 µm distance. Interestingly, reducing the distance will only increase the voltage since ions have to travel a certain distance before they gain sufficient energy to create secondary electron emission on impact with the cathode.

If you can touch electrodes to initially ignite the arc (by melting the metal with high currents as described above) and then take them apart, you can get a sizeable arc with lower voltages. This is how arc welding works. You need both voltage and high current to sustain such arcs, with voltage being roughly proportional to arc length. Typical welding voltages are 12-36V, which is enough to create an arc of several mm.

  • 1
    \$\begingroup\$ "Interestingly, reducing the distance will only increase the voltage since ions have to travel a certain distance before they gain sufficient energy to create secondary electron emission on impact with the cathode." :o Never heard of that before or don't remember. \$\endgroup\$
    – Rev
    Commented May 23, 2018 at 18:16
  • 1
    \$\begingroup\$ @Rev1.0 With electrodes being too close and insufficient voltage to form an arc you will get a glow discharge. If you don't limit the current, the glow discharge may eventually heat up the gas sufficiently to start the arc anyway. \$\endgroup\$ Commented May 24, 2018 at 11:25

Applied Physics answer #2

Do you need high voltage or current to produce a spark?

what is a spark? :

The light emitted by a spark does not come from the current of electrons itself, but from the material medium fluorescing in response to collisions from the electrons. When electrons collide with molecules of air in the gap, they excite their orbital electrons to higher energy levels. When these excited electrons fall back to their original energy levels, they emit energy as light. It is impossible for a visible spark to form in a vacuum. Without intervening matter capable of electromagnetic transitions, the spark will be invisible (see vacuum arc)

The spark energy can be very small because of extremely high density from an extremely small surface area. The charge field exponential rises with increasing force in the direction it is travelling. Colliding with a stationary similar charge never touches but is rapidly repelled to deflect its path and often branch in two different paths and continue towards the opposing polarity target.

Since velocity of the moving charge is very slow in conductors (see drift velocity) it’s surface area can be as small as the charged molecules which accelerate towards an opposing charge polarity in micro to milliseconds. Once reaching the conductor target the above mechanism as defined occurs which takes place in pico to nanoseconds and lasts until the stored energy is dissipated in air.

Experiment at Xmas time

We used to get Xmas 🌲 tree tinsel which is metalized plastic like plastic caps but stretch out like a short 40cm string . It could point horizontally towards the TV starting from 1 m distance and would stretch when closer then zap when the BDV of air ~1kV/mm from the tinsel was exceeded around 2~4 cm. That confirmed my charge voltage estimate. Yet the spark could barely be felt with probably Amp flowing in a nanosecond.

It is the air that detonates and not the conductors But the gap of current is so small that the welding electrode and target melts from the plasma hot gas at both ends.
The hot plasma medium becomes a superheated thermal and electrical conductor and a carrier medium for the transfer of electrode gas and particles to flow and weld the target metal.


A spark will be visible with very low energy but it must occur in air from with an impulse Current from charges stored in a cap.$$dt=ESR * C ~~~ E=1/2CV^2$$ is easily visible from shorting a 1uF 3.3V tantalum Cap or 3300 nanocoulombs or 5 microjoules.

The same spark can be created from 5 microjoules of breaking inductive energy stored in a sponontaneous break with high voltage if the break can be much fast than the drift velocity in the conductor. —-

One property of all high voltage insulators such as air is that they are dielectrics , which is a proportional constant of charge capacitance. We normalize the permeability of all other dielectrics such as Air which is also very close to 1.0 of a vacuum.

Yes a vacuum has an EM impedance which also breaks down at much higher levels in space unless there is an ion flow from Solar winds or worse, a Carrington Effect”

Also all dielectrics are electrical insulators and most are also thermal insulators except fluids like oil.

All insulators have a breakdown voltage even though air tends to lower the barrier threshold in kV/mm to breakdown due to mobile charged contaminants that collide and create an avalanche condition or a “Townsend Discharge” it can be made much worse or lower with a partial vacuum until the particles are so few there are none to collide , avalanche and flow current. Faraday characterized this arc with a great many experiments, so much that it inspired Paschen to develop the equation of air pressure vs breakdown threshold and inspire many others including Maxwell who read all of Faraday’s experiments and gave them more attention than the great German Mathematicians like Gauss who insisted on the effects of a charge at a distance, yet had great math properties, when there were clearly more effects going on at close range.

We know there are basically 3 charge properties, conductors, insulators and semiconductors. Surprise! Air becomes a semiconductor once the Inception Voltage is reached to cause a spark no matter how small. We call this in the power utility industry Partial Discharge Inception Voltage or PDIV which is only an optional factory test before breakdown voltage.

Wait a minute if it is a semiconductor , can we made a TRANSISTOR!out of it since avalanche effect on a gas is a negative resistance ?

No but you can make a vacuum tube of it and use an inert gas to avoid oxidation then you have a gas tube “semiconductor” But arcing is not good for audio vacuum tubes so you use the negative resistance or gm gain made more sensitive with heat and then HV bias it well below the blue corona effect that occurs from old age(due to electrode to gas contamination) corona is visible light but when interior to components before breakdown voltage (BDV) we call it Partial Discharge (PD) p.s. There are about 10 thousand PhD theses on Microsoft Academics or Google Scholar on this topic alone.

Other than the threshold changes somewhat linear with gap except at extremities like 50 um or 50 km then it is less linear.

But for practical purposes remember 1kV/mm or 10kV/cm for sharp conductors and about 3x this amount for smooth flat surfaces.

To behave like a TRIAC with a threshold of 1.3V the gap would have to start from zero like pulling out the plug of a motor and a long arc can be drawn until the lower threshold of holding current or some other force on air breaks the connection.

TRIACS also have a holding current threshold for DC even though we always consider the next “zero crossing “ of Ac current as the switch off time.


For this reason DC contacts in relays must be derated for current with inductive loads as the RESULT of breaking arc current can reach over 6000’C in air due to the oxygen and hydrogen content.

Finally —-

Simple Answer:

Yes AND No to both Voltage and current. You can make a spark with either high voltage or current OR low voltage or current,


Even from an AA battery cell or better a LiPo cell with an “ MOT transformer” will draw a big arc when disconnected yet it is still low voltage across the arc, yet very high voltage just before the arc starts since dry contacts break very fast (dt in ns) and we know V=LdI/dt but have contact bounce**

You cannot initiate an arc but you can stretch a big arc with the above after charging current for a few seconds on the primary

If already conducting , creating an insulator gap in some dielectric like air or SF6 or oil takes time for the electrons to get excited and leap across the gap ( microseconds) but then they turn into semiconductor mode and arc in picosecond to microsecond rise time depending on if we are talking about a void or contaminant in a plastic Y cap or XLPE HVAC power cable or a dust particle in oil or some humid air on a glass HV bushing or lightning. Then just like Triacs and Tunnel Diodes and gas tube protectors , they have a low negative resistance that depends on the current density. Which also makes them useful for arc High voltage generating oscillators as Tesla discovered and Transmitters as Marconi discovered and Faraday did all these experiments centuries ago.

  • \$\begingroup\$ as the lag as arc current can reach... looks like an editing mistake. Were you going to say something about voltage and current being out of phase with an inductive load? But then changed your mind because it's DC so there's no phase? \$\endgroup\$ Commented May 23, 2018 at 13:32
  • \$\begingroup\$ Thanks there were a bunch of typos of additions I made. I was going to get into ionization time but decided not to get too deep in describing activation time to collect E field of charges enough to trigger electron avalanche ejection \$\endgroup\$ Commented May 23, 2018 at 13:59
  • \$\begingroup\$ Can you (simply) explain what the amps does in a spark? \$\endgroup\$
    – JinSnow
    Commented Sep 26, 2019 at 18:36
  • \$\begingroup\$ Amps are the follow-on flow of high rate of charge flow ( the arc in air) when the breakdown voltage exceeds the V/um or kV/mm. The resistance drops due to ionization of air (SCR Effect) when the small gap occurs and electrical properties of air changes and then the storage capacitance of voltage may discharge with a glow until contact occurs or after disconnect of current from an inductive load.— This is due to the phase lead/lag properties of current respectively for C,L. The higher the Amps, the lower the ARC resistance \$\endgroup\$ Commented Sep 26, 2019 at 19:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.