# Spark Gaps

1. Oct 29, 2006

### Jdo300

Hello All,

I have another set of quick questions that may or may not have a quick answer . When a simple two-electrode spark gap discharges, what shape do the eclectic and magnetic fields take on? If I were to draw the electric and magnetic field vectors, would they be pointing radially out away from the direction of the flow of charge? or swirling around the current path like a wire? I know that the E and B fields are both at right angles to the direction of propagation (assuming I'm reading my physics book right). But I am trying to get a good 3D image in my head of what this looks like.

Thanks,
Jason O

2. Oct 30, 2006

### Jdo300

Okay, I got the first question answered. But I have a new one now. if I have a spark gap that is X cm long and I have a spark of X volts jump across it. Is there a way to calculate that induced B field flux dф/dt. at a point away from the spark?

Thanks,
Jason O

3. Nov 1, 2006

### Staff: Mentor

" I know that the E and B fields are both at right angles to the direction of propagation"

That's not right. The E field goes from the + electrode to the - electrode. In the axis of the spark gap, the E field lines are straight. Outside the spark gap, they curve a bit in going from the + electrode to the - electrode. Just like you plot the E field lines for a finite size capacitor.

There is no B field until the gap ionizes and fires. Then the B field circulates around the gap discharge current. After the gap fires, the B field goes away, and the E field drops down to a value below what the sustain voltage requires.

To calculate the magnetic flux, you use the usual equation for B field from a current, and you need the spark gap current versus time.

4. Nov 1, 2006

### Jdo300

Hi Berkeman,

Thanks for the correction. I was thinking that a spark gap behaved like an antenna because I was reading some early articles about transmitters back in the 1800's that used spark gaps to transmit broadband pulses for early wireless telegraphs. I just deduced that since that pulse could travel so far that there must be something there that is causing electromagnetic energy to be transmitted. If it is not, then what is happening that is allowing the pulse to be emitted?

The other question I want to know is how to calculate the current of the spark? I know that the time it occurs is extremely small, but how would one quantify it?

Thanks,
Jason O

5. Nov 1, 2006

### Staff: Mentor

No, you're correct. The changing B-field from the spark action creates a changing E-field as part of a small EM wave that propagates out from the spark itself. A spark gap is generally not a very effective radiator, because the gap is small (1mm or so) and the time constant is on the order of a nanosecond or so. With really high voltage like from a TV picture tube, you could probably fire a 35kV arc with a length closer to a few cm where the GHz bandwidth of the arc might make it a more efficient radiator.

I don't have any great info handy on how to calculate the ionization process that leads to the spark (I used to). Here is the hit list from a quick google search in case it helps.

6. Nov 1, 2006

### Jdo300

Ok, I made a mistake on that first question. After taking a second look at the radio transmitters, I saw that they put antennas on them so that is apparently what the pulses are transmitted with. But I’m asking about the spark gap because of an experiment with a Wimherst machine that I performed this past weekend. I was just cranking it over and making sparks with it when I decided that I would connect my digital Multimeter to these pegs on the base (which I thought were for transferring the power to an outside load). I wanted to see how much voltage was building up in the capacitors so I connected my Multimeter leads to the pegs and cranked it some more. As I expected, I saw 1000+ volt spikes show on the meter (which only went up to 1000V) every time a spark jumped the gap. I then took a closer look at the pegs and realized that they weren’t physically wired to any part of the generator!

So I disconnected the meter probes from the base and sat them on the table about 10 inches away and cranked the Wimherst again. My meter was still showing the 1000+ volt spikes. Then I decided to change the meter over to DC current and see what would happen. So I set it on the 20A scale and cranked away. Now my meter was showing me 14-18A reading as each spark arced across the gap. Could someone explain to me what is going on here? I remember hearing somewhere that spark gaps can interfere with electronics so my immediate assumption is that the sparks were bugging up my meter, but what do you all think?

Thanks,
Jason O

7. Nov 1, 2006

### Jdo300

Thanks for the handy links. I'll check them out. I should also mention that the electrodes on the Wimherst machine were about 3-4 inches apart when I did these experiments.

- Jason O

8. Nov 1, 2006

### Staff: Mentor

Good questions, Jason. Yes, the electric field pulses from the sparks were interfering with the proper operation of the meter, and it was displaying false info. This is a common test that is used to test the immunity of electronic devices to static electricity hits. The standard commercial test is called EN 61000-4-2, and involves injecting static shocks (in various ways) into the external metal pieces of products, and also into metal ground planes that the products are sitting on or next to.

http://www.conformity.com/0402an.pdf

9. Nov 1, 2006

### Jdo300

Hi,

Thanks for the great information. Is there also any magnetic component to the pulse or is it primarily electric in nature?

Thanks,
Jason O

10. Nov 1, 2006

### Staff: Mentor

For static shock hits and the associated spark, the main component is the fast current transient as it enters the internal protection circuitry via external metal contacts or connectors. However, when you hit a ground plane that a product is sitting on or next to, that generates a huge electric field transient (because you've just charged up that plane with the static hit), and that electric field transient is what confuses the device circuitry as a result of the plane hits.

Good product design should prevent any corruptions of device operation from direct or nearby plane hits. That's part of the immunity testing (the EN 61000-4-x series) that you need to do to a product in order to earn the "CE Mark" that is required to sell your electronics product in the EU. And it's a good idea to do the immunity test series no matter where you sell your products, since it helps to ensure reliable product operation in the field.

11. Nov 1, 2006

### Jdo300

Hi Berkeman,

Thanks for clearing that up for me :-).