spark plugs?

[Lerner]

We use automotive spark plugs as the trigger electrode in our spark-gap switches. We are making progress in getting the switches to all fire together, but we are beating up on our spark plugs a lot. The ends of the electrode seems to be melting. What we needs is a spark plug with a somewhat thicker electrode to spread out the current–something like 1/8″ in diameter rather than the 1/12″ we have. Or we could maybe use the iridium types, but we have only found those with very small electrode tips, which will make the problem worse. Anyone out there know a lot about spark plugs?

[Breakable]

I know absolutely nothing about cars, but after doing some research i think you might want to investigate the
Semi-surface-discharge configuration
you can see it here in the bottom right:
http://www.sparkplugs.com/glossaryImage.asp?imgID=471
Edit:
Alternativelly K22PB W20EPB
http://www.denso.com.sg/plug/img_ip19.jpg
from
http://www.denso.com.sg/tpl/pdtlineup_plug_lineup.html

[Aeronaut]

Breakable wrote: I know absolutely nothing about cars, but after doing some research i think you might want to investigate the
Semi-surface-discharge configuration
you can see it here in the bottom right:
http://www.sparkplugs.com/glossaryImage.asp?imgID=471
Edit:
Alternativelly K22PB W20EPB
http://www.denso.com.sg/plug/img_ip19.jpg
from
http://www.denso.com.sg/tpl/pdtlineup_plug_lineup.html

Breakable’s bottom link shows an iridium plug and a melting temperature that suggests it might work. If it doesn’t work, here’s another “high temperature” plug you can get in “The Zone”:
http://www.splitfire.com/troubleshooting.html

[KeithPickering]

Lerner wrote: We use automotive spark plugs as the trigger electrode in our spark-gap switches. We are making progress in getting the switches to all fire together, but we are beating up on our spark plugs a lot. The ends of the electrode seems to be melting. What we needs is a spark plug with a somewhat thicker electrode to spread out the current–something like 1/8″ in diameter rather than the 1/12″ we have. Or we could maybe use the iridium types, but we have only found those with very small electrode tips, which will make the problem worse. Anyone out there know a lot about spark plugs?

Electrode erosion is a common problem in automotive spark plugs too. You need a “cold” plug to help prevent this. “Cold” plugs typically have a larger insulator (on the combustion chamber side) to help carry heat away from the electrode better.

NGK kindly provides a guide to tell how to find a colder plug, here:
http://www.ngk.com/glossary.asp?kw=How+do+I+find+a+colder+or+hotter+plug?&manID=1&pt;=

The disadvantage of cold plugs in an engine is that carbon is more likely to build up on the electrodes. For the gas switch this should not be a problem, so I would go ahead and get the coldest plug you can find for your threading. That would be heat range 12 for NGK, or heat range 37 for Denso. You may have to hunt to find plugs that cold; racing engines (which run hot) need cold plugs, so you might try a racing supplier.

Another thought that occurred to me is that perhaps the SF6 is ionizing during the spark/switch process? If so, the presence of flourine ions would certainly explain a lot of the erosion you’re seeing. Perhaps a non-corrosive metal for the electrodes might help; platinum electrode spark plugs are available from some manufacturers.

[KeithPickering]

Lerner wrote: We use automotive spark plugs as the trigger electrode in our spark-gap switches. We are making progress in getting the switches to all fire together, but we are beating up on our spark plugs a lot. The ends of the electrode seems to be melting. What we needs is a spark plug with a somewhat thicker electrode to spread out the current–something like 1/8″ in diameter rather than the 1/12″ we have. Or we could maybe use the iridium types, but we have only found those with very small electrode tips, which will make the problem worse. Anyone out there know a lot about spark plugs?

Further update: NGK’s racing catalog is here:
http://www.ngksparkplugs.com/docs/racing_catalog.pdf

Not many 12-range plugs available, but most are available in range 11, including platinum and iridium electrodes. For comparison, my car would use an NGK plug with heat range 6, which is 5 ranges hotter.

[KeithPickering]

Hopefully-final-update:

NGK’s catalog claims that “nearly all insulator cracking and electrode melting can be prevented” with the semi surface discharge type of plug. Don’t have experience with them myself … just sayin’ …

I also note that the semi-surface discharge type plugs are available with nickel alloy electrodes only.

[Lerner]

The problem with the cold types is that this refers to their ability to conduct the heat away before the next pulse some milliseconds later. We have just one pulse, and it only last tens of nanoseconds, so heat conduction is not fast enough. Can you tell if any actually have large diameter conductors? Ours is .0955″.

[Aeronaut]

Maybe Wolfgang can make and install a custom electrode to arc across the entire grounding rim?

[Henning]

Found this link on Famulus’ Tweets: MACOR

MACOR is a machineable glass-ceramic developed and sold by Corning Incorporated. It is a white material that looks somewhat like porcelain. MACOR has excellent thermal characteristics, acting as efficient insulation, and stable up to temperatures of 1000 °C, with very little thermal expansion or outgassing. It can be machined into any desired shape using standard metalworking bits and tools.

MACOR has a density of 2.52 g/cm^3, and a thermal conductivity of 1.46 W/(m·K). Its low-temperature (25 to 300 °C) thermal expansion is 93×10^-7 m/(m·K). Its compressive strength is 50×103 lb/in^2. Nominal engineering properties are comparable to borosilicate glass.
MACOR is a very good insulator with excellent tolerance. Even with temperature changes, its low thermal expansion ensures that its shape changes very little.

Maybe usable for spark plugs insulators.

[skmurali]

Thank you for the information. I was aware of this material, and the visitors from PPPL also suggested this material. Though I knew this material was not suitable, we tried this material as the first version of our spark plug. Here is what happened. First the entire spark plug was maufactured from Maycor. The threads crumbled and the sparkplug broke in half during manufacturing. This was our Mark I spark plug. In the next version (Mark II) we made a sleeve out of maycor and the spark plugs (four of them) broke the very first time we fired the device.

[mchargue]

Mr Lerner;

From what I’m reading, you’re using a spark-plug, without the grounding arm, to initiate a plasma within the switch. The plasma then conducts, via the spark-plug’s center electrode, the current pulse over/along the plasma. Now you’re looking for a spark-plug with a larger center electrode to act both as a better heat-sink, and to lower the current density in the electrode. (with the current density used, I can see why they are beating themselves to death)

The real issue seems to be how do you start a plasma in the plasma switch without overloading the small electrodes that you need to assure that they fire together. Larger electrodes will conduct heat better, but they will not fire together in time as well as the smaller electrodes. In both cases, switch-to-switch timing is affected by the switch-to-switch gap mis-match due to the initial firing pulse ramping-up, as different gap distances will arc at different voltages. (times along the ramp)

If that’s the case, have you considered dispensing with the spark-plug altogether?

It may be easier to initiate the plasma using another means. One that would absolutely insure that all the plugs fired simultaneously, and could give you a way to use a larger diameter electrode. I had suggested this earlier, but the suggestion either withered, or was left off for good reason. Anyway, back to how I would do this.

How about using a LASER to initiate a plasma in the plasma switch? No issues with ramp-time as with the initial electrical pulse used to start the switch, and no issue with the switch-to-switch gap mis-match between the conducting electrodes that will arc at different voltages (times) as the initial electrical pulse ramps-up.

One LASER pulse can be routed to all switches simultaneously, (or tuned by varying the distance between source & destination) insuring that they all initiate a conductive plasma at the same time. Because the main conductive electrodes no longer need to be used to initiate the plasma, they can be made larger in diameter, with better attendant thermal characteristics, and a lower current density.

The issues would be the gas that’s excited by the LASER, and used to conduct the main capacitor bank’s current; finding an off-the-shelf LASER that can do the job; the selection of materials for the construction of the new conductors; and making the switch so that a LASER can be fired across it to start a plasma.

I think it likely that some of this research has already been done in the past, probably connected with lightning strike dampers, or some other current-dump system associated with power generation/switching systems.

From what I read, this is a critical issue; both so that experimentation can proceed, as well as for fielding a robust system in the future. I’m thinking that this means that the electrode heating/current density problem has to be solved, and that the switch-to-switch gap mis-match must be rendered moot as a maintenance issue.

What say you?
Patrick

[Henning]

Patrick: I think your proposal was ignored, because it was similar to Eric’s initial idea of using a diamond/laser switch. Diamonds are insulators, but conduct pretty well when ultra violet light is shined upon. This was considered to be too expensive, and no-one ever built a switch that big. So Eric and Murali fell back to ionizing spark plug switches (don’t know if it’s the right name). But what you’re now proposing is something different.

[JimmyT]

Henning wrote: Patrick: I think your proposal was ignored, because it was similar to Eric’s initial idea of using a diamond/laser switch. Diamonds are insulators, but conduct pretty well when ultra violet light is shined upon. This was considered to be too expensive, and no-one ever built a switch that big. So Eric and Murali fell back to ionizing spark plug switches (don’t know if it’s the right name). But what you’re now proposing is something different.

patentable?

[Henning]

JimmyT wrote:

Patrick: I think your proposal was ignored, because it was similar to Eric’s initial idea of using a diamond/laser switch. Diamonds are insulators, but conduct pretty well when ultra violet light is shined upon. This was considered to be too expensive, and no-one ever built a switch that big. So Eric and Murali fell back to ionizing spark plug switches (don’t know if it’s the right name). But what you’re now proposing is something different.

patentable?

Don’t think so. But it’s no easy part either. You need to ionize the gas enough to create plasma. That might require a fairly strong laser (I don’t know, just suspecting it). Then the glass to chamber holding the reaction gas where the laser passes through needs to stay clear, so the laser doesn’t break the glass because auf resudials blocking the sight.

BTW: I’m no expert in this.

[Henning]

So it would be something like this:

Attached files

[Author]

Posts