Dan MaloneyHey all
HI Greg, welcome!
Used to work in a lightning test lab. Here's 100kA going through a piggy bank - - I miss smiting things.
Poor piggy
Was that filled with copper?
Got what it deserved for working in finance.
How many joules in that pulse?
Yes, copper coins. Attempted a comedy sketch with the other tech working there, where we needed to smash the piggy bank after running out of tea and needing the funds. Probably good for the world that it's long lost as acting skills are at a minimum :O)
All of them?
About 60 kJ
Very low circuit resistance and inductance though
Whoa, that's quite a ka-pew. Voltage?
Max charging voltage was 60 kV - That lab was in Cardiff. There's another in Abingdon with much more energy
For reference, here's an onion passing just 30kA.
and any special measures like coaxial cables or waveguides or other high-frequency shenanigans to get the rise time shorter?
Never zapped food as the mess is hideous - and I'd be the cleaner...
The smell...
Just low inductance transmission lines. About 400mm wide with a 4mm gap filled with PTFE sheets
The guys in Boeing's test facility in Seattle used their kit to tenderize steak
any special design of the capacitors? (my high-freq-fu is weak...).
It worked, but they didn't find all of it - then there was a little smell
I saw somewhere a Marx generator to make shockwaves to break rocks into their component grains.
Yes, very low inductance GAEP capacitors with coaxial resistor assemblies and a coaxial spark gap
@Thomas Shaddack Yeah, Schlumberger has a rock fracturing rig like that, in a mobile truck.
@Greg Leyh Hiya Greg!
Hey SB!
@Greg Leyh Ready for some embarrassing questions?
There are no embarassing questions, only embarassing questioners. ;>
Hey Greg!
I wonder how difficult it'd be to make homemade low-inductance caps like these. Ordinary caps are easy(ish).
@Thomas Shaddack Those are usually wound in class 10 clean rooms.
Not at Norfolk Capacitors - grumble
OK, looks like it's time to start. I'm Dan and I'll be the mod today along with Dusan as we welcome Greg Leyh to the Hack Chat. We're going to talk lightning today, what it is and how to make your own.
Welcome, Greg! Can you tell us a little about how you got into the physics of lightning?
@Greg Leyh Remember this bitchin diode you gave me 35+ years ago?
I read that as "Bitcoin diode"...
@scubabear Oh yeah, wow... Been making any HVDC with it?
Listening now about the best ways to do that!
haha
Sadly Bitcoin probably needs thousands of diodes that size just to power all the mining rigs.
Maybe ask @Drake Anthony how to with the most prompt radiation possible.
@Dan Maloney For many years I’ve been intrigued by how lightning seems to defy classical electrical theory. Storm clouds can start an arc discharge with only 5-10% the voltage you would need in the lab using some unknown trick, possibly involving relativity.
My goal now is to build a machine large enough to re-create the starting conditions in storm clouds, so we can trigger this effect and discover how the trick works.
@Ingineer Just build an array of small modular reactors and be done with it. We have enough energy in thorium for thousands of years. And then there's fusion.
@Thomas Shaddack (agreed)
@Greg Leyh When you say 5-10%, are you talking lightning to ground or cloud to cloud or does it apply to both?
@scubabear Specifically the electric field intensity at the creation point of lightning. Air normally breaks down around 2,000,000 volts per meter, yet storm clouds can somehow start a lightning strike with less than 100,000 volts per meter.
@Greg Leyh first thank you for sharing your knowledge. I am excited to listen to your conversation. The hottest lighting stikes on earth are about 30,000 Kelvin (53,540 ºF). What is the temperature of your recreated lighting at the 1/3 scale coil and what is the expected your taller model at 120 ft. (36.5m). -Douglas from Tesla Science Center at Wardenclyffe.
I've been within 50 meters of a ground strike at least three times in my life. The last two times I heard a distinct clicking sound about a second before the main discharge, Sounded like a piezoelectric lighter spark. Always wondered about that.
How do you even begin to measure that, under what I'm sure are extremely harsh conditions?
A couple of Russian physicists wrote an interesting paper on this E-field discrepancy, and how reletivistic effects might explain it: http://nlpc.stanford.edu/nleht/Science/about/gurevich05phystoday.pdf
@Doug Borge I haven't measured the core temp of the arcs specifically, but the coil runs at way lower current [~500A peak] than a lightning strike.
FYI, I'll post a transcript after the chat in case anyone needs to refer back to links they may have missed.
why would a tesla coil be preferred over pulsed DC systems for researching lightning? It would seem pulsed DC would be a more accurate lightning model due to net charge transfer. is it because a tesla coil can achieve longer arcs with a relatively weaker field?
Is that 500A in the primary or secondary?
Ooh, cosmic rays... (Russian paper)
120 meter discharges from 5MV? why not do a tower like this instead?
Altthough, if we bootstrap the Tesla tower with a suitably sized Marx bank, the coil arcs can form a 'return stroke' of many 1000s of amps... the progression of which can be seen in the first few photos here: https://www.flickr.com/photos/bellmine/sets/72157690474474153/
Hello everyone
@Greg Leyh thanks ~500A would be for the 30' coil? Interesting to know the expected key outputs of things on your larger 35-meter coil. I'm trying to grasp the idea and magnitude of that coil in my head
Excellent use of Hello Kitty, BTW
wouldnt that multi kiloamp return stroke be enough to vaporize normal magnet wire? how will you protect against that?
@paradox That would be one way to get to Relativistic Breakdown voltages, but by far the most economical way to generate many millions of volts is the Tesla Coil, since most of its volume is empty space. DC machines, like Van De Graaf and Marx Generators, are many times more massive, and expensive, and they can’t be moved.
@Greg Leyh David M. Smith has done some work on terrestrial gamma-ray flashe (TGF) physics which may be relevant
@Drake Anthony Since the event we want to study [Relativistic Runaway Breakdown] comes and goes in about 1/100th the time of a single Tesla Coil cycle, the coil voltage itself is essentially constant during the event… just as if we had turned up the voltage of a DC supply, triggered the event, then turned the voltage back down.
The bi-polar output of a Tesla Coil also allows us the choice to run the experiment at either + or – polarities at will.
why haven't you chosen the QCWDRSSTC? they have much higher current discharges I thought
do they not scale well?
@Greg Leyh has lightning been produced in a lab using the atmospheric conditions present in a storm cloud? Air pressure, air density, water present in clouds, oxygen levels, etc.
oh cool, that makes sense!
@paradox The QCWDRSSTC doesn't scale economically to higher power levels -- that series Cpri is a bit of a pill. Also, any hiccups in the feedback scheme could be very costly. I've opted instead to go with a simpler, well proven resonant converter scheme that doesn't rely on feedback. Here's a paper from 1998 showing the overall topology, which I've adapted for a low coupling scenario.
http://www.lod.org/misc/Resonant_Converter_HVPS_1998.pdf
I watched Nasa's video about space lighting yesterday evening. https://science.nasa.gov/science-news/news-articles/lightning-across-the-solar-system What was interesting is that Jupiter has lighting that was discovered to be 10X in strengthening than here on earth. Can you share a link or details on the process of measuring lighting?
@dprice Not yet, afaik. I do think that it's not totally crazy to do though!
@Doug Borge Oh, I don't have a good one handy, though measuring the strength of 'narrow bipolar pulses' is one method.
[also sorry if I missed your questions, they get pushed off the screen too fast]
A Rogowski coil around a tower works rather well for measuring the current of the fast parts of a strike
if you're able to say, how close are you to the amount of money necessary to start on the bipolar twin 121ft tesla coils?
Books by Martin A. Uman are worth checking out for details on measurements
Yeah, Uman has some great books.
@Greg Leyh I think Paradox was focusing more on the QCW side. Why not ramp the supply? Is it just not economical to have ramping at such high powers?
How much power are you pumping in?
Thanks Greg. As humanity is looking to explore new planets it will be interesting to see how your research plays a role in planet selection and how NASA does fly over with rovers. Have you collaborated with them for and rover or satellite research?
@Bob Well, I've already started... 2020 was just too slow not to. Whether I'll be able to finish is another question. However, I've been finding some deals in the surplus markets that almost seem to hint there's forces 'out there' that want me to push forward.
@Greg Leyh Hello, can I ask if you think we could better understand lightings if we could get instruments in the storm cloud, or simulating conditions in a lab is the way to go?
do you have any ideas on the origins of lightning superbolts? the fact that there are bolts with energies that are several orders of magnitude above average is extremely interesting, but I can't find much info on the topic
I don't even want to know how much you'll be spending on the copper wire for the secondary. I'll let you know if I see any deals!
@Calvin Holmes Ramping the supply would involve active control, which would add a lot to the cost and complexity. Instead, the DC prime power source is fixed, and we ramp the duty factor of the main switches in the primary drivers, so they play multiple roles.
@Greg Leyh are you concerned about how loud the coil would be from its audible resonant frequency? a couple megawatts at such a low frequency would do some damage to hearing, i don't even think ear muffs could save you from that.
@Lightning Phil Looks like each tower will need 2.4MW RMS from prime power.
Wow, 2.4MW is no joke! "Tom, how are we gonna generate that kinda power?!?"
@Doug Borge NASA is an amazing org, but I don't think they're interested in helping build these towers. Several groups have shown interest in writing white papers once it's built though. ;>
If you build it, they will come?
2.4 MW, for how long periods? How much energy per run?
(A q-switched laser can give tens of megawatts from hundreds of millijoules...)
@vincent stuchly Yes indeed it would help. It's nearly impossible though, to float the instruments in the clouds at exactly the right place and right time... and also know what kinds of instrumetns you would need.
@Drake Anthony Do you mean 'bolts from the blue?'
Thanks Greg. The ABB IGBT modules such as (https://new.abb.com/products/5SNA1000N330300/hipak-igbt-module) can be used to make reasonably simple circuits that could ramp voltages or act as a controllable inverter. Good for 10ish KHz for a second or so even without cooling. Though usually use the 6kV versions which seem to be hiding
@bob The wire will be a big challenge, yes. If you finds any deals on 12AWG in 5000ft put-up let me know!
12 AWG, seriously?
Yep
that's way thicker than I expected.
@Greg Leyh are you concerned about how loud the coil would be from its audible resonant frequency? a couple megawatts at such a low frequency would do some damage to hearing, or maybe even cause nerves to fire in such a field, would it not?
12 AWG is about the thinnest I can get away with, based on the 1:3 scale coil results.
did the wire of the 1:3 scale coil heat up at all?
What's the limit? Resistance, or thermal dissipation?
@Dan Maloney Perhaps, once we find out who 'they' are. ;>
Bad time to need so much copper wire -- prices are soaring...
@Greg Leyh no I was referring to lightning strikes of unusually high energy, like referenced here https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2019JD030975
(bolts from the blue are cool though, my shop got rocked by one a year ago from a supercell thunderstorm over 20 miles away!)
Al wire is preferred actually
Would that be 8 AWG or 10 AWG then if Al magnet wire was avaliable?
@paradox It will be loud, sure. I'm guessing about 9x louder than the 1:3 scale coil, which is an earful, but survivable.
The Q calcs were done with 12AWG Al.
Reducing the weight of the secondary is critical, since the whole tower is supported by air.
are you not having any guywires?
Yes three guy lines.
What will they be made of?
@Thomas Shaddack The limit is reducing the unloaded Q below 60 or so.
@paradox Spectra line. Seems to have great long term HV properties.
@reece20beck No not noticably. But it also has a huge surface area.
how will you prevent surface tracking/racing sparks during operation?
Not surprising. Spectra/dyneema is UHMWPE, essentially a superlong-chain paraffin wax. Pure hydrocarbon, hydrophobic as it can be (if we don't count fluorocarbons).
Here's one winding section for the 1:3 scale coil. The full scale coil sections will be 15ft in diameter.
@paradox By trying to manage the surface field stress as best as possible, with shielding, etc.
How big is that then? Like 5' across?
It already looks huge!
@Dan Maloney Yes, at the base.
At this scale, sparks will probably be lazy in movement and hold plasma channels. This would also allow them to be influenced by slight winds- How do you prevent a streamer from just flying into the secondary and blowing up everything?
@paradox The sec can take hits, if they're short and keep moving. Beyond that, the toroid is contoured to resist strikes in that direction, as much as possible.
What is the toroid on the 1:3 scale made of? Almost looks inflatable
All of the odd items used to build the structure have to be voltage tested first, as they almost never come with any stress data, like bounce house material, velcro, etc.
@dan It's made of bounce house material, and is inflated, just like the main tower.
Cool!
How are you doing the testing?
It's the simplest, and cheapest, way to build a structure... to place all of the member in tension, and eliminate any compressive members.
@Thomas Shaddack I have been testing the 1:3 scale tower out on the tarmac at the abandoned Alameda Naval Air Station.
do they require you to have any permits to do that?
The city kind of flipped out about that last Feb, but now I have an actual permit. ;>
Use it to power a death ray and nobody will dare to annoy you about permits anymore. :D
Is there a generic term for "bounce house material"?
Lol, going Mythbuster style
Hope to get some more construction done this year, as money and obtainium finds permit!
How does the system maintain resonance without feedback? The paper linked seems to use feedback although I may be mistaken.
@charizardrekt In the paper, they feed back on output voltage to regulate. On the 1:3 scale coil, the Fres is constant enough that the control system takes one msmt before the pulse, and assumes it's constant throughout. There's an option to skew the freq, but it doesn
It doesn't make a noticable difference.
Interesting. Thanks!
We're up against the end of our allotted hour, which is when we like to let the host get back to work or whatever. The Hack Chat is always open, though, so feel free to stick around and continue the conversation. I'll just say an official "thank you" to Greg for spending time with us today, and say that I really enjoyed the discussion. And thanks to all for the great questions!
Sure, thanks for letting me nerd out about this stuff!
Much appreciated, it's really cool stuff.
30,000 K, really cool stuff!
Someday!
lol
@Greg Leyh Thanks for your time and HV craziness
Thanks Greg, this was very interesting.
Thanks all. And don't forget to drop by next week too, when we change things up a bit:
https://hackaday.io/event/178149-haptics-hack-chat
Haptics Hack Chat
Tim Szeto from Nanoport Technology will host the Hack Chat on Wednesday, April 7 at noon Pacific. Time zones got you down? Try our handy time zone converter. Of all our senses, the sense of touch is perhaps the most underappreciated.
Discussions
Become a Hackaday.io Member
Create an account to leave a comment. Already have an account? Log In.
Wow, your project is such a great. As I have learned something new by reading this project. Would like to share it with the team members of https://www.screens-people.com/product/7j92r-15-6-for-dell-inspiron-15-5565-5567-precision-7520-fhd-lcd-led-widescreen/
Are you sure? yes | no