Energetic Forum  
Facebook Twitter Google+ Pinterest LinkedIn Delicious Digg Reddit WordPress StumbleUpon Tumblr Translate Addthis Aaron Murakami YouTube 2020 ENERGY CONFERENCE - PRE-REGISTER NOW!!!!

2020 Energy Science & Technology Conference
PRE-REGISTER NOW!!!
http://energyscienceconference.com


Go Back   Energetic Forum > >
   

Renewable Energy Discussion on various alternative energy, renewable energy, & free energy technologies. Also any discussion about the environment, global warming, and other related topics are welcome here.

* NEW * BEDINI RPX BOOK & DVD SET: BEDINI RPX

Reply
 
Thread Tools
  #1  
Old 06-06-2013, 07:24 PM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Plasma Rocket Engine

My goal is to build a Plasma Rocket Engine and I am posting here to share results as I learn and to share information with other experimenters.

****** Warning ******
This project is experimental and the circuits contain LETHAL amounts of Electrical Current. Also, UV light is created which may permanently damage your eyes. Please do not attempt to replicate without proper safety precautions. I am not responsible for any injury or death that may result from your own experiments.
*******************

A Plasma Rocket Engine is different from an ION engine. A Plasma Rocket Engine uses plasma energy to heat up a gas (or vaporize a solid) and electrically propel that gas out of the engine. It is basically an electric engine. It has many other uses, such as burning waste or purifying water.

A solid fuel commonly used on Pulse Plasma Thrusters is Conductive Teflon. Teflon vaporizes at a low temperature. The teflon is vaporized, electrically charged by a spark plug, and accelerated via a rail gun setup to create thrust.

There are two main methods to create the plasma. One is to use a high current to create a hot plasma. The other is to use RF in 2-50 MHZ range to heat up a gas (similar to Magnetic Induction Heating).

I am starting with high current to create my plasma. The first challenge was to create a high Current Plasma Arc by tying together High Voltage with a High Current Capacitor Discharge. The three Plasma Arc Circuits I am attaching is how I accomplish this.

The first circuit showed how to obtain 3 electrodes by using a 50KV isolation capacitor.

The second circuit expanded on the first circuit by splitting the Neutral Electrode in two (by wire) and adding a Capacitor Discharge Circuit. This worked.

The third circuit expanded on the second by adding an air core isolation transformer along with the Capacitor Discharge Circuit. This worked too.


**** Test Result *****
Capacitor Discharge Circuits were dissapointing. They worked but I only get a pulse. To be really useful, I need a continous Plasma Stream.


**** Test Result ****
I got Continous Plasma by modifying the third circuit. By dropping the capacitor discharge circuit and directly connecting the AC to the air core isolation transformer, I get a Continous Plasma Discharge.

I still use the 50KV capacitor to drive the isolation transformer, but instead of merging the arc with the discharge capacitors, I directly connect the transformer to same AC source. Now I get to add the AC Current to the HV arc. WOW!!!!! The AC current I use is limited by the AC start Capacitor.


**** Update ****


What is really of significant importance, is the current limiting 30Uf AC Start Capacitor. This AC capacitor can be placed before (or after) the light dimmer switch. Wherever it is placed, it limits the current to the circuit by Capacitor Reactance.

Capacitor Reactance = 1 / ( 2 * Pi * Frequency * Uf / 1000000)

So a 30 Uf cap on a 60Hz circuit would be:
1 / (2 * 3.14 * 60 * 30 / 1000000) = 88.46

120Vac / 88.46 = 1.366 Amps.

So the 30 UF circuit limits the current to the coil and Plasma Arc to 1.366 Amps.

I had limited myself to a 30Uf capacitor, because, when I used larger caps, I blew out my ignition coil. But since I am now splitting my circuit between the Ignition Coil AND the Plasma ARC, I think I can upgrade to a larger value Current Limiting Start Cap. The limiting factor is now the 600 Watt light dimmer. I think I can pick a larger start cap and go as much as 4 Amps, unless I upgrade to a 1000 watt dimmer.


*** UPDATE ***

I changed to a 108Uf Start cap which puts the circuit at nearly 5 amps. The result was a larger plasma ball and burning sparks coming off the electrodes. More current = more powerful (and hotter) plasma.
Attached Images
File Type: jpg Plasma-ARC3.jpg (487.1 KB, 90 views)
File Type: jpg Plasma-ARC-SameAC-Source.jpg (510.9 KB, 69 views)
__________________
 
Reply With Quote
Sponsored Links

Download SOLAR SECRETS by Peter Lindemann
Free - Get it now: Solar Secrets

  #2  
Old 06-06-2013, 08:05 PM
Jeff Pearson Jeff Pearson is offline
Senior Member
 
Join Date: Jan 2012
Location: Taos NM
Posts: 188
coooool, this plasma thing has lots of tangents!!!
__________________
 
Reply With Quote
  #3  
Old 06-07-2013, 12:05 AM
mikrovolt mikrovolt is offline
Silver Member
 
Join Date: Aug 2010
Posts: 788
The drawings found in pdf called plasma reformer.
developed at MIT called plasmatron then government tested.
The MIT people sold out. one company in oregon called Umpqua
put out a few pictures and diagrams. Umpqua claims 30% increase fuel economy and emissions are very low.
__________________
 
Reply With Quote
  #4  
Old 06-07-2013, 09:05 PM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Added Snubber

I rebuilt my board using the 108Uf AC Start Capacitor to limit the current to 5 Amps. Turns out 5 Amps is powerful enough to erode electrodes and burn anything close through radiating heat.

I am updating my circuit to include a snubber on the coil to absorb spikes.

The attached are the new circuit and a picture of the 5 Amp board.
Attached Images
File Type: jpg Plasma-5Amp-Board.JPG (1.35 MB, 97 views)
File Type: jpg Plasma-5Amp.jpg (488.9 KB, 66 views)
__________________
 
Reply With Quote
  #5  
Old 06-16-2013, 09:22 PM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Replaced Dimmer Circuit

I pushed the circuit to the limit and burnt up the dimmer switch.

I found that I can replace the dimmer switch with a TRIAC based pulse circuit.

The TRIAC (NTE5646) is 10 Amp capable and has a built in Internal Trigger DIAC which simplifies the circuit. That plus an on/off switch, a 100K resistor, and a .1 UF 250V capacitor is all that is needed to have a solid state pulse circuit for an Ignition Coil.

The attached schematic shows how to hookup the TRIAC based pulse circuit.

*** Update ***

Tried different RC values for TRIAC to see which gave the smoothest spark gap. I didn't feel like updating the schematic. Got the smoothest result with 100k and .039UF 200v timing cap. Could use 39k & .1UF to get the same math result:

1T = 1000000 * .039 / 1000000 = .0039 Seconds Or:
1T = 39000 * .1 / 1000000 = .0039 Seconds

*** Update ***

I did some more testing, using different ignition coil and different AC start capacitor (To limit current). I found that the RC values used with the TRIAC are dependent of the coil and current provided by the AC start capacitor. Switching coils types and increasing current required different values for the capacitor and resistor used for the TRIAC gate timing. Too high of a resistor and it won't turn on. Too low of a capacitor and it won't turn on. You will have to test different values and decide on the right combination for your hardware.

If you increase the current available (larger value AC start Capacitor), then you need to slow the RC timer by increasing the values of the resistor and/or capacitor.
Attached Images
File Type: jpg Plasma-Pulse-Circuit.jpg (391.0 KB, 53 views)
__________________
 

Last edited by plasmahunt3r; 06-24-2013 at 11:00 PM.
Reply With Quote
  #6  
Old 06-16-2013, 11:32 PM
wayne.ct wayne.ct is offline
Silver Member
 
Join Date: Jan 2011
Posts: 552
Tell me more

This plasma generator seems to be very different from the spark gaps typical in this forum. Would someone like to explain and elaborate on this DANGEROUS circuitry?
__________________
There is a reason why science has been successful and technology is widespread. Don't be afraid to do the math and apply the laws of physics.
Reply With Quote
  #7  
Old 06-17-2013, 04:06 PM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Dangerous Plasma

This is serious stuff and is no way a toy.

This circuitry is closer to a plasma torch or a plasma welder. Depending on current, the plasma can produce thousands of degrees of heat. It can cut through steel and can crack molecules into their individual elements. That is why it can be used to decompose hazardous waste into safe elements like carbon, nitrogen, etc.

The higher the current, the more heat is generated. I limited myself to 5 amps for testing and that is hot enough. I used to have my plasma electrodes on wood, but the heat radiated, set the wood on fire. I am building a new ceramic base for the plasma electrodes.

Also current kills, so safety when testing is critical. I put on rubber gloves every time I test.

My goal is to build a plasma rocket engine and I know there are other experimenters out there interested in the same stuff. The circuits are not mine, although I modified them as a result of my testing.

Look at this site for a description on Electric Plasma Engine: My Electric Engine - Main

Look at this site for a description on a HV start box for a welder circuit: High-Frequency Start Box

Both these sites form the basis of my circuits.

I am not inventing anything new. MPD thrusters are currently on satelites. I am just learning how it is done.
__________________
 

Last edited by plasmahunt3r; 06-17-2013 at 05:06 PM.
Reply With Quote
  #8  
Old 06-18-2013, 01:33 AM
wayne.ct wayne.ct is offline
Silver Member
 
Join Date: Jan 2011
Posts: 552
How much thrust have you been able to document?

Thanks for the two references. That helped to round out my knowledge of the technology. Can you provide links to any videos of your experiments?
__________________
There is a reason why science has been successful and technology is widespread. Don't be afraid to do the math and apply the laws of physics.
Reply With Quote
  #9  
Old 06-18-2013, 04:17 AM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
No Video Yet

I don't know how to post video's yet. Suppose that is something I have to learn.

I am not planning on taking any video's until after I link the plasma with a pressurized gas and have achieved at least a Plasma ARC JET Thruster. I cannot forecast a date. There is a lot of research, testing, and construction to do before I get to that point. I am planning on spending 3 years and up to $10,000 on this project .

I can find pictures of thrusters, but actual construction details are elusive. There are a million details that I have to figure out. There is no "How-to" on building this. By the time I finish, this should become a how-to.

Part of this is the Power Electronics to produce plasma. Now that I have plasma, how much power do I need?

Part of this is Hardware Machining for the thruster, where the plasma arc and gas will come together.

Part of this is Materials Science, like coming up with ceramic electrical insulators or using metals as a fuel.

I have bought a lathe and can do some minor machining. I am running into some insulation issues when putting a metal electrode into a brass "T" fitting. I tried portland cement as an insulator, but portland cement is electrically conductive (about 27M ohms), and the arc jumped through the portland cement. Not where I wanted the arc. I tried a castable alumina ceramic but it is brittle and cannot handle 1800 psi pressure. Could use some ceramic help. I need a high pressure high voltage electrical insulator. A design change could mitigate lower strength materials.

Rignt now, too many problems to solve. If someone out there is working on electrical thruster hardware issues, I would sure like to know what you are doing.
__________________
 
Reply With Quote
  #10  
Old 06-19-2013, 01:02 AM
IndianaBoys's Avatar
IndianaBoys IndianaBoys is offline
Silver Member
 
Join Date: Aug 2008
Posts: 577
Quote:
Originally Posted by plasmahunt3r View Post
I tried portland cement as an insulator, but portland cement is electrically conductive (about 27M ohms), and the arc jumped through the portland cement. Not where I wanted the arc. I tried a castable alumina ceramic but it is brittle and cannot handle 1800 psi pressure. Could use some ceramic help. I need a high pressure high voltage electrical insulator. A design change could mitigate lower strength materials.
Hi Plasmahunt3r,

You could give Grancrete a whirl:
My Asymmetric Electrodynamic Machines

IndianaBoys
__________________
 
Reply With Quote
  #11  
Old 06-19-2013, 02:35 PM
Reiyuki's Avatar
Reiyuki Reiyuki is offline
Member
 
Join Date: Mar 2009
Posts: 51
1500psi? Maybe you should look at using multiple layers or a laminate, one layer for strength the other for dielectric. Concrete gathers its strength from compressive loads and not pressure-bearing ones; a concrete block could become a frag bomb.

smooth-on has a decent selection of epoxy putties,coatings, and sprays.


How about using an old small welding gas tank? Coat the inside in several layers of high-temp dielectric epoxy, then perhaps a coat of grill/muffler paint on top of that to smooth things out. Then you could get a high-pressure, high-temp, high-voltage seal to the inside by drilling+tapping a spark plug into one or more sides.
__________________
 
Reply With Quote
Sponsored Links
  #12  
Old 06-21-2013, 12:23 AM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Materials Science

Aluminum as a possible fuel for a Plasma Rocket:

From DR Goldschmidt, Inventor of Thermite:

Doctor Goldschmidt ........ knew that aluminum has a great affinity for
oxygen, or, in other words, .......... aluminum wants oxygen more than any other metal does.


So, if I understand this right: If I push enough current in a plasma arc, and heat an aluminum electrode to a sufficient vaporiziation point, in the presense of an oxidizing agent, reduction should occur, where aluminum will rob the oxygen from the other substance.

The oxidizing agent just has to contain oxygen for reduction to occur. In thermite, Iron oxide is the oxidization agent, but other oxygen bearing substances could be used. This means, that at a certain temperature, CO2 could be used as an oxidizing agent, and aluminum would rob the oxygen from the CO2.

In this case, the Plasma would be the initiator, and Aluminum/CO2 combustion would be closer to a normal rocket engine. The Plasma ARC may be required to sustain the Aluminum/CO2 combustion process. Combustion might cease once the Plasma Arc is terminated.
__________________
 

Last edited by plasmahunt3r; 06-24-2013 at 10:43 PM.
Reply With Quote
  #13  
Old 06-26-2013, 08:51 PM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Plasma Arc

I made a ceramic block for the electrodes out of portland cement and kaolin clay and had a local ceramic shop fire it for me. I had to abandon wood because it kept catching on fire.

I hope you can access photobucket ok. I put a picture there of the white plasma arc because I have reached my photo limit on this site. It is hard to capture, but in addition to the white plasma light, I also get a few sparks flying off the aluminum electrodes.

Plasma-White-ARC_zpsef3e8cd3.jpg Photo by plasmahunt3r | Photobucket
__________________
 
Reply With Quote
  #14  
Old 06-30-2013, 01:40 AM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
DC Plasma ARC - 5 Amp

I converted the PLASMA Arc from AC to DC using the dimmer replacement circuit with a Quadrac (TRIAC w/internal DIAC) and adding a full wave bridge.

Circuit posted at photobucket:

DC Plasma ARC Circuit Photo by plasmahunt3r | Photobucket

*** Update ***

Ran some test with DC Plasma vs AC Plasma. The AC plasma looks more like a bright white cloud. I guess this is because the polarity is reversing 60 times a second. The DC plasma puts out the bright white light, but is more condensed to the diameter of the electrodes. I guess that this is because the electrons are flowing from negative to positive direction.
__________________
 

Last edited by plasmahunt3r; 07-01-2013 at 08:31 PM.
Reply With Quote
  #15  
Old 07-01-2013, 08:36 PM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Plasma Magnetic Containment

*** Magnetic Containment Question - need help???? ***

In NASA studies regarding plasma rocket engines (VASIMR), they indicated the heated plasma will be contained in a magnetic field. I don't understand how this works.

When I put neodymium magnets close to the plasma, the plasma is attracted to the magnet and the plasma moves towards the magnet. This occurs on either magnet polarity. So, if my test show that Plasma is attracted to a magnetic field, how can they contain the plasma in a magnetic field? I am confused. My test results don't jive with NASA's presentations. Is there a trick to accomplish magnetic containment of plasma? I know they are using superconductor magnets, but the magnetic reaction should be the same.
__________________
 
Reply With Quote
  #16  
Old 07-19-2013, 03:19 AM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Plasma Containment

Courtesy of RMCYBERNETICS.COM More Electromagnetism - Coils - RMCybernetics

I think I have the solution to plasma containment: called a Penning Trap (Not Tested Yet).

A Penning trap is a set of electromagnets or coils that will create a controlled non-uniform magnetic field. The diagram shows how the magnetic field is slightly less intense in the centre of the trap. This 'bubble' in the field lines is where anon neutral plasma can be contained. A non neutral plasma is simply a plasma that has an overall positive or negative charge, such as a collection of electrons. A charged 'particle' such as an electron would require an accelerating force to move it across the magnetic field lines. Without this external force the electrons in a penning trap will tend to spiral back and forth within the area of weakest field. The converging field lines at each end act like a 'magnetic mirror' allowing the plasma to remain contained.

I don't know if the magnetic field can be pulsed or has to remain stable.
Attached Images
File Type: jpg PlasmaContain.jpg (7.7 KB, 21 views)
__________________
 
Reply With Quote
  #17  
Old 09-06-2013, 01:00 AM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Micro Plasma Thruster

It has been awhile since I had an update. Spent the last two months learning ceramics to produce ceramic pipe fitting to isolate the Positive (center) and Negative (Copper Case) electrodes.

I have built a "Micro Plasma Thruster" using my 5 amp 120v (600 Watt) Plasma circuit.

Micro Plasma Thruster1 Photo by plasmahunt3r | Photobucket

It works similar to a candle, by using the plasma arc to burn a polyurethane plastic fuel. I call the center Positive Anode "The Wick", because it's similarity to a candle.

Micro Thruster 2 Photo by plasmahunt3r | Photobucket
__________________
 

Last edited by plasmahunt3r; 09-22-2013 at 01:57 AM.
Reply With Quote
  #18  
Old 09-22-2013, 01:55 AM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Redo Micro Plasma Thruster

The first test was successful, but revealed a problem. The Plasma Arc used the shiortest path from Cathode to Anode, which resulted in an uneven fuel burn.

I fixed that problem by adding a coil to produce a magnetic field, which causes the Plasma Arc to rotate around the Anode and the Cathode. Instead of providing a separate power supply for the coil, I had the Positive go thru the coil before connecting to center Anode. So, when the Plasma Arc is energized, the same power energizes the magnetic coil. Neat.

Here is an end view of the new Micro Plasma Thruster. Notice the white substance between the center Anode and the Cathode Case. This is Polyurethane Plastic. The Plasma Arc burns the fuel, in a circular motion, similar to an "End Burn Solid Rocket Motor". No oxidizer. Fuel Burn occurs only when Plasma Arc is on.

Micro Plasma Thruster Photo by plasmahunt3r | Photobucket

This is the completed Micro Plasma Thruster:

Micro Plasma Thruster Photo by plasmahunt3r | Photobucket

This Thruster runs on my 5 Amp Plasma Circuit.
__________________
 
Reply With Quote
  #19  
Old 10-16-2013, 08:53 PM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Upgrade to 5 amp DC Plasma Circuit

I did some testing, and I was able to remove the "Air Core Transformer", however, I left the secondary as a Choke Coil to choke off AC from the HV side. See attached circuit:

DC Plasma Arc Circuit Photo by plasmahunt3r | Photobucket

This works, because, the 50KV .001 uf Cap blocks DC, so it stops the high current side from backing up into the coil. The choke blocks the HV AC from backing up into the DC source.

I am new to chokes, so if anyone has some experience building chokes, I would like your advice. I am currently building a new choke using a 1/5 inch PVC pipe filled with "Iron Oxide" as a ferrite source.

This version of the circuit should work with the "Water Sparkplug" forums. It works well with aluminum electrodes but fails on standard spark plugs (Autolite and Bosch). Standard spark plugs have and internal resistance built in. I know people are trying to get "High Current Plasma Sparks". This merged version (merging High Voltage with High Current) of the plasma spark should fit the bill, if a proper spark plug can be found that doesn't have internal resistor. They can use any ignition coil driver they want.
__________________
 

Last edited by plasmahunt3r; 10-16-2013 at 10:33 PM.
Reply With Quote
  #20  
Old 10-17-2013, 04:35 PM
wrtner wrtner is offline
Silver Member
 
Join Date: Oct 2008
Posts: 950
Quote:
Originally Posted by plasmahunt3r View Post
*** Magnetic Containment Question - need help???? ***

In NASA studies regarding plasma rocket engines (VASIMR), they indicated the heated plasma will be contained in a magnetic field. I don't understand how this works.
The NTRS site is down but these people are talking in terms of superconducting magnets:
Our Engine | Ad Astra Rocket

Maybe the magnetic field needs to be huge.
__________________
 
Reply With Quote
  #21  
Old 10-18-2013, 03:26 PM
wrtner wrtner is offline
Silver Member
 
Join Date: Oct 2008
Posts: 950
This search may help. It represents "VASIMR" in the title and "Magnetic" in the text of the abstract:

http://ntrs.nasa.gov/search.jsp?N=0&Ntk=Title|Abstract&Ntt=vasimr|magne tic&Ntx=mode%20matchall|mode%20matchall

Thwe advanced search is very good:
NASA Technical Reports Server (NTRS) - Advanced Search
__________________
 
Reply With Quote
Sponsored Links
  #22  
Old 10-19-2013, 03:36 PM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Upgraded circuit to 11AMP DC

I changed out the AC start capacitor to a larger size (233-280 UF). Based on the capacitor reactance formula, the current should be around 11 Amps.

Since the AC start capacitor had a range of 233-280, I used an average of 256 for my current calculation:

Capacitor Reactance = 1 / ( 2 * Pi * Frequency * Uf / 1000000)

So a 256 Uf cap on a 60Hz circuit would be:
1 / (2 * 3.14 * 60 * 256 / 1000000) = 10.367

120Vac / 10.367 = 11.575 Amps.

So the 233-280 UF circuit limits the current to the coil and Plasma Arc to around 11 Amps.

Here is a picture of the upgraded circuit working.

Plasma-10Amp-DC1_zpsfa2d1ccc.jpg Photo by plasmahunt3r | Photobucket

It is still not powerful enough to vaporize aluminum, but I am noticing pitting on the aluminum electrode ends. So some aluminum is vaporizing over time, but not at a rate fast enough to support combustion. Those aluminum pits are the occasional sparks I see fly off the electrodes while the circuit is running.

I also sprayed water into the plasma arc and the water spray looks like sparks emerging from the plasma arc. Still needs more current to combust water.
__________________
 
Reply With Quote
  #23  
Old 10-29-2013, 02:03 PM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Upgraded circuit to 13 Amp DC

I changed out the AC start capacitor to a larger size (270-324). Based on the capacitor reactance formula, the current should be around 13 Amps.

Since the AC start capacitor had a range of 270-324, I used an average of 297 for my current calculation:

Capacitor Reactance = 1 / ( 2 * Pi * Frequency * Uf / 1000000)

So a 297 Uf cap on a 60Hz circuit would be:
1 / (2 * 3.14 * 60 * 297 / 1000000) = 8.935784

120Vac / 8.935784 = 13.43 Amps.

So the 270-324 UF circuit limits the current to the coil and Plasma Arc to around 13 Amps.

Although, this is a powerful arc, it is still not powerful enough. Since I am plugging into a 15 amp household outlet, I have reached the limit of this circuit. I could probably squeeze out another amp, but it won't make a difference. I can see that I am probably going to need more than 30 Amps to do the job. The 14ga wire to my AC outlet plug is starting to get warm.


I HAVE REACHED THE LIMIT ON THIS CIRCUIT DESIGN> A NEW DESIGN IS NEEDED.


I have started working on a new circuit design, that utilizes two blocking capacitors, and then can use a separate DC power supply to provide the amps. It shows promise, but still needs tweeking.

I tried one attempt, using a 1500 watt 120/240 step/up transformer to charge two 1800uf capacitors in parallel to 324 volts. The DC circuit is a totally different circuit from the HV side and ties together with the HV at the Plasma Arc. The blocking capacitors isolate the voltages until the HV arc starts. It partially worked.

When I turned it on, the capacitor discharge had a loud pop and a large white pulse. The cap discharge sounded like a gun shot and I was expecting my neighbors to call the cops.

I was intending for the step transformer to continue providing power for the Plasma Arc, but the fuse blew. This design has promise to provide the Amps, but needs tweeks. The Plasma Arc is essentially a short of the DC circuit.

Since it uses a different DC source, it gives me options for providing the Amps:
A. Could use an isolation or step/up transformer
B. Could use two car batteries is series.
C. Lithium batteries provide 20 times their current rating for a short period of time. This means a 4 Amp drill battery could provide 80 Amps for a short period.
D. I have a microwave transformer. I could cut out the secondary and replace it with as much 10ga wire I can fit into it. From the heater element coil on the transformer, 1 turn should get about 1 volt. If I can fit 20-30 turns of a heavier guage wire onto the secondary, I can get some volts and some serious amps.
__________________
 

Last edited by plasmahunt3r; 10-29-2013 at 02:11 PM.
Reply With Quote
  #24  
Old 10-29-2013, 08:54 PM
dward dward is offline
Member
 
Join Date: Nov 2012
Posts: 46
have you thought about electro-static containment? From the Fusor groups online, and from Eric Dollards math, I gather that electrostatic containment might work better. It might require a complete redesign, but it's just a thought. I too am working on space flight technologies and right now am trying to figure out ways to combat reentry heating on a capsule that uses Hiddink style pulsed capacitors.

I don't know if you're doing this for fun or if like me you have a goal of private, personal spaceflight but I am finding that the technical problems are solvable, where as the bureaucratic problems are proving impossible. I've been contemplating a move to a NON OST signatory state, simply because I cannot design around a world full of politicians and lawyers.

The legal side of space flight is a muddy mess...
__________________
 
Reply With Quote
  #25  
Old 11-26-2013, 05:47 PM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Plasma Dual Coil

I finally got a Plasma Arc hot enough to burn 1/4 inch aluminum electrodes.

Set up two isolated circuits and tied together at two electrodes for the Plasma Arc. The Plasma Arc burned aluminum (on the positive electrode).

The left circuit is a dual ignition coil setup with two blocking capacitors.

The Plasma electrodes are the Arc is in the middle.

The right circuit is a step-up transformer to produce 318V DC (no load) and charges two 450V 1800 Uf capacitors. Both circuits tie together at the plasma electrodes.

Here is a picture of both circuits producing the Plasma:

http://i1351.photobucket.com/albums/...ps7b9b682b.jpg

Here is a pic of the melted electrode. The positive aluminum electrode formed a mushroom shape on it's end and radiating some aluminum wiskers. I am amazed that electrons have enough force to flatten out the aluminum positive electrode end. This result shows that electrons are moving (forcefully) from negative to positive.

http://i1351.photobucket.com/albums/...ps2ef1449e.jpg

So what temp did I reach? I found a .mil document that reads: Ermakov, et al16 embedded a thermocouple into an aluminum
particle, and measured ignition temperatures of ~2000-2100 K, concluding that ignition occurs due to the
failure of the oxide shell integrity, but not necessarily due to melting



2000-2100K = 3140-3320F. Not bad


*** WARNING ***
This circuit contains lethal amounts of current, temperatures at the Plasma Arc that exceed 3000F, and UV light from ARC is intense & dangerous
*** End Warning***


Here is the circuit for the isolated Plasma Arc. The left side is the High Voltage Side and can run standalone if a high voltage source is needed. The right side is the High Current DC side and can serve as a power supply for other circuits. Both circuits come together at the Plasma Arc Electrodes. The High Current DC side is an open circuit until the High Voltage initiates the Plasma Arc. The Plasma ARC closes the DC circuit and allows the Capacitors to discharge into the ARC, maintained by the Step Up Transformer. The step-up stansformer is an "Autotransformer" and has no polarity; so both sides are connected to AC Hot wire. Also, while I limit the current from the transformer, the instantaneous discharge from the two parallel 1800Uf capacitors exceed the planned current for an initial burst of power. This power burst sounds like a gunshot when the Plasma Arc starts. I am not sure how to calculate this added current.

http://i1351.photobucket.com/albums/...psc2d3275f.png

I think the initial joule calculation for each electrolytic cap at start of Plasma Arc is:

If I charge 1800UF cap to 318 volts (318 * 318 * 1800 / 2,000,000) = 91.01 Joules per Capacitor """Over 50 Joules is FATAL""" . I don't know the time period of discharge.


Note:
On the DC side, the current limiting capacitor from the transformer must "NOT BE ON THE NEUTRAL SIDE". Cut off non polarized European plug and replaced with 3 prong USA AC polarized plug to ensure correct operation.
__________________
 

Last edited by plasmahunt3r; 11-29-2013 at 12:14 AM.
Reply With Quote
  #26  
Old 01-08-2014, 02:49 AM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Need Help: Magnetic Induction Heater Circuit

In order to use RF to produce plasma, I decided to use a crystal controlled oscillator to get frequency control of my RF Induction coil.

I plan to use 13.56 Mhz for the plasma oscillator, but the RF Power MOSFET's I need are $90 each. So I decided to test the circuitry at 32.768 Khz, where I can use lower cost MOSFET's and IGBT's to work out the bugs.

The basic steps are:
A. Crystal controlled CD4069 IC to produce 32.768 Khz
B. This frequency is passed to a MOSFET Driver IC MIC4420ZT (5 pin TO-220)
C. The output pin of MIC4420ZT is used to run a Toroid Gate Drive Transformer.
D. I built the Gate Drive Transformer on a .75 inch Toroid, 7T primary, 13T secondary. This transformer passes 12.5 VAC to the gate of the IGBT. The Gate Drive Transformer is used to provide isolation between the low voltage frequency driver from the high voltage power amp.
E. The IGBT forms a Class E power Amplifier. Every time it turns on, it charges the Choke Coil. When the IGBT turns off, the Choke Coil discharges into the Induction Coil and the series resonant tank capacitor.

Even though, I confirmed that I am running the IGBT at 32.768 KHZ, the induction coil is running at 57 Khz (without a load). When I insert a rod to be heated in the coil, the frequency drops to around 26 KHZ.

From this I conclude that the Induction Coil Frequency is based on the Inductor, the Series Resonant Tank Capacitor, and the load (item to be heated). The shunt capacitor has little effect on the frequency (it does alter the frequency a little).

Here is the circuit I am using for Magnetic Induction Heating:
http://i1351.photobucket.com/albums/...ps4561e2b9.png

I need precise frequency control. I can see that using a tank circuit is NOT the best way to maintain precise control of the frequency. I get precise control up to the IGBT, but the resultant frequency of the Induction Coil is hard to control.

DOES ANYONE HAVE ANY SUGGESTIONS ON HOW TO MAINTAIN EXACT FREQUENCY CONTROL THROUGH THE INDUCTION HEATING COIL???

I might try a half bridge (or possibly a "H" bridge) circuit. I could use some help. Anyone out there a Ham Radio expert? Someone with some RF experience would be useful.
__________________
 

Last edited by plasmahunt3r; 01-08-2014 at 06:49 AM.
Reply With Quote
  #27  
Old 01-08-2014, 03:42 AM
citfta's Avatar
citfta citfta is online now
Gold Member
 
Join Date: Jun 2008
Posts: 1,323
Tune to the load.

Hi Plasmahunt3r,

I worked on induction heaters for a while. Whenever you change the load you have to adjust the circuit back to resonance again to match the driving signal. Our machines had meters on them to show us when we had the best transfer of power to the load. You would adjust the circuit to resonance by changing the value of the 2.2 uf cap. In the real world we had a series of caps that could be selected in combination until we got resonance. You would do this by putting a .1 uf and a .2 uf and a .5 uf and a 1 uf and maybe also another 1 uf and a 2 uf. Wire all of them so that one end of each is connected to ground and the other end is connected to a single pole switch. The other side of each of the switches would then go to the induction coil. By selecting different combinations of values of the caps you should be able to get the circuit into resonance with your load in the induction coil.

It is not a problem to get the driving signal to stay on frequency but if the circuit output is not tuned to resonance then you will not get a good transfer of power. Ham radio guys (I am one) have to retune their antenna matching circuit any time they make a large change in the transmitting frequency. We have to do this in order to get the most power to go into the antenna and not be reflected back into the transmitter. I hope this info is some help to you.

Respectfully, Carroll
__________________
 
Reply With Quote
  #28  
Old 01-09-2014, 01:11 AM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Other RF Plasma Options

Thankyou citfta. Your suggestion on multiple tuning caps is an option if I stay with inductively coupled RF plasma.

I want to look at other options.

First of all, the gate drive transformer maintains the 32.768 KHZ frequency. I am looking in to why gate transformer maintains the frequency and the Induction Coil doesn't maintain the frequency. The difference, is how the circuits are wired.

The gate drive IC I am using (MIC4420ZT), is a MOSFET Half Bridge. On the positive pulse, the capacitor charges through the primary coil, and on the negative pulse, the capacitor discharges through the primary coil, simulating AC through the coil. I don't see why I cannot use a similar approach to the induction coil, by creating a half bridge and use the capacitor as an energy storage/blocking capacitor (not as a tank capacitor).

Also, I need to consider another option. Instead of Inductively Coupled Plasma (by using an Induction Coil), maybe I should consider Capacitively Coupled Plasma.

The switch to Capacitively Coupled Plasma, would involve eliminating the Induction coil and leaving the capacitor. The choke coil will still charge like normal when the IGBT is on, but then pass AC through the capacitor when the IGBT is off. It is hard to find circuit example where people have done Capacitive lyCoupled Plasma. I can only find Youtube examples of it working.

I think I could use a pipe as a hollow electrode if I remove the induction coil from the Class E power Amp and connect a pipe to my circuit (at 13.56 MHZ frequency). In this case, the pipe would be an antenna with the gas to be energized, passing through the center. I don't know how much current I would need to supply for this to work. Goal is to heat the plasma gas (>3000F), not just ionize it.

Any ideas or suggestions???
__________________
 

Last edited by plasmahunt3r; 01-27-2014 at 03:05 PM. Reason: Remove possible copywritten image
Reply With Quote
  #29  
Old 01-13-2014, 06:51 PM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
I have some news on chokes.

I had a problem with the Class E Power Amp portion of my circuit. An IGBT (or a MOSFET) transistor would fail after a while. I did some research on switch mode power supplies, and found that I should include a second winding on the choke to stop the choke from magnetically saturating, causing the primary current to rise excessively, destroying the transistor.

The 'extra' winding of a forward converter's transformer ensures that at the start of a switch conduction, the net magnetization of the transformer core is zero. If there were no extra winding, then after a few cycles the transformer core would magnetically saturate, causing the primary current to rise excessively, so destroying the switch (ie transistor).
The diode on the secondary that is connected between the 0V line and the junction of the inductor is often called the 'flywheel diode'.


This "Bifilar" dual winding choke, is wound exactly like the toroid's found in "Joule Thief" circuits. Twist the end of one coil with the beginning of the other coil. The connection of the two coils go to the positive source. You can verify the coils with an induction meter. Each Coil read 2.3mH independently. When the two output wires are connected together (trying to read both coils simultaneously), the net inductor reading is "ZERO".


Here is an updated circuit showing new choke coil. Also, I added a diode (parallel to gate transistor) to shorten "Turn-Off Time" of transistor:

http://i1351.photobucket.com/albums/...g?t=1389582390

Here is another explanation of the auxiliary coil choke:

The core of the transformer has been magnetized by the magnetization current. The remnants of the core magnetism would cause the core to saturate within a few switching cycles. So it has to be demagnetized after each switching period. Various techniques have been used for this purpose. During the shutdown phase the magnetizing current is fed back via Flywheel Diode through the auxiliary winding. The number of turns used for the auxiliary winding is usually the same as that used for the primary winding. This means that the same time is required for demagnetization as for magnetization. The circuit can therefore be driven at a maximum duty cycle of 50%
__________________
 
Reply With Quote
  #30  
Old 01-20-2014, 01:03 AM
plasmahunt3r plasmahunt3r is offline
Member
 
Join Date: May 2013
Posts: 47
Testing Results of Class E Power Amp

I have been trying different component values and placement for the Class E Power Amp stage. Basically "Trial and Error". I hate the error part, because each error cost me a $4.00 Transistor. Adds up after a while.

First of all, the Shunt Capacitor's purpose is to limit the voltage rise from the choke, while the transistor is in the "OFF" state. The .039 Uf Shunt Cap works fine. I tried a .005uf Shunt, and it blew the transistor. Another $4. If the Shunt Capacitor is too large (1uF), the induction coil doesn't heat. I read somewhere, that you can run without a Shunt Capacitor, but the voltage rise may be 8 times the input voltage. The transistor would have to be able to handle that greatly increased voltage rise from the choke inductor.

The placement of the "Tuning Capacitor" matters.

Originally,I had it following the inductor and the frequency readings were 57Khz with "No Load", and reduced to 29Khz when the steel rod is placed in the center.

I thought in series resonance, the sequence doesn't matter. This is incorrect.

Place the "Tuning Capacitor" before the induction coil. Now the frequency is stable at around 26 Khz and only reduces .3 Khz with the steel rod load. Heating seems to be about the same. By careful sizing of tuning capacitor, I can match the crystal driver frequency.

Also, the input current matters. This circuit is based on 4 amp circuit by using an AC capacitor to limit current (Capacitor Reactance from earlier posts). If you increase the current available to 8 amps, then a larger value transistor is required, and all component values will have to be retuned.

Here is an updated circuit of the Class E Magnetic Induction Heating Circuit:

http://i1351.photobucket.com/albums/...ps1ab98a06.png
__________________
 

Last edited by plasmahunt3r; 01-20-2014 at 01:27 AM.
Reply With Quote
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off



Please consider supporting Energetic Forum with a voluntary monthly subscription.

Choose your voluntary subscription

For one-time donations, please use the below button.


All times are GMT. The time now is 12:17 PM.


Powered by vBulletin® Version 3.8.8
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
Search Engine Optimisation provided by DragonByte SEO v1.4.0 (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Shoutbox provided by vBShout v6.2.8 (Lite) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
2007-2015 Copyright - Energetic Forum - All Rights Reserved

Bedini RPX Sideband Generator

Tesla Chargers