I believe that both gases and liquids are considered fluids. The former compressible, the latter - not so much.
Since it's on topic, plasmas are fluids too.
Thanks for the feedback
What I was trying to convey, was for the fluid to be at high pressure before it entered the jets. I believe this would maximize the velocity of the fluid droplets being expelled from the nozzles and result in maximum disruption of the boundary fluid layer (ie turbulence) on the thermal interface.
Regardless, would the plasma loudspeaker topology I described (fully insulated w/ ceramics, minus the electrodes) be safe in a home environment (Other than issues associated with Ozone production)? I've considered directing a stream of Helium gas into the area surrounding the electrodes to minimize ozone production. I believe plasmatronics did something similar with their tweeter.
It probably won't be anywhere near safe in the vicinity of pets, children or laypeople. Remember that people will be attracted to it and will want to poke their fingers, or a screwdriver, inside of it. If you make it foolproof, there'll always be a bigger fool.
It's kind of like trying to make a high power tesla coil safe for home use. Come to think of it, I once saw a video (on youtube?) of a coiler who modulated his TC with sound. Of course, there were lots of added noises from the spark gap, transformers, and from sparks hitting various parts...
Kenneth
It's kind of like trying to make a high power tesla coil safe for home use. Come to think of it, I once saw a video (on youtube?) of a coiler who modulated his TC with sound. Of course, there were lots of added noises from the spark gap, transformers, and from sparks hitting various parts...
Kenneth
This is why significant efforts must be directed towards maintaining a stable plasma. If a stable plasma is formed, we might assume very little distortion.
I'm interested in a minimum of 110dB. However, I would prefer 120dB. The transducer should be capable of 20dB transient peaks. If we assume an average loudness of 90dB, we will require 110dB. I would prefer the transducer to be capable of 10dB beyond this level to allow the listener to be a safe distance from it considering the high voltage involved.
As far as bandwidth, I would like its output capabilities to extend down to 500hz to cover a majority of the audio bandwidth. A lower bandwidth limit of 300hz would be VERY desirable as it would extend response into the modal region where Ray acoustics no longer apply. Sources can easily be combined since the acoustic field is tightly coupled.
However, I'm not sure if these constraints are realistic. The plasma will have to be capable of significant displacement. I believe another gas (ex. Helium) must be allowed to flow past the electrodes to keep ozone production at an absolute minimum.
Couldn't significant heat transport be achieved silently with a dielectric fluid cooling system? Alternatively, water could be used as long as the thermal interface possessed a very high dielectric strength. I believe Beryllium oxide would be the material of choice. Heat pipes or heat spreaders utilizing carbon-graphene foam could be used to transport thermal energy to the thermal interface (Beryllium Oxide) by a 2 phase liquid-vapor cooling system. If such a system was achieved, I believe the limitations related to heat dissipation could be primarily ignored.
Last edited:
I'm sure if a full-range plasma speaker were possible, someone would have done it by now...
I might be able to measure the transducer within my Universities anechoic chamber (Herrick Labs) if I was able to persuade some of the engineering faculty. I'm hoping demonstration of a prototype which improves upon current designs might impress them enough to make that possible. In addition, I might be able to gain access to some interesting electrical measurement equipment.
What parameters would you allow to vary if you were interested in conducting experiments?
Ozone production is an obvious problem with plasma loudspeakers. If a solution for this is not provided, they could never be considered a realistic option. We might test a variety of gases which are allowed to flow over the electrodes. Helium would of course be tested. I'm very curious as to how the distribution of Helium affects the plasma and the acoustic response.
At this point in my research, I believe significant gains can be realized by maximizing the Q of the electrical resonance. The optimal components of the resonant circuit should be capable of extremely high voltages, be maximally linear, and have an absolute minimum of resistance.
I believe a water cooled variable vacuum tube capacitor would satisfy these constraints. A vacuum will be significantly more linear than any physical dielectric and be capable of significantly higher voltages. The fact that it is adjustable is obviously very desirable since it allows the opportunity for a variety of experiments.
An air-core transformer could be used to step-up the voltage. I believe this could function as the inductor in the resonant circuit. If the transformer was insulated and cooled with a dielectric fluid, we might assume linear behavior at high power levels. One of the graduate students informed me that Purdue has a significant quantity of Alumina in addition to an induction furnace which I might be able to get access to.
I assume the coupling coefficient would have a significant effect on the leakage inductance. I believe this would be highly dependent upon the geometry of the coils and their spatial relationship. Assuming construction isn't radically complex, a variety of configurations could be tested.
However, I am still concerned that significant resistance might be encountered in copper wire. The skin effect, which will be present at 27Mhz, further exacerbates this problem. The final result of this may be a significant amount of electrical energy dissipated as heat. This is obviously undesirable since we are already very limited in terms of available power.
A minimum "holding current" will be required for a stable plasma. I believe a stable plasma would result in low distortion since it allows us to assume a Maxwellian distribution. A transient arc may produce quite chaotic and undesirable effects.
Power = V*I
Creating a potential which exceeds the dielectric breakdown of air obviously necessitates extremely high voltages. The distance between the electrodes will define the required voltage for plasma formulation. As a result, current will be inherently limited by this constraint. Assuming a stable plasma is required, we must be able to achieve the necessary "holding current". This will place a constraint on the widest achievable air gap between the electrodes. Thus, the power dissipated in the inductor will directly limit the range of electrode distances we can test.
As a result, I would like to achieve an absolute minimum of resistance within the inductor. Superconducting electromagnetic coils obviously meet these constraints. However, I believe this would result in great expense and we might dismiss them on these grounds alone.
Out of curiosity, does anyone have any idea how much superconducting coils actually cost? High temperature superconductors could be used to keep the cooling requirements at an absolute minimum. I assume they would only be a viable option if the energy required for supercooling the coils was less than the energy dissipated as heat in the copper coils. However, I have no idea how much something like that would cost or even where the parts would be sourced.
Assuming copper is the only realistic option, would narrow bands of thin copper foil wrapped around an insulator in a helical pattern minimize the skin effect and result in a minimum of resistance?
What other inductor types exist which may satisfy the aforementioned constraints? Are adjustable inductors capable of the same levels of performance as fixed types?
I assume the optimal electrode material is already well defined. I believe Tungsten satisfies these constraints (electrical conductor capable of withstanding extremely high temperatures). The diameter of the electrodes should not approach the highest desired acoustic wavelength to minimize interference.
Should I seriously consider impedance matching techniques at the connections between the components (capacitor, inductor, electrodes) of the circuit? At audio frequencies, I would not expect this to be a significant topic for consideration. However, at 27Mhz, I expect the requirements for a minimum of reflected power to be quite different.
At this point, it is not intuitively obvious to me what type of connections should be used between the thin bands of copper foil (or wire) and the electrodes. Could anybody offer further insight on this?
What properties of the plasma would you seek to measure?
What equipment would be required to measure these effects?
Thanks,
Thadman
See: "Atlas Shrugged" by Ayn Rand
Ref: "Project X"
You'll need some Rearden Metal to make the horns. I hear it's on allocation...
;-)
Michael
BTW, some of the very highest power transmitting tubes are cooled by phase change impingement (spray evaporation) cooling, pretty much as you describe.
Ref: "Project X"
You'll need some Rearden Metal to make the horns. I hear it's on allocation...
;-)
Michael
BTW, some of the very highest power transmitting tubes are cooled by phase change impingement (spray evaporation) cooling, pretty much as you describe.
Use litze coils to combat the skin effect.
I wonder what would the average DIY'ers wife have to say about speakers requiring fluid cooling and inert gas circulation?
Look at how old Cray vectorsupercomputers were cooled for some good ideas.
There is a better and cheaper alternative to superconductor technology: just buy the musicians you like to listen to.
Also I think you are forgetting that the voltage across the arc will be much less than the voltage required to get the arc started. You might want use a transformer with a magnetic shunt to automatically limit current and drop voltage once current is drawn.
The breakdown voltage for a given gap can be made smaller than you are assuming. Just use sharp pointy electrodes. The local gradient of the electric field at the point will be quite high, and will more easily create ions to get the breakdown going.
Kenneth
I wonder what would the average DIY'ers wife have to say about speakers requiring fluid cooling and inert gas circulation?
Look at how old Cray vectorsupercomputers were cooled for some good ideas.
There is a better and cheaper alternative to superconductor technology: just buy the musicians you like to listen to.
Also I think you are forgetting that the voltage across the arc will be much less than the voltage required to get the arc started. You might want use a transformer with a magnetic shunt to automatically limit current and drop voltage once current is drawn.
The breakdown voltage for a given gap can be made smaller than you are assuming. Just use sharp pointy electrodes. The local gradient of the electric field at the point will be quite high, and will more easily create ions to get the breakdown going.
Kenneth
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.