I'm in the same boat. If I had the time and money to thoroughly research the project I'd jump right on it. I did build a hastily constructed version of a design in AudioExpress(I don't remember which issue) It worked ok as a transducer, but not great, probably because I used junk parts and strayed from the original design so I could use what was on hand. Pretty cool nontheless. The fun part is using the corona to cut things.
P.S. I like your sig.
P.S. I like your sig.
plasma plans
Hi,
I built one of these last year. It was a great fun project. I used a sort of synthesis of Colin Joye's scheme and Ulhrich Haumann (they may both be misspelled). The EL/PL509 or 519 was the tube of choice for both and I think it is the tube that was used in the last commercial products. I screen modulated mine from the output of a small 6V6 PP fed through a line transformer run backwards to modulate the screen.
Anyway, I think the reason they are more like works in progress is that there are many ways and many tubes that would work fine. The secret is in the tuning of the coils and capacitance. That makes a huge difference in sound quality and efficiency. I emphasize HUGE. Until some one makes a standard tank circuit, and then it's not DIY, it will always be that way.
I do encourage you to try it, it is a cheap fun project. When you want to make it a permanent part of your system is when the real work will start. You will have to figure out a good shielding scheme for the RFI and build in the types of safety precautions that are necessary for a fire generating beast that runs on high voltage and current.
There is a lot of information to be gleaned from old patents if you want to search there. The original one, from 1952 or 3, used a pair of 807s. The 6146 have been used and I was thinking a 4-65 or 4-125 might make a good option. Read up on radio transmitters and vacuum tube tesla coils, they have a lot more in common than anything out of the audio world.
If you have any specific questions, ask and I will see if I can dig up my notes and literature.
By the way, they do sound incredible and it is on my short list of things to revisit.
Enjoy,
Michael
Hi,
I built one of these last year. It was a great fun project. I used a sort of synthesis of Colin Joye's scheme and Ulhrich Haumann (they may both be misspelled). The EL/PL509 or 519 was the tube of choice for both and I think it is the tube that was used in the last commercial products. I screen modulated mine from the output of a small 6V6 PP fed through a line transformer run backwards to modulate the screen.
Anyway, I think the reason they are more like works in progress is that there are many ways and many tubes that would work fine. The secret is in the tuning of the coils and capacitance. That makes a huge difference in sound quality and efficiency. I emphasize HUGE. Until some one makes a standard tank circuit, and then it's not DIY, it will always be that way.
I do encourage you to try it, it is a cheap fun project. When you want to make it a permanent part of your system is when the real work will start. You will have to figure out a good shielding scheme for the RFI and build in the types of safety precautions that are necessary for a fire generating beast that runs on high voltage and current.
There is a lot of information to be gleaned from old patents if you want to search there. The original one, from 1952 or 3, used a pair of 807s. The 6146 have been used and I was thinking a 4-65 or 4-125 might make a good option. Read up on radio transmitters and vacuum tube tesla coils, they have a lot more in common than anything out of the audio world.
If you have any specific questions, ask and I will see if I can dig up my notes and literature.
By the way, they do sound incredible and it is on my short list of things to revisit.
Enjoy,
Michael
Are plasma tweeters significantly better than good ESL panels?
audiobot, I don't have a specific question because my question is about getting started; all I do know is that I want to use them as tweeters in the speaker setup I'll be putting together this summer, so it's not just about playing around with HV (I already do this other ways
), but it's mainly about sound quality.
Certainly, if you do revisit the project, make sure to share your designs with us
BTW I was also thinking about the Ionic Cloud. Why not just put that kind of design in a sealed enclosure to avoid the ozone and nitric oxides problem? I seem to remember sealed chamber ESL speakers, althogh for a different reason (argon instead of air to prevent arcing, or something like that).
audiobot, I don't have a specific question because my question is about getting started; all I do know is that I want to use them as tweeters in the speaker setup I'll be putting together this summer, so it's not just about playing around with HV (I already do this other ways
), but it's mainly about sound quality.Certainly, if you do revisit the project, make sure to share your designs with us
BTW I was also thinking about the Ionic Cloud. Why not just put that kind of design in a sealed enclosure to avoid the ozone and nitric oxides problem? I seem to remember sealed chamber ESL speakers, althogh for a different reason (argon instead of air to prevent arcing, or something like that).
sound quality
I've not had much experience with different ESL speakers. I will say the the plasma tweeter had a sound quality I have not heard before. In it's simplest form it is a pure point source and that does somehow seem to make a unique difference.
If you are familiar with HV work and have a variable power supply that can give you a 150mA or so, you should be able to knock up a working model in an afternoon or two. Then you can decide for yourself if it is worth pursuing to the extent of being able to add it into your system.
I didn't notice any problems with NO2 in the hundred or so hours I ran mine. Ozone production was a different story. It may seem funny, but when the system was tuned and playing the loudest, lowest distortion, etc the ozone production seemed to be at a minimum, almost indetectable and similar to one of those ionic air fresheners. Overall sound output level is a problem you will have to deal with, they are not power efficient to say the least. On the other hand I am judging from the standpoint of running them almost full range most of the time I was playing. If horn loaded, my rough experiments, they are loud, but lose the point source effect. I do think horn loading could be used to minimize ion migration as well, it that remains a concern with you.
I will try to dig out some old response files and maybe I can draw a schematic. Like I said though, it is a piece of cake to toss up a prototype. Probably some other people on the forum have had more experience with these than I have, though.
Michael
I've not had much experience with different ESL speakers. I will say the the plasma tweeter had a sound quality I have not heard before. In it's simplest form it is a pure point source and that does somehow seem to make a unique difference.
If you are familiar with HV work and have a variable power supply that can give you a 150mA or so, you should be able to knock up a working model in an afternoon or two. Then you can decide for yourself if it is worth pursuing to the extent of being able to add it into your system.
I didn't notice any problems with NO2 in the hundred or so hours I ran mine. Ozone production was a different story. It may seem funny, but when the system was tuned and playing the loudest, lowest distortion, etc the ozone production seemed to be at a minimum, almost indetectable and similar to one of those ionic air fresheners. Overall sound output level is a problem you will have to deal with, they are not power efficient to say the least. On the other hand I am judging from the standpoint of running them almost full range most of the time I was playing. If horn loaded, my rough experiments, they are loud, but lose the point source effect. I do think horn loading could be used to minimize ion migration as well, it that remains a concern with you.
I will try to dig out some old response files and maybe I can draw a schematic. Like I said though, it is a piece of cake to toss up a prototype. Probably some other people on the forum have had more experience with these than I have, though.
Michael
Re: sound quality
Yeah, but it's a good sounding prototype I'm after
If you have a good schematic, please post one, or comment on existing schematics such as
http://www.ece.villanova.edu/~cdanjo/schematics.html
or
http://www.angelfire.com/electronic/cwillis/tweeter.html
Yeah, but it's a good sounding prototype I'm after
If you have a good schematic, please post one, or comment on existing schematics such as
http://www.ece.villanova.edu/~cdanjo/schematics.html
or
http://www.angelfire.com/electronic/cwillis/tweeter.html
plans
Hi,
As you can see at the following link this commercially available plasma tweeter is based on the EL519, so I would say a good sounding result is possible.
http://www.plasmatweeter.de/acapella.htm
This site has a lot of good background and design info also.
http://www.plasmatweeter.de/
I will try to find my drawings, they are not in electronic form so I will have to find time to convert them before they can be posted, and data sometime today. For now let me say, the plasma speaker/amplifier is fundamentally different, in my view, from other tube audio projects. It is a self oscillating Class C amplifier. As such, execution is everything. A picofarad here or there, which has little or no effect in the audio range, has a huge effect because you are working with a carrier in the MHz range. That means that in the abscence of a commercially available tank circuit, the coil must be tuned empirically. You can achieve excellent sonic results and the fact that there have been commercial products available show that the other obstacles (RFI, heat, ozone) can be overcome. But if you are unwilling to experiment this is probably not a project for you.
If you can get a high quality audio signal to the amp (transmitter)you can get quality out. My speaker sounded fantastic to my ears. The easiest and down and dirty way to do this that I found was to use a stepup transformer cap coupled to the screen supply of the EL519. The rest of the circuit in front of the 519 is just preamp, crossover and RFI protection, you can pick any design you want.
Colin Joye, who wrote the Audio Express article, believed that the best results were obtained with a low DCR coil oscillating around 27MHz. I have not seen the article, but I did have private communication with Mr. Joye prior to it's publication. Willis used a much higher DCR coil around 27MHz and also claimed good results. To me this just shows there are many paths to the same end. From my own experiments (from memory) I obtained the best result with a moderate DCR coil oscillating at a much lower frequency, 2-12MHz. I tried the 6CL6, per Mr. Willis' design, to drive the screen and it worked fine but I can't say it was vastly superior to the transformer coupling. The main thing is tuning the circuit so that the phase angle of current conduction is offset from the voltage phase angle such that the conduction takes place when there is no voltage drop across the tube. In early experiments I had the plates glowing bright red at only about 600V and 60mA. Later, with more tuning, I could pass 200mA with a plate voltage of over 700 and the tube seemed just fine with it. There is a technical article on the Svetlana site about using the EL519 as a radio transmitter that has some useful information.
I will warn you that my Radio Shack voltmeter was useless for measuring the circuit, turn it on anywhere near the thing and it would start switching from volts to amps to ohms and unknown modes, even turning off and on, all on it's own. I even had problems getting the SPL meter to work. This is of course with no shielding whatsoever, just experiments you understand. On the plus side, if you have a scope, all you have to do is hang a probe off of it and you will receive the signal. Very handy for checking modulation level, carrier frequency and watching for overloads on the transients of the audio signal.
If you want to try a really over the top design download the patents dealing with the Hill Plasmatronic. It was a masterpiece of design and dealt with the known problems of intermodulation distortion (which I did not really find to cause any objectionable effects) and ozone production.
It is a fun project and if you are willing to put in the time and thought you will end up with a truly unique speaker that few have heard and even fewer own. The only real downside is that if you don't finish the project, as I haven't, you will find it very hard to go back to an ordinary tweeter.
Michael
Hi,
As you can see at the following link this commercially available plasma tweeter is based on the EL519, so I would say a good sounding result is possible.
http://www.plasmatweeter.de/acapella.htm
This site has a lot of good background and design info also.
http://www.plasmatweeter.de/
I will try to find my drawings, they are not in electronic form so I will have to find time to convert them before they can be posted, and data sometime today. For now let me say, the plasma speaker/amplifier is fundamentally different, in my view, from other tube audio projects. It is a self oscillating Class C amplifier. As such, execution is everything. A picofarad here or there, which has little or no effect in the audio range, has a huge effect because you are working with a carrier in the MHz range. That means that in the abscence of a commercially available tank circuit, the coil must be tuned empirically. You can achieve excellent sonic results and the fact that there have been commercial products available show that the other obstacles (RFI, heat, ozone) can be overcome. But if you are unwilling to experiment this is probably not a project for you.
If you can get a high quality audio signal to the amp (transmitter)you can get quality out. My speaker sounded fantastic to my ears. The easiest and down and dirty way to do this that I found was to use a stepup transformer cap coupled to the screen supply of the EL519. The rest of the circuit in front of the 519 is just preamp, crossover and RFI protection, you can pick any design you want.
Colin Joye, who wrote the Audio Express article, believed that the best results were obtained with a low DCR coil oscillating around 27MHz. I have not seen the article, but I did have private communication with Mr. Joye prior to it's publication. Willis used a much higher DCR coil around 27MHz and also claimed good results. To me this just shows there are many paths to the same end. From my own experiments (from memory) I obtained the best result with a moderate DCR coil oscillating at a much lower frequency, 2-12MHz. I tried the 6CL6, per Mr. Willis' design, to drive the screen and it worked fine but I can't say it was vastly superior to the transformer coupling. The main thing is tuning the circuit so that the phase angle of current conduction is offset from the voltage phase angle such that the conduction takes place when there is no voltage drop across the tube. In early experiments I had the plates glowing bright red at only about 600V and 60mA. Later, with more tuning, I could pass 200mA with a plate voltage of over 700 and the tube seemed just fine with it. There is a technical article on the Svetlana site about using the EL519 as a radio transmitter that has some useful information.
I will warn you that my Radio Shack voltmeter was useless for measuring the circuit, turn it on anywhere near the thing and it would start switching from volts to amps to ohms and unknown modes, even turning off and on, all on it's own. I even had problems getting the SPL meter to work. This is of course with no shielding whatsoever, just experiments you understand. On the plus side, if you have a scope, all you have to do is hang a probe off of it and you will receive the signal. Very handy for checking modulation level, carrier frequency and watching for overloads on the transients of the audio signal.
If you want to try a really over the top design download the patents dealing with the Hill Plasmatronic. It was a masterpiece of design and dealt with the known problems of intermodulation distortion (which I did not really find to cause any objectionable effects) and ozone production.
It is a fun project and if you are willing to put in the time and thought you will end up with a truly unique speaker that few have heard and even fewer own. The only real downside is that if you don't finish the project, as I haven't, you will find it very hard to go back to an ordinary tweeter.
Michael
here you go
http://gilmore.chem.northwestern.edu/ionovac1.gif
http://gilmore.chem.northwestern.edu/ionovac2.gif
I have a couple of pairs of these plus some spare parts.
Without the horn the efficiency is just unusable.
http://gilmore.chem.northwestern.edu/ionovac1.gif
http://gilmore.chem.northwestern.edu/ionovac2.gif
I have a couple of pairs of these plus some spare parts.
Without the horn the efficiency is just unusable.
Thanks, Kevin. These look similar to most of the DIY designs I've found, and it appears many commercial ones too.
The Hill speakers are unique in that they do not use a corona discharge, but a lower voltage helium plasma between electrodes. I came across this page:
http://members.cox.net/jayphil/plasma.html
The curves shown here look flatter and show lower frequency response than what I've seen for corona discharge tweeters. What I find most interesting is his claim that they do not use RF to pump the plasma, so that's why I was trying to find info on the driving circuit. I tried contacting the webmaster of that site, but he said he was busy with other stuff and didn't provide any info.
I wonder if this approach would work with air plasma, as messing with helium tanks doesn't look fun or cheap.
The Hill speakers are unique in that they do not use a corona discharge, but a lower voltage helium plasma between electrodes. I came across this page:
http://members.cox.net/jayphil/plasma.html
The curves shown here look flatter and show lower frequency response than what I've seen for corona discharge tweeters. What I find most interesting is his claim that they do not use RF to pump the plasma, so that's why I was trying to find info on the driving circuit. I tried contacting the webmaster of that site, but he said he was busy with other stuff and didn't provide any info.
I wonder if this approach would work with air plasma, as messing with helium tanks doesn't look fun or cheap.
Well, at least I found Hill's patent (thanks to someone at audioasylum). It's 4,219,705. Fascinating. PDF version can be retrieved using http://free.patentfetcher.com/
I am curious what the DC bias voltage and the amplitude of the AC signal is in Hill's speakers. Obviously the AC low peak cannot be nearly as much as the DC bias, since the total voltage would drop too low on the low peaks and extinguish the plasma.
I tried an experiment: amplifier output through a 12V step-down transformer BACKWARDS to get a couple of hundred V, then biased the transformer output with rectified mains (340V). I hooked up a lit propane torch to +, and held the - electrode at the flame end. The low voltage means small gap, and with unsteady hands I was mostly getting noisy arcs. However, on occasion I was able to get sound for at most a second at a time. I was about to build a voltage multiplier ladder to try higher voltage, but the arcs are almost shorts and overheated the small transformer -- POP! and that was the end of my experiments. Also, my eyes are messed up from the UV emitted by the arcs.
It's probably hard to get a stable air (or in my case, mostly CO2) glow discharge at atmospheric pressure even with sufficient voltage and shaping of the gas flow, but my search on Google revealed that researchers have used microhollow cathode discharge (MHCD) to get stable plasma at these conditions, so that may be one way to improve performance instead of getting mostly arcs.
Edit: I forgot to high-pass the signal, which must have made thing less stable. Also, if anyone else should try this, the torch should be on low output as fast flow disturbs the discharge.
I tried an experiment: amplifier output through a 12V step-down transformer BACKWARDS to get a couple of hundred V, then biased the transformer output with rectified mains (340V). I hooked up a lit propane torch to +, and held the - electrode at the flame end. The low voltage means small gap, and with unsteady hands I was mostly getting noisy arcs. However, on occasion I was able to get sound for at most a second at a time. I was about to build a voltage multiplier ladder to try higher voltage, but the arcs are almost shorts and overheated the small transformer -- POP! and that was the end of my experiments. Also, my eyes are messed up from the UV emitted by the arcs.
It's probably hard to get a stable air (or in my case, mostly CO2) glow discharge at atmospheric pressure even with sufficient voltage and shaping of the gas flow, but my search on Google revealed that researchers have used microhollow cathode discharge (MHCD) to get stable plasma at these conditions, so that may be one way to improve performance instead of getting mostly arcs.
Edit: I forgot to high-pass the signal, which must have made thing less stable. Also, if anyone else should try this, the torch should be on low output as fast flow disturbs the discharge.
I have come across these two papers on stable air plasmas using a microhollow cathode discharge as the cathode (i.e. the cathode is a discharge itself): "Direct Current Glow Discharges in Atmospheric Air" (IEEE Transactions on Plasma Science, Vol. 30, No. 1, February 2002) and "Formation of Large-volume, High-pressure Plasmas in Microhollow
Cathode Discharges" (Applied Physics Letters Vol. 82, No. 19, 12 May 2003).
I'm going to try to build the system described in the latter article, though drilling clean submillimeter holes in the copper/mica sandwitch doesn't seem easy. With this type of desing, it is possible to have large glow discharges by increasing voltage without transitioning to arcs.
The problems with using air instead of helium as in Hill's design are that it is probably more difficult to get a good response curve and linearity (I'm not sure if different gasses in air have differing thermal relaxation behavior), and the generation of O3 and especially NO2 (too bad it's NO2 instead of N2O, the latter being laughing gas
). The first problem can be dealt with circuit and mechanical desing. The thermal relaxation is affected by the cooling plates (see Hill's patent), for example. The second problem can be either dealt with a catalyst coated cage, or by using a natural gas flame (much cheaper than helium), or by reversing the fan in Hill's patent diagram to suck out the gases through a duct.
Any design, including Hill's (and even the corona discharge type tweeters), however, suffers from a serious problem that I don't see any way to avoid: UV emission. There's no UV shield I can think of that won't affect the sound. Even when horn loaded this is a problem, since you have to stand in front of the horn.
Cathode Discharges" (Applied Physics Letters Vol. 82, No. 19, 12 May 2003).
I'm going to try to build the system described in the latter article, though drilling clean submillimeter holes in the copper/mica sandwitch doesn't seem easy. With this type of desing, it is possible to have large glow discharges by increasing voltage without transitioning to arcs.
The problems with using air instead of helium as in Hill's design are that it is probably more difficult to get a good response curve and linearity (I'm not sure if different gasses in air have differing thermal relaxation behavior), and the generation of O3 and especially NO2 (too bad it's NO2 instead of N2O, the latter being laughing gas
). The first problem can be dealt with circuit and mechanical desing. The thermal relaxation is affected by the cooling plates (see Hill's patent), for example. The second problem can be either dealt with a catalyst coated cage, or by using a natural gas flame (much cheaper than helium), or by reversing the fan in Hill's patent diagram to suck out the gases through a duct.Any design, including Hill's (and even the corona discharge type tweeters), however, suffers from a serious problem that I don't see any way to avoid: UV emission. There's no UV shield I can think of that won't affect the sound. Even when horn loaded this is a problem, since you have to stand in front of the horn.
I bought a 2.1kV 0.7A MOT from eBay; when I'll get it I'll post more results.
BTW there's a photo of the PlasmaSonic ion headphones:
http://www.audioasylum.com/cgi/m.pl?forum=general&n=337109&highlight=&r=&session=
These are also corona discharge, but it would suggest that full range glow discharge drivers are workable for headphones, and could probably put the Stax OmegaII to shame
BTW there's a photo of the PlasmaSonic ion headphones:
http://www.audioasylum.com/cgi/m.pl?forum=general&n=337109&highlight=&r=&session=
These are also corona discharge, but it would suggest that full range glow discharge drivers are workable for headphones, and could probably put the Stax OmegaII to shame
I got a 1.5 kW MOT; one end of the HV winding is connected to the transformer case. Would there be any problems with disconnecting that end so I can use a bridge rectifier?
Also, where do I buy 3 kV rectifiers and capacitors? I know I can put regular ones in series but that's not as good due to uneven distribution of the voltage, and using resistor dividers as some people do to fix that results in a lot of components and a mess.
The heater winding puts out about 3 V RMS. I'm wondering if I use a full wave doubler to get 6 V and rectify that, if it can work as heater supply for tubes. I don't know how to figure out how much current it can supply (the winding only has about three turns).
Also, where do I buy 3 kV rectifiers and capacitors? I know I can put regular ones in series but that's not as good due to uneven distribution of the voltage, and using resistor dividers as some people do to fix that results in a lot of components and a mess.
The heater winding puts out about 3 V RMS. I'm wondering if I use a full wave doubler to get 6 V and rectify that, if it can work as heater supply for tubes. I don't know how to figure out how much current it can supply (the winding only has about three turns).
Prunce, interesting journey. Thanks.
Is the use of helium rather than air possibly also to prevent electrode burn out? Could something like argon be profitably used? Since it's heavy, one could essentially curtain the electrodes from above. I think the tradeoff might be the energy necessary to make the gas jiggle because of its higher density.
Is the use of helium rather than air possibly also to prevent electrode burn out? Could something like argon be profitably used? Since it's heavy, one could essentially curtain the electrodes from above. I think the tradeoff might be the energy necessary to make the gas jiggle because of its higher density.
Well, all the corona-type ionic tweeters have an electrode in air at similar temperatures. Electrodes shold be tungsten or something similar. However, one of my thoughts is to use C02, that is, burning natural gas (butane or propane). Fully burned natural gas shouldn't contain but trace amounts of any oxidizers. That has another use, too, since the injected gas needs to be heated (see Hill's patent). The amount of natural gas used would still make this much cheaper than helium, especially if plumbed in (I'd have to run a line from the kitchen though... )
)
Stable DC glow discharge, 3 kV at 100 mA:
http://diyaudio.com/forums/showthread.php?postid=469140#post469140
I'll try full power (2 x 200 mA) when I get the rest of the filter capacitors and chokes. MHCD not shown in the pictures, just plain electrodes, but I did try it with good results.
Now I need to figure out what sort of amplifier will drive this...
http://diyaudio.com/forums/showthread.php?postid=469140#post469140
I'll try full power (2 x 200 mA) when I get the rest of the filter capacitors and chokes. MHCD not shown in the pictures, just plain electrodes, but I did try it with good results.
Now I need to figure out what sort of amplifier will drive this...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.