Plasmatronics Speakers: Tube Replacements

[Tonehack]

Hello All and Happy New Year!

I have been ruminating on a solid state replacement for the hard to find 6MJ6 tubes used in the plasma driver. I see a way to make it from high voltage JFETs or enhancement mode MOSFETs, or perhaps SiC JFETs. I do wonder if you owners of speakers are looking for something like this. I am certainly not looking to start a business making these; I just want to know if I should put much effort into this. Please let me know your thoughts.

Tony Salsich

[brian segura]

Hello.  I would be very interested in seeing what your solution would be

Regards,
Brian

--
You received this message because you are subscribed to the Google Groups "Hill Plasmatronics" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hill-plasmatronics+unsub...@googlegroups.com.
To post to this group, send email to hill-plasmatronics@googlegroups.com.
Visit this group at https://groups.google.com/group/hill-plasmatronics.
For more options, visit https://groups.google.com/d/optout.

[Jay Philippbar]

Tony,

I like your idea and have looked into this at one time.  The Si MOSFET's probably have too much capacitance for stable operation.  I remember about 20 years ago designing a 20kV linear driver (MOSFET) for driving a CO2 laser.  The problem was that we couldn't get things stable enough and there were relaxation oscillations that couldn't be quenched.  The laser tube had about 1,200 volts drop at 20mA which is very similar to one cathode of our Plasmatronics!  The negative resistance plasma load makes life difficult.  I ended up using a single vacuum tube design (with excellent results) as the unloaded power supply voltage was about 20kV which was needed to get the tube to strike over during startup.  The lowest current rated 4,500 volt MOSFET from IXYS may work.  Yes, the SiC FET's look good and I design circuits using mostly the 1,700 volt devices (highest rated voltage).  Do you think it be best to use a single device that can handle the entire 4kV from the power supply (unloaded voltage)?  I have worked with the technical persons at both IXYS and CREE (now Wolfspeed).  I have datasheets for research devices that are up to 10kV rated however I have signed an NDA so I cannot disclose anything and this time.  One note is that these very high voltage devices are also high current and would not work in our application.  Unfortunately switchmode power conversion is their main area of interest.  GaN devices would be best however they are struggling to get past about 650 volts.  I like the idea of replacing the tubes as a good circuit would further improve linearity.

I have an extra pair of plasma units that need major work and I had thought about converting them over to fully solid-state design with fully updated electronics.  I had also looked into linear servo valves for precisely controlling the helium flow rate.  Tony, could you use a much lower unloaded power supply voltage, say around 1,500 volts and then use a high voltage ignitor circuit to get the plasma started?  This is commonly used in metal halide lamp ballasts.  The you could us a single 1,700 volt SiC device.

I am not sure if this image below will appear, however this is one of my high speed, high voltage drivers that's used in laser electronics (Pockel's Cell driver).  It is switching 4,400 volts into a 21pF load in about 17nS.  There is no overshoot or ringing.

Happy New Year to all,

Jay Philippbar

www.analogdesign.com

www.fastanalog.com

---

On 1/1/2017 8:55 AM, Tonehack wrote:

Hello All and Happy New Year!

I have been ruminating on a solid state replacement for the hard to find 6MJ6 tubes used in the plasma driver. I see a way to make it from high voltage JFETs or enhancement mode MOSFETs, or perhaps SiC JFETs. I do wonder if you owners of speakers are looking for something like this. I am certainly not looking to start a business making these; I just want to know if I should put much effort into this. Please let me know your thoughts.

Tony Salsich

--
You received this message because you are subscribed to the Google Groups "Hill Plasmatronics" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hill-plasmatron...@googlegroups.com.
To post to this group, send email to hill-plas...@googlegroups.com.

[Tonehack]

Jay - I agree that there is one part from IXYS that has a chance of working: the 4500V, 200mA part has fairly low Coss and also has the SOA for the job. All the rest of their parts are not likely to work. As for a lower bus with a 'striker' circuit, i am pretty sure that the higher bus was needed to allow for the stabilizing resistor (25kOhm) voltage drop. I played with the value of that resistor and the supply voltage many long years ago and was concerned with the stability of the circuit. Also, my goal (perhaps too lofty) is to come up with a direct tube replacement - no mods to the speaker electronics. This will take a while to be sure. 

A 3-4kV JFET would be a real boon. I will check with Infineon (Wolfspeed) and USiC to see if they have something along those lines. SiC really should be able to do the voltage.

Thanks for the feedback, Brian. 

Tony

[brian segura]

Tony:

FYI...When it comes to ignition of the helium to plasma.  I am using Jay's current sink circuit.  It had ignited perfectly when the helium was purged.

Regards,
Brian

--

You received this message because you are subscribed to the Google Groups "Hill Plasmatronics" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hill-plasmatronics+unsub...@googlegroups.com.
To post to this group, send email to hill-plasmatronics@googlegroups.com.

[John Mayberry]

The answer is yes.

 

Hope all is well.

 

John

--

You received this message because you are subscribed to the Google Groups "Hill Plasmatronics" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hill-plasmatron...@googlegroups.com.
To post to this group, send email to hill-plas...@googlegroups.com.