-
Notifications
You must be signed in to change notification settings - Fork 1
Future Work
gideonpowell edited this page Sep 5, 2022
·
2 revisions
This page describes future work and improvements that can be made to the Plasma Speaker demo.
The Power Stage was copied from Franzoli and has had very little modification so far. It could use some work.
- The large power MOSFET (Q2 - IRFP260MPBF) could be replaced by a MOSFET with a lower drain-source resistance. This would improve efficiency and reduce heat.
- Re-design the gate driver (Q1 and D2). Switching the power MOSFET on and off requires a spike of high current. A gate driver is typically used between the output of an IC and the gate of a power MOSFET to source that current. It essentially acts as a current amplifier, since the IC may not be able to supply enough current. The BJT Q1 (in combination with D2) appears to be used as a gate driver, but it still sources it's current from the IC, which defeats the point. Also, if you connect an oscilloscope to the gate of the power MOSFET, you will see that the square wave has a slow rise or fall time (I can't remember which). The longer this state of transition, the more power will be consumed in the MOSFET. Redesigning the gate driver could also reduce the rise/fall time and help reduce heat.
The oscillator was copied from Franzoli and it is a mess. It could use a complete re-design.
- Why are there so many resistors and capacitors all over the place? The cap on the 12V line is a good idea, but what about the rest of them? This stage should just output a square wave (with a frequency controlled by VR1).
- What frequency should it run at? Franzoli said it works best at 22.5kHz, but we haven't tested that.
- Add "pin 1" markings for the ICs. Typically this is done with a small dot in the silkscreen next to pin 1. This will make it easier when you need to install an IC (or rather, the IC socket).
- Use a standard 1/4 watt resistor for R2. The exact value of this resistor is not important, it will just affect the corner frequency of the high-pass filter. Anything close to 80 ohms will do.