Let's say for example it's audio, driving a loudspeaker. Fairy Nuff, we could use class D, but has anyone tried a high power 2C C DAC to drive a speaker? No DC required.
(A 2C C ladder is the capacitance version of R 2R)
Let's say for example it's audio, driving a loudspeaker. Fairy Nuff, we could use class D, but has anyone tried a high power 2C C DAC to drive a speaker? No DC required.
(A 2C C ladder is the capacitance version of R 2R)
What would drive the taps?
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
Let's say maybe some FET things, maybe half bridge drivers. On or off, high or low.
If the DAC is going to drive the loudspeaker directly the speaker would need to have close to zero impedance otherwise the DAC would be non-monotonic and would sound awful. John
Sorry, unclear. I mean from digital data using a serial to parallel chip. Otherwise an ADC providing the switching for the tap drivers.
To drive the speaker directly, I'd expect gigantic caps, horrible efficiency, and lots of distortion.
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
Any sensible class-D amplifier uses feedback - it isn't just a DAC. Without feedback, the power supplies need to be so good that it is more work than building a conventional amplifier that is better than the one you are trying to build. Also, any mismatch or non-ideality in the rise and fall times will create distortion, as will the inductor, and none of that gets corrected without feedback. I expect some audiophool will try it, when they get bored with their open-loop class-A triode amplifiers.
Also, switching voltages across capacitors that are connected together without inductors in series will dissipate lots of energy in the switches, so you won't even get good efficiency.
Thanks, all good points. I did a simulation with near perfect switches so no dissipation there, and just a bit in the capacitors. However the power supply is doing lots and lots of kilowatts.
A silly idea, but a lesson learned.
An ohmic driver, like a mosfet, charging or discharging a cap, will dissipate as much energy in the driver as is delivered into or out of the cap. There's a simple mental experiment to prove that.
Switching an ideal voltage source into an ideal cap gets interesting
There are capacitive DACs, but low level, inside ICs. The SiO2 caps in ICs are extraordinarily good.
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
Have something to add? Share your thoughts — no account required.
Ask the community — no account required