As mentioned yesterday in Class D instead of class AB, I tried making a class D amplifier out of the one-op-amp amplifier I had built before, adding a comparator comparing that value with a triangle wave from my FG-500 function generator. The output of the comparator feeds into the gates of a pair of FETs arranged as a cMOS inverter. The slew rate of the the MCP6002 or MCP6004 was causing problems: I couldn’t use a high enough frequency on the triangle wave to be inaudible, and the FETs were spending too much time in the intermediate region where they had higher resistance (and more power dissipation).
This morning I replaced the MCP6002 op amp used as a comparator with an LM311, which is designed for use as a comparator. Its output just turns on a bipolar transistor between two pins, so I needed to use a pull-up resistor. I chose to use 100Ω, as a compromise between getting a fast rise time and dissipating too much power when the transistor is turned on.
This design worked fine as a class D amplifier, and the FETs for the output stage barely got warm. I’m planning to restructure the power amp lab around class D amplifiers rather than class AB, since the design is much simpler to do (no worries about the threshold voltages of the FETs moving around or thermal runaway). Class D is also more typical of cheap audio power amplifiers these days, so students should become familiar with it, even if it is not much used in bioinstrumentation.
I’ll probably have the students use the LM2903, rather than the LM311, since it seems to be the cheapest through-hole comparator available from DigiKey, and does not have the complications of the balance and strobe pins of the LM311 (nor the pair of output pins). I’ll order a bunch from DigiKey, and check that the design works ok with them as well. The LM2903 does have only a 20mA output current, rather than the 45mA output current for the LM311, which would mean a 330Ω pullup resistor, rather than 150Ω for the LM311 with a 6.6v supply (hmm, my 100Ω resistor was exceeding the spec a bit). That means somewhat slower rise times for the gates of the FETs, which is probably the limiting factor on the efficiency of the amplifier. I suppose I could add a class A stage with a bipolar transistor to drive the FET gates if I needed faster rise times, but I think that the current design is working well enough for a 1-week lab exercise, so I won’t mess with it unless I need to.
If this were a more advanced class on amplifiers, we’d probably go through all the major classes of amplifier design, but for a first circuits class, I think that using op amps, comparators, Schmitt triggers, instrumentation amps, and class D amplifiers is probably enough.