bcarso
Well-known member
I'll have to look at THAT ;-)
BTW an aside: because a customer was unable to get stock on an integrated power amp for a powered subwoofer that needs to be shipping with agency approvals in Sept. (ouch), I cranked out a simple DMOS output power amp design in record time. I ended up using a current feedback topology tailored to the modest bandwidth requirements of the woofer. That's how I came to notice the A06/A56 close Vbe match, at least according to the simulator.
In the end some alternative parts were found so I won't need to use it, but it looked quite decent in a mid-fi way in simulation. It also caused Circuitmaker conniption fits when I attempted fourier analysis, but I'm used to that now. Even the transient analysis was baulky and required a number of tricks to run. When fourier fails I throw in a cascaded notch filter I have handy and thwart the program. It's also feasible to throw in narrow bandpass filters and analyze the notch residue, although you can already get a pretty good idea of what is going on just looking at them in the time domain. In this case they confirmed my suspicions that this would not be a very pretty-sounding amp in the audio midrange (ugly high-order artifacts), but would manage with a sufficiently band-limited signal.
BTW an aside: because a customer was unable to get stock on an integrated power amp for a powered subwoofer that needs to be shipping with agency approvals in Sept. (ouch), I cranked out a simple DMOS output power amp design in record time. I ended up using a current feedback topology tailored to the modest bandwidth requirements of the woofer. That's how I came to notice the A06/A56 close Vbe match, at least according to the simulator.
In the end some alternative parts were found so I won't need to use it, but it looked quite decent in a mid-fi way in simulation. It also caused Circuitmaker conniption fits when I attempted fourier analysis, but I'm used to that now. Even the transient analysis was baulky and required a number of tricks to run. When fourier fails I throw in a cascaded notch filter I have handy and thwart the program. It's also feasible to throw in narrow bandpass filters and analyze the notch residue, although you can already get a pretty good idea of what is going on just looking at them in the time domain. In this case they confirmed my suspicions that this would not be a very pretty-sounding amp in the audio midrange (ugly high-order artifacts), but would manage with a sufficiently band-limited signal.
