On the other hand the input bias current looks like trouble for an integrator. The loop gain of an integrator, which I am guessing is a big part of the filter circuit you mentioned, is highly frequency dependent so maybe it matters in that circuit. My dim recall and shaky understanding tell me that the use of feedback dramatically reduces the impact of the input impedance of the device itself: the overall input impedance of an amplifier will be something like that 300k multiplied by the loop gain, which is usually a large number. I have used these for sample and holds in the past, in a pinch, when I had run out of LF444 (my preferred opamp for S&H applications, with its typical bias current of 10pA).
Incidentally, the TL064 has typical input bias current of 30pA, less than half the 072. At 300k input impedance, that seriously affects how you design, because this sets up a divider with the input resistor which changes the gain significantly, I should think. I would say that the first "golden rule" of opamps (the inputs admit no current) barely applies to the 5532. So, the 5532 has input impedance 3 million times lower, and bias current about 3000 times higher, than 072. Here is how the two stack up ("typical" values from datasheets): Well, I'm not sure, but I'd guess it's the input impedance.
Is it the input impedance that's the trouble, or the input bias current? (Not a rhetorical question - I'd love to understand this properly.) Sometimes, the most important aspect of an opamp is its input impedance. Sketch-n-Etch wrote:A friend recently tried to replace TL072 with NE5532 in a filter circuit, and the circuit basically didn't work.
0 kommentar(er)
