Hello! I have a question regarding the input impedance of an op amp in a non-inverting configuration. I need to make a voltage follower with an input impedance of 500kohms. I was looking at the datasheet for NE5532 and it says that the input impedance is typically 300k. What does that exactly mean? I thought that the input impedance in a non-inv configuration is determined by the value of the bias resistor. According to the specs the bias input current is 200nA, so this shouldn't be a problem with a bias resistor of 500k? (200nA * 500kOhms = 0.1V drop across it. I think that's low enough or is it?) With a bias resistor of 500 k, what would the actual input resistance - 500k or 500k in parallel with the op amp's 300k?
I take it that the input impedance quoted is the differential one - between + and - inputs. It seems about right because LTPs in input stages typically run with collector currents in the range of 100's of uA. This would give an Re (equivalent emitter resistance) of the order of 100R or so. With the input transistors having Hfe around 1000 (estimated from the bias current) it would mean each is looking like 100k into the base terminal.
In a non-inverting configuration the feedback means you end up with the common-mode input impedance which is at least two orders of magnitude higher, because the Re becomes the tail current source's dynamic impedance. So to all intents and purposes, you'll just see the bias resistor as the input impedance. There is the CM input capacitance to take into account though at higher frequencies.
In a non-inverting configuration the feedback means you end up with the common-mode input impedance which is at least two orders of magnitude higher, because the Re becomes the tail current source's dynamic impedance. So to all intents and purposes, you'll just see the bias resistor as the input impedance. There is the CM input capacitance to take into account though at higher frequencies.
.
Because this thread has been referenced from other posts:
300k is the input impedance (Zin) of the NE5532 itself. Not the Zin of any given circuit.
With all op amps the use of feedback increases the effective circuit Zin to several time that of the op amp itself. So yes the bias resistor sets the effective circuit Zin. In buffer configuration (100% feedback) The NE5532 has a Zin of several megohms.
BUT the NE5532 has a worst-case bias current of 1000nA = 1 uA = 0.001mA. Across high-value resistors and amplified this can create serious DC offset errors. To pick a number out of the air, keep DC offset under 20mV.
If a circuit doesn't have a lot of gain this DC offset might not be a problem as long as you use a blocking capacitor at the output--perhaps 470uF.
Otherwise a TL072 or OPA2134 (among others) would be a better choice because both have extremely low bias currents.
The TL072 is the economy way to go. It costs 50-cents, but it's not the best audio quality you can get.
The OPA2134 costs $3.50 on eBay, but it's recognized as an excellent audio chip.
They'll both work in the same circuit, so a builder's wise move might be to first work things out with the TL072, then do the actual build with the OPA2134. Or the TL072 might be all that's needed anyway, it's a standard chip for use with electric guitars.
.
Because this thread has been referenced from other posts:
300k is the input impedance (Zin) of the NE5532 itself. Not the Zin of any given circuit.
With all op amps the use of feedback increases the effective circuit Zin to several time that of the op amp itself. So yes the bias resistor sets the effective circuit Zin. In buffer configuration (100% feedback) The NE5532 has a Zin of several megohms.
BUT the NE5532 has a worst-case bias current of 1000nA = 1 uA = 0.001mA. Across high-value resistors and amplified this can create serious DC offset errors. To pick a number out of the air, keep DC offset under 20mV.
If a circuit doesn't have a lot of gain this DC offset might not be a problem as long as you use a blocking capacitor at the output--perhaps 470uF.
Otherwise a TL072 or OPA2134 (among others) would be a better choice because both have extremely low bias currents.
The TL072 is the economy way to go. It costs 50-cents, but it's not the best audio quality you can get.
The OPA2134 costs $3.50 on eBay, but it's recognized as an excellent audio chip.
They'll both work in the same circuit, so a builder's wise move might be to first work things out with the TL072, then do the actual build with the OPA2134. Or the TL072 might be all that's needed anyway, it's a standard chip for use with electric guitars.
.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.