OPA2607 Datasheet, PDF (9/13 Pages) Burr-Brown (TI) – Dual, High Output, Current-Feedback OPERATIONAL AMPLIFIER

The second major consideration, touched on in the previous
paragraph, is that the signal-source impedance becomes part
of the noise-gain equation and will have slight effect on the
bandwidth through Equation 1. The values shown in Figure
4 have accounted for this by slightly decreasing R_F to re-
optimize the bandwidth for the noise gain. In the example of
Figure 4, the R_M value combines in parallel with the external
50Ω source impedance, yielding an effective driving imped-
ance of 50Ω || 75Ω = 30.0Ω. This impedance is added in
series with R_G for calculating the noise gain, which gives
NG = 7.72 (instead of NG = 9.00 with a 0Ω source). This
value, along with the R_F of Figure 3 and the inverting input
impedance of 33Ω, are inserted into Equation 3 to get
R_F = 1186Ω.
Note that the non-inverting input in this bipolar-supply
inverting application is connected directly to ground. It is
often suggested that an additional resistor be connected to
ground on the non-inverting input to achieve bias-current
error cancellation at the output. The input bias currents for
a current-feedback op amp are not generally matched in
either magnitude or polarity. Connecting a resistor to ground
on the non-inverting input of the OPA2607 in the circuit of
Figure 4 will actually provide additional gain for that input’s
bias and noise currents, but will not decrease the output DC
error since the input bias currents are not matched.
NOISE PERFORMANCE
Wideband current-feedback op amps generally have a higher
output noise than comparable voltage-feedback op amps. The
OPA2607 offers an excellent balance between voltage and
current noise terms to achieve low output noise. The inverting
current noise (15pA/√Hz) is significantly lower than competi-
tive solutions, while the input voltage noise (1.7nV/√Hz) is
lower than most unity gain stable, wideband, voltage-feedback
op amps. This low input voltage noise was achieved at the
price of higher non-inverting input current noise (11pA/√Hz).
As long as the AC source impedance looking out of the non-
inverting node is less than 100Ω, this current noise will not
contribute significantly to the total output noise. The op amp
input voltage noise and the two input current noise terms
combine to give low output noise under a wide variety of
operating conditions. Figure 5 shows the op amp noise analy-
sis model with all the noise terms included. In this model, all
noise terms are taken to be noise voltage or current density
terms in either nV/√Hz or pA/√Hz.
E_NI
1/2
R_S
I_BN
OPA2607
E_O
OUTPUT CURRENT AND VOLTAGE
The OPA2607 provides outstanding output voltage and
current capabilities. Under no-load conditions at 25°C, the
output voltage typically swings within 0.8V of either supply
rail; the guaranteed swing limit is within 1.1V of either rail.
Into a 5Ω load (the minimum tested load), it is guaranteed to
deliver more than ±180mA.
E_RS
÷4kTR_S
4kT
R_G
R_F
÷4kTR_F
R_G
I_BI
4kT = 1.6 x 10^–20J
at 290°K
FIGURE 5. Op Amp Noise Analysis Model.
DISTORTION PERFORMANCE
The OPA2607 provides good distortion performance into a
100Ω load on ±12V supplies. Relative to alternative solutions,
it provides exceptional performance into lighter loads.
Increasing the load impedance improves distortion directly.
Remember that the total load includes the feedback network—
in the non-inverting configuration (Figure 1) this is the sum of
R_F + R_G, while in the inverting configuration it is just R_F.
Also, providing an additional supply decoupling capacitor
(0.1µF) between the supply pins (for bipolar operation) im-
proves the 2nd-order distortion slightly (3 to 6dB).
In most op amps, increasing the output voltage swing in-
creases harmonic distortion directly. As the typical perfor-
mance curves show, the spurious intermodulation powers do
not increase as predicted by a traditional intercept model.
As the fundamental power level increases, the dynamic
range does not decrease significantly. For 2 tones centered
at 1MHz, with 10dBm/tone into a matched 50Ω load
(i.e. 2Vp-p for each tone at the load, which requires 8Vp-p
for the overall 2-tone envelope at the output pin), the typical
performance curves show 85dBc difference between the
test-tone power and the 3rd-order intermodulation spurious
levels.
The total output spot noise voltage can be computed as the
square root of the sum of all squared output noise voltage
contributors. Equation 4 shows the general form for the
output noise voltage using the terms shown in Figure 5.
(4)
( ) ^{( ) ( ) E}O ⁼
E
2
NI
+
I _BN R _S
2 _{+ 4kTRS}
NG² +
I _BI R _F
2 _{+ 4kTRFNG}
Dividing this expression by the noise gain (NG = (1+R_F/R_G))
will give the equivalent input referred spot noise voltage at the
non-inverting input as shown in Equation 5.
(5)
( ) E_N =
E_NI2 +
I _BN R _S
2
+
4kTR_S
+


I _BI R _F
NG


2
+
4 kTR _F
NG
Evaluating these two equations for the OPA2607 circuit and
component values shown in Figure 1 will give a total output
spot noise voltage of 27nV/√Hz and a total equivalent input
spot noise voltage of 3.4nV/√Hz. This total input referred
spot noise voltage is higher than the 1.7nV/√Hz specifica-
®
9
OPA2607