AD844
UNDERSTANDING THE AD844
The AD844 can be used in ways similar to a conventional op
amp while providing performance advantages in wideband ap-
plications. However, there are important differences in the inter-
nal structure which need to be understood in order to optimize
the performance of the AD844 op amp.
Open Loop Behavior
The closed loop transresistance is simply the parallel sum of R1
and Rt. Since R1 will generally be in the range 500
Ω
to 2 kΩ
and Rt is about 3 MΩ the closed loop transresistance will be
only 0.02% to 0.07% lower than R1. This small error will often
be less than the resistor tolerance.
When R1 is fairly large (above 5 kΩ) but still much less than
Rt, the closed loop HF response is dominated by the time con-
stant R1Ct. Under such conditions the AD844 is over-damped
and will provide only a fraction of its bandwidth potential. Be-
cause of the absence of slew rate limitations under these condi-
tions, the circuit will exhibit a simple single pole response even
under large signal conditions.
In Figure 26, R3 is used to properly terminate the input if de-
sired. R3 in parallel with R2 gives the terminated resistance. As
R1 is lowered, the signal bandwidth increases, but the time
constant R1Ct becomes comparable to higher order poles in the
closed loop response. Therefore, the closed loop response be-
comes complex, and the pulse response shows overshoot. When
R2 is much larger than the input resistance, R
IN
, at Pin 2, most
of the feedback current in R1 is delivered to this input; but as
R2 becomes comparable to R
IN
, less of the feedback is absorbed
at Pin 2, resulting in a more heavily damped response. Conse-
quently, for low values of R2 it is possible to lower R1 without
causing instability in the closed loop response. Table I lists
combinations of R1 and R2 and the resulting frequency re-
sponse for the circuit of Figure 26. Figure 13 shows the very
clean and fast
±
10 V pulse response of the AD844.
Figure 25 shows a current feedback amplifier reduced to essen-
tials. Sources of fixed dc errors such as the inverting node bias
current and the offset voltage are excluded from this model and
are discussed later. The most important parameter limiting the
dc gain is the transresistance, Rt, which is ideally infinite. A fi-
nite value of Rt is analogous to the finite open loop voltage gain
in a conventional op amp.
The current applied to the inverting input node is replicated by
the current conveyor so as to flow in resistor Rt. The voltage
developed across Rt is buffered by the unity gain voltage follower.
Voltage gain is the ratio Rt/ R
IN
. With typical values of Rt = 3 MΩ
and R
IN
= 50
Ω,
the voltage gain is about 60,000. The open loop
current gain is another measure of gain and is determined by the
beta product of the transistors in the voltage follower stage (see
Figure 28); it is typically 40,000.
Figure 25. Equivalent Schematic
The important parameters defining ac behavior are the trans-
capacitance, Ct, and the external feedback resistor (not shown).
The time constant formed by these components is analogous to
the dominant pole of a conventional op amp, and thus cannot
be reduced below a critical value if the closed loop system is to
be stable. In practice, Ct is held to as low a value as possible
(typically 4.5 pF) so that the feedback resistor can be maximized
while maintaining a fast response. The finite R
IN
also affects the
closed loop response in some applications as will be shown.
The open loop ac gain is also best understood in terms of the
transimpedance rather than as an open loop voltage gain. The
open loop pole is formed by Rt in parallel with Ct. Since Ct is
typically 4.5 pF, the open loop corner frequency occurs at about
12 kHz. However, this parameter is of little value in determining
the closed loop response.
Response as an Inverting Amplifier
Figure 26. Inverting Amplifier
Table I.
Gain
R1
R2
BW (MHz)
GBW (MHz)
Figure 26 shows the connections for an inverting amplifier. Un-
like a conventional amplifier the transient response and the
small signal bandwidth are determined primarily by the value of
the external feedback resistor, R1, rather than by the ratio of
R1/R2 as is customarily the case in an op amp application. This
is a direct result of the low impedance at the inverting input. As
with conventional op amps, the closed loop gain is –R1/R2.
–1
–1
–2
–2
–5
–5
–10
–10
–20
–100
+100
1 kΩ
500
Ω
2 kΩ
1 kΩ
5 kΩ
500
Ω
1 kΩ
500
Ω
1 kΩ
5 kΩ
5 kΩ
1 kΩ
500
Ω
1 kΩ
500
Ω
1 kΩ
100
Ω
100
Ω
50
Ω
50
Ω
50
Ω
50
Ω
35
60
15
30
5.2
49
23
33
21
3.2
9
35
60
30
60
26
245
230
330
420
320
900
REV. C
–7–
AD [ ANALOG DEVICES ]
