LMH6504MM [NSC]

Wideband, Low Power, Variable Gain Amplifier; 宽带,低功耗,可变增益放大器
LMH6504MM
型号: LMH6504MM
厂家: National Semiconductor    National Semiconductor
描述:

Wideband, Low Power, Variable Gain Amplifier
宽带,低功耗,可变增益放大器

模拟IC 信号电路 放大器 光电二极管
文件: 总19页 (文件大小:755K)
中文:  中文翻译
下载:  下载PDF数据表文档文件
June 2004  
LMH6504  
Wideband, Low Power, Variable Gain Amplifier  
General Description  
Features  
The LMH 6504 is a wideband DC coupled voltage con-  
VS = 5V, TA = 25˚C, RF = 1 K, RG = 100, RL = 100, AV  
trolled gain stage followed by a high-speed current feedback  
Op Amp which can directly drive a low impedance load. Gain  
adjustment range is 80 dB for up to 10 MHz by varying the  
gain control input voltage, VG.  
= AVMAX = 9.7V/V, Typical values unless specified.  
n −3 dB BW  
n Gain control BW  
n Adjustment range ( 10 MHz)  
150 MHz  
150 MHz  
80 dB  
<
Maximum gain is set by external components, and the gain  
can be reduced all the way to cut-off. Power consumption is  
110 mW with a speed of 150 MHz and a gain control band-  
width (BW) of 150 MHz. Output referred DC offset voltage is  
less than 55 mV over the entire gain control voltage range.  
Device-to-device gain matching is within 0.42 dB at maxi-  
mum gain. Furthermore, gain is tested and guaranteed over  
a wide range. The output current feedback Op Amp allows  
n Output offset voltage  
n Gain matching (limit)  
55 mV  
0.42 dB  
7V to 12V  
1500 V/µs  
11 mA  
60 mA  
2.2V  
4.4 nV/  
2.6 pA/  
−45dBc  
n Supply voltage range  
n Slew rate (inverting)  
n Supply Current (no load)  
n Linear Output Current  
n Output Voltage Swing  
n Input Noise Voltage  
n Input Noise Current  
n THD (20 MHz, RL = 100, VO = 2 VPP  
n Replacement for CLC5523  
>
high frequency large signals (Slew Rate  
1500 V/µs) and  
can also drive a heavy load current (60 mA). Near ideal input  
characteristics (i.e. low input bias current, low offset, low pin  
3 resistance) enable the device to be easily configured as an  
inverting amplifier as well (see Application Information sec-  
tion for details).  
)
Applications  
n Variable attenuator  
n AGC  
n Voltage controlled filter  
n Video imaging processing  
To provide ease of use when working with a single supply,  
VG range is set to be from 0V to +2V relative to the ground  
pin potential (pin 4). VG input impedance is high in order to  
ease drive requirement. In single supply operation, the  
ground pin is tied to a "virtual" half supply.  
LMH6504 gain control is linear in dB for a large portion of the  
total gain control range. This makes the device suitable for  
AGC applications. For linear gain control applications, see  
LMH6503 data sheet.  
The combination of minimal external components and small  
outline packages (SO8 and MSOP8) allows the LMH6504 to  
be used in space-constrained applications.  
Typical Application  
20084302  
AVMAX = 9.7 V/V  
20084311  
Gain vs. VG  
LMH is a trademark of National Semiconductor Corporation.  
© 2004 National Semiconductor Corporation  
DS200843  
www.national.com  
Absolute Maximum Ratings (Note 1)  
If Military/Aerospace specified devices are required,  
please contact the National Semiconductor Sales Office/  
Distributors for availability and specifications.  
Junction Temperature  
150˚C  
Soldering Information:  
Infrared or Convection (20 sec)  
Wave Soldering (10 sec)  
235˚C  
260˚C  
ESD Tolerance (Note 4):  
Human Body  
1000V  
100V  
Operating Ratings (Note 1)  
Supply Voltages (V+ - V)  
Machine Model  
7V to 12V  
Input Current  
10 mA  
Operating Temperature Range  
−40˚C to +85˚C  
Output Current  
120 mA (Note 3)  
Thermal Resistance:  
8 -Pin SOIC  
(θJC  
)
(θJA)  
Supply Voltages (V+ - V)  
Voltage at Input/ Output pins  
Storage Temperature Range  
12.6V  
60  
165  
235  
V+ +0.8V, V−0.8V  
−65˚C to 150˚C  
8-Pin MSOP  
65  
Electrical Characteristics(Note 2)  
Unless otherwise specified, all limits guaranteed for TJ = 25˚C, VS  
=
5V, AVMAX = 9.7 V/V, RF = 1k, RG = 100, VIN  
=
0.1V, RL = 100, VG = +2V. Boldface limits apply at the temperature extremes.  
Min  
(Note 6)  
Typ  
(Note 6)  
Max  
(Note 6)  
Symbol  
Parameter  
Conditions  
Units  
Frequency Domain Response  
<
BW  
GF  
-3dB Bandwidth  
Gain Flatness  
VOUT 1 VPP  
150  
58  
MHz  
MHz  
<
VOUT 4 VPP, AVMAX = 100  
<
VOUT 1 VPP  
40  
0.9V VG 2V, 0.2 dB  
<
Att Range Flat Band (Relative to Max Gain)  
Attenuation Range (Note 13)  
0.2 dB Flatness, f 30 MHz  
26  
9.5  
150  
dB  
<
0.1 dB Flatness, f 30 MHz  
BW  
Gain control Bandwidth  
VG = 1V (Note 12)  
MHz  
Control  
CT (dB)  
Feed-through  
VG = 0V, 30 MHz  
(Output/Input)  
−53  
dB  
dB  
<
<
GR  
Gain Adjustment Range  
f
f
10 MHz  
30 MHz  
80  
73  
Time Domain Response  
tr, tf  
Rise and Fall Time  
0.5V Step  
2.1  
20  
ns  
%
OS %  
SR  
Overshoot  
Slew Rate (Note 5)  
4V Step, Non Inverting  
4V Step, Inverting  
800  
1500  
V/µs  
Distortion & Noise Performance  
HD2  
HD3  
THD  
En tot  
IN  
2nd Harmonic Distortion  
3rd Harmonic Distortion  
Total Harmonic Distortion  
Total Equivalent Input Noise  
Input Noise Current  
2VPP, 20 MHz  
−47  
–55  
−45  
4.4  
dBc  
>
>
f
f
1 MHz, RSOURCE = 50Ω  
nV/  
pA/  
1 MHz  
2.6  
DG  
Differential Gain  
f = 4.43 MHz, RL = 100Ω  
0.45  
0.13  
%
DP  
Differential Phase  
deg  
www.national.com  
2
Electrical Characteristics(Note 2) (Continued)  
Unless otherwise specified, all limits guaranteed for TJ = 25˚C, VS  
=
5V, AVMAX = 9.7 V/V, RF = 1k, RG = 100, VIN  
=
0.1V, RL = 100, VG = +2V. Boldface limits apply at the temperature extremes.  
Min  
Typ  
Max  
Symbol  
Parameter  
Conditions  
(Note 6)  
(Note 6)  
(Note 6)  
Units  
DC & Miscellaneous Performance  
GACCU  
G Match  
K
Gain Accuracy  
VG = 2.0V  
0
0.33  
0.45  
3.9  
dB  
dB  
(See Application Note)  
<
<
0.8V VG 2V  
Gain Matching  
VG = 2.0V  
0.42  
(See Application Note  
<
<
0.8V VG 2V  
+2.8/−4.2  
1.01  
Gain Multiplier  
0.920  
0.965  
V/V  
(See Application Notes)  
Input Voltage Range  
0.916  
1.02  
V
IN NL  
IN L  
RG Open  
3.2  
V
V
RG = 100Ω  
0.48  
0.40  
4.8  
0.68  
I
RG Current  
Pin 3  
6.8  
mA  
RG_MAX  
4.0  
IBIAS  
Bias Current  
Pin 2 (Note 7)  
−1.4  
−3.5  
µA  
−3.7  
TC IBIAS  
RIN  
Bias Current Drift  
Input Resistance  
Input Capacitance  
VG Bias Current  
Pin 2 (Note 8)  
Pin 2  
–190  
7
pA/˚C  
MΩ  
pF  
CIN  
Pin 2  
2.8  
0.9  
10  
IVG  
Pin 1, VG = 2V (Note 7)  
Pin 1 (Note 8)  
Pin 1  
µA  
TC IVG  
VG Bias Drift  
pA/˚C  
MΩ  
pF  
R
VG Input Resistance  
VG Input Capacitance  
Output Voltage Range  
25  
VG  
C
Pin 1  
2.8  
2.2  
VG  
VOUT  
L
RL = 100Ω  
2.0  
1.7  
V
VOUT NL  
ROUT  
RL = Open  
DC  
3.1  
0.12  
80  
Output Impedance  
Output Current  
IOUT  
VOUT  
=
4V from Rails  
60  
mA  
40  
<
<
VO  
Output Offset Voltage  
0V VG 2V  
10  
–76  
–88  
11  
55  
mV  
dB  
OFFSET  
70  
+PSRR  
−PSRR  
IS  
+Power Supply Rejection Ratio  
(Note 9)  
Input Referred, 1V change,  
VG = 2.2V  
–65  
–65  
−Power Supply Rejection Ratio  
(Note 9)  
Input Referred, 1V change,  
VG = 2.2V  
dB  
Supply Current  
No Load  
8.5  
15  
mA  
6.5  
16  
3
www.national.com  
Electrical Characteristics(Note 2) (Continued)  
Note 1: Absolute Maximum Ratings indicate limits beyond which damage to the device may occur. Operating Ratings indicate conditions for which the device is  
intended to be functional, but specific performance is not guaranteed. For guaranteed specifications, see the Electrical Characteristics.  
Note 2: Electrical Table values apply only for factory testing conditions at the temperature indicated. Factory testing conditions result in very limited self-heating of  
>
the device such that T = T . No guarantee of parametric performance is indicated in the electrical tables under conditions of internal self-heating where T  
T .  
J
A
J
A
Note 3: The maximum output current (I  
) is determined by device power dissipation limitations or value specified, whichever is lower.  
OUT  
Note 4: Human body model, 1.5 kin series with 100 pF. Machine Model, 0in series with 200 pF  
Note 5: Slew rate is the average of the rising and falling slew rates.  
Note 6: Typical values represent the most likely parametric norm. Bold numbers refer to over temperature limits.  
Note 7: Positive current corresponds to current flowing into the device.  
Note 8: Drift determined by dividing the change in parameter distribution at temperature extremes by the total temperature change.  
+
Note 9: +PSRR definition: [|V  
/V | / A ], −PSRR definition: [|V  
/V | / A ] with 0.1V input voltage. V  
is the change in output voltage with offset shift  
OUT  
OUT  
V
OUT  
V
subtracted out.  
Note 10: Gain/Phase normalized to low frequency value at 25˚C.  
Note 11: Gain/Phase normalized to low frequency value at each setting.  
Note 12: Gain control frequency response schematic:  
20084316  
Note 13: Flat Band Attenuation (Relative To Max Gain) Range Definition: Specified as the attenuation range from maximum which allows gain flatness specified  
<
(either 0.2dB or 0.1dB), relative to A  
gain. For example, for f 30 MHz, here are the Flat Band Attenuation ranges:  
VMAX  
0.2 dB: 19.7 dB down to -6.3 dB = 26 dB range  
0.1 dB: 19.7 dB down to 10.2 dB = 9.5 dB range  
Connection Diagram  
8-Pin SOIC  
20084301  
Top View  
Ordering Information  
Package  
Part Number  
LMH6504MA  
LMH6504MAX  
LMH6504MM  
LMH6504MMX  
Package Marking  
Transport Media  
NSC Drawing  
8-Pin SOIC  
LMH6504MA  
95 Units/Rail  
M08A  
2.5k Units Tape and Reel  
1k Units Tape and Reel  
3.5k Units Tape and Reel  
8-Pin MSOP  
A93A  
MUA08A  
www.national.com  
4
Typical Performance Characteristics Unless otherwise specified: VS  
=
5V, TA = 25˚C,  
VG = VGMAX, RF = 1 k, RG = 100, VIN = 0.1V, input terminated in 50. RL = 100, Typical values.  
Frequency Response Over Temperature Frequency Response for Various VG  
20084303  
20084304  
Frequency Response (AVMAX = 2)  
Inverting Frequency Response  
20084344  
20084346  
Frequency Response for Various VG (AVMAX = 100)  
(Large Signal)  
Frequency Response for Various Amplitudes  
20084364  
20084345  
5
www.national.com  
Typical Performance Characteristics Unless otherwise specified: VS  
=
5V, TA = 25˚C,  
VG = VGMAX, RF = 1 k, RG = 100, VIN = 0.1V, input terminated in 50. RL = 100, Typical values. (Continued)  
Gain Control Frequency Response  
IS vs. VS  
20084333  
20084321  
IS vs. VS  
Input Bias Current vs. VS  
20084322  
20084320  
PSRR  
AVMAX vs. Supply Voltage  
20084334  
20084323  
www.national.com  
6
Typical Performance Characteristics Unless otherwise specified: VS  
=
5V, TA = 25˚C,  
VG = VGMAX, RF = 1 k, RG = 100, VIN = 0.1V, input terminated in 50. RL = 100, Typical values. (Continued)  
Feed through Isolation for Various AVMAX  
Gain Variation Over entire Temp Range vs. VG  
20084341  
20084312  
IRG vs. VIN  
Gain vs. VG  
20084318  
20084311  
#
Gain vs. VG Including Limits  
Output Offset Voltage vs. VG (Typical Unit 1)  
20084363  
20084325  
7
www.national.com  
Typical Performance Characteristics Unless otherwise specified: VS  
=
5V, TA = 25˚C,  
VG = VGMAX, RF = 1 k, RG = 100, VIN = 0.1V, input terminated in 50. RL = 100, Typical values. (Continued)  
#
#
Output Offset Voltage vs. VG (Typical Unit 2)  
Output Offset Voltage vs. VG (Typical Unit 3)  
20084330  
20084328  
Distribution of Output Offset Voltage  
Output Noise Density vs. Frequency  
20084308  
20084361  
Output Noise Density vs. Frequency  
Output Noise Density vs. Frequency  
20084337  
20084338  
www.national.com  
8
Typical Performance Characteristics Unless otherwise specified: VS  
=
5V, TA = 25˚C,  
VG = VGMAX, RF = 1 k, RG = 100, VIN = 0.1V, input terminated in 50. RL = 100, Typical values. (Continued)  
Input Referred Noise Density vs. Frequency  
Output Voltage vs. Output Current (Sinking)  
20084365  
20084336  
Output Voltage vs. Output Current (Sourcing)  
Distortion vs. Frequency  
20084331  
20084342  
HD vs. POUT  
THD vs. POUT  
20084309  
20084343  
9
www.national.com  
Typical Performance Characteristics Unless otherwise specified: VS  
=
5V, TA = 25˚C,  
VG = VGMAX, RF = 1 k, RG = 100, VIN = 0.1V, input terminated in 50. RL = 100, Typical values. (Continued)  
THD vs. POUT  
THD vs. Gain  
20084310  
20084339  
THD vs. Gain  
Differential Gain & Phase  
20084340  
20084335  
VG Bias Current vs. VG  
Output Impedance  
20084362  
20084314  
www.national.com  
10  
Typical Performance Characteristics Unless otherwise specified: VS  
=
5V, TA = 25˚C,  
VG = VGMAX, RF = 1 k, RG = 100, VIN = 0.1V, input terminated in 50. RL = 100, Typical values. (Continued)  
Step Response Plot  
Step Response Plot  
20084315  
20084317  
Gain vs. VG Step  
20084332  
11  
www.national.com  
The amount of current which the input buffer can source/sink  
into RG is limited and is specified in the IRG_MAX spec. This  
sets the maximum input voltage:  
Application Information  
GENERAL DESCRIPTION  
The key features of the LMH6504 are:  
Low power  
Broad voltage controlled gain and attenuation range  
(From AVMAX down to complete cutoff)  
Eq. 2  
As the IRG_MAX limit is approached (with increasing input  
voltage or with lowering of RG), the device harmonic distor-  
tion will increase. Changes in RF will have a dramatic effect  
on the small signal bandwidth. The output amplifier of the  
LMH6504 is a current feedback amplifier (CFA) and its band-  
width is determined by RF. As with any CFA, doubling the  
feedback resistor will roughly cut the bandwidth of the device  
in half. For more about CFA’s, see the basic tutorial, OA-20,  
“Current Feedback Myths Debunked”, or a more rigorous  
analysis, OA-13, “Current Feedback Amplifier Loop Gain  
Analysis and Performance Enhancements”.  
Bandwidth independent, resistor programmable gain  
range (RG)  
Broad signal and gain control bandwidths  
Frequency response may be adjusted with RF  
High impedance signal and gain control inputs  
Refer to Figure 1 below. The LMH6504 combines a closed  
loop input buffer (“X1” Block), a voltage controlled variable  
gain cell (“MULT” Block) and an output amplifier (“CFA”  
Block). The input buffer is a transconductance stage whose  
gain is set by the gain setting resistor, RG. The output  
amplifier is a current feedback op amp and is configured as  
a transimpedance stage whose gain is set by, and is equal  
to, the feedback resistor, RF. The maximum gain, AVMAX, of  
the LMH6504 is defined by the ratio: K · RF / RG where “K” is  
the gain multiplier with a nominal value of 0.965. As the gain  
control input (VG) changes over its 0 to 2V range, the gain is  
adjusted over a range of about 80 dB relative to the maxi-  
mum set gain.  
OTHER CONFIGURATIONS  
1) Single Supply Operation  
The LMH6504 can be configured for use in a single supply  
environment. Doing so requires the following:  
a) Bias pin 4 and RG to a “virtual half supply” somewhere  
close to the middle of V+ and V- range. The other end of  
RG is tied to pin 3. The “virtual half supply” needs to be  
capable of sinking and sourcing the expected current flow  
through RG.  
b) Ensure that VG can be adjusted from 0V to 2V above the  
“virtual half supply”.  
c) Bias the input (pin 2) to make sure that it stays within the  
range of 1.8V above V- to 1.8V below V+ (see “Input  
voltage Range” specification in the Electrical Character-  
istics table). This can be accomplished by either DC  
biasing the input and AC coupling the input signal, or  
alternatively, by direct coupling if the output of the driving  
stage is also biased to half supply.  
Arranged this way, the LMH6504 will respond to the current  
flowing through RG. The gain control relationship will be  
similar to the split supply arrangement with VG measured  
referenced to pin 4. Keep in mind that the circuit described  
above will also center the output voltage to the “virtual half  
supply voltage”.  
20084347  
2) Arbitrarily Referenced Input Signal  
FIGURE 1. LMH6504 Typical Application and Block  
Diagram  
Having a wide input voltage range on the input (pin 2)  
(+/-3.2V typical), the LMH6504 can be configured to control  
the gain on signals which are not referenced to ground (e.g.  
Half Supply biased circuits, etc.). We will call this node the  
“reference node”. In such cases, the other end of RG (the  
side not tied to pin 3) can be tied to this reference node so  
that RG will “look at” the difference between the signal and  
this reference only. Keep in mind that the reference node  
needs to source and sink the current flowing through RG.  
SETTING THE LMH6504 MAXIMUM GAIN  
Eq. 1  
GAIN ACCURACY  
Although the LMH6504 is specified at AVMAX = 9.7V/V, the  
recommended AVMAX varies between 2 and 100. Higher  
gains are possible but usually impractical due to output  
offsets, noise and distortion. When varying AVMAX several  
tradeoffs are made:  
Gain accuracy is defined as the actual gain compared  
against the theoretical gain at a certain VG (results ex-  
pressed in dB) (See Figure 2).  
Theoretical gain is given by:  
RG: determines the input voltage range  
RF: determines overall bandwidth  
www.national.com  
12  
GAIN PARTITIONING  
Application Information (Continued)  
If high levels of gain are needed, gain partitioning should be  
considered:  
Eq. 3  
@
Where K = 0.965 (nominal) N = 0.96V & VC = 80mV room  
temperature  
For a VG range, the value specified in the tables represents  
the worst case accuracy over the entire range. The "Typical"  
value would be the worst case difference between the "Typi-  
cal gain" and the "Theoretical gain". The "Max" value would  
be the worst case difference between the actual gain and the  
"Theoretical gain" for the entire population.  
20084352  
GAIN MATCHING  
Gain matching as the limit on gain variation at a certain VG  
(expressed in dB) (see Figure 2) and is specified as "Max"  
only (no "Typical"). For a VG range, the value specified  
represents the worst case matching over the entire range.  
The "Max" value would be the worst case difference between  
the actual gain and the typical gain for the entire population.  
FIGURE 3. Gain Partitioning  
The maximum gain range for this circuit is given by the  
following equation:  
Eq. 4  
The LMH6624 is a low noise wideband voltage feedback  
amplifier. Setting R2 at 909and R1 at 100produces a  
gain of 20 dB. Setting RF at 1000as recommended and RG  
at 50, produces a gain of about 26 dB in the LMH6504. The  
total gain of this circuit is therefore approximately 46 dB. It is  
important to understand that when partitioning to obtain high  
levels of gain, very small signal levels will drive the amplifiers  
to full scale output. For example, with 46 dB of gain, a 20 mV  
signal at the input will drive the output of the LMH6624 to 200  
mV, the output of the LMH6504 to 4V. Accordingly, the  
designer must carefully consider the contributions of each  
stage to the overall characteristics. Through gain partitioning  
the designer is provided with an opportunity to optimize the  
frequency response, noise, distortion, settling time, and  
loading effects of each amplifier to achieve improved overall  
performance.  
LMH6504 GAIN CONTROL RANGE AND MINIMUM GAIN  
20084351  
Before discussing Gain Control Range, it is important to  
understand the issues which limit it. The minimum gain of the  
LMH6504, theoretically, is zero, but in practical circuits is  
limited by the amount of feedthrough, here defined as the  
gain when VG = 0V. Capacitive coupling through the board  
and package as well as coupling through the supplies will  
determine the amount of feedthrough. Even at DC, the input  
signal will not be completely rejected. At high frequencies  
feedthrough will get worse because of its capacitive nature.  
At frequencies below 10 MHz, the feed through will be less  
FIGURE 2. LMH6504 Gain Accuracy & Gain Matching  
Defined  
than −60 dB and therefore, it can be said that with AVMAX  
20 dB, the gain control range is 80 dB.  
=
13  
www.national.com  
improvement comes about because in the non-inverting con-  
figuration, the slew rate of the overall amplifier is limited by  
the input buffer. In the inverting circuit, the input buffer re-  
mains at a fixed voltage and does not affect slew rate.  
Application Information (Continued)  
LMH6504 GAIN CONTROL FUNCTION  
In the plot, Gain vs. VG, we can see the gain as a function of  
the control voltage. The “Gain (V/V)” plot, sometimes re-  
ferred to as the S-curve, is the linear (V/V ) gain. This is a  
hyperbolic tangent relationship and is given by Equation 3.  
The “Gain (dB)” plots the gain in dB and is linear over a wide  
range of gains. Because of this, the LMH6504 gain control is  
referred to as “linear-in-dB.”  
TRANSMISSION LINE MATCHING  
One method for matching the characteristic impedance of a  
transmission line is to place the appropriate resistor at the  
input or output of the amplifier. Figure 5 shows a typical  
circuit configuration for matching transmission lines.  
For applications where the LMH6504 will be used at the  
heart of a closed loop AGC circuit, the S-curve control char-  
acteristic provides a broad linear (in dB) control range with  
soft limiting at the highest gains where large changes in  
control voltage result in small changes in gain. For applica-  
tions requiring a fully linear (in dB) control characteristic, use  
the LMH6504 at half gain and below (VG 1V).  
AVOIDING OVERDRIVE OF THE LMH6504 GAIN  
CONTROL INPUT  
20084356  
There is an additional requirement for the LMH6504 Gain  
Control Input (VG): VG must not exceed +2.3V (with 5V  
supplies). The gain control circuitry may saturate and the  
gain may actually be reduced. In applications where VG is  
being driven from a DAC, this can easily be addressed in the  
software. If there is a linear loop driving VG, such as an AGC  
loop, other methods of limiting the input voltage should be  
implemented. One simple solution is to place a 2.2:1 resis-  
tive divider on the VG input. If the device driving this divider  
is operating off of 5V supplies as well, its output will not  
exceed 5V and through the divider VG can not exceed 2.3V.  
FIGURE 5. TRANSMISSION LINE MATCHING  
The resistors RS, RI, RO, and RT are equal to the character-  
istic impedance, ZO, of the transmission line or cable. Use  
CO to match the output transmission line over a greater  
frequency range. It compensates for the increase of the op  
amp’s output impedance with frequency.  
MINIMIZING PARASITIC EFFECTS ON SMALL SIGNAL  
BANDWIDTH  
The best way to minimize parasitic effects is to use surface  
mount components and to minimize lead lengths and com-  
ponent distance from the LMH6504. For designs utilizing  
through-hole components, specifically axial resistors, resis-  
tor self-capacitance should be considered. Example: the  
average magnitude of parasitic capacitance of RN55D 1%  
metal film resistors is about 0.15 pF with variations of as  
much as 0.1 pF between lots. Given the LMH6504’s ex-  
tended bandwidth, these small parasitic reactance variations  
can cause measurable frequency response variations in the  
highest octave. We therefore recommend the use of surface  
mount resistors to minimize these parasitic reactance ef-  
fects.  
IMPROVING THE LMH6504 LARGE SIGNAL  
PERFORMANCE  
Figure  
4
illustrates an inverting gain scheme for the  
LMH6504.  
RECOMMENDATIONS  
Here are some recommendations to avoid problems and to  
get the best performance:  
Do not place a capacitor across RF. However, an appro-  
priately chosen series RC combination could be used to  
shape the frequency response.  
20084354  
Keep traces connecting RF separated and as short as  
possible  
FIGURE 4. Inverting Amplifier  
Place a small resistor (20-50) between the output and  
CL  
The input signal is applied through the RG resistor. The VIN  
pin should be grounded through a 25resistor. The maxi-  
mum gain range of this configuration is given in the following  
equation:  
Cut away the ground plane, if any, under RG  
Keep decoupling capacitors as close as possible to the  
LMH6504.  
Connect pin 2 through a minimum resistance of 25.  
ADJUSTING OFFSETS AND DC LEVEL SHIFTING  
Offsets can be broken into two parts: an input-referred term  
and an output-referred term. These errors can be trimmed  
using the circuit in Figure 6. First set VG to 0V and adjust the  
Eq. 5  
The inverting slew rate of the LMH6504 is much higher than  
that of the non-inverting slew rate. This 2X performance  
www.national.com  
14  
Application Information (Continued)  
USING THE LMH6504 IN AGC APPLICATIONS  
trim pot R4 to null the offset voltage at the output. This will  
eliminate the output stage offsets. Next set VG to 2V and  
adjust the trim pot R1 to null the offset voltage at the output.  
This will eliminate the input stage offsets.  
In AGC applications, the control loop forces the LMH6504 to  
have a fixed output amplitude. The input amplitude will vary  
over a wide range and this can be the issue that limits  
dynamic range. At high input amplitudes, the distortion due  
to the input buffer driving RG may exceed that which is  
produced by the output amplifier driving the load. In the plot,  
Distortion vs. Gain, total harmonic distortion (THD) is plotted  
over a gain range of nearly 35 dB for a fixed output amplitude  
of 0.25 VPP in the specified configuration, RF = 1k, RG  
=
100. When the gain is adjusted to -15 dB (i.e. 35 dB down  
from AVMAX), the input amplitude would be 1.41 VPP and we  
can see the distortion is at its worst at this gain. If the output  
amplitude of the AGC were to be raised above 0.25 VPP, the  
input amplitudes for gains 40 dB down from AVMAX would be  
even higher and the distortion would degrade further. It is for  
this reason that we recommend lower output amplitudes if  
wide gain ranges are desired. Using a post-amp like the  
LMH6714/ 6720/ 6722 family or LMH6702 would be the best  
way to preserve dynamic range and yield output amplitudes  
much higher than 100 mVPP. Another way of addressing  
distortion performance and its limitations on dynamic range,  
would be to raise the value of RG. Just like any other high-  
speed amplifier, by increasing the load resistance, and  
therefore decreasing the demanded load current, the distor-  
tion performance will be improved in most cases. With an  
increased RG, RF will also have to be increased to keep the  
same AVMAX and this will decrease the overall bandwidth. It  
may be possible to insert a series RC combination across RF  
in order to counteract the negative effect on BW when a  
large RF is used.  
20084357  
FIGURE 6. OFFSET ADJUST CIRCUIT  
DIGITAL GAIN CONTROL  
Digitally variable gain control can be easily realized by driv-  
ing the LMH6504’s gain control input with a digital-to-analog  
converter (DAC). Figure 7 illustrates such an application.  
This circuit employs National Semiconductor’s eight-bit  
DAC0830, the LMC8101 MOS input op-amp (Rail-to-Rail  
Input/Output), and the LMH6504 VGA. With VREF set to 2V,  
the circuit provides up to 80 dB of gain control in 256 steps  
with up to 0.05% full scale resolution. The maximum gain of  
this circuit is 20 dB.  
AUTOMATIC GAIN CONTROL (AGC) #1  
Fast Response AGC Loop  
The AGC circuit shown in Figure 8 will correct a 6 dB input  
amplitude step in 100 ns. The circuit includes a two op-amp  
precision rectifier amplitude detector (U1 and U2), and an  
integrator (U3) to provide high loop gain at low frequencies.  
The output amplitude is set by R9. Some notes on building  
fast AGC loops: Precision rectifiers work best with large  
output signals. Accuracy is improved by blocking DC offsets,  
as shown in Figure 8.  
20084358  
FIGURE 7. Digital Gain Control  
15  
www.national.com  
Application Information (Continued)  
20084359  
FIGURE 8. Automatic Gain Control Circuit #1  
Signal frequencies must not reach the gain control port of the  
LMH6504, or the output signal will be distorted (modulated  
by itself). A fast settling AGC needs additional filtering be-  
yond the integrator stage to block signal frequencies. This is  
provided in Figure 8 by a simple R-C filter (R10 and C3);  
better distortion performance can be achieved with a more  
complex filter. These filters should be scaled with the input  
signal frequency. Loops with slower response time (longer  
integration time constants) may not need the R10 – C3 filter.  
U2 is configured to provide negative feedback. U2 generates  
a rectified gain control signal that works against an adjust-  
able bias level which may be set by the potentiometer and  
RB. CI integrates the bias and negative feedback. The result-  
ant gain control signal is applied to the U1 gain control input  
VG. The bias adjustment allows the U1 output to be set at an  
arbitrary level less than the maximum output specification of  
the amplifier. Rectification is accomplished in U2 by driving  
both the amplifier input and the gain control input with the U1  
output signal. The voltage divider that is formed by R1 and  
R2, sets the rectifier gain.  
Checking the loop stability can be done by monitoring the VG  
voltage while applying a step change in input signal ampli-  
tude. Changing the input signal amplitude can be easily  
done with an arbitrary waveform generator.  
AUTOMATIC GAIN CONTROL (AGC) #2  
Figure 9 illustrates an automatic gain control circuit that  
employs two LMH6504’s. In this circuit, U1 receives the input  
signal and produces an output signal of constant amplitude.  
www.national.com  
16  
Application Information (Continued)  
20084360  
FIGURE 9. Automatic Gain Control Circuit #2  
CIRCUIT LAYOUT CONSIDERATIONS & EVALUATION  
BOARD  
the output (e.g. 100and 39 pF in series tied between the  
LMH6504 output and ground). CL can also be isolated from  
the output by placing a small resistor in series with the output  
(pin 6).  
A good high frequency PCB layout including ground plane  
construction and power supply bypassing close to the pack-  
age are critical to achieving full performance. The amplifier is  
sensitive to stray capacitance to ground at the I- input (pin 7);  
keep node trace area small. Shunt capacitance across the  
feedback resistor should not be used to compensate for this  
effect. Capacitance to ground should be minimized by re-  
moving the ground plane from under the body of RG. Para-  
sitic or load capacitance directly on the output (pin 6) de-  
grades phase margin leading to frequency response  
peaking.  
Component parasitics also influence high frequency results.  
Therefore it is recommended to use metal film resistors such  
as RN55D or leadless components such as surface mount  
devices. High profile sockets are not recommended.  
National Semiconductor suggests the following evaluation  
board as a guide for high frequency layout and as an aid in  
device testing and characterization:  
Device  
Package  
Evaluation Board  
Part Number  
CLC730066  
The LMH6504 is fully stable when driving a 100load. With  
reduced load (e.g. 1k.) there is a possibility of instability at  
very high frequencies beyond 400 MHz especially with a  
capacitive load. When the LMH6504 is connected to a light  
load as such, it is recommended to add a snubber network to  
LMH6504  
SOIC  
The evaluation board is shipped when a device sample  
request is placed with National Semiconductor.  
17  
www.national.com  
Physical Dimensions inches (millimeters) unless otherwise noted  
8-Pin SOIC  
NS Package Number M08A  
8-Pin MSOP  
NS Package Number MUA08A  
www.national.com  
18  
Notes  
LIFE SUPPORT POLICY  
NATIONAL’S PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT  
DEVICES OR SYSTEMS WITHOUT THE EXPRESS WRITTEN APPROVAL OF THE PRESIDENT AND GENERAL  
COUNSEL OF NATIONAL SEMICONDUCTOR CORPORATION. As used herein:  
1. Life support devices or systems are devices or  
systems which, (a) are intended for surgical implant  
into the body, or (b) support or sustain life, and  
whose failure to perform when properly used in  
accordance with instructions for use provided in the  
labeling, can be reasonably expected to result in a  
significant injury to the user.  
2. A critical component is any component of a life  
support device or system whose failure to perform  
can be reasonably expected to cause the failure of  
the life support device or system, or to affect its  
safety or effectiveness.  
BANNED SUBSTANCE COMPLIANCE  
National Semiconductor certifies that the products and packing materials meet the provisions of the Customer Products  
Stewardship Specification (CSP-9-111C2) and the Banned Substances and Materials of Interest Specification  
(CSP-9-111S2) and contain no ‘‘Banned Substances’’ as defined in CSP-9-111S2.  
National Semiconductor  
Americas Customer  
Support Center  
National Semiconductor  
Europe Customer Support Center  
Fax: +49 (0) 180-530 85 86  
National Semiconductor  
Asia Pacific Customer  
Support Center  
National Semiconductor  
Japan Customer Support Center  
Fax: 81-3-5639-7507  
Email: new.feedback@nsc.com  
Tel: 1-800-272-9959  
Email: europe.support@nsc.com  
Deutsch Tel: +49 (0) 69 9508 6208  
English Tel: +44 (0) 870 24 0 2171  
Français Tel: +33 (0) 1 41 91 8790  
Email: ap.support@nsc.com  
Email: jpn.feedback@nsc.com  
Tel: 81-3-5639-7560  
www.national.com  
National does not assume any responsibility for use of any circuitry described, no circuit patent licenses are implied and National reserves the right at any time without notice to change said circuitry and specifications.  

相关型号:

LMH6504MM/NOPB

IC SPECIALTY ANALOG CIRCUIT, PDSO8, MSOP-8, Analog IC:Other
NSC

LMH6504MM/NOPB

LMH6504 Wideband, Low Power, Variable Gain Amplifier
TI

LMH6504MMX

Wideband, Low Power, Variable Gain Amplifier
NSC

LMH6504MMX

SPECIALTY ANALOG CIRCUIT, PDSO8, MSOP-8
TI

LMH6504MMX/NOPB

Wideband, Low Power, Variable Gain Amplifier 8-VSSOP -40 to 85
TI

LMH6505

Wideband, Low Power, Linear-in-dB, Variable Gain Amplifier
NSC

LMH6505

Wideband, Low Power, Linear-in-dB, Variable Gain Amplifier
INTERSIL

LMH6505

宽带、低功耗、dB 线性可变增益放大器
TI

LMH6505MA

Wideband, Low Power, Linear-in-dB, Variable Gain Amplifier
NSC

LMH6505MA

Wideband, Low Power, Linear-in-dB, Variable Gain Amplifier
INTERSIL

LMH6505MA/NOPB

宽带、低功耗、dB 线性可变增益放大器 | D | 8 | -40 to 85
TI

LMH6505MAX

Wideband, Low Power, Linear-in-dB, Variable Gain Amplifier
NSC