ADL5505ACBZ-P7 [ADI]

450 MHz to 6000Mhz TruPwr Detector; 450 MHz至6000Mhz TruPwr检测器
ADL5505ACBZ-P7
型号: ADL5505ACBZ-P7
厂家: ADI    ADI
描述:

450 MHz to 6000Mhz TruPwr Detector
450 MHz至6000Mhz TruPwr检测器

电源电路 电源管理电路 信息通信管理 PC
文件: 总20页 (文件大小:1458K)
中文:  中文翻译
下载:  下载PDF数据表文档文件
450 MHz to 6000 MHz  
TruPwr Detector  
ADL5505  
FEATURES  
FUNCTIONAL BLOCK DIAGRAM  
VPOS  
True rms response detector  
Excellent temperature stability  
ADL5505  
INTERNAL  
FILTERING  
0.2ꢀ dB rms detection accuracy vs. temperature  
Over 3ꢀ dB input power dynamic range, inclusive of crest factor  
RF bandwidths from 4ꢀ0 MHz to 6000 MHz  
ꢀ00 Ω input impedance  
100  
RFIN  
RMS CORE  
VRMS  
BUFFER  
Single-supply operation: 2.ꢀ V to 3.3 V  
Low power: 1.8 mA at 3.0 V supply  
RoHS-compliant part  
COMM  
Figure 1.  
APPLICATIONS  
10  
Power measurement of W-CDMA, CDMA2000, QPSK-/QAM-  
based OFDM (LTE and WiMAX), and other complex  
modulation waveforms  
RF transmitter or receiver power measurement  
1
0.1  
0.01  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
INPUT (dBm)  
Figure 2. Output vs. Input Level, Supply = 3.0 V, Frequency = 1900 MHz  
GENERAL DESCRIPTION  
The ADL5505 is a TruPwr™ mean-responding (true rms) power  
detector for use in high frequency receiver and transmitter signal  
chains from 450 MHz to 6000 MHz. Requiring only a single  
supply between 2.5 V and 3.3 V, the detector draws less than  
1.8 mA. The input is internally ac-coupled and has a nominal  
input impedance of 500 Ω. The rms output is a linear-responding  
dc voltage with a conversion gain of 1.86 V/V rms at 900 MHz.  
The on-chip modulation filter provides adequate averaging for  
most waveforms. An on-chip, 100 Ω series resistance at the  
output, combined with an external shunt capacitor, creates a low-  
pass filter response that reduces the residual ripple in the dc output  
voltage.  
The ADL5505 offers excellent temperature stability across a  
30 dB range and near 0 dB measurement error across temperature  
over the top portion of the dynamic range. In addition to its  
temperature stability, the ADL5505 offers low process variations  
that further reduce calibration complexity.  
The ADL5505 is a highly accurate, easy to use means of  
determining the rms of complex waveforms. It can be used for  
power measurements of both simple and complex waveforms  
but is particularly useful for measuring high crest factor (high  
peak-to-rms ratio) signals, such as W-CDMA, CDMA2000,  
WiMAX, WLAN, and LTE waveforms.  
The power detector operates from −40°C to +85°C and is  
available in an 4-ball, 0.8 mm × 0.8 mm wafer-level chip scale  
package. It is fabricated on a high fT silicon BiCMOS process.  
Rev. 0  
Information furnished by Analog Devices is believed to be accurate and reliable. However, no  
responsibility is assumed by Analog Devices for its use, nor for any infringements of patents or other  
rights of third parties that may result from its use. Specifications subject to change without notice. No  
license is granted by implication or otherwise under any patent or patent rights of Analog Devices.  
Trademarks and registeredtrademarks arethe property of their respective owners.  
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.  
Tel: 781.329.4700  
Fax: 781.461.3113  
www.analog.com  
©2010 Analog Devices, Inc. All rights reserved.  
 
 
ADL5505  
TABLE OF CONTENTS  
Features .............................................................................................. 1  
Basic Connections...................................................................... 14  
RF Input Interfacing................................................................... 14  
Linearity....................................................................................... 15  
Output Drive Capability and Buffering................................... 16  
Selecting the Output Low-Pass Filter ...................................... 16  
Power Consumption .................................................................. 17  
Device Calibration and Error Calculation.............................. 17  
Calibration for Improved Accuracy......................................... 18  
Drift Over a Reduced Temperature Range ............................. 18  
Device Handling......................................................................... 18  
Land Pattern and Soldering Information................................ 18  
Evaluation Board........................................................................ 18  
Outline Dimensions....................................................................... 20  
Ordering Guide .......................................................................... 20  
Applications....................................................................................... 1  
Functional Block Diagram .............................................................. 1  
General Description......................................................................... 1  
Revision History ............................................................................... 2  
Specifications..................................................................................... 3  
Absolute Maximum Ratings............................................................ 6  
ESD Caution.................................................................................. 6  
Pin Configuration and Function Descriptions............................. 7  
Typical Performance Characteristics ............................................. 8  
Circuit Description......................................................................... 13  
RMS Circuit Description and Filtering ................................... 13  
Filtering........................................................................................ 13  
Output Buffer.............................................................................. 13  
Applications Information .............................................................. 14  
REVISION HISTORY  
4/10—Revision 0: Initial Version  
Rev. 0 | Page 2 of 20  
 
ADL5505  
SPECIFICATIONS  
TA = 25°C, VS = 3.0 V, COUT = open, light condition ≤ 600 lux, 75 Ω input termination resistor, unless otherwise noted.  
Table 1.  
Parameter  
Test Conditions  
Input RFIN  
Min Typ  
Max  
Unit  
FREQUENCY RANGE  
RF INPUT (f = 450 MHz)  
Input Impedance  
RMS Conversion  
450  
6000 MHz  
Input RFIN to output VRMS  
No termination  
510||1.01  
Ω||pF  
Dynamic Range1  
Continuous wave (CW) input, −40°C < TA < +85°C  
Delta from 25°C  
0.25 dꢀ Error2  
25  
16  
36  
40  
dꢀ  
dꢀ  
dꢀ  
dꢀ  
dꢀm  
dꢀm  
V/V rms  
V
V
V
0.25 dꢀ Error3  
1 dꢀ Error3  
2 dꢀ Error3  
Maximum Input Level  
Minimum Input Level  
Conversion Gain  
Output Intercept4  
Output Voltage, High Input Power  
Output Voltage, Low Input Power  
Temperature Sensitivity  
0.25 dꢀ error3  
1 dꢀ error3  
VRMS = (gain × VIN) + intercept  
15  
−22  
1.88  
0.008  
0.755  
0.082  
PIN = 5 dꢀm, 400 mV rms  
PIN = −15 dꢀm, 40 mV rms  
PIN = 0 dꢀm  
+25°C < TA < +85°C  
−40°C < TA < +25°C  
Input RFIN to output VRMS  
No termination  
0.0027  
0.0024  
dꢀ/°C  
dꢀ/°C  
RF INPUT (f = 900 MHz)  
Input Impedance  
370||0.80  
Ω||pF  
RMS Conversion  
Dynamic Range1  
Continuous wave (CW) input, −40°C < TA < +85°C  
Delta from 25°C  
0.25 dꢀ Error2  
26  
dꢀ  
0.25 dꢀ Error3  
17  
dꢀ  
1 dꢀ Error3  
36  
dꢀ  
2 dꢀ Error3  
40  
dꢀ  
Maximum Input Level  
Minimum Input Level  
Conversion Gain  
0.25 dꢀ error3  
1 dꢀ error3  
VRMS = (gain × VIN) + intercept  
15  
−23  
1.86  
dꢀm  
dꢀm  
V/V rms  
V
1.6  
2.2  
+0.1  
Output Intercept4  
Output Voltage, High Input Power  
Output Voltage, Low Input Power  
Temperature Sensitivity  
−0.1 +0.009  
0.748  
PIN = 5 dꢀm, 400 mV rms  
PIN = −15 dꢀm, 40 mV rms  
PIN = 0 dꢀm  
V
0.083  
V
+25°C < TA < +85°C  
−40°C < TA < +25°C  
0.0026  
0.0024  
dꢀ/°C  
dꢀ/°C  
Rev. 0 | Page 3 of 20  
 
ADL5505  
Parameter  
Test Conditions  
Min Typ  
270||0.67  
Max  
Unit  
RF INPUT (f = 1900 MHz)  
Input Impedance  
RMS Conversion  
Input RFIN to output VRMS  
No termination  
Ω||pF  
Dynamic Range1  
Continuous wave (CW) input, −40°C < TA < +85°C  
Delta from 25°C  
0.25 dꢀ Error2  
21  
16  
36  
40  
dꢀ  
dꢀ  
dꢀ  
dꢀ  
dꢀm  
dꢀm  
V/V rms  
V
V
V
0.25 dꢀ Error3  
1 dꢀ Error3  
2 dꢀ Error3  
Maximum Input Level  
Minimum Input Level  
Conversion Gain  
Output Intercept4  
Output Voltage, High Input Power  
Output Voltage, Low Input Power  
Temperature Sensitivity  
0.25 dꢀ error3  
1 dꢀ error3  
VRMS = (gain × VIN) + intercept  
15  
−22  
1.82  
0.007  
0.727  
0.079  
PIN = 5 dꢀm, 400 mV rms  
PIN = −15 dꢀm, 40 mV rms  
PIN = 0 dꢀm  
+25°C < TA < +85°C  
−40°C < TA < +25°C  
Input RFIN to output VRMS  
No termination  
0.0017  
−0.0026  
dꢀ/°C  
dꢀ/°C  
RF INPUT (f = 2600 MHz)  
Input Impedance  
240||0.58  
Ω||pF  
RMS Conversion  
Dynamic Range1  
Continuous wave (CW) input, −40°C < TA < +85°C  
Delta from 25°C  
0.25 dꢀ Error2  
14  
dꢀ  
0.25 dꢀ Error3  
11  
dꢀ  
1 dꢀ Error3  
35  
dꢀ  
2 dꢀ Error3  
40  
dꢀ  
Maximum Input Level  
Minimum Input Level  
Conversion Gain  
0.25 dꢀ error3  
1 dꢀ error3  
VRMS = (gain × VIN) + intercept  
15  
dꢀm  
dꢀm  
V/V rms  
V
V
V
−22  
1.77  
0.005  
0.700  
0.075  
Output Intercept4  
Output Voltage, High Input Power  
Output Voltage, Low Input Power  
Temperature Sensitivity  
PIN = 5 dꢀm, 400 mV rms  
PIN = −15 dꢀm, 40 mV rms  
PIN = 0 dꢀm  
+25°C < TA < +85°C  
−40°C < TA < +25°C  
Input RFIN to output VRMS  
No termination  
0.0016  
0.0042  
dꢀ/°C  
dꢀ/°C  
RF INPUT (f = 3500 MHz)  
Input Impedance  
210||0.48  
Ω||pF  
RMS Conversion  
Dynamic Range1  
Continuous wave (CW) input, −40°C < TA < +85°C  
Delta from 25°C  
0.25 dꢀ Error2  
5
dꢀ  
0.25 dꢀ Error3  
5
dꢀ  
1 dꢀ Error3  
33  
dꢀ  
2 dꢀ Error3  
39  
dꢀ  
Maximum Input Level  
Minimum Input Level  
Conversion Gain  
0.25 dꢀ error3  
1 dꢀ error3  
VRMS = (gain × VIN) + intercept  
13  
dꢀm  
dꢀm  
V/V rms  
V
V
V
−21  
1.61  
0.001  
0.630  
0.065  
Output Intercept4  
Output Voltage, High Input Power  
Output Voltage, Low Input Power  
Temperature Sensitivity  
PIN = 5 dꢀm, 400 mV rms  
PIN = −15 dꢀm, 40 mV rms  
PIN = 0 dꢀm  
+25°C < TA < +85°C  
−40°C < TA < +25°C  
0.0046  
0.0085  
dꢀ/°C  
dꢀ/°C  
Rev. 0 | Page 4 of 20  
ADL5505  
Parameter  
Test Conditions  
Min Typ  
80||0.42  
Max  
Unit  
RF INPUT (f = 6000 MHz)  
Input Impedance  
Input RFIN to output VRMS  
No termination  
Ω||pF  
RMS Conversion  
Dynamic Range1  
Continuous wave (CW) input, −40°C < TA < +85°C  
1 dꢀ Error3  
23  
dꢀ  
2 dꢀ Error3  
33  
dꢀ  
Maximum Input Level  
Minimum Input Level  
Conversion Gain  
0.25 dꢀ error3  
1 dꢀ error3  
VRMS = (gain × VIN) + intercept  
11  
dꢀm  
dꢀm  
V/V rms  
V
V
V
−17  
0.77  
0.002  
0.298  
0.032  
Output Intercept4  
Output Voltage, High Input Power  
Output Voltage, Low Input Power  
Temperature Sensitivity  
PIN = 5 dꢀm, 400 mV rms  
PIN = −15 dꢀm, 40 mV rms  
PIN = 0 dꢀm  
+25°C < TA < +85°C  
−40°C < TA < +25°C  
Pin VRMS  
0.0103  
0.0138  
dꢀ/°C  
dꢀ/°C  
VRMS OUTPUT  
Output Offset  
No signal at RFIN  
VS = 3.0 V, RLOAD ≥ 10 kΩ  
10  
2.5  
3
3
3
100  
3.3  
mV  
V
mA  
μs  
Maximum Output Voltage  
Available Output Current  
Pulse Response Time  
Power-Up Response Time5  
POWER SUPPLIES  
COUT = open, 10 dꢀ step, 10% to 90% of settling level  
COUT = open, 0 dꢀm at RFIN  
μs  
Operating Range  
Quiescent Current6  
−40°C < TA < +85°C  
No signal at RFIN  
2.5  
V
mA  
1.8  
1 The available output swing and, therefore, the dynamic range are altered by the supply voltage; see Figure 8.  
2 Error referred to delta from 25°C response; see Figure 13, Figure 14, Figure 15, Figure 19, Figure 20, and Figure 21.  
3 Error referred to best-fit line at 25°C; see Figure 10, Figure 11, Figure 12, Figure 16, Figure 17, and Figure 18.  
4 Calculated using linear regression.  
5 The response time is measured from 10% to 90% of settling level; see Figure 30 and Figure 31.  
6 Supply current is input level-dependent; see Figure 27.  
Rev. 0 | Page 5 of 20  
ADL5505  
ABSOLUTE MAXIMUM RATINGS  
Table 2.  
Stresses above those listed under Absolute Maximum Ratings  
may cause permanent damage to the device. This is a stress  
rating only; functional operation of the device at these or any  
other conditions above those indicated in the operational  
section of this specification is not implied. Exposure to absolute  
maximum rating conditions for extended periods may affect  
device reliability.  
Parameter  
Rating  
Supply Voltage, VS  
3.5 V  
VRMS  
0 V to VS  
RFIN  
1.25 V rms  
15 dꢀm  
150 mW  
260°C/W  
125°C  
−40°C to +85°C  
−65°C to +150°C  
Equivalent Power, Referred to 50 Ω  
Internal Power Dissipation  
θJA (WLCSP)  
Maximum Junction Temperature  
Operating Temperature Range  
Storage Temperature Range  
ESD CAUTION  
Rev. 0 | Page 6 of 20  
 
 
ADL5505  
PIN CONFIGURATION AND FUNCTION DESCRIPTIONS  
VPOS  
RFIN  
1
2
4
3
VRMS  
COMM  
ADL5505  
TOP VIEW  
(BALL SIDE DOWN)  
Not to Scale  
Figure 3. Pin Configuration  
Table 3. Pin Function Descriptions  
Pin No. Mnemonic Description  
1
2
3
4
VPOS  
RFIN  
COMM  
VRMS  
Supply Voltage. The operational range is 2.5 V to 3.3 V.  
Signal Input. This pin is internally ac-coupled after internal termination resistance. The nominal input impedance is 500 Ω.  
Device Ground.  
RMS Output. This pin is a rail-to-rail voltage output with limited current drive capability. The output has an internal  
100 Ω series resistance. High resistive loads and low capacitance loads are recommended to preserve output swing  
and allow fast response.  
Rev. 0 | Page 7 of 20  
 
ADL5505  
TYPICAL PERFORMANCE CHARACTERISTICS  
TA = 25°C; VS = 3.0 V; COUT = open; light condition ≤ 600 lux; 75 Ω input termination resistor; colors: black = +25°C, blue = −40°C,  
red = +85°C; unless otherwise noted.  
10  
3
450MHz  
450MHz  
900MHz  
900MHz  
1900MHz  
2600MHz  
3500MHz  
5000MHz  
6000MHz  
1900MHz  
2600MHz  
3500MHz  
5000MHz  
6000MHz  
2
1
1
0
0.1  
–1  
–2  
–3  
0.01  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
INPUT (dBm)  
INPUT (dBm)  
Figure 4. Output vs. Input Level; Frequencies = 450 MHz, 900 MHz, 1900 MHz,  
2600 MHz, 3500 MHz, 5000 MHz, 6000 MHz; Supply = 3.0 V  
Figure 7. Linearity Error vs. Input Level; Frequencies = 450 MHz, 900 MHz,  
1900 MHz, 2600 MHz, 3500 MHz, 5000 MHz, 6000 MHz; Supply = 3.0 V  
2.0  
10  
450MHz  
2.5V  
900MHz  
2.7V  
3.0V  
3.3V  
1.8  
1900MHz  
2600MHz  
3500MHz  
5000MHz  
6000MHz  
1.6  
1.4  
1.2  
1.0  
0.8  
0.6  
0.4  
0.2  
0
1
0.1  
0.01  
–25  
0
0.1  
0.2  
0.3  
0.4  
0.5  
0.6  
0.7  
0.8  
0.9  
1.0  
–20  
–15  
–10  
–5  
0
5
10  
15  
INPUT (V rms)  
INPUT (dBm)  
Figure 8. Output vs. Input Level; Supplies =  
2.5 V, 2.7 V, 3.0 V, and 3.3 V; Frequency = 900 MHz  
Figure 5. Output vs. Input Level (Linear Scale); Frequencies = 450 MHz, 900 MHz,  
1900 MHz, 2600 MHz, 3500 MHz, 5000 MHz, 6000 MHz;Supply= 3.0V  
2.5  
2.0  
1.5  
1.0  
0.5  
0
100  
80  
60  
40  
20  
0
700  
600  
500  
400  
300  
200  
100  
0
1.4  
1.2  
1.0  
0.8  
0.6  
0.4  
0.2  
0
SHUNT CAPACITANCE  
SHUNT RESISTANCE  
0
1
2
3
4
5
6
0.5  
1.0  
1.5  
2.0  
2.5  
3.0  
FREQUENCY (GHz)  
FREQUENCY (GHz)  
Figure 6. Conversion Gain and Intercept vs. Frequency; Supply = 3.0 V;  
Temperatures = −40°C, +25°C, and +85°C  
Figure 9. Input Impedance vs. Frequency; Supply = 3.0 V  
Rev. 0 | Page 8 of 20  
 
 
 
 
 
ADL5505  
3
2
3
2
1
1
0
0
–1  
–2  
–3  
–1  
–2  
–3  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
15  
15  
INPUT (dBm)  
INPUT (dBm)  
Figure 10. Output Temperature Drift from +25°C Linear Reference  
for 50 Devices at −40°C, +25°C, and +85°C; Frequency = 450 MHz  
Figure 13. Output Delta from +25°C Output Voltage for  
50 Devices at −40°C and +85°C; Frequency = 450 MHz  
3
3
2
2
1
1
0
0
–1  
–2  
–3  
–1  
–2  
–3  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
–25  
–20  
–15  
–10  
–5  
0
5
10  
INPUT (dBm)  
INPUT (dBm)  
Figure 11. Output Temperature Drift from +25°C Linear Reference  
for 50 Devices at −40°C, +25°C, and +85°C; Frequency = 900 MHz  
Figure 14. Output Delta from +25°C Output Voltage for  
50 Devices at −40°C and +85°C; Frequency = 900 MHz  
3
3
2
2
1
1
0
0
–1  
–2  
–3  
–1  
–2  
–3  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
–25  
–20  
–15  
–10  
–5  
0
5
10  
INPUT (dBm)  
INPUT (dBm)  
Figure 12. Output Temperature Drift from +25°C Linear Reference  
for 50 Devices at −40°C, +25°C, and +85°C; Frequency = 1900 MHz  
Figure 15. Output Delta from +25°C Output Voltage for  
50 Devices at −40°C and +85°C; Frequency = 1900 MHz  
Rev. 0 | Page 9 of 20  
 
 
 
 
ADL5505  
3
3
2
2
1
1
0
0
–1  
–2  
–1  
–2  
–3  
–3  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
15  
15  
INPUT (dBm)  
INPUT (dBm)  
Figure 16. Output Temperature Drift from +25°C Linear Reference  
for 50 Devices at −40°C, +25°C, and +85°C; Frequency = 2600 MHz  
Figure 19. Output Delta from +25°C Output Voltage for  
50 Devices at −40°C and +85°C; Frequency = 2600 MHz  
3
3
2
2
1
1
0
0
–1  
–2  
–3  
–1  
–2  
–3  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
–25  
–20  
–15  
–10  
–5  
0
5
10  
INPUT (dBm)  
INPUT (dBm)  
Figure 17. Output Temperature Drift from +25°C Linear Reference  
for 50 Devices at −40°C, +25°C, and +85°C; Frequency = 3500 MHz  
Figure 20. Output Delta from +25°C Output Voltage for  
50 Devices at −40°C and +85°C; Frequency = 3500 MHz  
3
2
3
2
1
1
0
0
–1  
–2  
–3  
–1  
–2  
–3  
–25  
–20  
–15  
–10  
–5  
0
5
10  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
INPUT (dBm)  
INPUT (dBm)  
Figure 18. Output Temperature Drift from +25°C Linear Reference  
for 50 Devices at −40°C, +25°C, and +85°C; Frequency = 6000 MHz  
Figure 21. Output Delta from +25°C Output Voltage for  
50 Devices at −40°C and +85°C; Frequency = 6000 MHz  
Rev. 0 | Page 10 of 20  
 
 
 
ADL5505  
3
2
3
2
CW  
CW  
PICH, 4.7dB  
12.2kbps, DPCCH (–5.46dB, 15kSPS) + DPDCH (0dB, 60kSPS), 3.4dB CF  
144kbps, DPCCH (–11.48dB, 15kSPS) + DPDCH (0dB, 480kSPS), 3.3dB CF  
768kbps, DPCCH (–11.48dB, 15kSPS) + DPDCH1 + 2 (0dB, 960kSPS), 5.8dB CF  
DPCCH (–6.02dB, 15kSPS) + DPDCH (–4.08dB, 60kSPS) +  
HS-DPCCH (0dB, 15kSPS), 4.91dB CF  
PICH + FCH (9.6kbps), 4.8dB CF  
PICH + FCH (9.6kbps) + DCCH, 6.3dB CF  
PICH + FCH (9.6kbps) + DCCH + SCH (153.6kbps), 7.6dB CF  
PICH + FCH (9.6kbps) + SCH (153.6kbps), 6.7dB  
DPCCH (–6.02dB,15kSPS) + DPDCH (–11.48dB, 60kSPS) +  
HS-DPCCH (0dB, 15kSPS), 5.34dB CF  
DPCCH (–6.02dB, 15kSPS) + HS-DPCCH (0dB, 15kSPS), 5.44dB CF  
1
1
0
0
–1  
–2  
–3  
–1  
–2  
–3  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
INPUT (dBm)  
INPUT (dBm)  
Figure 25. Error from CW Linear Reference vs. Input with Various  
CDMA2000 Reverse Link Waveforms at 1900 MHz, COUT = Open  
Figure 22. Error from CW Linear Reference vs. Input with Various  
W-CDMA Reverse Link Waveforms at 900 MHz, COUT = Open  
3
3
CW  
CW  
16QAM RB1  
16QAM RB10  
TEST MODEL 1 WITH 16 DPCH, 1 CARRIER  
TEST MODEL 1 WITH 32 DPCH, 1 CARRIER  
2
1
16QAM RB100  
QPSK RB1  
QPSK RB10  
QPSK RB100  
2
1
TEST MODEL 1 WITH 64 DPCH, 1 CARRIER  
TEST MODEL 1 WITH 64 DPCH, 2 CARRIERS  
TEST MODEL 1 WITH 64 DPCH, 3 CARRIERS  
TEST MODEL 1 WITH 64 DPCH, 4 CARRIERS  
0
0
–1  
–2  
–3  
–1  
–2  
–3  
–25  
–20  
–15  
–10  
–5  
0
5
10  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
INPUT (dBm)  
INPUT (dBm)  
Figure 26. Error from CW Linear Reference vs. Input with Various  
LTE Reverse Link Waveforms at 2600 MHz, COUT = Open  
Figure 23. Error from CW Linear Reference vs. Input with Various  
W-CDMA Forward Link Waveforms at 2200 MHz, COUT = Open  
15  
14  
13  
12  
11  
10  
9
3
CW  
BPSK, 11dB CF  
QPSK, 11dB CF  
2
1
16QAM, 12dB CF  
64QAM, 11dB CF  
8
0
7
6
2.5V 3.3V 3.0V  
COLD  
ROOM  
HOT  
–1  
–2  
–3  
5
4
3
2
1
0
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
0
0.1  
0.2  
0.3  
0.4  
0.5  
0.6  
0.7  
0.8  
0.9  
1.0  
INPUT (V rms)  
INPUT (dBm)  
Figure 24. Error from CW Linear Reference vs. Input with Various  
802.16 OFDM Waveforms at 3500 MHz, 10 MHz Signal BW, and  
256 Subcarriers for All Modulated Signals, COUT = Open  
Figure 27. Supply Current vs. Input Level; Supplies = 2.5 V, 3.0 V, and 3.3 V;  
Frequency = 900 MHz; Temperatures = −40°C, +25°C, and +85°C  
Rev. 0 | Page 11 of 20  
 
 
ADL5505  
PULSED RFIN  
PULSED VPOS  
400mV rms RF INPUT  
400mV rms RF INPUT  
250mV rms  
160mV rms  
70mV rms  
250mV rms  
160mV rms  
70mV rms  
VRMS  
VRMS  
4μs/DIV  
4μs/DIV  
Figure 28. Output Response to Various RF Input Pulse Levels; Supply = 3.0 V;  
Frequency = 900 MHz; COUT = Open  
Figure 30. Output Response to Supply Gating at Various RF Input Levels;  
Supply = 3.0 V; Frequency = 900 MHz; COUT = Open  
PULSED RFIN  
PULSED VPOS  
400mV rms RF INPUT  
400mV rms RF INPUT  
250mV rms  
160mV rms  
250mV rms  
160mV rms  
70mV rms  
70mV rms  
VRMS  
VRMS  
10μs/DIV  
10μs/DIV  
Figure 29. Output Response to Various RF Input Pulse Levels; Supply = 3.0 V;  
Frequency = 900 MHz; COUT = 100 nF  
Figure 31. Output Response to Supply Gating at Various RF Input Levels;  
Supply = 3.0 V; Frequency = 900 MHz; COUT = 100 nF  
Rev. 0 | Page 12 of 20  
 
 
ADL5505  
CIRCUIT DESCRIPTION  
The ADL5505 employs two-stage detection. The critical aspect  
of this technical approach is the concept of first stripping the  
carrier to reveal the envelope and then performing the required  
analog computation of rms.  
FILTERING  
An important aspect of rms-dc conversion is the need for  
averaging (the function is root-mean-square). The on-chip  
averaging in the square domain has a corner frequency of  
approximately 180 kHz. and is sufficient for common modu-  
lation signals, such as CDMA-, CDMA2000-, W-CDMA-, and  
QPSK-/QAM-based OFDM (for example, LTE, WLAN, and  
WiMAX).  
RMS CIRCUIT DESCRIPTION AND FILTERING  
The rms processing is executed using a proprietary translinear  
technique. This method is a mathematically accurate rms  
computing approach and allows for achieving unprecedented  
rms accuracies for complex modulation signals irrespective of  
the crest factor of the input signal. An integrating filter capaci-  
tor performs the square-domain averaging. The VRMS output  
can be expressed as  
Adequate filtering ensures the accuracy of the rms measure-  
ment; however, some ripple or ac residual can still be present on  
the dc output. To reduce this ripple, an external shunt capacitor  
can be used at the output to form a low-pass filter with the on-  
chip, 100 ꢀ resistance (see the Selecting the Output Low-Pass  
Filter section).  
T2  
VI2N ×dt  
T1  
VRMS = A ×  
OUTPUT BUFFER  
T2 T1  
A buffer takes the internal rms signal and amplifies it accor-  
dingly before it is output on the VRMS pin. The output stage  
of the rms buffer is a common source PMOS with a resistive  
load to provide a rail-to-rail output. The buffer has a 100 Ω  
on-chip series resistance on the output, allowing for easy low-  
pass filtering.  
where A is a scaling parameter that is determined by the on-chip  
resistor ratio.  
There are no other scaling parameters involved in this computa-  
tion, which means that the rms output is inherently free from  
any sources of error due to temperature, supply, and process  
variations.  
Rev. 0 | Page 13 of 20  
 
 
 
 
ADL5505  
APPLICATIONS INFORMATION  
BASIC CONNECTIONS  
Resistive Tap RF Input  
Figure 34 shows a technique for coupling the input signal into  
the ADL5505 that can be applicable when the input signal is  
much larger than the input range of the ADL5505. A series  
resistor combines with the input impedance of the ADL5505  
to attenuate the input signal. Because this series resistor forms  
a divider with the frequency-dependent input impedance, the  
apparent gain changes greatly with frequency. However, this  
method has the advantage of very little power being tapped off  
in RF power transmission applications. If the resistor is large  
compared with the impedance of the transmission line, the  
VSWR of the system is relatively unaffected.  
Figure 32 shows the basic connections for the ADL5505. The  
device is powered by a single supply between 2.5 V and 3.3 V,  
with a quiescent current of 1.8 mA. The VPOS pin is decoupled  
using 100 pF and 0.1 μF capacitors.  
Placing a single 75 ꢀ resistor at the RF input provides a  
broadband match of 50 Ω. More precise resistive or reactive  
matches can be applied for narrow frequency band use (see  
the RF Input Interfacing section).  
The ac residual can be reduced further by increasing the output  
capacitance, COUT. The combination of the internal 100 Ω output  
resistance and COUT produces a low-pass filter to reduce output  
ripple of the VRMS output (see the Selecting the Output Low-  
Pass Filter section for more details).  
RF TRANSMISSION LINE  
+V = 2.5V TO 3.3V  
S
RFIN  
VPOS  
VRMS  
1
4
VRMS  
R
SERIES  
0.1µF  
100pF  
RFIN  
R
C
OUT  
OUT  
ADL5505  
ADL5505  
Figure 34. Attenuating the Input Signal  
2
RFIN  
COMM  
3
R10  
75ꢀ  
The resistive tap or series resistance, RSERIES, can be expressed as  
R
SERIES = RIN (1 − 10ATTN/20)/(10ATTN/20  
where:  
IN is the input impedance of RFIN.  
ATTN is the desired attenuation factor in decibels.  
)
(1)  
Figure 32. Basic Connections for ADL5505  
RF INPUT INTERFACING  
R
The input impedance of the ADL5505 decreases with increasing  
frequency in both its resistive and capacitive components (see  
Figure 9). The resistive component varies from 370 ꢀ at 900 MHz  
to about 245 ꢀ at 2600 MHz.  
For example, if a power amplifier with a maximum output power  
of 28 dBm is matched to the ADL5505 input at 5 dBm, then a  
−23 dB attenuation factor is required. At 900 MHz, the input  
resistance, RIN, is 370 Ω.  
A number of options exist for input matching. For operation  
at multiple frequencies, a 75 ꢀ shunt to ground, as shown in  
Figure 33, provides the best overall match. For use at a single  
frequency, a resistive or a reactive match can be used. By plotting  
the input impedance on a Smith Chart, the best value for a  
resistive match can be calculated. (Both input impedance and  
input capacitance can vary by up to 20ꢁ around their nominal  
values.) Where VSWR is critical, the match can be improved  
with a series inductor placed before the shunt component.  
RF TRANSMISSION LINE  
R
SERIES = (370 Ω)(1 − 10−23/20)/(10−23/20) = 4856 Ω  
(2)  
Thus, for an attenuation of −23 dB, a series resistance of approx-  
imately 4.87 kΩ (the nearest available standard resistor value)  
is needed.  
DIRECTIONAL  
COUPLER  
50  
ATTN  
RFIN  
75ꢀ  
ADL5505  
Figure 33. Input Interfacing to Directional Coupler  
Rev. 0 | Page 14 of 20  
 
 
 
 
 
 
 
 
ADL5505  
Multiple RF Inputs  
Output Swing  
Figure 35 shows a technique for combining multiple RF input  
signals to the ADL5505. Some applications can share a single  
detector for multiple bands. Three 16.5 Ω resistors in a T network  
combine the three 50 Ω terminations (including the ADL5505  
with the shunt 75 Ω matching component). The broadband resis-  
tive combiner ensures that each port of the T network sees a  
50 Ω termination. Because there are only 6 dB of isolation from  
one port of the combiner to the other ports, only one band  
should be active at a time.  
At 900 MHz, the VRMS output voltage is nominally 1.86 × the  
input rms voltage (a conversion gain of 1.86 V/V rms). The output  
voltage swings from near ground to 2.4 V on a 3.0 V supply.  
Figure 8 shows the output swings of the ADL5505 to a CW input  
for various supply voltages. Only at the lowest supply voltages  
(2.5 V and 2.7 V) is there a reduction in the dynamic range as  
the input headroom decreases.  
Output Offset  
The ADL5505 has a 1 dB error detection range of about 30 dB,  
as shown in Figure 10 to Figure 12 and Figure 16 to Figure 18.  
The error is referred to the best-fit line defined in the linear region  
of the output response (see the Device Calibration and Error  
Calculation section for more details). Below an input power of  
−16 dBm, the response is no longer linear and begins to lose  
accuracy. In addition, depending on the supply voltage, saturation  
may limit the detection accuracy above +14 dBm. Choose cali-  
bration points in the linear region, avoiding the nonlinear  
ranges at the high and low extremes.  
BAND 1  
DIRECTIONAL  
COUPLER  
50ꢀ  
16.5ꢀ  
16.5ꢀ  
BAND 2  
RFIN  
DIRECTIONAL  
COUPLER  
50ꢀ  
75ꢀ  
16.5ꢀ  
ADL5505  
Figure 35. Combining Multiple RF Input Signals  
Figure 36 shows a distribution of the output response vs. the  
input for multiple devices. The ADL5505 loses accuracy at low  
input powers as the output response begins to fan out. As the  
input power is reduced, the spread of the output response  
increases along with the error.  
LINEARITY  
Because the ADL5505 is a linear responding device, plots of output  
voltage vs. input voltage result in a straight line (see Figure 4  
and Figure 5) and the dynamic range in decibels (dB) is not  
clearly visible. It is more useful to plot the error on a logarith-  
mic scale, as shown in Figure 7. The deviation of the plot from  
the ideal straight line characteristic is caused by input stage  
clipping at the high end and by signal offsets at the low end.  
However, offsets at the low end can be either positive or  
negative; therefore, the linearity error vs. input level plots (see  
Figure 7) can also trend upwards at the low end. Figure 10 to  
Figure 12 and Figure 16 to Figure 18 show error distributions  
for a large population of devices at specific frequencies over  
temperature.  
10  
1
0.1  
0.01  
0.001  
It is also apparent in Figure 7 that the error at the lower portion  
of the dynamic range tends to shift up as frequency is increased.  
This is due to the calibration points chosen: −14 dBm and +8 dBm  
(see the Device Calibration and Error Calculation section).  
0.0001  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
INPUT (dBm)  
Figure 36. Output vs. Input Level Distribution of 50 Devices;  
Frequency = 900 MHz; Supply = 3.0 V  
The input impedance of the ADL5505 varies with frequency,  
decreasing the actual voltage across the input stage as the  
frequency increases and, thus, reducing the conversion gain.  
Similarly, conversion gain is less at frequencies near 450 MHz  
because of the small on-chip coupling capacitor. The dynamic  
range is near constant over frequency, but with a decrease in  
conversion gain as frequency is increased.  
Although some devices follow the ideal linear response at very  
low input powers, not all devices continue the ideal linear regres-  
sion to a near 0 V y-intercept. Some devices exhibit output  
responses that rapidly decrease, and some flatten out.  
With no RF signal applied, the ADL5505 has a typical output  
offset of 10 mV (with a maximum of 100 mV) on VRMS.  
Rev. 0 | Page 15 of 20  
 
 
 
ADL5505  
250  
200  
150  
100  
50  
OUTPUT DRIVE CAPABILITY AND BUFFERING  
The ADL5505 is capable of sourcing a VRMS output current of  
approximately 3 mA. The output current is sourced through the  
on-chip, 100 Ω series resistor; therefore, any load resistor forms  
a voltage divider with this on-chip resistance. It is recommended  
that the ADL5505 VRMS output drive high resistance loads to  
preserve output swing. If an application requires driving a low  
resistance load (as well as in cases where increasing the nominal  
conversion gain is desired), a buffering circuit is necessary.  
SELECTING THE OUTPUT LOW-PASS FILTER  
The internal filter capacitor of the ADL5505 provides averaging  
in the square domain but leaves some residual ac on the output.  
Signals with high peak-to-average ratios, such as W-CDMA or  
CDMA2000, can produce ac residual levels on the ADL5505  
VRMS dc output. To reduce the effects of these low frequency  
components in the waveforms, some additional filtering is  
required.  
0
1
10  
100  
1000  
C
CAPACITANCE (nF)  
OUT  
Figure 38. Effect of COUT on Response Time  
The turn-on time and pulse response are strongly influenced by  
the sizes of the output shunt capacitor. Figure 39 shows a plot of  
the output response to an RF pulse on the RFIN pin, with a 0.1 μF  
output filter capacitor. The falling edge is particularly dependent on  
the output shunt capacitance, as shown in Figure 39.  
The output of the ADL5505 can be filtered directly by placing a  
capacitor between VRMS (Pin 4) and ground. The combination  
of the on-chip, 100 Ω output series resistance and the external  
shunt capacitor forms a low-pass filter to reduce the residual ac.  
PULSED RFIN  
Figure 37 show the effects on the residual ripple vs. the output  
filter capacitor value at two communication standards with high  
peak-to-average ratios. Note that there is a trade-off between ac  
residual and response time. Large output filter capacitances  
increase the turn-on and pulse response times, as shown in  
Figure 38.  
400mV rms RF INPUT  
250mV rms  
160mV rms  
70mV rms  
400  
W-CDMA FORWARD LINK (4.6dB CF)  
W-CDMA REVERSE LINK (11.7dB CF)  
350  
VRMS  
300  
250  
200  
150  
100  
50  
1ms/DIV  
Figure 39. Output Response to Various RF Input Pulse Levels; Supply = 3 V;  
Frequency = 900 MHz; Square-Domain Filter Open; COUT = 0.1 μF  
To improve the falling edge of the enable and pulse responses,  
a resistor can be placed in parallel with the output shunt capacitor.  
The added resistance helps to discharge the output filter capacitor.  
Although this method reduces the power-off time, the added  
load resistor also attenuates the output (see the Output Drive  
Capability and Buffering section).  
0
1
10  
100  
1000  
COUT CAPACITANCE (nF)  
Figure 37. AC Residual vs. COUT, W-CDMA Reverse Link (11.7 dB CF) Waveform  
and W-CDMA Forward Link (4.6 dB CF) Waveform  
Rev. 0 | Page 16 of 20  
 
 
 
 
 
 
 
ADL5505  
Once gain and intercept are calculated, an equation can be  
written that allows calculation of an (unknown) input power  
PULSED RFIN  
based on the measured output voltage.  
400mV rms RF INPUT  
V
IN = (VVRMS Intercept)/Gain  
(5)  
For an ideal (known) input power, the law conformance error of  
the measured data can be calculated as  
250mV rms  
160mV rms  
70mV rms  
ERROR (dB) =  
20 × log [(VVRMS, MEASURED Intercept)/(Gain × VIN, IDEAL)] (6)  
Figure 41 shows a plot of the error at 25°C, the temperature  
at which the ADL5505 is calibrated. Note that the error is not 0;  
this is because the ADL5505 does not perfectly follow the ideal  
linear equation, even within its operating region. The error at  
the calibration points is, however, equal to 0 by definition.  
3
VRMS  
1ms/DIV  
Figure 40. Output Response to Various RF Input Pulse Levels,  
Supply = 3 V, Frequency = 900 MHz; Square-Domain Filter Open;  
COUT = 0.1 μF with Parallel 1 kΩ  
POWER CONSUMPTION  
2
The quiescent current consumption of the ADL5505 varies  
linearly with the size of the input signal from approximately  
1.8 mA for no signal up to 8.5 mA at an input level of 0.7 V rms  
(10 dBm, referred to 50 Ω) as shown in Figure 27. There is  
little variation in supply current across power supply voltage  
or temperature.  
1
+25°C  
+85°C  
0
–40°C  
–1  
In applications requiring power saving, it is recommended that the  
ADL5505 be disabled while idle by removing the power supply to  
the device.  
–2  
–3  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
DEVICE CALIBRATION AND ERROR CALCULATION  
INPUT (dBm)  
Because slope and intercept vary from device to device, board-  
level calibration must be performed to achieve high accuracy.  
In general, calibration is performed by applying two input power  
levels to the ADL5505 and measuring the corresponding output  
voltages. The calibration points are generally chosen to be within  
the linear operating range of the device. The best-fit line is  
characterized by calculating the conversion gain (or slope) and  
intercept using the following equations:  
Figure 41. Error from Linear Reference vs. Input at −40°C, +25°C, and  
+85°C vs. +25°C Linear Reference, 1900 MHz Frequency, 3.0 V Supply  
Figure 41 also shows error plots for the output voltage at −40°C  
and +85°C. These error plots are calculated using the gain and  
intercept at 25°C. This is consistent with calibration in a mass  
production environment where calibration at temperature is not  
practical.  
Gain = (VVRMS2 VVRMS1)/(VIN2 VIN1  
Intercept = VVRMS1 − (Gain × VIN1  
where:  
)
(3)  
(4)  
)
V
INx is the rms input voltage to RFIN.  
V
VRMSx is the voltage output at VRMS.  
Rev. 0 | Page 17 of 20  
 
 
 
 
ADL5505  
CALIBRATION FOR IMPROVED ACCURACY  
DRIFT OVER A REDUCED TEMPERATURE RANGE  
Another way of presenting the error function of the ADL5505  
is shown in Figure 42. In this case, the decibel (dB) error at hot  
and cold temperatures is calculated with respect to the transfer  
function at ambient temperature. This is a key difference in  
comparison to Figure 41, in which the error was calculated  
with respect to the ideal linear transfer function at ambient  
temperature. When this alternative technique is used, the  
error at ambient temperature becomes equal to 0 by definition  
(see Figure 42).  
Figure 43 shows the error over temperature for a 1.9 GHz input  
signal. The error due to drift over temperature consistently  
remains within 0.15 dB and only begins to exceed this limit  
when the ambient temperature rises above +65°C and drops  
below −20°C. For all frequencies using a reduced temperature  
range, higher measurement accuracy is achievable.  
1.00  
–40°C  
–20°C  
0°C  
+15°C  
+35°C  
+55°C  
+75°C  
–30°C  
–10°C  
+5ºC  
+25°C  
+45°C  
+65°C  
+85°C  
0.75  
0.50  
0.25  
0
This plot is a useful tool for estimating temperature drift at a  
particular power level with respect to the (nonideal) response  
at ambient temperature. The linearity and dynamic range tend  
to be improved artificially with this type of plot because the  
ADL5505 does not perfectly follow the ideal linear equation  
(especially outside of its linear operating range). Achieving  
this level of accuracy in an end application requires calibration  
at multiple points in the operating range of the device.  
–0.25  
–0.50  
–0.75  
–1.00  
In some applications, very high accuracy is required at just one  
power level or over a reduced input range. For example, in a  
wireless transmitter, the accuracy of the high power amplifier  
(HPA) is most critical at or close to full power. The ADL5505  
offers a tight error distribution in the high input power range,  
as shown in Figure 42. The high accuracy range, beginning  
around 6 dBm at 1900 MHz, offers 8 dB of 0.15 dB detection  
error over temperature. Multiple point calibration at ambient  
temperature in the reduced range offers precise power  
measurement with near 0 dB error from −40°C to +85°C.  
3
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
INPUT (dBm)  
Figure 43. Typical Drift at 1.9 GHz for Various Temperatures  
DEVICE HANDLING  
The wafer level chip scale package consists of solder bumps  
connected to the active side of the die. The part is Pb-free and  
RoHS compliant with 95.5ꢁ tin, 4.0ꢁ silver, and 0.5ꢁ copper  
solder bump composition. The WLCSP can be mounted on  
printed circuit boards using standard surface-mount assembly  
techniques; however, caution should be taken to avoid damaging  
the die. See the AN-617 Application Note, MicroCSP Wafer  
Level Chip Scale Package, for additional information. WLCSP  
devices are bumped die; therefore, the exposed die may be  
sensitive to light, which can influence specified limits. Lighting  
in excess of 600 lux can degrade performance.  
2
1
+25°C  
+85°C  
0
LAND PATTERN AND SOLDERING INFORMATION  
–40°C  
–1  
Pad diameters of 0.20 mm are recommended with a solder paste  
mask opening of 0.30 mm. For the RF input trace, a trace width  
of 0.30 mm is used, which corresponds to a 50 ꢀ characteristic  
impedance for the dielectric material being used (FR4). All traces  
going to the pads are tapered down to 0.15 mm. For the RFIN  
line, the length of the tapered section is 0.20 mm.  
–2  
–3  
–25  
–20  
–15  
–10  
–5  
0
5
10  
15  
INPUT (dBm)  
Figure 42. Error from +25°C Output Voltage at −40°C, +25°C, and +85°C After  
Ambient Normalization, 1900 MHz Frequency, 3.0 V Supply  
EVALUATION BOARD  
Figure 44 shows the schematic of the ADL5505 evaluation board.  
The board is powered by a single supply in the 2.5 V to 3.3 V  
range. The power supply is decoupled by 100 pF and 0.1 μF  
capacitors.  
Note that the high accuracy range center varies over frequency  
(see Figure 13 to Figure 15 and Figure 19 to Figure 21).  
The RF input has a broadband match of 50 Ω using a single  
75 ꢀ resistor at R7B. More precise matching at spot frequencies  
is possible (see the RF Input Interfacing section).  
Table 4 details the various configuration options of the evaluation  
board. Figure 45 shows the layout of the evaluation board.  
Rev. 0 | Page 18 of 20  
 
 
 
 
 
 
ADL5505  
C7B  
(OPEN)  
R3B  
0  
ADL5505  
1
2
4
3
VRMS  
VRMSB  
VPOS  
VPOSB  
R2B  
(OPEN)  
C4B  
(OPEN)  
R5B  
(OPEN)  
C9B  
(OPEN)  
C2B  
0.1µF  
C1B  
100pF  
COMM  
RFIN  
RFINB  
C8B  
(OPEN)  
R6B  
N)  
R7B  
75ꢀ  
(OPE  
(P1 – B12)  
Figure 44. Evaluation Board Schematic  
Figure 45. Layout of Evaluation Board, Component Side  
Table 4. Evaluation Board Configuration Options  
Component  
Description  
Default Condition  
VPOSꢀ, GNDꢀ  
Ground and supply vector pins.  
Not applicable  
C1ꢀ, C2ꢀ, C7ꢀ, C8ꢀ,  
C9ꢀ  
Power supply decoupling. Nominal supply decoupling of 0.1 μF and 100 pF.  
C1ꢀ = 100 pF (Size 0402)  
C2ꢀ = 0.1 μF (Size 0402)  
C7ꢀ = C8ꢀ = open (Size 0805)  
C9ꢀ = open (Size 0402)  
R7ꢀ  
RF input interface. The 75 Ω resistor at R7ꢀ combines with the ADL5505 internal  
input impedance to give a broadband input impedance of around 50 Ω.  
R7ꢀ = 75 Ω (Size 0402)  
C4ꢀ, R2ꢀ, R3ꢀ  
Output filtering. The combination of the internal 100 Ω output resistance and C4ꢀ  
produce a low-pass filter to reduce output ripple of the VRMS output. The output  
can be scaled down using the resistor divider pads, R2ꢀ and R3ꢀ.  
R3ꢀ = 0 Ω (Size 0402)  
R2ꢀ = open (Size 0402)  
C4ꢀ = open (Size 0402)  
P1, R5ꢀ, R6ꢀ  
Alternate interface. The end connector, P1, allows access to various ADL5505 signals. P1 = not installed  
These signal paths are only used during factory test and characterization.  
R5ꢀ = R6ꢀ = open (Size 0402)  
Rev. 0 | Page 19 of 20  
 
 
 
ADL5505  
OUTLINE DIMENSIONS  
0.660  
0.600  
0.540  
0.860  
0.820 SQ  
0.780  
0.381  
0.356  
0.331  
SEATING  
PLANE  
2
1
A
B
BALL A1  
IDENTIFIER  
0.280  
0.260  
0.240  
0.40  
BALL PITCH  
TOP VIEW  
(BALL SIDE DOWN)  
BOTTOM VIEW  
(BALL SIDE UP)  
0.230  
0.200  
0.170  
0.050 NOM  
COPLANARITY  
Figure 46. 4-Ball Wafer Level Chip Scale Package [WLCSP]  
(CB-4-2)  
Dimensions shown in millimeters  
ORDERING GUIDE  
Model1  
Temperature Range Package Description  
Package Option  
Branding Ordering Quantity  
ADL5505ACꢀZ-P7  
ADL5505ACꢀZ-P2  
ADL5505-EVALZ  
–40°C to +85°C  
–40°C to +85°C  
4-ꢀall WLCSP, 7Pocket Tape and Reel Cꢀ-4-2  
4-ꢀall WLCSP, 7Pocket Tape and Reel Cꢀ-4-2  
Evaluation board  
3R  
3R  
3,000  
250  
1 Z = RoHS Compliant Part.  
©2010 Analog Devices, Inc. All rights reserved. Trademarks and  
registered trademarks are the property of their respective owners.  
D0ꢀ799-0-4/10(0)  
Rev. 0 | Page 20 of 20  
 
 

相关型号:

ADL5506

30 MHz to 4.5 GHz, 45 dB RF Detector
ADI

ADL5506-EVALZ

30 MHz to 4.5 GHz, 45 dB RF Detector
ADI

ADL5506ACBZ-R7

30 MHz to 4.5 GHz, 45 dB RF Detector
ADI

ADL5506WACBZ-R7

30 MHz to 4.5 GHz, 45 dB RF Detector
ADI

ADL5511

Envelope and TruPwr RMS Detector
ADI

ADL5511-EVALZ

Envelope and TruPwr RMS Detector
ADI

ADL5511ACPZ

DC to 6 GHz Envelope and TruPwr RMS Detector
ADI

ADL5511ACPZ-R7

Envelope and TruPwr RMS Detector
ADI

ADL5513

1 MHz to 4 GHz, 80 dB Logarithmic Detector/Controller
ADI

ADL5513-ACPZ-R2

1 MHz to 4 GHz, 80 dB Logarithmic Detector/Controller
ADI

ADL5513-ACPZ-R7

1 MHz to 4 GHz, 80 dB Logarithmic Detector/Controller
ADI

ADL5513-ACPZ-WP

1 MHz to 4 GHz, 80 dB Logarithmic Detector/Controller
ADI