Successful VSA Analysis Using an Oscilloscope

How to make vector signal analysis measurements using the Agilent 89600 vector signal analysis software and Agilent scopes

The 89600 VSA software shown in this document has been replaced with the new 89600B software. To find out how you can see through signal complexity with the new 89600B VSA software visit www.agilent.com/find/89600B

Table of Contents

An Organized Approach to Making Scope/VSA Measurements............................................... 1 Step I: Set-up........................................................................................................................................ 2 Step II: Evaluate the signal ............................................................................................................. 10 Step III: Optimize result accuracy ................................................................................................. 12 Summary ................................................................................................................................................. 12 Appendix 1: Aliasing and Under-, Over-, and Nyquist Sampling ............................................... 13 Selecting VSA oscilloscope sampling modes for MIMO signals ......................................... 13 Baseband vs. IF sampling ............................................................................................................... 13 Trade-offs of baseband sampling...................................................................................................... 14 Sub-sampling ..................................................................................................................................... 14 Agilent 89600 VSA Sampling Modes ............................................................................................... 15 Appendix 2: Additional Information ................................................................................................. 16 Related Literature List ........................................................................................................................... 17

An Organized Approach to Making Scope/VSA Measurements

Many RF engineers are reluctant to use a scope to analyze a signals spectrum and modulation characteristics because of the potential that signal aliasing will corrupt the measurements. But good quality frequency domain measurements are achievable with a good scope, good VSA software, and an organized approach to the measurements that account for aliasing with a selection of sampling modes. Using this approach allows the designer to make RF-type measurements at baseband, on multi-channel designs, or with a scope simply because it is the most convenient measurement tool available. This application note outlines an organized approach to making spectrum and modulation domain measurements with a scope and VSA software. This method has proven to be successful for reducing set-up errors and increasing the quality of the measurement results. The steps in this method are given here and explained in the following pages.

Scope/VSA software measurements

Step I: Set-up a. Select a sampling mode b. Check for aliased signals c. Trigger on the signal (burst signals only) Step II: Evaluate the signal 1. Spectrum and time 2. Basic modulation analysis 3. Advanced modulation analysis Step III: Optimize the results This measurement approach includes techniques to check for, reduce or eliminate the effects of signal aliasing in a measurement. Some of these techniques involve a trade-off between measurement speed and measurement accuracy. The approach outlined here will help you manage that trade-off to optimize your results. If you are not familiar with aliasing, over-sampling, under-sampling and Nyquist-sampling, please refer to Appendix 1 for explanations of these key concepts.

Step I: Set-up

The 89600 VSA software sets the scope sample rate automatically based on three parameters the user enters: signal center frequency, signal span, and scope sampling mode. The sampling modes the VSA software uses to control scopes are: Minimize mode, Maximize mode, User Rate mode and Full Rate mode. All of the following set up steps assume that you have already set the analyzer to accept input from an Inniium or InniiVision scope. This is done by clicking on Utilities>Hardware>ADC1[tab] and then selecting either Inniium or 6000Scope.

a. Select a sampling mode

Minimize mode (Signal discovery)

2Fs 3Fs 4Fs Maximize mode (Signal discovery - wideband)

Fs

Scope sample mode Minimize (Max undersample)

EVM Measurement Signal aquire length (noise oor) speed

(Recommended)

Fs User rate mode (Nice for Nyquist)

2Fs

Maximize (least undersample) User Rate (Any rate, nice for Nyquist)

Fs/2 2Fs Full rate mode (Baseband)

Full Rate (No aliases)

Fs/2

2Fs

Figure 1. Description of the various sampling modes available for use with the 89600 VSA and supported Agilent oscilloscopes. The relative advantages for each mode are also shown.

Minimize mode
Always start with the Minimize sample mode. Its speed, large acquisition time length, and exibility in parameter settings makes it ideal for verifying that the signal is at the expected center frequency, has the expected span, and the scopes input range is set so the signal is at full scale without overload. If any of these parameters are wrong, further measurements will be difcult. The Minimize mode is also useful for the initial set-up of triggering, time domain measurements and modulation domain analysis. Select the 89600 VSA software Minimize sample mode by clicking on Utilities >Hardware >ADC1[tab] >Congure >Sample Mode>Edit[button] and then selecting Minimize.

Figure 2. Hardware conguration dialog box for Minimize mode.

This mode is susceptible to signal aliasing and has the highest noise oor of the four sampling modes. It is not recommended for measurements requiring accuracy. Once the signal and set-up are veried, you can switch to more accurate sample modes. For best measurement accuracy use the User Rate or Full Rate sample modes. Because these modes can update slowly, sometimes less than once per minute, it is best to use them only after the signal has been veried and the measurement parameters have been set-up using the fast update rate of the Minimize mode.

Maximize mode
If your signal bandwidth is 100 MHz or greater, select Maximize mode rather than Minimize mode to verify your signal. It provides somewhat slower update rates and shorter acquisition lengths than the Minimize mode, but is better suited to wider signals and still offers good exibility. Select the Maximize sample mode by clicking on: Utilities >Hardware >ADC1 (tab) >Congure >Sample Mode>Edit[button] and then select Maximize.

Figure 3. Hardware conguration dialog box with Maximize sample mode selected.

The Maximize mode is also susceptible to signal aliasing and has the second highest noise oor of the four sampling modes, so it is not recommended for measurements requiring accuracy. Use the User Rate or Full Rate modes for best measurement accuracy.

User Rate mode

The User Rate sample mode has the second lowest noise oor of the four sampling modes, and is therefore one of the most accurate modes. But this mode can update slowly, sometimes less than once per minute, so it is best to use it only after the signal and measurement parameter set-up have been veried using the Minimize or Maximize modes. Select the User Rate sample mode by clicking on: Utilities >Hardware >ADC1[tab] >Congure >Sample Mode>Edit[button] and then selecting User Rate. Next, input the desired sample rate by double-clicking on User Sample Rate in the Hardware Conguration menu and entering the rate in Hz.

Figure 4. Hardware conguration dialog box with User Sample Rate mode selected. Note that you must also set the specic sample rate desired.

Use the User Rate mode to set specic sampling rates, for example, the Nyquist sampling rate, in order to meet specic measurement needs. Most scopes support only a specic set of sample rates. Many use a 5, 4, 2, 1 sequence (i.e. 500MSa/s, 400MSa/s, 200MSa/s, 100 MSa/s, 50MSa/s, etc). In the User Rate mode, the VSA software selects the rate that is closest to the frequency the user enters in the User Sample Rate menu. In Figure 5, the VSA software will select the 500MHz sample rate because it is the rate closest to the requested rate.

Figure 5. User Sample Rate conguration. Although the dialog box shows that the user set this to 512 MHz, the analyzer software will automatically set the scope sample rate to 500 MHz, the closest supported sample rate.

The User Rate mode is susceptible to signal aliasing. Using a low pass accessory lter and the alias checker macro (see the following section) is recommended to achieve the best results. Refer to Appendix 2 for help nding an accessory lter vendor.

Full Rate mode

This mode offers the lowest noise oor and therefore the most accurate measurements. It is also the least susceptible to signal aliasing. However, it has the slowest update rate and the least exibility in parameter setting ranges. It is best used when the signal of interest is exceptionally wide, aliasing is a severe problem, or the very best EVM accuracy is required. Using a low pass accessory lter is recommended to achieve the best results. Whether or not the measurement requires use of a low pass accessory lter depends on the model and measurement conguration. See the scope performance guides mentioned in Appendix 2 for more information. Select the Full Rate sample mode by clicking on: Utilities > Hardware > ADC1(tab) > Congure > Sample Mode. Click the Edit button and then select Full Rate.

Figure 6. Full Rate sample mode selected in Hardware Conguration dialog box.

Some settings in this mode, particularly modulation analysis measurements, can take several minutes to show results. If you are going to do modulation analysis in the Full Rate mode, it is best to use the Minimize mode rst to test the set-up parameters, then switch to Full Rate when all parameters are properly set.

b. Check for aliased signals

Signal aliasing will occur in the Minimize, Maximize and User Rate modes. Aliased signals can reduce accuracy and interfere with measurements if they are large enough. The 89600 VSA software includes an Alias Checker macro that searches the analysis span for potential signal aliasing, highlights the potential locations, and indicates how they could impact a measurement. Be sure to use this checker with the three vulnerable modes. To install the Alias Exposure Zone Checker: In the 89600 VSA software tool bar click on Utilities>Macros>Recall. Navigate to the Sample Macros subdirectory in the directory where you installed the VSA application (for example, C:\Program Files\Agilent\89600 VSA\Examples\Macros). Select RecallSetupAfterAliasCheck.vbs and AliasChecker.vbs, and click OK. Right click in the VSA applications tool bar. You will see a popup menu showing the various toolbars. Select the Macros toolbar. This will make the Alias Checker icons appear on your VSA application toolbar.

To use the Alias Checker click on the icon. The macro will check the entire span to determine if alias products are possibly present.

Although the Alias Checker macro found an alias exposure zone (orange box), the total power in the alias zone is below the orange threshold, indicating that any aliased signals will be more than 40 dB below the signal of interest.

Figure 7. Full span spectrum trace after running the Alias Checker macro.

The Alias Checker calculates the frequencies of the alias exposure zones for any combination of center frequency, span, and oscilloscope sample rate. It then measures the power in each zone and compares it to the power of the desired signal in the span of the measurement setup at the time the checker was started. An unwanted signal in an exposure zone is a source of interference if it is larger than the desired signals amplitude minus 40 dB. This 40 dBc threshold is sufcient for 1 to 2 percent EVM measurements. When the macro stops, a full-span spectrum trace will appear within the trace. It shows the desired signal plus any unwanted signals. Superimposed is an orange trace showing boxes for each exposure zone frequency range. Any unwanted signal falling inside the zone box is a source of alias interference. If the height of any box is higher than the orange reference line, the power in that zone is greater than the 40 dBc threshold of interference. When you are through with the checker, click on the measurement. icon to return to your

It is also recommended that you use an external low pass lter to reduce the effects of out-of-band signal aliasing. See the Inniium and InniiVision app notes listed in Appendix 2 for more information on the suggested stop band frequencies for various scope models.

c. Trigger on the signal (burst signals only)

The VSA software offers two triggering modes: Burst Hi Duty Cycle for signals like those used in wireless communication that have off times that are relatively short compared to their on time Low Duty Cycle for signals like radar signals that have relatively long off times compared to their on time These modes are set by clicking on: Utilities >Hardware >ADC1[tab] >Congure[button] >Hold-off Type>Edit[button] and then selecting the mode.

Figure 8: Choose the trigger Holdoff Type which best suits your signal.

Once the Hold-off mode is selected, the trigger level can be set using the VSA softwares trigger controls located at: Input>Trigger. Version 10.0 of the 89600 VSA software adds decimation and resampling capabilities to the scope driver contained in the Hardware Connectivity option (Option 300). These capabilities process the signal data passed from the scope in small segments rather than as one large data block. This increases the maximum raw data block size the VSA software can handle and increases the measurement speed on large data blocks required for modulation quality measurements on many wireless communication standards. However, this processing can make trigger level setting with the VSA software more difcult, as the trigger may occur within a segment of data not passed on to the software. Because of this, set up your triggering with the resampling done in the measurement as follows: To set up a reliable trigger level: 1. Disable decimation and resampling in the scope driver and enable it in the measurement by clicking: Utilities >Hardware >ADC1[tab] >Congure[button] >Resample Location> Edit[button] and selecting Measurement in the drop down box.

2. Select Raw Main Time trace format to display the signal by double clicking the trace title (title above upper left corner of the trace) and selecting Channel 1 from left-hand column and then Raw Main Time from the alphabetical list of data.

3. In the same trace, set the units for the vertical axis to volts by double clicking on the label for that axis and selecting Linear Mag.

These settings disable the VSA software frequency domain processing of the data coming from the scope, making it easier to set the trigger point in the time domain. Note: Make sure to change the Resample Location setting to Driver for best measurement speed and longest time capture after the trigger level is set up and working reliably.

Step II: Evaluate the signal

Once the scope and VSA software are set up, signal evaluation can start. When measuring and troubleshooting digitally modulated signals, it is usually best to follow this sequence: Signal troubleshooting steps: 1. Basic spectrum and vector (combined frequency and time) measurements 2. Basic digital modulation analysis (constellation and basic I/Q parameters) 3. Advanced and/or standard specic analysis This sequence of measurements is useful because it reduces the chance that important signal problems will be missed and because getting parameters like center frequency and span wrong make it much harder, sometimes impossible, to demodulate the signal for evaluation.

1. Spectrum and time domain measurements

Figure 9. Beginning with a correctly set up signal in the time and frequency domain will help ensure good measurements later.

These measurements give the basic parameters of the signal in frequency and time domain so that correct demodulation can take place in the next step. Along with verifying the center frequency, bandwidth, and power of the signal, this step is the place to evaluate symbol timing, adjacent channel power, and other spectral characteristics of the signal.

10

2. Basic digital demodulation

These measurements evaluate the quality of the constellation. Along with a display of the constellation, they include parameters such as overall EVM, I/Q offset, I/Q gain imbalance, frequency error, and symbol clock error.

Figure 10. Up to 6 simultaneous displays (4 shown here) of user-selected information lets you see a wide range of basic digital demodulation info, especially useful for multi-channel analysis.

3. Advanced digital demodulation

These measurements are used to investigate the causes of errors uncovered in the basic modulation parameters, particularly EVM errors. These include dynamic parameters such as error vector frequency, error vector time, and pilot phase error.

Figure 11. Advanced demodulation measurements and displays let you closely investigate the signal and allow you to set special measurement troubleshooting parameters.

11

Step III: Optimize result accuracy

In general the higher the scope sample rate you use making spectrum and modulation measurements, the lower the noise oor for the measurement and the better the EVM performance. Figure 12 illustrates this point.
Statistic Results of Under Sampling Experiments -20 -25 -30 dB -35 -40 -45 -50
-(Worst SNR) -(Average SNR) -(Best SNR) Worst EVM Average EVM Best EVM

10

12

14

16

18

20

Sampling Rate (GHz)

Figure 12. Higher scope sample rates allow you to use processing gain to improve measurement performance.

As indicated earlier in this document, this improved accuracy comes at the price of slower update rates and shorter acquisition time length, sometimes much slower and much shorter. The best way to handle this trade-off is to verify the signal parameters and measurement set-up using the higher speed Minimize or Maximize sampling modes. After signal and measurement parameter set-up is veried, switch to the Full Rate mode to get your most accurate reading.

Summary

Using a scope as a measurement front end to the 89600 VSA software is especially useful for multi-channel analysis, such as MIMO measurements, or when super-wideband signal analysis is needed, as with UWB measurements made for Wireless USB and other formats. Also, sometimes a scope may be the most easily available instrument, or you may be making other measurements with it, particularly in the baseband section of your designs. This note provides some helpful hints for setting up triggering and can help you determine which sample mode to use out of the several modes provided by the VSA software for use with Agilent oscilloscopes.

12

Appendix 1: Aliasing and Under-, Over-, and Nyquist Sampling

Selecting VSA oscilloscope sampling modes for MIMO signals
Oscilloscopes offer a number of important advantages for MIMO signal analysis. Optimizing their performance to match your measurement needs is straightforward with the intelligent driver provided in the 89600 VSA software, especially if some signal analysis sampling basics are kept in mind. Using an oscilloscope for measuring MIMO signals offers several advantages including synchronized coherent sampling of 2-4 channels and a more cost effective solution for 3-4 channel measurement. The Agilent 89600 VSA software, when combined with high-performance deep-memory oscilloscopes, provides the best optimization of choices for MIMO signal analysis. The information which follows applies to all applications, not just MIMO analysis.

Baseband vs. IF sampling

Most modern spectrum and signal analyzers and vector signal analyzers digitize a bandlimited version of the signal under test at a down-converted intermediate frequency (IF) and at a sampling rate related to the signal bandwidth itself rather than its center frequency. The down-conversion, ltering, and sampling are shown in a simplied form in Figure 13, where several frequency conversion and ltering stages are represented as a single one, ahead of the ADC block.

RF input attenuator Input signal Pre-selector, or low-pass lter

Mixer

IF lter ADC DSP

Local oscillator

Reference oscillator

Sweep generator

Display

Figure 13. Block diagram of RF spectrum/signal analyzer with digital IF section.

This heterodyne down-conversion scheme uses low distortion RF circuits to preserve the signal quality while mixing down in frequency the selected signal analysis band and ltering it so that the signal can be digitized at a comparatively low frequency. The lower sampling rate and limited signal bandwidth provides the best measurement accuracy and dynamic range. The low sample rate also has the benet of reducing the amount of data (the number of samples/second and thus the number of samples/measurement) that must be processed for a measurement result, and allows for longer measurement time lengths for a given amount of acquisition memory. With an oscilloscope as the acquisition device, the RF signal is sampled directly, without frequency conversion or bandpass ltering. Instead of IF sampling by the ADC, the signal is thus sampled at baseband. Subsequent ltering and frequency conversion are performed digitally, through digital ltering and resampling algorithms.

13

Trade-offs of baseband sampling

This baseband sampling approach places much higher sample rate and sample memory demands on the oscilloscope. For example, measuring a 2.5 GHz signal with a 10 MHz bandwidth with a signal analyzer requires an IF sample rate of approximately 25 MHz (twice the signal bandwidth [Nyquist] + a guard band for anti-alias lter roll off), regardless of the center frequency of the signal. Measuring the same signal using baseband sampling with a scope requires a sample rate of approximately 6.25 GHz, a rate 250 times higher than IF sampling. The sample memory required is similarly large when compared to IF sampling. Fortunately these sampling and memory requirements are readily handled by modern oscilloscopes and measurement software. Oscilloscopes such as the Agilent Inniium Series can be equipped with the high sample rates and large memories required for baseband sampling at Nyquist rates.

Sub-sampling

A second approach to measuring RF signals involves a technique referred to as undersampling, sub-sampling, sub-Nyquist, or harmonic sampling. For the RF engineer, sampling can be understood as a frequency mixing process and sub-sampling is therefore equivalent to harmonic mixing. Just as with harmonic mixing, the benets of operation at much lower sampling/local oscillator frequencies (lower sample rate, smaller memory) come with the drawback of possible signal aliasing. Aliasing is most frequently understood as a sampling phenomenon where signals at a high frequency (greater than twice the sampling frequency) fold over or mix into the sampled frequency range, appearing as lower frequencies called aliases. A graphic example is shown in Figure 14.

Actual signal

Reconstructed alias signal

ADC sample points

Figure 14. Graphic illustration of aliasing.

Figure 14 shows aliasing in the form of a discrete signal, but in sampled systems it is important to understand that aliasing also applies to noise. Therefore, the drawbacks of under-sampling for RF signal analysis include both the possibility of false signals in the measurement and the inclusion of noise from outside the band of interest. The greater the degree of under-sampling, the higher the number of possible aliases and the greater the amount of apparent noise added to the analysis band. This is directly analogous to a higher multiple of harmonic mixing in microwave analysis, and the consequences (absent an analog lter such as a preselector) are the same.

14

Agilent 89600 VSA Sampling Modes

A sophisticated oscilloscope driver in the VSA offers the user the choice of intelligent under-sampling, user-rate sampling, and full-rate sampling. The benets and limitations of the sampling modes are described in Table 1.
Table 1. Comparison of VSA sampling modes

Sampling mode Benets

Minimize Fastest measurement update rate Longest maximum time record length (demodulation result length and main time length) No aliases in the selected center frequency/span region Provides performance, measurement speed, and time/demod length choices between Minimize and Full Rate modes Best EVM, dynamic range performance Maximum sample processing in oscilloscope to improve accuracy, dynamic range Optimize performance, time/demod length, and update rate to user requirements

Limitations
Lowest signal/noise ratio, dynamic range Greatest possibility for aliases in the measurement

Maximize

EVM, dynamic range performance not as good as Full Rate mode Maximum time record and demod result length not as long as minimize mode Shortest maximum time record length (demodulation result length and main time length) Slowest measurement update rate VSA software does not attempt to avoid alias exposure regions

Full Rate

User Rate

The benets and limitations described in Table 1 illustrate the reasoning behind the suggested sequence of sampling modes described earlier in this note. For more detailed information on signal analysis and sampling techniques please consult the additional information sources listed below.

15

Appendix 2: Additional Information

For information on sampling, aliasing, and signal analysis in general see section 2 of Agilent application note 243 The Fundamentals of Signal Analysis, starting on page 29, available at www.agilent.com by searching for literature number 5952-8898E. For information specic to the use of the Agilent 89600 VSA software and Inniium oscilloscopes see the sampling modes section of the Agilent application note titled Agilent Inniium Oscilloscopes Performance Guide Using 89600 Vector Signal Analyzer Software. The application note is available at www.agilent.com by searching for literature number 5988-4096EN. Similar information is available in the 89600 VSA online help facility by searching hardware conguration parameters (Inniium) For information specic to the use of the Agilent 89600 VSA software and InniiVision scopes, see the sampling modes section of the Agilent application note titled Agilent InniiVision 6000 and 7000 Series Oscilloscopes Performance Guide Using 89600 Vector Signal Analyzer Software, literature number 5989-4523EN. For information on RF signal analysis and the signal processing of spectrum/signal analyzers see chapter 2 of Agilent application note 150 Spectrum Analyzer Basics, available at www.agilent.com by searching for literature number 5952-0292. More information on the scopes used in the examples in this application note is available at www.agilent.com/nd/scope. This note requires the 8000, 80000 or 90000 Series Inniium scopes. Complete information on the Agilent 89600 VSA software is available at www.agilent.com/nd/89600. This note requires VSA software v10.0 or greater. For information on accessory low pass lters that can be used to reduce aliasing go to: Mini-Circuits, Inc., www.minicircuits.com. There is also more information on the specic stopband frequencies for each scope model suggested in the Inniium application note, 5988-4096EN and the InniiVision application note, 5989-4523EN.

16

Related Literature List

89600 Vector Signal Analyzer CD, literature number 5980-1989E 89600 Vector Signal Analysis Software 89601A/89601AN/ 89601N12,Technical Overview, literature number 5989-1679EN 89600 Vector Signal Analysis Software 89601A/89601AN/ 89601N12, Data Sheet, literature number 5989-1786EN Agilent Infiniivision Series Oscilloscopes Performance Guide Using 89600 Vector Signal Analyzer Software, literature number 5989-4523EN Hardware Measurement Platforms for the Agilent 89600 Series Vector Signal Analysis Software, Data Sheet, literature number 5989-1753EN 89600S Vector Signal Analyzers, VXI Configuration Guide, literature number 5968-9350E 89650S Wideband Vector Signal Analyzer System with High Performance Spectrum Analysis, Technical Overview, literature number 5989-0871EN 89650S Wideband Vector Signal Analyzer System with High Performance Spectrum Analysis, Configuration Guide, literature number 5989-1435EN Understanding Time and Frequency Domain Interactions in the Agilent Technologies 89400 Series Vector Signal Analyzers, literature number 5962-9217EN 89607A WLAN Test Suite Software, Technical Overview, literature number 5988-9574EN 89604A Distortion Test Suite Software, Technical Overview, literature number 5988-7812EN

17

www.agilent.com

Agilent Email Updates

www.agilent.com/find/emailupdates Get the latest information on the products and applications you select.

For more information on Agilent Technologies products, applications or services, please contact your local Agilent office. The complete list is available at:

www.lxistandard.org LAN eXtensions for Instruments puts the power of Ethernet and the Web inside your test systems. Agilent is a founding member of the LXI consortium.

Agilent Channel Partners

www.agilent.com/find/channelpartners Get the best of both worlds: Agilents measurement expertise and product breadth, combined with channel partner convenience.

Agilent Advantage Services is committed to your success throughout your equipments lifetime. We share measurement and service expertise to help you create the products that change our world. To keep you competitive, we continually invest in tools and processes that speed up calibration and repair, reduce your cost of ownership, and move us ahead of your development curve.
www.agilent.com/find/advantageservices

www.agilent.com/find/contactus Americas Canada Brazil Mexico United States Asia Pacific Australia China Hong Kong India Japan Korea Malaysia Singapore Taiwan Other AP Countries

(877) 894 4414 (11) 4197 3500 01800 5064 800 (800) 829 4444 1 800 629 485 800 810 0189 800 938 693 1 800 112 929 0120 (421) 345 080 769 0800 1 800 888 848 1 800 375 8100 0800 047 866 (65) 375 8100

www.agilent.com/quality

Europe & Middle East Belgium 32 (0) 2 404 93 40 Denmark 45 70 13 15 15 Finland 358 (0) 10 855 2100 France 0825 010 700*
*0.125 /minute

Germany Ireland Israel Italy Netherlands Spain Sweden United Kingdom

49 (0) 7031 464 6333 1890 924 204 972-3-9288-504/544 39 02 92 60 8484 31 (0) 20 547 2111 34 (91) 631 3300 0200-88 22 55 44 (0) 118 9276201

For other unlisted Countries: www.agilent.com/find/contactus

Revised: October 14, 2010

Product specifications and descriptions in this document subject to change without notice. Agilent Technologies, Inc. 2011 Printed in USA, February 11, 2011 5990-3276EN