NIST Calibration Services: Pulse Waveform Measurements

Electromagnetic Measurements

Pulse Waveform Measurements

Technical Contacts:

Paul Hale (65100S, 65200S and 65400S)
Tel: 303/497-5367
E-mail: hale@boulder.nist.gov

John Lomax (65100S, 65200S and 65400S)
Administration and Logistics
Tel: 303/497-3842
Fax: 303/497-4286
E-mail: john.lomax@boulder.nist.gov

Thomas Nelson (65250S, 65500S and 65501S)
Tel: 301/975-2986 (office)
Tel: 301/975-2000 X72416 (lab)
Fax: 301/926-3972
E-mail: thomas.nelson@nist.gov

Denise D. Prather (65250S, 65500S and 65501S)
Administration and Logistics
Tel: 301/975-4221
E-mail: denise.prather@nist.gov

Please contact the administration and logistics staff before shipping instruments or standards to the address listed below.

Mailing Address for 65100S, 65200S and 65400S:
National Institute of Standards and Technology
325 Broadway, MC 815.01
Boulder, CO 80305-3328

Mailing Address for 65250S, 65500S and 65501S:
National Institute of Standards and Technology
100 Bureau Drive, Stop 8170
Gaithersburg, MD 20899-8170

Service ID Number	Description of Services	Fee ($)
65100S	Impulse Spectrum Amplitude (50 )	At Cost
65200S	Fast Repetitive Pulse Transition Parameters (50 )	At Cost
65250S	Repetitive Pulse Waveform Measurements, Including Settling Parameters	At Cost
65400S	Pulse Time Delay Interval	At Cost
65500S	Peak-to-Peak Detector Calibration at One Frequency Selected from Those Given in Table 9.23 at 1.2 V	At Cost
65501S	Additional Frequency for Peak-to-Peak Detector in 65500S	At Cost

Fees are subject to change without notice.

Pulse Waveform Measurements-General Information

NIST offers special-test services for measuring the pulse parameters of the output of pulse generators and the step response of sampling oscilloscopes (samplers) and real-time oscilloscopes. These are broken down into five categories: impulse spectrum amplitude, fast repetitive pulse transition parameters, fast repetitive pulse settling parameters, network impulse response, and pulse time delay interval. All of the special test services except the fast repetitive pulse settling parameters service are performed using the NIST Waveform Parameter Analysis System (WPAS). The WPAS consists of a pulse generator, a calibrated wide-band sampling oscilloscope (input impedance 50 , and nominal bandwidth of dc to greater than 70 GHz, equivalent to a step response transition duration of less than 5 ps), and dedicated microcomputer system interfaced to the oscilloscope. The fast repetitive pulse settling parameters test service uses the NIST Sampling Comparator System (SCS), a NIST-designed sampling comparator oscilloscope, controlled by a microcomputer. The SCS has a 50 input impedance, nominal bandwidth of dc to 2.5 GHz (equivalent to a step response transition duration of 140 ps), and settling < 1% within 1 ns and to 0.1 % within 10 ns. All of these special tests are performed at cost, and by prearrangement. References pertaining to these four services are located at the end of this section.

Impulse Spectrum Amplitude (65100S)

In response to calibration needs from the electromagnetic interference (EMI) community, NIST has a special-test service to calibrate the broadband spectrum amplitude output from impulse generators. Such generators can then be used as transfer standards of broadband impulsive noise for field calibration of spectrum analyzers and field intensity meters. The NIST special-test service uses the time-domain measurement/frequency-domain deconvolution computational method for calibration of impulse generators. The WPAS wideband sampling oscilloscope is used to measure the time-domain waveform from the impulse generator. The spectrum amplitude, S(f), versus frequency is then computed using the fast Fourier transform (FFT). NIST will provide 50 to 200 data points over a wide frequency range for a single fee. NIST impulse spectrum amplitude measurement service capabilities are shown in Table 9.20 .

Table 9.20 . NIST Impulse Generator Spectrum Amplitude Measurement Service Capability

Parameter	Limits	Notes
Maximum impulse amplitude without attenuators	600 mV	1, 2
Maximum impulse amplitude with external attenuators	1.2 kV	2, 3, 6
S(f) range	- 15 dB (µV/MHz) <[S(f)-S _o ] <+5 dB (µV/MHz)	4, 5
S(f), expanded uncertainty	0.1 dB	4, 5
Frequency range	10 MHz to 10 GHz	4, 5
Frequency spacing	f >10 MHz
Frequency, expanded uncertainty	Nominally 0.3%
Load impedance	50.0
Load impedance expanded uncertainty	Nominally 0.5	5, 7
Trigger pulse amplitude	> 100 mV	7
Trigger pulse transition duration	< 5 ns	8
Trigger to impulse delay interval	> 24 ns	8
Trigger to impulse jitter	< 10 ps rms	8
Notes: 1. Impulse generator with an adjustable-amplitude impulse output will be calibrated with the generator adjusted to give a peak amplitude in the range of 300 mV to 500 mV. 2. Impulse generators with fixed output amplitudes greater than 600 mV will have the impulse attenuated to a level of 300 mV to 500 mV by 50 wideband coaxial attenuators. 3. Either customer-supplied or NIST attenuators may be used. 4. Data will not be given in the first spectrum null or at frequencies above. Typically about 100 data points are supplied. 5. Only for impulse amplitudes less than 600 mV. 6. If external attenuators and/or a delay-line network are used, then the uncertainties associated with the attenuator and/or network calibration are added to these values. 7. Load impedance uncertainty depends upon input impedance of external attenuators when used. 8. If the impulse generator does not supply a trigger output or if the trigger output does not have the proper characteristics, then a delay-line network will be used to provide a suitable trigger pulse.

Fast Repetitive Pulse Transition Parameters, 50 (65200S)

NIST offers a special-test service for measuring the pulse parameters of the output of pulse generators and the step response of sampling oscilloscopes (samplers) and real-time oscilloscopes. These parameters are measured with the NIST WPAS described above. This service is optimized for measuring the durations of very fast pulse transitions (transition durations less than 350 ps, i.e., 3 dB attenuation bandwidths greater than 1 GHz). The parameters, ranges, and estimated uncertainty limits for this service are listed in Table 9.21 .

Table 9.21 . Uncertainty for Calibration of Fast Repetitive Pulse Transition Parameters

Parameter	Parameter Range	Typical Expanded Uncertainty
Pulse Amplitude (A)	- 400 mV < A < 400 mV	1 mV + 1.4 A*¹
Transition Duration (t_d)	5 ps <t_d <100 ns	1.0 ps + 0.1 t*²
Pulse Duration (t_p) (between 50 % reference level instants)	10 ps < t_p <100 ns	1.8 ps + 0.14 t
Pulse Overshoot	< 50 %	2 %
Pulse Undershoot	< 50 %	2 %

Restrictions and Notes:

*¹ A is the amplitude discretization interval and is calculated using the full-scale amplitude range set on the sampler (for example, the full scale amplitude range is 100 mV for an amplitude sensitivity setting of 10 mV/div and a full scale display of 10 vertical divisions) and effective number of bits of the analog-to-digital converter at the input of the sampler. The effective number of bits is based on the actual number of bits of the converter and signal averaging where the noise level exceeds the range of the least significant bit of the converter.

*² t is the sampling interval, that is, the interval between sampling instances, used during acquisition of the DUT waveform. For example, a waveform epoch of 1 ns where the waveform contains 1000 elements gives a sampling interval of 1 ps.

1. Customer's device must generate a repetitive pulse with repetition rate between 100 Hz and 1 GHz. Alternatively, NIST can provide a range of trigger signals, however, in this case, the customer's pulse generator must be able to be driven and provide an output signal between 100 Hz and 1 GHz.

2. Customer's device must have a nominal output impedance of 50 ohms.

3. Customer's device must have a precision coaxial output connector, such as, an SMA, APC-7, Type N, APC-3.5, etc. Note, the NIST WPAS input uses a 1.85 mm (female) coaxial connector. Consequently, the measured response of the submitted device using any other coaxial connector will contain the effects of any necessary adapters.

4. Maximum input signal amplitude (including overshoot and undershoot) measurable without attenuators is 800 mV. For larger pulse amplitudes, the customer shall supply an attenuator to decrease the pulse amplitude to 800 mV or less. The permissible dc offset is ± 500 mV.

5. Minimum pulse transition duration must be greater than or equal to 5 ps.

6. Pulse duration are measured only for rectangular pulses or impulse-like pulses.

Measurements of other pulse parameters or parameter ranges may be provided by special arrangement. Consulting and advisory services also are available.

Repetitive Pulse Waveform Measurements, Including Settling Parameters (65250S)

NIST offers a special-test service for measurement of repetitive pulse waveforms whose major frequency components are below 1 GHz. Waveform measurement data can be provided on diskette, along with a report of measurement uncertainties as a function of the duration from the mesial (50%) point of the pulse transition. When required, certain derived waveform parameters can also be provided. For step-like waveforms, these include waveform settling errors, with respect to a defined reference level. For impulse-like waveforms, pulse energy into an ideal 50 load can be provided. The waveforms are measured with the NIST Sampling Comparator System described above. Waveforms within ± 2 V into 50 can be accommodated directly. Higher amplitudes require the use of external attenuators. Both 50 and 2 k attenuators are available for amplitudes up to 20 V peak; however, the 2 k attenuator substantially reduces the bandwidth of the measurement system. Typical measurement epochs range from 10 ns to 1 µs, and record lengths range from 1000 to 4000 samples.

Representative uncertainties for settling parameter measurements are listed in Table 9.22 .

Table 9.22 . Uncertainty for Measurement of Repetitive Pulse Settling Parameters

Pulse Amplitude (V)	Duration from Mesial Point (ns)	Typical Expanded Uncertainty (% of pulse amplitude)
0.25	1	1.0
	2	0.3
	5	0.1
	10	0.1
	100	0.05
	1000	0.02
0.5 to 2.0	1	0.5
	2	0.2
	4	0.1
	5	0.06
	6	0.05
	8	0.03
	10	0.02
	20	0.02
	50	0.02
	100	0.01
	1000	0.01

Restrictions and Notes :

1. All measurements are performed with a 50 input impedance. The input connector is a female SMA type. The sampling probe is connected directly to the output connector of the waveform source; no intervening cables are used unless they are specifically provided for this purpose by the customer.

2. The settling error at time t(measured from the mesial point) is defined as the largest absolute difference between the waveform and the reference level occurring in the interval from time t to the end of the data record.

3. Unless otherwise requested, the reference level is the final dc or steady state value of the final level. This level is measured by inputting a steady state logic level to the generator under test corresponding to the final level.

4. Short term settling can also be measured with respect to the final level in a specified time epoch, if requested. In this case, long term settling error-the difference between the value at the end of the specified epoch and the dc value-also will be reported.

5. Pulse generators that are internally clocked must provide a separate trigger output pulse. For best results, this should lead the wave-form under test by at least 35 ns. If the trigger pulse leads by less than 35 ns, the waveform measurement will begin one cycle later, with a resulting increase in jitter and time-quantization errors. If the pulse generator can be clocked externally, NIST will provide the clock signal and the necessary trigger output signal, when required.

6. The clock pulse requirements should be specified including high level, low level, repetition rate, and duty cycle. Repetition rates between 10 kHz and 10 MHz are preferred. Measurements of other pulse parameters or parameter ranges may be provided by special arrangement. Consulting and advisory services also are available.

Pulse Time Delay Interval (65400S)

NIST offers a special-test service for measuring pulse time delay interval, using the NIST WPAS described above. The pulse time delay interval range is 10 ps to 100 ns with typical expanded uncertainties of (2 t + 1 ps).

Restrictions:

Customer's device must utilize precision coaxial connectors for both delay ports, e.g., SMA, APC-7, Type N, APC-3.5, etc. Note, the NIST WPAS input uses a 2.4 mm (female) coaxial connector. Consequently, the measure response of the submitted device using any other coaxial connector will contain the effects of any necessary adapters.

Measurements for other ranges and configurations may be made by special arrangement. Consulting and advisory services are available.

Peak-to-Peak Detectors (65500S-65501S )

Measurements on peak-to-peak detectors are performed from 100 kHz to 500 MHz. The quantity measured by this service is the rf-ac difference defined as the percentage of difference between the rf and ac input voltages required to produce zero dc detector outputs. A 50 kHz ac reference signal is applied instead of dc. The services available are specified in Table 9.23.

Table 9.23. Measurement Ranges and Uncertainties for Peak-to-Peak Detector Services

Frequency (MHz)	Applied Peak-to-Peak Voltage for "0" Detector Output (V)	Relative Expanded Uncertainty (%)
0.1, 0.3, 1.0	1.2	0.08
3, 10	1.2	0.14
30	1.2	0.24
50	1.2	0.58
100, 200, 300	1.2	1.20
400	1.2	1.30
500	1.2	2.20

References-Impulse Spectrum Amplitude

Spectrum Amplitude Definition, Generation, and Measurement , J. R. Andrews and M. G. Arthur, Natl. Bur. Stand. (U.S.), Tech. Note 699 (Oct. 1997).

Impulse Generator Spectrum Amplitude Measurement Techniques , J. R. Andrews, IEEE Trans. Instrum. Meas., IM-25 (4), 280 (Dec. 1976).

Pulse Techniques and Apparatus, Part 1: Pulse Terms and Definitions; Part 2: Pulse Measurements and Analysis, General Considerations, IEC Publications 469-1 and 469-2, Intl. Electrotech. Com. (IEC), Switzerland (1974).

References-Fast Repetitive Pulse Transition Parameters

Temperature Effects on the High-speed Response of Digitizing Sampling Oscilloscopes , D. R. Larson and N. G. Paulter, NCSL International, 2000 Workshop and Symp., Toronto, Canada (July 2000).

The Effect of Offset Voltage on the Kick-out Pulses Used in the Nose-to-nose Sampler Impulse Response Characterization Method , D. R. Larson and N. G. Paulter, Instrum. and Meas. Tech. Conf., IMTC 2000, Baltimore, MD, pp. 1425-1428 (May 2000).

Time-Base Nonlinearity Determination Using Iterated Sine-Fit Analysis , G. N. Stenbakken and J. P. Deyst, IEEE Trans. Instrum. Meas., Vol. 47, pp. 1056-1061 (Oct. 1998).

Multiparameter Optimization of Inverse Filtering Algorithms, T. Dabozi and I. Kollar, IEEE Trans. Instrum. Meas., Vol. 45, pp. 417-421 (Apr. 1996).

Broadband Sampling Oscilloscope Characterization with the "Nose-to-nose" Calibration Procedure: A Theoretical and Practical Analysis, J. Verspecht, IEEE Trans. Instrum. Meas., Vol. 44, pp. 991-997 (Dec. 1995).

Guidelines for Evaluating and Expressing the Uncertainty of NIST Measurement Results , B. N. Taylor and C. E. Kuyatt, NIST Tech. Note 1297, U.S. Dept. of Commerce (1994).

Dynamic Calibration of Waveform Recorders and Oscilloscopes Using Pulse Standards , W. L. Gans, IEEE Trans. Instrum. Meas, 39 (6), 952 (Dec. 1990).

Characterizing High-speed Oscilloscopes, K. Rush, S. Draving, and J. Kerley, IEEE Spectrum, pp. 38-39 (Sept. 1990).

The Measurement and Deconvolution of Time Jitter in Equivalent-time Waveform Samplers , W. L. Gans, IEEE Trans. Instrum. Meas., Vol. IM-32, pp. 126-133 (Mar. 1983).

References-Repetitive Pulse Waveform Measurements, Including Settling Parameters

A Custom Integrated Circuit Comparator for High-Performance Sampling Applications , O. B. Laug, T. M. Souders, and D. R. Flach, IEEE Trans. Instrum. Meas. 41 (6), 850 (Dec. 1992).

Characterization of a Sampling Voltage Tracker for Measuring Fast, Repetitive Signals , T. M. Souders, H. K. Schoenwetter, P. S. Hetrick, IEEE Trans. Instrum. Meas. IM-36 (4), 956 (Dec. 1987).

References-Pulse Time Delay Interval

Date created: 06/30/1999
Last updated: 05/19/2009