Browse Prior Art Database

Accurate, Time Stable Measurement Technique for Observing Repetitive Noisy Picosecond Pulses

IP.com Disclosure Number: IPCOM000081265D
Original Publication Date: 1974-Apr-01
Included in the Prior Art Database: 2005-Feb-27
Document File: 3 page(s) / 41K

Publishing Venue

IBM

Related People

Elliott, BJ: AUTHOR

Abstract

The above-referenced patent allows accurate measurement of repetitive picosecond pulse waveforms. The distorting effects of slow-time drift and fast-time jitter are effectively eliminated, by using multiple sampling methods and time drift control. However, in the patent, the data acquisition rate is intrinsically slow (the second sampler takes only one sample per analog output waveform of the first sampler). And, the system is limited to relatively large signals (> 20 mV), which implies that a reasonable input signal-to-noise ratio (say 5 to 1) is required as well as low Y vertical position drift.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 3

Accurate, Time Stable Measurement Technique for Observing Repetitive Noisy Picosecond Pulses

The above-referenced patent allows accurate measurement of repetitive picosecond pulse waveforms. The distorting effects of slow-time drift and fast- time jitter are effectively eliminated, by using multiple sampling methods and time drift control. However, in the patent, the data acquisition rate is intrinsically slow (the second sampler takes only one sample per analog output waveform of the first sampler). And, the system is limited to relatively large signals (> 20 mV), which implies that a reasonable input signal-to-noise ratio (say 5 to 1) is required as well as low Y vertical position drift.

In the present arrangement, the data rate is typically 256 to 1024 times faster; while signals which are 3 orders of magnitude smaller (approx. 20 Mu volts) and much below the input noise level (approx. 5mV) can be observed; Y drift effects are also eliminated.

The present approach uses a commercial signal averager instead of a second sampler to process the output analog signal from the first sampler. FABRI-TEK* Model 1010, or the ENHANCETRON** Model 801, Signal Averagers which are commercially available, take a sequence (256 to 1024) of amplitudes at successive, narrow time "slots" of the analog waveform, which has already been slowed down from the picosecond time domain (i.e., for the signal input) to the millisecond domain by the sampler which is so designated in Fig. 1. The signal averager improves the signal-to-noise ratio of the input signal by up to approx. 70 db, provided the noise is incoherent and provided the sampler has very high time stability, i.e., the time drift is sufficiently small. In practice, commercially available samplers are not sufficiently time stable and so for the fastest waveforms and for greatest accuracy, corrective techniques must be employed. Two modes are shown: (i) when the signal is large enough (> 20 mV) (Fig. 1), (ii) when the signal is too...