Wave transmission lines and networks – Coupling networks – Frequency domain filters utilizing only lumped parameters
Reexamination Certificate
2002-09-09
2004-05-11
Wamsley, Patrick (Department: 2819)
Wave transmission lines and networks
Coupling networks
Frequency domain filters utilizing only lumped parameters
C333S175000
Reexamination Certificate
active
06734759
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to compensation of droop caused in a transmission path.
Integrated Circuits (IC) generally need to be tested to assure proper operation. This—in particular—is required during IC development and manufacturing. In the latter case, the ICs are usually tested before final application. During test, the IC, as device under test (DUT), is exposed to various types of stimulus signals, and its responses are measured, processed and usually compared to an expected response of a good device. Automated test equipments (ATE) usually perform these tasks according to a device-specific test program. Examples for ATE are the Agilent 83000 and 93000 families of Semiconductor Test Systems of Agilent Technologies. Details of those families are also disclosed e.g. in EP-A-859318, EP-A-864977, EP-A-886214, EP-A-882991, EP-A-1092983, U.S. Pat. Nos. 5,499,248, 5,453,995.
Signals generally experience some degradation due to more or less lossy transmission paths, usually referred to as ‘Droop Effect’. In particular when reaching higher frequencies beyond several hundred MHz, such signal degradation becomes more and more important and has to be considered in design and applications.
In digital test and measurement applications, e.g. ATE applications, it has to be made sure that the rise time at the output of a transmission path stimulated by a voltage step at the input is sufficiently lower than the rise time of pulses to be measured. Otherwise, pulse rise times cannot be measured accurately.
SUMMARY OF THE INVENTION
It is an object of the invention to improve higher frequency measurement applications. The object is solved by the independent claims. Preferred embodiments are shown by the dependent claims.
According to the present invention, a filter is applied between a digital signal source and a signal receiver for providing compensation for droop caused in a transmission path between the signal source and the signal receiver.
The filter is designed to provide a high pass characteristics substantially approximating or following in a relevant frequency range an attenuation function substantially proportional to e
−k{square root over (f)}
or—when denoting attenuation in dB—substantially proportional to the square root of the frequency. Thus, droop effects dominated by skin effect in the relevant frequency range can be efficiently compensated.
In a preferred embodiment, the filter comprises a plurality of different filter stages each substantially following an attenuation characteristic characterized by asymptotic behaviors for higher and for lower frequencies and a transition behavior between the two asymptotic behaviors. The difference between the attenuation values of the asymptotes for higher and for lower (e.g. DC-attenuation) frequencies represents a measure for the attenuation of each stage and shall be referred to as the ‘stage attenuation’. It is clear, however, that this stage attenuation need not necessarily represent the maximum possible attenuation of the stage. In particular overswing behaviors, e.g. towards the asymptotes, might cause higher attenuation values than the stage attenuation.
A center frequency can be assigned to each transition behavior representing a frequency in the center of the transition behavior. The center frequency can be determined e.g. by a reversal or turning point in the transition, a mean frequency of the transition behavior range, or as the point in frequency where the stage has half of its stage attenuation. However, it is to be understood that the center frequency only represents a tool for characterizing the attenuation behavior over frequency of each stage, but is not to be interpreted in a sense e.g. of providing precisely the ‘center’ of the transition behavior range.
The plurality of filter stages is preferably designed such that the transition behaviors are distributed over the relevant frequency range. Preferably, the center frequencies of the stages are distributed over the relevant frequency range. In one embodiment, the relevant frequency range is divided into a plurality of sub-ranges and each sub-range will be dominated by the transition behavior of one stage. It is clear that a certain overlapping of the transition behaviors of different stages might occur, in particular dependent on the number and width of the sub-ranges. A higher overlapping will generally occur with increasing number of stages/sub-ranges.
The plurality of filter stages is further preferably designed such that the stage attenuation increases with increasing center frequency.
The attenuation characteristics of the different filter stages superimpose to the attenuation characteristics of the (entire) filter.
While it is clear that approximation of the square root behavior can be improved with increasing number of stages, it has to be considered that also parasitic effects generally increase with increasing number of stages. In a preferred embodiment, 3 stages are provided already allowing to sufficiently approximating the square root behavior.
In a preferred embodiment, the filter comprises a plurality of stages each having the same schematics or arrangement of components but with different component values. Thus, the characteristics off all stages are the same in principle, and the individual characteristic can be adjusted by selecting the component values. This significantly fosters designing of the filter. Preferably, the stages are arranged in series, however parallel or even mixed arrangements can also be applied.
Preferably, each stage comprises two resistors (preferably with substantially same resistance value) in series, with a third resistor being coupled between the two. A capacitor is coupled parallel to the two resistors, and an inductor is connected in series with the third resistor (preferably between the third resistor and ground). This T- or star-arrangement of resistors represents a standard attenuator circuit known in the art. Such attenuator circuit, when designed for an environment of characteristic impedance Z and terminated at its output with impedance Z, will attenuate signals from its input to output by a defined, frequency-independent amount and will not reflect any portion of the signal present at its input. Other equivalent topologies for the attenuator are commonly referred to as “Pi” and “Bridged T”.
Preferably, the capacitor and inductor values are chosen (matched?) such that L=Z
2
C. This will ensure that the capacitor and the inductor contribute to the transition behavior of the stage at the same frequency and that, over the entire frequency range, the whole stage will not reflect any portion of the signal present at its input.
The values of the resistors determine the DC-attenuation (i.e. the asymptote for lower frequencies), while the matched values of capacitor and inductor determine the transition and thus the center frequency of each stage. Such stages are preferably arranged in series.
In another preferred embodiment, the stages of the filter are designed in a combined arrangement. Preferably, the filter comprises two resistors (preferably with substantially same resistance value) in series with a third resistor coupled between the two. An inductor is connected in series with the third resistor thus representing a first stage. A first capacitor in series with a resistor is coupled in parallel to the two resistors thus representing a second stage. A second capacitor with or without a resistor in series is further coupled in parallel to the two resistors thus representing a third stage. This T- or star-arrangement of resistors represents the same standard attenuator circuit as explained above. The values of the resistors each determine the DC-attenuation (i.e. the asymptote for lower frequencies), while the values of the capacitors and the inductor each determine the transition and thus the center frequency of each stage. Contrary to the above-illustrated embodiment, it is therefore not possible to match each capacitor to an inductor (and vice versa) such that reflectio
Agilent Technologie,s Inc.
Perman & Green LLP
Wamsley Patrick
LandOfFree
Droop compensation filter having high pass characteristic does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Droop compensation filter having high pass characteristic, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Droop compensation filter having high pass characteristic will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3186076