__he Present invention relates to mass flow measurement of fluids and has particular advantage in connection with measurement of jet aircraft fuels, rocket fuels and oxidizers, cryogens generally. The invention incorporates our discovery that intrinsic charge generation transfer phenomena of such fluids and also of fluid borne solid particles media (where the particles are essentially non-conductive) can be used effectively as a means of measuring velocity of the fluid and our further invention and discovery of method steps and ap¬ paratus for making such effective utlization.

The triboelectric effect is the mechanism which gene¬ rates static electric charge when two materials rub against each other. A stored charge can be generated in particles and fluids by flow alone in a flowing stream and the charged stream can provide a readable current or voltage signal downstream.

It is a principal object of the invention to provide means for measuring velocity and mass flow of fluids as des¬ cribed above.

Further objects of the invention comprise the realization of reliable, accurate and calibratable measurements of such fluids utilizing the triboelectric effect.

SOMMAEY OP THE INVENTION

When a cryogenic liquid flows through a pipe, the fluid

SUBSTϊTUTc StfEΞT

randomly picks up charges from the pipe walls through the triboelectric effect. These charges attach themselves to atoms or molecules in the fluid and move with it without relative motion for moderate periods of time. The fluid thus becomes a moving randomly charged dielectric medium which will produce a unique voltage-versus-ti e signal in sensors capable of measuring it.

Similar charge development effects apply to gas borne particle streams. A pair of sensor electrodes, located in the walls of the pipe, (one upstream of the other) measure the voltage induced as the charged sections of fluid move past them. Since the charges move with the fluid, the voltage waveform at the downstream sensor is mainly a time-delayed version of the same, random voltage waveform at the upstream sensor. This sameness is true only within a narrow range of distance whose length depends on several characteristics of the fluid and the charge applied thereto. In order to deter¬ mine the time delay between the two signals, the voltages from the sensors are sampled by A ) converters and store in micro- porcessor memory, live time delay between the two signals is then determined by performing a digital cross-correlation on the two sets of data in memory. Given the time delay, the velocity is easily computed, since the sensor spacing is known. Cross-correlation is used to find the time delay between two signals which are time-displaced with respect to each other. The correlation calculation consists of taking samples of the two signals and processing them by shifting one rela¬ tive to the other, multiplying them together, and taking the average of the result. If the input signal is random (i.e., not periodic), the value of the cross-correlation will remain small until the imposed correlation delay cancels the original time separation. When this occurs, the cross-correlation will reach a maximum since the two signals are then identical and the correlation process effectively multiplies the signal fcy itself, thus calculating the mean squared value of the origi-

nal signal. This peak is the maximum possible result of a correlation and its location (on a scale of time shifts of the waveforms developed at the two sensors relative to each other) thus identifies the time delay (or flow movement from the upstream sensor to downstream sensor) for velocity calcula¬ tion. The availability of inexpensive high speed digital multipliers and analog to digital converters make digital cross-correlation, per se, a simple and straightforward pro¬ cess. Other objects, features and advantages will be apparent from the following detailed description of preferred embodi¬ ments thereof taken in connection with the accompanying drawings, in which:

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram representation of apparatus made in accordance with a preferred embodiment of the inven¬ tion; showing two sensors in a pipe and related block diagram of controls;

FIG. 2 is a cur rent- time trace taken as to two spaced sensor elements of the FIG. 1 apparatus; and

FIG. 3 is a flow chart of the time shifting of wave¬ forms associated with the two sensors.

DETAILED DESCRIPTION OF PREFERRED EM__CD_C__-__7ES

Referring now to FIG. 1, there is shown a preferred em- bodiment of apparatus for practice of the invention comprising a pipe form of flow path 10 (for fluid flow F) with spaced electrodes 12 and 14 therein spaced by a distance D. Read-out circuitry 16 is provided for the electrodes. Each electrode is electrically isolated from surrounding structure of the pipe to form a control capacitor with the fluid comprising the dielectric.

The smoothed, analog signal obtainable through the cir-

cuitry 16 is fed to analog/digitial converters and to a cen¬ tral signal processing facility etc. including digital multi¬ pliers, memory, high speed I/ϋ and comparator segments.

FIG. 2 is a flow chart of the cross-correlation processing method using voltage or current (via a load resistor) estab¬ lished in circuit "pipelines" of the FIG. 1 apparatus. The integrated amplitudes over a time range waveforms of the two electrodes are sampled and multiplied, then relatively shifted and multiplied in a process that is repeated iteratively until a preset number of shifts is made to cover all of adef ined range of elapsed time. For instance a data value at the upstream can be multiplied by data values at the second elec¬ trodes as taken 1, 2 , 3 , 4, 5, 6, 7, 8, 9, 10 milliseconds later and stored in memory. The memory product store will have a triangular form with a peak as determined by elapsed time. For examples:

(a) 1 2 3 4 5 6 7 8 9 10 (b) 0

In case (a) the elapsed time was 5 milliseconds, and in case (b) 3 milliseconds, both as indicated by the triangle peaking for multiplication product.

The method utilizes the fact that maximum coincidence between waveforms produces maximum multiplication product.

FIG.3 shows waveforms of a first waveform WF-1 comprising an electrical read-out of voltage (or current via an appro- priate load resistor) developed at the downstream electrode. Each such waveform has peaks (A), (B), (C) with strong corre¬ lation between the corresponding such peaks of the two wave¬ forms, over a short period of elapsed time, a few milliseconds to a few seconds depending on the flow conditions involved. Over such a short period velocity variation is negligible and a delta (^__.) T between cross-correlation peaks (A), (B), (C) of the two waveforms will, therefore, appear as a virtual constant or a linearly (or newly linearly) changing quantity

₍ ._-«-_- ^, " ^• '"-"fl*^? .* '- i____Λ-____> »

which can be averaged, to afford a reliable readout using inexpensive electronic readout equipment, simple programming instructions and with modest data bufering and storage re¬ quirements. It will now be apparent to those skilled in the art that other embodiments, improvements, details, and uses can be made consistent with the letter and spirit of the foregoing disclosure and within the scope of this patent, which is limited only by the following claims, construed in accordance with the patent law, including the doctrine of equivalents.

SUBSTITUTE £ _ _T