System for measuring the flow rate of liquid in a microfluidic pipe

Abstract

A system for measuring the flow rate in a microfluidic pipe having n pressure sensors arranged in series on a pipe and measuring the pressure Pi of the liquid passing through same. These sensors being separated from one another by pipe portions Ri. Each pipe portion having a hydraulic resistance Rhi, thus making it possible to measure the pressure variations or head loss Pi between two consecutive sensors of the liquid flowing successively through these hydraulic resistances. The comparison between the estimated flow rates Di=Pi/Ri, making it possible to determine the fouling of the microfluidic pipe.

Claims

1. A flow rate measurement system in a microfluidic pipe, comprising: n pressure sensors, with n greater than or equal to 3, arranged in series on the microfluidic pipe to measure a pressure of a liquid through the n pressure sensors, the n pressure sensors being separated from each other by n1 pipe portions of the microfluidic pipe, each pipe portion having a calibrated hydraulic resistance Rhi, respectively, each pressure sensor measures the pressure of the liquid through said each pressure sensor, a pressure sensor is arranged respectively at the two ends of said each pipe portion of the microfluidic pipe configured in series to measure pressure variations Pi between two pressure sensors at both ends of an i-th portion of the microfluidic pipe; and a processor-based computer specially configured to: measure pressure variations or pressure losses between two successive pressure sensors; calculate an apparent liquid flow rate Di in said each pipe portion according to the following relationship: $\mathrm{Di}=\mathrm{Pi}/\mathrm{Rhi};$ calculate a proportional difference Dk,j in the apparent liquid flow rates between two pipe portions according to the following relationship: $\mathrm{Dk},j=.Math.\mathrm{Dk}-\mathrm{Dj}.Math./\left(\left(\mathrm{Dk}+\mathrm{Dj}\right)/2\right);$ comparing the proportional difference in the apparent liquid flow rates between any two pipe portions to a threshold value; and wherein the processor-based computer further comprises a blockage detector configured to generate an alert signal when the proportional difference in the apparent liquid flow rates between any two pipe portions is greater than a threshold value, the alert signal indicating at least one of a partial blockage and a clogging of the microfluidic pipe.

2. The system of claim 1, wherein the processor-based computer is configured to calculate an average of the apparent liquid flow rates.

3. The system of claim 1, wherein the processor-based computer is configured to calculate an average of the apparent liquid flow rates over a largest set of the apparent liquid flow rates such that any two flow rates of the largest set have a lower proportional difference in the apparent liquid flow rate than the threshold value.

4. The system of claim 1, wherein at least one of the pipe portions has a hydraulic diameter different from that of the other pipe portions.

5. The system of claim 4, wherein two pipe portions have a hydraulic diameter ratio greater than or equal to 2.

6. A method for measuring a blockage of a microfluidic pipe, comprising: circulating a fluid through n1 pipe portions of the microfluidic pipe, n being greater than or equal to 3, arranged in series and each pipe portion having a calibrated hydraulic resistance Rhi, respectively, a pressure sensor among n pressure sensors being arranged respectively at the two ends of said each pipe portion of the microfluidic pipe configured in series to measure pressure variations Pi between two pressure sensors at both ends of an i-th portion of the microfluidic pipe; calibrating by injecting a clean liquid at a predetermined flow rate and measuring responses of the n pressure sensors with a processor-based computer to determine a pressure drop and to determine the calibrated hydraulic resistance Rhi of each of the n1 pipe portions; calculating an apparent liquid flow rate Di in said each pipe portion according to the following relationship: $\mathrm{Di}=\mathrm{Pi}/\mathrm{Rhi};$ calculating a proportional difference Dk,j in the apparent liquid flow rates, according to the following relationship: $\mathrm{Dk},j=.Math.\mathrm{Dk}-\mathrm{Dj}.Math./\left(\left(\mathrm{Dk}+\mathrm{Dj}\right)/2\right);$ calculating an average of the apparent liquid flow rates Dmean according to the following relationship: $\mathrm{Dmean}=.Math.\left(\mathrm{Di}\right)/\left(n-1\right);$ and comparing the proportional difference in the apparent liquid flow rates between any two pipe portions to a threshold value; and generating alert signal when at least one proportional difference is greater than the threshold value indicating a blockage of one of the two portions of microfluidic pipe.

7. The method of claim 6, further comprises determining a largest subset of the apparent liquid flow rates such that the proportional difference between any two apparent liquid flow rates of the largest subset is less than the threshold value; and calculating an average of the largest subset of the apparent liquid flow rates.

8. The method of claim 6, further comprises generating a pressure set point to regulate the apparent liquid flow rate in said each pipe portion by a controller in accordance with the average of the apparent liquid flow rates.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The invention will be better understood with the help of the following exemplary embodiments, given without limitation, together with the figures which represent it in the particular case of three pressure sensors and two portions of pipe (n=3):

(2) FIG. 1A is a schematic illustration of the system and method according to the invention;

(3) FIG. 1B a graphical representation of flow rate curves inside two pipe portions in FIG. 1A;

(4) FIG. 2 is a diagram of an example of the microfluidic flow rate system according to the invention; and

(5) FIG. 3 represents another embodiment of the invention.

(6) Identical, similar, or analogous elements have the same reference from one figure to another.

DETAILED DESCRIPTION OF THE EMBODIMENTS

(7) FIG. 1A schematically represents a module illustrating the principle underlying the system and method according to the invention. In FIG. 1A, a liquid flow rate 4 is sent into the microfluidic channel R1 through the pressure sensor 1 at the inlet of R1 then recovered at the outlet of R1 after passing through the pressure sensor 2. The microfluidic channels R1 and R2 are connected in series, R2 being connected to the output of the sensor 2 so as to recover the liquid flow rate from R1 and lead it to the pressure sensor 3 then to the liquid outlet 5. In this embodiment, the pipe portions R1, R2 are thus produced with the help of microfluidic channels.

(8) By means of a specially configured processor-based computer 10, the pressure differences P1 and P2 are measured between respectively the two ends of the microfluidic channels R1 and R2 having a common end at the sensor 2 and used to calculate the flow rate values of the fluid D1 and D2 in the channels R1 and R2, respectively. Under normal circumstances, in which no clogging and/or blockage of the microfluidic channels R1 and R2 has occurred, the two fluid flow rate values D1 and D2 will be equal. On the other hand, if there is a partial blockage and/or clogging of the device, this will produce a proportional difference D=|D2D1|/((D2+D1)/2) between the two calculated fluid flow rate values (D1D2) due to the change in effective hydraulic resistance of the affected channel, from which the fluid flow rate is calculated. This proportional difference D will be calculated with the help of the computer 10 specially configured to calculate P1, P2, D1 and D2 from the pressure measurements from the sensors, then compare D to a set value Dthreshold, and generate a signal alarm (optical, electrical, mechanical, etc.) which will be operated by the user, so as to allow him, if he wishes, to clean in order to measure the liquid flow rate more accurately in the microfluidic pipe circuit.

(9) Of course, there are different possible formulas for calculating the proportional difference in flow rate which are within the reach of those skilled in the art, in particular that according to which the difference in flow rates DkDj is divided by the sum of the flow rates only or more generally by X or a 1/X times the sum of the flow rates (X being an integer).

(10) Indeed, when a liquid containing debris is injected into successive pipe portions, the first pipe portions upstream of the liquid flow tend to be blocked more quickly than the pipe portions downstream of the flow. This partial blocking of the first pipe portions causes the hydraulic resistance Rhi of the first pipe portions to increase more quickly than those located downstream of the flow. Thus, we can see in the flow rate curve in FIG. 1B that clogging of the portion R1 leads to a virtual increase in the flow rate D1 relative to D2 over time, which alerts the user that the flow measurement is probably wrong. This invention therefore allows the user to check the consistency of the successive flow rates measured in the restrictions to ensure that a blockage does not occur in the system at a pipe portion.

(11) In the case where the clogging was due to the continuous deposition of a layer of material onto the walls of the channel, it is nevertheless possible that the hydraulic resistances vary concomitantly and therefore that the user is not able to see the clogging of pipes when comparing the flow rate values. In this case, it will be preferable to use channels with different hydraulic diameters. By hydraulic diameter it is meant the equivalent diameter that a circular pipe with an equivalent pressure loss would have.

(12) As the hydraulic resistance varies non-linearly with the hydraulic diameter of a pipe, opting for pipes with different hydraulic radii will allow, in the event of homogeneous clogging of the walls (linear reduction over time of the hydraulic diameter), to increase the hydraulic resistance and therefore the flow rate measured much more quickly in the pipe having a smaller hydraulic diameter. In particular, the use of two pipes, one of which has a hydraulic diameter at least 2 times smaller than the other, will make it possible to distinguish a homogeneous clogging, that is constant over time, with certainty by causing a relative variation in hydraulic resistance at least twice higher for the smaller hydraulic diameter pipe than for the larger hydraulic diameter pipe. We will thus detect flow rate difference of the order of the error caused by a homogeneous clogging. In the case of a microfluidic application, we could for example use two capillaries in series with respective internal diameters of 50 M and 150 m or two micro-channels of rectangular section, of constant height for example of 100 m and respective widths of 50 m and 200 m, to achieve these two pipe sections with significantly different hydraulic resistance.

(13) If, however, equivalent clogging was to modify the hydraulic resistances in several pipe portions in a similar manner, it will then be preferable to increase the number of these pipe portions and the pressure sensors. This general diagram is represented in FIG. 1A by the element 29 with n pressure sensors and n1 pipe portions arranged in series. By multiplying the number of flow rates measured for each of the n1 portions, the probability of having concomitant equivalent clogging is reduced and the user will therefore benefit from an even more reliable flow sensor.

(14) If the hydraulic resistance of the n1 pipe portions is not known, or varies according to the manufacturing method, it will be possible for the operator to carry out a calibration step by injecting a clean liquid (ultrapure water) at a known flow rate (for example with the help of a syringe pump) and measure with the computer the response of the n pressure sensors so as to deduce the pressure drop and therefore the hydraulic resistance of each of the n1 pipe portions. These hydraulic resistances can thus be stored in the computer memory to be used when measuring an unknown flow rate. Likewise, if a flow rate sensor has suffered a partial blockage of one of its pipes, the initial calibration step will correct the resistances of the contaminated flow rate sensor to allow subsequent reuse, for example after a cleaning step. This will therefore allow the user to make substantial savings because s/he can reuse a sensor that has been partially contaminated.

(15) FIG. 2 represents an exemplary embodiment of the microfluidic rate flow system according to the invention. The flow rate of microfluidic liquid 4 arrives at the first pressure sensor 1. This delivers a measurement of the pressure P1 via the connection 21 to the computer 10 specially configured to measure with the subassembly 22 the pressure difference P1 between the pressure P1 and the pressure P2 measured with the help of the pressure sensor 2 at the outlet of the microfluidic pipe R1 then calculate the flow rate of the fluid D1 in the pipe R1, transmitted to the computer 10 by the connection 23. The subassembly 22 transmits the value D1 via the connection 12 to the blockage detector subassembly 11.

(16) The microfluidic liquid from R1 at the sensor 2 then flows in the microfluidic pipe R2 to the third pressure sensor 3 and the liquid flow output 5. The pressure sensor 3 delivers via the connection 24 a pressure signal P3 to the subassembly 26 similar to the subassembly 22 while the sensor 2 is also connected via the connection 25 to the subassembly 26 to which it also delivers a pressure signal P2, the subassembly 26 transmitting the value D2 via the connection 13 to the blockage detector subassembly 11, which calculates the proportional difference of the flow rate values D (from D1 and D2) and compares it with Dthreshold.

(17) The computer is then able to deliver an alert signal 14 to the operator in the event that this maximum threshold is of exceeded. This alert makes it possible in particular to avoid leaving the initiative to the operator to assess the moment when the flow rates are too different and makes the use of this type of sensor more robust, particularly in production lines. The computer will also be able to carry out any form of signal processing and analysis of (averaging, integral, derivative) to detect a drift below the alert threshold but making it possible to prevent a future error and trigger preventive maintenance on the flow rate sensor.

(18) The value D2 is also sent via the connection 15 to the input 17 of the subassembly 19 which also receives on its input 18 the value D1 via the connection 16, the computer 10 performing thanks to its subassembly 19 the average of the amplitudes of D1 and D2 and delivering at the output 20 a measurement of the average flow rate of the microfluidic liquid, more precise than D1 and D2: this is also one of the advantages of the invention for which the values D1 and D2 have a double detection function of a pipe blockage and measurement of an average flow rate. It should also be noted that the operator may prefer to configure the alert at the computer level so that it is triggered when the absolute value of the flow rate difference is greater than an absolute threshold, for example 1 L/min, for a sensor that must operate over a range from 0 to 100 L/min.

(19) FIG. 3 shows an exemplary embodiment of the invention in which the measurements of pressure variations P1 and P2 are used, hence the variations in flow rate of the fluid rate D1 and D2, to generate a signal S representing the average flow rate in the microfluidic pipe in the case where a blockage is not detected by the blockage detector subassembly 11. This signal S is used by a control device 26 (subassembly of the computer 10) of the PID type (for Proportional Integral Derivative) in order to generate a desired pressure signal P to a pressure controller 27 (for example the Elveflow OB1 controller) which will control the pressure of the gaseous upper part in the reservoir 28 for the liquid injected into the microfluidic pipe 4, thus making it possible to control the flow rate of the microfluidic liquid.

(20) The measurement S of the flow rate sensor can thus be used to quickly read the liquid flow rate at a given instant and be entered into a PID control loop in order to adjust the pumping pressure applied in the reservoir 28, and thus control the flow rate in the microfluidic system.

(21) Preferably, the pressure detector at the downstream end of a pipe portion Ri is the same as the pressure detector at the inlet of the successive pipe portion Ri+1, that is to say, there is preferably a single pressure sensor between two portions of successive hydraulic resistances.

(22) The invention works by continuously reading the integrated pressure sensors and calculating the values Pi and Di with the help of a specially configured processor-based computer.

(23) Before the values Di are averaged, the difference between them is calculated. If the proportional difference Dk,j between any two flow rate values Dk and Dj among all the values Di is greater than Dthreshold (for example, 2%), it is probable that the clogging or blockage of the flow path of the micro-channel is present. This threshold is determined by the cumulative error of successive pressure sensors (e.g. measurement accuracy 0.2%), whose signals constitute the calculation of an Pi and a Di, in addition to any uncertainty linked to the dimensions of the integrated micro-channels (e.g.: a dimensional tolerance of 0.5% of the micro-machining process, because the accuracy of the flow rate calculation also depends on the accuracy of the dimensions of the micro-channels). In cases where the maximum difference between the fluid flow rate values exceeds this threshold, for example a LED light outside the sensor will turn on (controlled by the specially configured computer) to indicate the probable blockage to the user. Preferably the specially configured computer will deliver a warning message suggesting the user to perform a sensor cleaning protocol to resolve the clogging problem.

(24) These arithmetic operations (to determine the values Pi, Di, Dk,j and Dmean) will be carried out several times per second by the specially configured computer and then communicated to a software interface for controlling and recording data (not shown on the figures).