Monitoring automatic transmission fluid condition in real time

The present invention relates to fluid condition monitoring utilizing a sensor providing an electrical signal indicating in real time the chemical condition of the fluid to be monitored. Sensors of this type are known to employ impedance spectroscopy techniques; and, an example of such a sensor is that shown and described in U.S. Pat. No. 6,278,281 Bauer, et al granted Aug. 21, 2001 in which a pair of spaced electrodes are sequentially excited at a relatively low frequency for determining the electrochemical interaction at the electrode surface and at a relatively high frequency for determining the bulk impedance of the fluid. The current is measured at both excitation frequencies and the impedance computed for each current measurement and the impedance differential computed which enables determining the fluid condition by a comparison of the computed impedance differential with that determined for known conditions of the fluid as determined by chemical analysis.

It has been proposed to use such devices for monitoring fluid condition in power transmissions and for real time monitoring of lubricating oil in combustion engines. A further example of such a sensor application for engines is that shown and described in U.S. Pat. No. 6,377,052, McGinnis, et al granted Apr. 23, 2002 in which the spaced electrodes are spirally wound on a dipstick for insertion into the engine crankcase.

Devices of the aforesaid type employing impedance spectroscopy may utilize the electrode arrangements of the type employing interdigitated planar arrays of electrodes or the aforementioned spiral arrangement or concentric radially spaced tubular electrodes such as for example those taught in U.S. Pat. No. 6,433,560 issued to Hansen et al. and granted Aug. 13, 2002.

The aforesaid Bauer, et al. patent describes in FIG. 15 thereof the impedance determined at the aforesaid high and low frequencies for automatic transmission fluid in the new condition and after a limited number of vehicle miles in service.

However, since actual vehicle service conditions depend upon the type of vehicle operation and the loading and environment during such operation, it has long been desired to provide a sensor which can provide over the service life of the vehicle a real time indication of the fluid condition based upon the actual chemical characteristics of the fluid and to indicate the amount or percentage of estimated remaining useful life (RUL) based upon the current condition of the fluid.

The present invention provides a technique for generating a continuously varying electrical signal in real time indicative of the chemical condition of the fluid being monitored and employs algorithms based upon sensor readings in samples of fluid of known chemical conditions to provide a database for comparison with the real time electrical signal for providing an indication of the remaining useful life (RUL) based upon the current condition of the fluid.

The present invention provides algorithms for determining the RUL of automatic transmission fluid, particularly fluid of the type comprising solvent dewaxed paraffinic oil in real time based upon differential impedance techniques. The present invention employs a pair of spaced electrodes configured preferably as concentrically disposed radially spaced annular electrodes for improved dispersion of the fluid over the electrode surfaces. The present invention utilizes any of three parameters derived from chemical analysis of the fluid, namely total acid number (TAN) per ASTM D-664, delta Oxidation per ASTM e-168 (ΔOX) and HPDSC induction time per ASTM D-5483 (MIN).

FIG. 1 is a pictorial block diagram of the sensor of the present invention deployed in a fluid filled transmission casing;

FIGS. 2A and 2B are a block flow diagram of the system operation for determining RUL based on TAN;

FIGS. 3A and 3B are a flow diagram similar to FIG. 2 for MIN;

FIGS. 4A and 4B are a block diagram similar to FIG. 2 for ΔOX;

FIG. 5 is a perspective view of the probe of the present invention;

FIG. 6 is a graph plotting differential impedance values as a function of TAN;

FIG. 7 is a graph similar to FIG. 6 plotting values of differential impedance as a function of ΔOX;

FIG. 8 is a graph plotting values of differential impedance as a function of MIN and,

FIG. 9 is a graph of changes with temperature of values of ΔZ.

Referring to FIGS. 1 and 5 , the sensor of the present invention is indicated generally at 10 and includes a probe assembly indicated generally at 12 immersed in fluid denoted ATF contained in an automatic transmission housing or casing 14 . The present invention is particularly useable with ATF of the type having solvent dewaxed heavy paraffinic oil as the essential ingredient

The sensor 12 may comprise any of those known in the art, as for example, an interdigitated planar array or spirally configured electrode pair; however, in the presently preferred practice the probe 12 comprises a pair of concentrically disposed radially spaced tubular or annular electrodes 16 , 18 retained in closely spaced concentric or nested arrangement by end caps 20 , 22 . Inner tubular electrode 16 has a connector terminal 24 extending axially therefrom and outwardly through a clearance slot 26 formed in header 22 ; and, similarly outer electrode 18 has a connector terminal 28 extending outwardly through slot 30 formed in cap 22 .

A temperature sensor, which may comprise a thermistor device, indicated by reference numeral 32 is disposed such that the sensing element thereof is exposed to the ATF within the casing 14 .

The inner electrode terminal 24 is connected along line 34 to receive an excitation signal from driver 36 . The connector terminal 28 is connected along line 36 to a current sensor 40 .

In the presently preferred practice of the invention, the probe 12 has the concentric electrodes 16 , 18 spaced radially a distance of about 0.15 mm for an inner electrode 18 having a diameter of about 6 mm and a length of about 38 mm. It will be understood however that other diameters and lengths may be employed to provide about the same surface area exposed between the electrodes. In the presently preferred practice the electrodes 16 , 18 are formed of stainless steel; however, other electrode materials may be employed which are compatible with the ATF. In the present practice, the invention has been found particularly suitable for use with ATF comprising solvent dewaxed heavy paraffinic oil but the invention may be employed with other types of ATF.

Referring again to FIG. 1 , the excitation driver 36 receives an input along line 42 from an oscillator 44 which is powered by an on-board vehicle supply such as the 12 Volt DC supply 46 which also supplies the excitation driver 36 along line 48 . A microcomputer 50 is powered by the power supply 46 along line 52 ; and, the microcomputer receives an input along line 54 from the current sensor 40 and a temperature input along line 56 from sensor 32 and provides an output along line 58 to an alarm or readout device 60 .

In the present practice of the invention, the oscillator 44 provides a low frequency alternating voltage of a frequency not greater than about 0.1 Hertz (100 milliHertz) and a relatively high frequency alternating voltage at a frequency not less than about 7.5 Hertz.

The microcomputer 50 is programmed with lookup tables based upon data taken from laboratory chemical aging of the ATF and determining the differential impedance at successive intervals. The fluid samples were tested to determine any one of three known test parameters, namely Total Acid Number per ASTM D-664 (TAN), delta Oxidation per ASTM E168 (ΔOX) and HPDSC induction time per ASTM D-5483 (MIN). The data is then plotted for each of the parameters and curves drawn therebetween as displayed respectively in FIGS. 6 through 8 . The graphs include data points taken for ATF stressed by laboratory oxidation aging tests, such as an Aluminum Beaker Oxidation Test (ABOT) per Southwest Research Institute, San Antonio, Tex. procedure BJ110-4 and some ATF fluid drained from vehicles in actual road service. It will be noted from FIGS. 6 through 8 that the linear approximations may be made for the data; and, algorithms for the slope used to calculate the respective chemical parameter for valves of ΔZ TC are indicated on the graphs.

Referring to FIG. 2 , the operation of the system circuitry is denoted generally by reference numeral 62 ; and, upon user initiation at step 64 the system proceeds to step 68 . At step 68 the system inquires as to whether the fluid temperature T f is within desired limits T MIN , T MAX ; and, if the answer is negative the system proceeds to abort or Stop. However, if the determination at step 68 is affirmative, the system proceeds to step 70 and excites the probe 32 with an alternating voltage at a relatively high frequency and measures the current I HI . The system then proceeds to step 72 and computes and stores the impedance Z HI from the measured current I HI .

The system then excites the probe 32 with a relatively low frequency alternating voltage and measures the current I LO and proceeds to step 76 and computes and stores the impedance Z LO from the measured current I LO .

The system then proceeds to step 78 and computes the vector {right arrow over (ΔZ T )} by subtracting {right arrow over (Z LO )} from {right arrow over (Z HI )} yielding the vector result {right arrow over (ΔZ T )}. The system then proceeds to step 80 and computes the absolute value of ΔZ T and proceeds to step 82 and determines ΔZ TC the temperature compensated impedance differential from a lookup table of values of ΔZ TC versus temperature, which table is compiled by taking data points from a temperature correction curve. Typical curves for such temperature compensation are shown in FIG. 9 .

It will be seen that a family of three plots; namely on upper graph:

ΔZ TC =−1.59 E +04 *T +1.92 E +06,

a lower graph:

ΔZ TC =−9.98 E +03 *T +1.18 E +06

and a middle graph:

ΔZ TC =−1.16 E +04 *T +1.39 E +06

plotted by interpolating between the upper and lower graph are presented in FIG. 9 . It will be noted that the shapes and intercepts of the three graphs are similar; and, thus provide a region from which ΔZ TC may be computed.

The system then proceeds to store the value determined at step 82 as ΔZ TC 1 at step 84 ; and, after a suitable time delay of not more than about 10 seconds at step 86 the system proceeds to step 88 and repeats steps 70 through 82 and stores the result as ΔZ TC 2 at step 90 .

The system then proceeds to step 92 , computes the change in ΔZ denoted ΔΔZ by subtracting ΔZ TC 1 from ΔZ TC 2 and proceeds to step 94 and inquires as to whether ΔΔZ is positive.

If the determination at step 94 is affirmative the system proceeds to step 98 and computes TAN from a lookup table of values of TAN versus ΔZ TC based upon the algorithm from FIG. 6 :

ΔZ TC =2.77 E +05*TAN−2.7 E +05

If however, the determination at step 94 is negative, the system proceeds to step 96 and repeats steps 70 through 94 .

After completion of any of step 98 , the system proceeds to step 100 and stores the determined value as TAN 1 and proceeds to step 102 for a time delay ΔT of not less than about one hour. The system then proceeds to step 104 and repeats steps 70 through 102 and stores the result as TAN 2 at step 106 . The system then proceeds to step 108 and computes the rate of decay ψ by subtracting TAN 1 from TAN 2 and dividing the differential by ΔT. The system then proceeds to step 110 and recalls a stored value TAN EOL and then proceeds to step 112 and computes the remaining useful life in hours (RUL) by subtracting TAN EOL from TAN 2 and dividing the differential by ψ. The system then displays the computed value of RUL at step 114 .

Referring to FIG. 3 , the flow diagram for determining RUL from the parameter HPDSC induction time per ASTM D-5483 (MIN) is shown wherein the system, upon initiation at step 116 proceeds to read the fluid temperature at step 118 from the sensor 32 . The system then proceeds to step 120 and asks whether the temperature read at T f read at step 118 is between the limits T MIN , T MAX ; and, if the determination is negative the system proceeds to abort or Stop. However, if the determination at step 120 is affirmative the system proceeds to step 122 and excites the probe at the relatively high frequency alternating voltage and measures the current I HI . The system then proceeds to step 124 , computes and stores the impedance Z HI computed from the current measured at step 122 .

The system then proceeds to step 126 and excites the probe at the relatively low frequency alternating voltage and measures the current I LO . The system then proceeds to step 128 and computes and stores the impedance Z LO from the current measured in step 126 .

The system then proceeds to step 130 , computes the vector difference {right arrow over (ΔZ t )} by subtracting {right arrow over (Z LO )} from {right arrow over (Z HI )} and proceeds to step 132 and determines the absolute value ΔZ T .

The system then proceeds to step 134 and determines ΔZ TC from a lookup table of ΔZ T versus Temperature which lookup table is determined from data points taken from curves such as those shown in FIG. 9 which identify the change in the differential impedance with temperature for samples of fluid of known condition. The procedure is the same as for step 82 .

The system then proceeds to step 136 and stores the value ΔZ TC 1 computed at step 134 and proceeds to step 138 for a time delay of not more than about ten seconds. The system then proceeds to step 140 and repeats steps 122 through 134 and stores the computed value ΔZ TC 2 at step 142 .

The system then proceeds to compute the change in ΔZ TC denoted ΔΔZ by subtracting ΔZ TC 1 from ΔZ TC 2 . The system then proceeds to step 146 and asks whether ΔΔZ is positive; and, if the answer is affirmative the system proceeds to step 148 and repeats steps 122 through 146 . However, if the determination at step 146 is negative, the system proceeds to step 150 and asks whether ΔZ TC is equal to or greater than 3.4E+05. If the query in step 150 is answered in the affirmative, the system proceeds to step 152 and determines MIN from a lookup table of values of MIN versus ΔZ TC compiled from the graph of FIG. 8 using the algorithm:

ΔZ TC =−2.78 E +05*MIN+2.95 E +06.

However, if the determination at step 150 is negative the system proceeds to step 154 and determines MIN from a lookup table of values of ΔZ TC versus MIN compiled from the graph of FIG. 8 using the algorithm:

ΔZ TC =−1.98 E +04*MIN+5.26 E +05.

Upon completion of one of the steps 152 or 154 the system proceeds to step 156 and stores the determined value of MIN as MIN 1 and proceeds to step 158 for a time delay ΔT of not less than about one hour and then proceeds to step 160 and repeats steps 122 through 154 . The value of MIN determined at step 160 is then stored as MIN 2 at step 162 and the system proceeds to step 164 and computed the rate of decay by ψ determined by subtracting MIN 1 from MIN 2 and dividing the differential by ΔT.

The system then proceeds to step 168 to get a stored value of MIN EOL and proceeds to step 170 and computed the remaining useful life RUL by subtracting MIN EOL from MIN 2 and dividing the differential by ψ as determined in step 164 . The system then proceeds to display the computed value of RUL at step 172 .

Referring to FIG. 4 , the flow diagram for determining RUL from the parameter Delta Oxidation per ASTM E-168 (ΔOX) is shown wherein the system, upon user initiation at step 174 , proceeds to read the fluid temperature at step 176 and then proceeds to step 178 to determine if temperature T f is within the limits T MIN , T MAX . If the determination at step 178 is negative the system aborts or proceeds to Stop. However if the determination at step 178 is affirmative, the system proceeds to step 180 to excite the probe 32 with a relatively high frequency alternating voltage and measures the resultant current I HI . The system then proceeds to step 182 , computes the impedance Z HI from the measured current and stores the computed value. The system then proceeds to step 184 and excites the probe 12 with a relatively low frequency alternating voltage and measures the resultant current I LO and proceeds to step 186 and computes and stores the impedance Z LO from the measured current I LO .

The system then proceeds to step 190 and computes the impedance vector differential {right arrow over (ΔZ t )} by subtracting {right arrow over (Z LO )} from {right arrow over (Z HI )} and then determines the absolute value of the computed differential ΔZ T at step 192 .

The system then proceeds to step 194 and determines the temperature compensated value ΔZ TC , compiled from data points taken from curves such as shown FIG. 9 from the lookup table of values of ΔZ TC versus temperature and, the system then proceeds to step 196 and stores the computed value as ΔZ TC 1 .

The system then proceeds to step 198 and provides a time delay of not more than about 10 seconds and then proceeds to step 200 and repeats steps 180 through 194 and stores the computed value as ΔZ TC 2 at step 202 .

The system then computes the change ΔΔZ in the differential impedance ΔZ by subtracting ΔZ TC 1 from ΔZ TC 2 at step 204 and proceeds to step 206 and asks the question whether ΔΔZ is positive. If the answer to the query in step 206 is negative, the system proceeds to step 208 and repeats steps 180 through 204 . If the query in step 206 is answered in the affirmative, the system proceeds to step 210 and asks whether ΔZ TC is equal to or less than 3.40E+05. If the determination in step 210 is affirmative, the system proceeds to step 212 and determines from a lookup table the values of ΔOX versus ΔZ TC compiled from the graph of FIG. 7 using the algorithm:

ΔZ TC =1.35 E +04 *ΔOX +1.32 E +05.

If the system answers in the negative at step 210 , the system proceeds to step 214 and determines ΔOX from a lookup table of values of ΔOX versus ΔZ TC based on the graph of FIG. 7 using the algorithm:

ΔZ TC =2.67 E +04 *ΔOX −6.86 E +04.

After completing one of the operations 214 , 212 , the system proceeds to step 216 and stores the result as ΔOX 1 and proceeds to execute a time delay of not less than about one hour at step 218 . The system then proceeds to repeat steps 180 through 214 at step 220 and stores the result as ΔOX 2 at step 224 .

The system then proceeds to step 226 and computes the difference ψ of the values of ΔOX by subtracting ΔOX 1 from ΔOX 2 and dividing the result by ΔT. The system then proceeds to step 228 and recalls a stored value of ΔOX EOL and proceeds to step 230 to compute the remaining useful life (RUL) by subtracting ΔOX EOL from ΔOX 2 and dividing the result by the computed value of ψ and proceeds to step 232 to display the value of RUL.

Irrespective of which of TAN, ΔOX or MIN is calculated, if ΔZ TC is measured equal or greater than 6.5×10 5 , the ATF fluid is deemed to have reached the end of its useful life.

Although the invention has hereinabove been described with respect to the illustrated embodiments, it will be understood that the invention is capable of modification and variation and is limited only by the following claims.