Astm ds57 1975

LOW-TEMPERATURE PUMPABILITY CHARACTERISTICS OF ENGINE OILS IN FULL-SCALE ENGINES # Prepared by the Data Analysis Panel of ASTM Committee D-2 R&D Division Vll-C on Non-Newtonian, Low-

LOW-TEMPERATURE PUMPABILITY CHARACTERISTICS OF ENGINE OILS

IN FULL-SCALE ENGINES

#

Prepared by the Data Analysis Panel

of ASTM Committee D-2 R&D Division Vll-C

on Non-Newtonian, Low-Shear Phenomena

ASTM DATA SERIES PUBLICATION 57

List Price $16.00 05-057000-12

AMERICAN SOCIETY FOR TESTING AND MATERIALS

1916 Race Street, Philadelphia, Pa 19103

©by AMERICAN SOCIETY FOR TESTING AND MATERIALS 1975 Library of Congress Catalog Card Number: 75-24603

NOTE The Society is not responsible, as a body, for the statements and opinions advanced in this publication

Printed in Long Island City, N.Y

September 1975

Data Series DS 57 American Society for Testing and Materials

Related ASTM Publications

Multicylinder Test Sequences for Evaluating Automotive Engine Oils, STP315F (1972), $8.25, 04-315060-12

Shear Stability of Multigrade Crankcase Oil—IP Fleet Tests, DS 49-S1

(1974), $4.00, 05-049001-12 Single Cylinder Engine Tests for Evaluating Performance of Crankcase Lubricants, STP 509 (1972), $5.00, 04-509000-12

Editorial Staff

Jane B Wheeler, Managing Editor Helen M Hoersch, Associate Editor Charlotte E Wilson, Senior Assistant Editor Ellen J McGlinchey, Assistant Editor

SUMMARY

Low-temperature engine oil pumpability data have been obtained on thirteen ASTM Pumpability Reference Oils in seven full-scale test engines Border- line Pumping Temperatures based on gallery oil pressure traces were deter- mined for all thirteen Reference Oils in four of the test engines, and for nine of the Reference Oils in all seven test engines Data were also obtained as to the type of flow failure occurring (air-binding or flow- limited) and on rocker arm oiling times

The results indicate substantial differences in pumpability among both test oils and test engines Possible reasons for these differences are analyzed, and a model, based on engine oil pump inlet system dimensions,

is proposed This model accounts for some of the differences found in engine severity and the type of flow failure which occurs

The pumpability data reported should provide the basis for development of

a laboratory bench test to predict engine oil pumpability Results from various bench tests and engine pumping rigs and their correlation with the full-scale engine data will be summarized in a later report

INTRODUCTION AND BACKGROUND

The current ASTM Committee D-2 Research and Development Division VII-C Task Force Program to evaluate the low-temperature pumpability properties

of thirteen ASTM reference oils in seven full-scale test engines, five engine pumpability rigs, and a variety of bench tests was initiated at the request of SAE

In 1970, a Task Group of SAE Fuels and Lubricants Subcommittee 2 completed

an inquiry to determine the extent of the low-temperature pumpability

problem Based on the results of the survey, Subcommittee 2 requested (1)* ASTM to "develop a simple laboratory test to measure the ability of an oil to flow to the engine oil pump inlet in a manner relatable to engine conditions, and to consider the relationship between low-temperature oil properties and the supply from the pump to critical engine parts."

In response to the SAE request, ASTM Task Force members conducted ex-

ploratory tests in engines, engine pumping rigs, and laboratory bench rigs

in an effort to obtain enough information to design a comprehensive coop- erative program The ASTM Task Force concluded early in the program that they would have to obtain extensive full-scale engine data to serve as the basis for evaluating the ability of bench tests and engine pumping rigs to predict engine oil pumpability

* Numbers in parentheses designate References at end of report

DS57-EB/Sep 1975

Copyright © 1975 by ASTM International www.astm.org

A literature search conducted early in 1970 revealed that no usable engine pumpability data existed In 1961, R G Moyer (2) conducted tests on an automotive oil pump rig His results showed that pumping failures occurred

by air binding, and that an oil's pumpability in his rig could not be

predicted by its 0°F (-17.8°C) viscosity, ASTM Pour Point, or channel point

In 1963, T W Selby (3) evaluated the pumpability problem and concluded that existing shear rates in different parts of the engine and the vis-

cosities of the Vl-improved oils at the existing shear rates determined

where the oil failed to flow fast enough to satisfy the engine's demands The first-engine results were reported in 1971 by Smith and Graham (4),

who showed that rocker arm oiling times could not be predicted by ASTM Pour Point, Brookfield viscosity, or Cold Cranking Simulator viscosity They found that some used oils were more pumpable than new oils and that engine design was an important factor in pumpability Boone, Crouse, and McLaughlin (5), however, found some indications that a tilt-can pour test could be

used to predict an oil's rocker arm oiling time, but found no correlation between polymer type and low-temperature fluidity In 1972, Stewart and Spohn (6) motored an engine at 1650 rpm in a cold room at 0°F (-17.8°C) and -20°F (-28.9°C) using twelve commercial oils and nine laboratory blends

to determine rocker arm oiling times and times to achieve normal gallery oil pressure They showed that oil temperature, oil pressure, engine speed, valve lifter type, flapper valve plate removal, and fuel dilution affected rocker arm oiling time, and that the time required to achieve normal gallery oil pressure was affected by oil temperature, oil screen removal, engine speed, and VI improver type They also found that neither the rocker arm oiling time nor the time to achieve normal gallery pressure could be pre- dicted by Cold Cranking Simulator viscosity, Brookfield viscosity, tilt- bottle pour time, or ASTM Pour Point

Because of the confusion and lack of agreement which existed in 1972 con- cerning the scope and causes of the low-temperature pumpability problem,

a joint ASTM-SAE Symposium was organized and held in May of 1973 to help reconcile the differences and to further define the problem Four papers (7-10) were presented in which engine studies, laboratory tests, and theo- retical considerations were used to point out specific examples of low-

temperature pumpability problems and to identify several important vari- ables

Following the Symposium in 1973, the ASTM Low-Temperature Engine Oil

Pumpability Task Force finalized their plans for a cooperative program and selected thirteen Pumpability Reference Oils (PRO's) These oils were

supplied to all of the participants for their various evaluations

The ASTM Task Force has completed the full-scale engine evaluations of

the thirteen PRO's, but the bench tests and the engine pumping rig tests are not yet completed Therefore, this report summarizes only the full- scale engine data A follow-up report will summarize the bench test and engine pumping rig data and evaluate their ability to predict engine oil pumpability in full-scale engines

Considerable test work by many different laboratories was required to obtain the data summarized in this report The laboratories which have partici- pated in the various phases of the pumpability program are summarized in Appendix A, Table A-l The Data Analysis Panel members are also listed in this table

PROGRAM ASTM Pumpability Reference Oils (PRO's)

To ensure the availability of oils which could be used in engine tests and later in laboratory bench tests, a series of Pumpability Reference Oils was established The following criteria were used in formulating the Ref- erence Oils:

• Inclusion of commonly used types of viscosity index

(VI) improvers commercially available in 1973

• Inclusion of Vl-improved oils which cover the widest

SAE viscosity grade ranges commercially available

• Inclusion of one set of oils containing the same VI

improver, but different base stock wax levels

• Inclusion of two oils containing no VI improver, and

formulated so as to be Newtonian in the temperature

range of interest

• Inclusion of a commercial oil known to fail by air

binding in a specific engine

• Inclusion of one oil blended specifically to be soak-

time sensitive

• Inclusion of a standard additive package in all non-

commercial oils formulated for this program (excluding

the Newtonian oils)

The thirteen PRO's along with their SAE viscosity grade, coded VI improver type, and base stock wax level are shown in Table 1 New oil inspection data, including kinematic viscosities, Cold Cranking Simulator viscosities, Brookfield viscosities, and ASTM Pour Points are shown in Table 2

engine is shown in Table 3 A wide variety was included to ensure that

pumpability data obtained would be representative of the commercial market, and to minimize the possibility of overlooking pumpability effects asso- ciated with a specific engine type

Diagrams and/or photographs of the overall lubrication scheme and the engine oil pump inlet system for each of the six different engine types are in- cluded in Appendix B

Pumpability Criteria

Definition of Pumping Conditions - In the results to be presented and discussed, PRO's were run in each of the engines at various temperatures and the results classified in terms of the gallery oil pressure-time re- sponse as illustrated in Figure 1

The pumpability condition was classified as:

• Normal if gallery oil pressure was greater than 20 psig

(138 kPa) at all times after one minute of operation

• Borderline if gallery oil pressure was equal to or less

than 20 psig (138 kPa) but greater than 6 psig (41 kPa)

at any time after one minute of operation

• No Pump if gallery oil pressure was 6 psig (41 kPa) or

less at any time after one minute of operation

Some typical pressure-time curves for different oil-engine combinations

are also shown in Figure 1 to illustrate the three types of pumpability In all cases, the minimum pressure after one minute of test, not the shape of the curve, determines the pumpability category

The choice of 20 psig (138 kPa) as the upper limit for the borderline region was based on reports of engine noise occurring at oil pressures below 20 psig (138 kPa) The value of 6 psig (41 kPa) was chosen as the lower limit since pressures below this will usually activate the oil pressure indicator light and may cause engine damage A time of one minute was chosen in all cases since this was thought to be a reasonable time for full engine oil pressure

to be developed.*

Determination of Borderline Pumping Temperature - The Borderline Pump- ing Temperature (BPT) for any oil-engine combination is the temperature at which the minimum gallery pressure at any time after one minute of test

is 20 psig (138 kPa) It is usually determined by interpolation from the pressure-time traces of tests at several temperatures

Rather than superimposing oil pressure-time traces on a grid such as out- lined in Figure 1 and estimating the BPT, a more definitive method was used

Figures 2A and 2B show two types of pressure-time responses In each,

curves for two temperatures are shown, one curve illustrating a borderline condition and the other a normal pumpability condition Data such as shown

by the curves of Figure 2A were treated by plotting the minimum pressure after one minute, defined as shown in Figure 2A, as a function of test

temperature for several different temperatures A smooth curve such as

shown in Figure 3A was drawn through the resulting points, and the temper- ature corresponding to 20 psig (138 kPa) was read from the curve This

temperature corresponds to the Borderline Pumping Temperature (BPT) The BPT determined from Figure 3A is -19°F (-28°C) For those tests in which the "normal" pressure-time trace showed a continuous rise with no dip,

the pressure at one minute was taken as the minimum

Pressure-time responses such as those in Figure 2B, which showed only a

steep rise in pressure and which varied with temperature mainly by the

time at which the steep rise occurred, were treated by plotting the time to reach 20 psig (138 kPa) at each temperature as a function of temperature From a smooth curve drawn through the data points, as illustrated in Fig- ure 3B, the BPT corresponding to a time of one minute was determined The BPT determined from Figure 3B is -25°F (-32°C)

All BPT's in each of the engines were determined by one of the two methods illustrated in Figures 3A or 3B

Test Procedure

The general procedure is outlined briefly here Further details are pro- vided in Appendix C

Each engine was mounted on a test stand in a cold chamber in such a way that

it could be motored at 1600 rpm for the test After the engine was flushed twice and charged with the oil to be tested, it was motored at 1600 rpm for ten minutes, or until the temperature reached 180°F (82.2°C) It was then soaked at the test temperature for 16 hours

For the actual pumpability test, the engine was motored at 1600 rpm for ten minutes, during which oil pressures and temperatures were taken at various points Visual observation of the oil-sump surface to determine the type

of pumpability failure, as well as a determination of rocker arm oiling

times were included for most of the engines when possible Since the primary objective of the engine tests was to determine the Borderline Pumping Tem- perature for each oil, this procedure was repeated at various temperatures until at least one borderline pumping condition was achieved Usually three

to five tests were required to reach or bracket the BPT

RESULTS AND DISCUSSION Engine Borderline Pumping Temperatures and Pumpability Failure Types

A summary of the Borderline Pumping Temperatures and failure types deter- mined in this program is shown in Table 4 In addition, the number of

failures, by each of the two failure types is listed for each oil in the

two right hand columns The results in this table were the basis for all statistical analyses included in the report

A complete summary of each test run on each oil in each engine is given in Tables D-l through D-7 of Appendix D These data were used to determine the BPT's listed in Table 4 Also included in Appendix D is a separate

listing of the flow-limited and air-binding results, Tables D-8 and D-9 While these tables were derived directly from Table 4, separation of the two failure types may be useful in the future development of laboratory

tests

Ranking of Oils

The results in Table 4 indicate wide differences in the BPT's among oils

in the same engine as well as among engines However, the relative pump- ability characteristics of the PRO's are nearly the same in six of the

seven engines, as may be seen by ranking the oils as in Figure 4 For

example, PRO's 05, 06, and 16 have poor pumpability performance in all

engines except 0-240, while PRO's 01 and 15 have good pumpability perform- ance in all engines Thus, while absolute values of the BPT vary greatly among engines, the relative pumpability ranking is essentially the same

in six of the engines Oils which fail by air binding are denoted by a

box in Figure 4

Analyses of Pumpability Data

Since all thirteen PRO's were not evaluated in all seven engines, the data were analyzed in two groups

The BPT's obtained for the nine oils run in all seven engines are listed

in Table 5 along with an analysis of variance of the data The engine

data are listed from left to right in order of decreasing severity based

on the nine-oil averages for each engine The analysis of variance re-

sults show that both the oils and the engines are highly significant factors with regard to pumpability Prior to this program, differences among engines were assumed to be considerably smaller than differences among oils

The BPT's obtained for the thirteen oils that were evaluated in only four

of the engines are shown in Table 6 along with an analysis of variance

of the data Again, the oils and the engines are highly significant factors Supplemental statistical analyses, to determine significant differences

among Reference Oil and engine BPT averages, were also performed These results are included in Appendix E

Estimation of Seven-Engine Averages

Since all thirteen of the PRO's were not evaluated in all seven engines, a method was devised to estimate seven-engine average BPT's for those four oils that were evaluated in only four engines A linear regression

analysis was conducted relating the four-engine average data to the

seven-engine average results for the nine oils evaluated in all seven

engines As shown in Figure 5, the two results correlated very well

The equation derived from this regression was used to calculate seven-

engine averages from the four-engine averages for the four oils that were evaluated in only four engines These estimated values are shown in

Table 7 along with the original four-engine and seven-engine average re-

sults No information, however, as to the type of flow failure which

occurs can be obtained by this estimation procedure

Correlating Individual Engine Results with the Estimated Seven-Engine Averages The BPT's for each engine were regressed against the estimated seven-engine average results to determine the degree of correlation The results of

these regression analyses are shown in Table 8 The correlations are also shown graphically in Figures 6 through 12 Of the seven engines, only the most severe engine (L-235) and the mildest engine (0-240) failed to correlate well with the estimated seven-engine averages From the graphs it can be seen that PRO-05 generally deviated from the regression line more than any other oil However, for engines L-235 and 0-240 (the most severe and mildest engines) several other oils also contributed substantially to their poor

correlation coefficients The regression lines plotted in Figures 6 through

12 are those determined by including all of the oils tested in each engine

in the regression analyses

Agreement Between Duplicate Engines

Nine of the PRO's were run in duplicate N-235 engines at two different

laboratories The Borderline Pumping Temperatures obtained at each of the two laboratories are listed in Table 9 along with the difference between

the BPT values obtained for each oil The average difference between the results obtained in the duplicate engines was 2°F (1°C), and in no case

was the difference greater than 4°F (2°C) While the data obtained in this program are insufficient to determine the reproducibility of the test

method employed, the small differences in Table 9 suggest that the larger differences shown in Tables 4 to 7 are probably due to real differences

in the ability of the engines to pump the oils, and not to differences

among laboratories

Effect of Soak Time on Pumpability

Because other investigators (7,9) had reported differences in pumpability and low-temperature viscosity due to differences in soak time at a con-

stant temperature, one of the oils, PR0-11, was blended specifically to

be soak-time-sensitive Borderline Pumping Temperatures after 16 and

64-hour soak periods were determined in five of the seven engines, and

the results are included in Table 10 While BPT's after 64 hours are

greater than those after 16 hours, the differences were not as large as

expected This is probably because PRO-11 was not as soak-time-sensitive

as had originally been hoped for, its Brookfield viscosity after 64 hours being only two or three times greater than that after 16 hours (12)

Both McMillan and Murphy (7) and Spohn and Stewart (11) found that dif-

ferences in pumpability of 10°F (5.5°C) or more can occur because of

soak time differences when the ratio of the Brookfield viscosity after

a long soak period to that after a short soak period approaches ten Thus, for some oils, soak time effects may be greater than those found here with PRO-11

Relationship Between Pumpability and SAE Viscosity Grade

One measure of the general significance of the pumpability results of this program is available from a comparison of the BPT data with the Cold Crank- ing Simulator (CCS) viscosities The latter are related to engine cranking resistances and form the basis for the SAE winter grade classifications, 5W, 10W, and 20W This comparison is made in Figure 13 for one of the four engines (engine N-114) in which all thirteen PRO's were tested The results

in Figure 13, which are typical of those obtained in the other three engines also, show clearly that the CCS viscosity is not a reliable indication of low-temperature pumpability Consequently, SAE viscosity grade, which is determined by the CCS viscosity at 0°F (-17.8°C), is also inadequate for predicting pumpability

Another way of viewing the significance of the pumpability results is to compare the BPT's obtained in each of the engines with the corresponding engine manufacturers' recommendations Making such a comparison indicates that for ten of the eleven SAE 10W and 20W oils, the Borderline Pumping Tem- peratures were below the minimum ambient temperatures recommended in the

1975 owners' manuals However, one oil, PRO-05, had unsatisfactory pump- ability characteristics in six of the seven engines Thus, pumpability could limit an oil's acceptability if pumpability standards were presently available

Pumpability Failure Types

Throughout this report a distinction has been made between the two types of pumpability failure, air binding and flow limited While such a distinction may not seem important in the sense that, no matter what the failure type, engine oil pressure is lost, an understanding of the conditions surround- ing a failure is important in assessing the factors pertinent to the

selection of a laboratory test to predict pumpability For this reason, care was taken to identify the type of pumpability failure which occurred

in the various engines

Description of Pumpability Failure Types - As indicated in Table 4, two types of flow failure were obtained - air binding and flow limited These conditions can be interpreted with the aid of Figure 14 To ensure

a continuous supply of oil to the oil pump, the oil must be able to flow both t£ the oil screen and through the screen and pump inlet tube

Air binding occurs when the first of these two requirements, flow to the oil screen, is not met Oils subject to this condition become so gelled that they cannot be drawn to the oil screen fast enough to prevent a hole from forming between the oil-sump surface and the oil screen This con- dition once encountered usually persists until the oil slumping into the

hole combined with warm oil returning from the overhead system is sufficient

to fill the void and prevent air from being drawn into the pump

A flow-limited condition occurs when the second of these two requirements (flow through the screen and pump inlet tube) is insufficient In this case, the screen remains completely covered, but the oil is so viscous

that it cannot flow through the restrictions imposed by the screen and

inlet tube at a rate sufficient to satisfy pump demands This results in

a lower than normal pump outlet pressure, and usually persists until warmer, less viscous oil begins returning from the overhead system

Typical Curves - Oil pressure-time traces, typical of the various types

of pumpability failure just described, are shown in Figure 15 At a suit- ably high temperature, all oils perform as indicated by the curve labeled

"satisfactory," the pressure rising rapidly to a normal operating level which is maintained throughout the test At lower temperatures, however, air-binding or flow-limited conditions airse, depending on the combined oil and engine characteristics

Two types of flow-limited behavior are depicted in Figure 15 The Flow- Limited-I curve is typical of the pressure response which occurs in all of the engines investigated except L-235 An initial peak pressure is reached, after which the pressure drops because of inlet tube restrictions After reaching some minimum level, the pressure slowly returns to its normal

operating level

In the Flow-Limited-II case, typical of runs in the L-235 engine, no peak

or minimum pressure occurs The borderline condition is obtained, not

when the pressure falls back into the borderline region, but rather when the pressure does not reach 20 psig (138 kPa) after one minute of operation While the curves for the Flow-Limited-I and Flow-Limited-II cases illus- trated in Figure 15 appear quite different, both arise because of restric- tions within the oil screen and pump inlet tube, rather than by flow to the oil screen

The air-binding curve shown in Figure 15 has the same overall character- istics as the Flow-Limited-I curve, in that a peak pressure is reached fol- lowed by a rapid decline to some unsatisfactory minimum level The rapid pressure fluctuations, however, are characteristic of the air-binding case only, occurring as the pump receives oil and air intermittently These fluctuations were used to distinguish between air-binding and flow-limited conditions when visual observations were not made

Relating Pumpability to Inlet System Geometry

The results of this study have shown that both oil and engine effects are important factors in pumpability considerations While several investi- gations have already shed light on the variables controlling the oils'

influence on pumpability, engine effects have been relatively ignored

As mentioned at the outset, the key to good engine oil pumpability is keep- ing an adequate supply of engine oil available at the pump inlet Thus,

strictly from geometrical considerations of an oil pump inlet system, three factors are important:

• oil inlet screen mesh size

• oil inlet tube dimensions (length and diameter)

• oil pump location (submerged or not)

These factors are listed in Table 11 for each of the six different engines investigated Fine mesh screens, long or small diameter inlet tubes, and oil pumps situated out of the oil are all conditions which restrict oil

flow to the pump and would seem to promote a flow-limited condition On the other hand, coarse screens, short or large diameter inlet tubes, and sub- merged oil pumps offer less resistance and would tend to promote air bind- ing, provided of course that the oil becomes structured enough that it

cannot "slump" to the screen fast enough to prevent a hole from forming

The overall effect of these interacting variables may be estimated if it

is assumed that the pump inlet tube is limiting and that screen restrictions are negligible compared to those of the inlet tube (probably a good assump- tion because of the screen bypass which opens under conditions of severe

restriction) The resistance of the inlet tube to flow may be estimated

from Poiseuille's law for flow in a circular tube:

where Q • volumetric flowrate

AP = pressure drop across tube

D = tube inside diameter

V • fluid viscosity

L =•= tube length

If it is further assumed that the oils are Newtonian (this assumption is

implicit in the derivation of Equation 1 and will be discussed further later), and that the pressure drop across the inlet tube is the same for all of the engines, Equation 1 becomes

where k is a constant at a given temperature, and R is the resistance to

flow defined as

Equation 2 suggests that flowrates from various systems (or the relative ease of pumpability) can be estimated by comparing R values Such a com- parison is made in Table 12, in which the values of R for the six different engines included in the program are compared with their average BPT's from Table 5 This comparison is shown, graphically in Figure 16, from which

the increasing ease of pumpability with decreasing R is generally evident (i.e., larger flows or lower BPT's with smaller R values) The L-235

engine, with the largest resistance, also is the most severe from a pump- ability standpoint, while the 0-240 engine, with the smallest resistance,

is the least severe The discrepancies evident in Table 12 and Figure 16 are probably due to one or more of the following:

1) Effects of the oil inlet screen may not be negligible

2) Oil pump location was not considered in the analysis -

initial tube condition (filled or void of oil) may be

important

3) Oil level or height above screen was not considered,

and may have influenced the type of failure and the

BPT

4) The number and severity of bends in the inlet system

were not taken into account, nor were entrance and

exit effects in the inlet tube

5) Non-Newtonian characteristics may have been more im-

portant in some systems than others Differences in

shear stress may have caused variations in apparent

viscosity, even under the same pressure drop and tem-

perature conditions

In spite of these qualifications, however, the data in Table 12 suggest that the resistance concept may be useful in evaluating and predicting engine pumpability performance

An engine's R value may also be indicative of its sensitivity to air binding

As previously discussed, long, small diameter tubes (high R values) are more likely to bring about a flow-limited condition, whereas short, large diameter tubes (low R values) are more prone to air binding While oil effects can- not be completely separated from engine effects in this matter, the results

in Table 4 are in accord with this interpretation The L-235 engine, with the largest R value, always failed by becoming flow limited; the D-444

engine, with the next largest R value, failed by air binding for only one

of nine oils The other engines, with lower R values, failed by air binding for several oils

Rocker Arm Oiling Times

The rocker arm oiling times obtained from five of the seven engines are

summarized in Appendix F, Tables F-l to F-5 One of the engines, M-114,

11

had an overhead camshaft and therefore had no rocker arms, and one laboratory was not equipped to obtain RAO times

The average time required for oil to reach the rocker arms varied from almost eight minutes to less than three minutes, depending on engine type How- ever, the temperatures in this program were chosen so as to obtain the

pumpability limit based on gallery oil pressure, and covered a fairly narrow range Thus, the RAO time data are not well suited for analysis

Frequently, no relationship between RAO time and test temperature was evident from the data Figure 17 is typical of such results However, some of the data did show a well defined relationship and other data suggested a general trend In those cases where some relationship between RAO time and tempera- ture existed, the relationship depended on both oil and engine type Ex- amples of how this temperature sensitivity differed for the same oil in

different engines is shown in Figures 18 and 19 for two Newtonian oils and

in Figures 20 and 21 for a Vl-improved oil An example of differing tem- perature sensitivity for different oils in the same engine is shown in Fig- ure 22

Extending the cold-soak period of the test from 16 hours to 64 hours did not affect the relationship between rocker arm oiling time and test temper- ature for the soak-time-sensitive oil (PRO-11) in three engines, but did seem to have an effect in a fourth engine These comparisons are also

shown in Figures 20 and 21

CONCLUSIONS

Based on low-temperature pumpability data obtained with thirteen Pumpability Reference Oils in seven engines, the following conclusions were reached:

• The Borderline Pumping Temperature and failure mode data

developed in this program can serve as the basis for

evaluating the ability of bench tests to predict engine

oil pumpability at low temperatures

• Engine design and oil characteristics both significantly

affect low-temperature pumpability

• Pumpability failures occur when there is insufficient

flow either

a) to the oil screen, or b) between the screen and oil pump

• SAE Viscosity Grade is not a reliable indicator of low-

temperature engine oil pumpability

Times for oils to reach the rocker arms can be as much

as three times longer in some engines than in others

RECOMMENDATION

The full-scale engine pumpability data developed in this program should be the criteria for selecting a pumpability bench test

13

5 E F Boone, W W Crouse, and J J McLaughlin, "Low Temperature

Fluidity of Multigrade Motor Oils," Paper 710140 presented at SAE

Automotive Engineering Congress, Detroit, January 1971

6 R M Stewart and C R Spohn, "Some Factors Affecting the Cold Pump- ability of Crankcase Oils," SAE Transactions, Vol 81 (1972), Paper

720150

7 M L McMillan and C K Murphy, "The Relationship of Low-Temperature Rheology to Engine Oil Pumpability," Special SAE Publication, Viscometry and Its Application to Automotive Lubricants (SP-382), 1973, pp 7-24

8 R M Stewart and M F Smith, Jr., "Proposed Laboratory Methods for Predicting the Low-Temperature Pumpability Properties of Crankcase

Oils," Special SAE Publication, Viscometry and Its Application to

Automotive Lubricants (SP-382), 1973, pp 25-34

9 A R Nolf, "Engine Oil Pumpability and Related Properties," Special SAE Publication, Viscometry and Its Application to Automotive Lubricants (SP-382), 1973, pp 35-44

10 F F Tao and W E Waddey, "Low-Shear Viscometry and Cold Flow

Mechanism-Engine Oils," Special SAE Publication, Viscometry and Its Application to Automotive Lubricants (SP-382), 1973, pp 45-55

11 C R Spohn and R M Stewart, "Cold Pumpability Characteristics of Engine Oils Predicted by a Bench Test," SAE Paper 740541, presented at Fleet Week Meeting, June 1974

12 Summary Status Report of RDD VII-C, Task Force on Brookfield Vis-

cometry, December 1974

Table 1 Formulation Characteristics of ASTM Pumpabillty Reference Oils

Newtonian Newtonian (1)

(2)

(3)

Oils PRO-02, PRO-04, and PRO-14 were omitted from the test program PRO-STD - Pumpability Reference Oil - standard additive package ARO-STD - ASTM Reference Oil - standard additive package

Reference oil ARO-104 used by ASTM RDD VII-B in their Shear Stability Program

(4) Minimal antiwear, antioxidant treatment only

210°F (98.9°C) 11.25

-20°F (-28.9°C)

5 300

0°F (-17.8°C)

1 260

-40'F (-40°C)

118 200

-30°F (-34.4°C)

30 100

-20°F (-28.9°C)

8 930

-10'F (-23.3°C)

3 720

0°F (-17.8°C)

Table 3 Test Engines

Engine

Code*

Engine Displacement Litre In.3 Engine Type Oiling System Rocker Arm

Relative Oil Pick-Up Tube Length

Long

* Two N-235 engines were run by different laboratories in this program Where distinction between the two is required, they will be designated N-235-1 and N-235-2

17

00

Table 4 Summary of Borderline Pumping Temperatures and Failure Types

Borderline Pumping Temperature , % and Failure Type*

Table 5

Borderline Pumping Temperatures

of Nine PRO's Run in Seven Engines

Borderline Pumping Temperature, °F

Seven- Engine Engine: L-235 0-125 D-444 N-235-1 N-235-2 N-114 0-240 Average

-7 -5 -10

-18.3

-12 -20.1

-17 -23.1

8

6

48

782.3 329.3 26.7

29.3** 12.4**

Grand Mean

Error Std Dev

-19.3°F 5.16°F

* Estimated from incomplete data using average pressure-time slope from other oils

** Significant at the 99% confidence level

19

Table 6

Borderline Pumping Temperatures

of Thirteen PRO's Run In Four Engines

Borderline Pumping Temperature, °F Engine:

Table 7

Seven-Engine Average BPT's for All Thirteen PRO's

Borderline Pumping Temperature Four-

Engine Average

ll

Seven-Engine Average Decimal Rounded**

21

Table 8

Correlation of Individual Engine Results with Seven-Engine Average Results Regression Model: Y = A + BX

Where: Y = the seven-engine average BPT's and

Table 10 Soak Time Effects for PRO-11

Table 11 Oil Pump Inlet System Characteristics

Oil Level - Screen Characteristics Oil Pump Inlet Tube ID Length Oil : Pump Height Scr Above

mm in

een

FIGURE 2A Typical pressure-time curves for determining Borderline

Pumping Temperature (typical for all engines except L-235)

FIGURE 2B Typical pressure-time curves for determining Borderline

Pumping Temperature (typical for Engine L-235)

29

FIGURE 3A Determination of Borderline Pumping Temperature based

on minimum oil pressure after 60 seconds

FIGURE 3B Determination of Borderline Pumping Temperature based

on time to reach 20 psig (138 kPa)

31

(• DENOTES AIR BINDING)

12 07,09 03,08,15

FOUR-ENGINE AV£RAGE BORDERLINE PUMPING TEMPERATURE, °F

i I

+5 +5

FIGURE 5 Correlation between seven-engine average and four-engine

average Borderline Pumping Temperatures

33