Application Of Variane In Compressor

ANALYSIS OF VARIANCE FOR COMPRESSOR SWASH-PLATE DESIGN WITH 3 DIFFERENT COATING MATERIALS

1. INTRODUCTION

In this project I am going to present the 3 sample data on the strength of the swash-plate with different 3 different types of advanced material coatings that a compressor supplier uses to achieve higher durability and lower warranty returns.

1.1 Compressor Swash-Plate Function

A swash plate is used in a rotary valve AC compressor [used for automotive application] wherein pistons are fitted into a cylinder bores having their centers on a circle are reciprocated by swash-plate secured to a rotary shaft at a tilt-angle. A swash-plate chamber accommodating the swash-plate functions as a passage for the refrigerant gas. The refrigerant gas enters into the system through an annular passage and the refrigerant gas sucked functions as a by-pass therefor, diminishes the flow resistance of the refrigerant gas, and facilitates the cooling of the pistons with a result of reducing the operation noise produced from the engaging portion of the pistons and the swash-plate. Please see figure 1 and 2 for the cross-sction of the compressor and the location of swash-plate inside the compressor.

Figure 1: Cross-section of AC Compressor showing the Oil Flow

Figure 2: AC Compressor showing Swash-Plate and Pistons

1.2 Swash-Plate Issues On Low Oil

Seizing of the swash-plate occurs when the refrigerant oil is low, which result in seizing of the compressor thereby creating a customer complaint. Around 10 % of the compressor warranty returns are due to seizing of the compressor. Improving the durability of the swash-plate results in less seizures of the AC compressor and there by increasing the customer satisfaction. In order to increase the durability of the swash-plate, compressor suppliers use several kinds of advanced material coatings that help keep the compressor running even under low oil condition. The coatings act as a lubricant and high temperatures at pressures.

1.3 Durability Testing of Coatings on Swash-Plate

In order to determine how long a swash-plate can perform under low-oil conditions, the supplier has developed a standard test called “Component Dry-Lock test”. This test uses adds a constant load [400 lbf] at a rotational speed of 10.4 m/s under low-oil condition and measures the time [in seconds] it takes for the swash-plate to seize. For each surface coating of the swash-plate the test is carried out at least 30 times to get a good estimate of seizure data. Please see figure 3 for test set-up.

Testing is carried out, by applying a constant load of 440 lbf on swash-plate that rotates at 10.4 m/s on its axis. Time is recorded as long as the swash-plate keeps rotating. The longer the swash-plate rotates the better. The coating on the swash-plate which acts as a solid lubricant helps the swash-plate to rotate easily. Different coatings have different lubricant properties.

Test Conditions:

Load: 440 lbf

Speed: 10.4 m/s

Environment: R134a [refrigerant oil]

Figure 3: Test setup to measure the durability of the Swash-Plate

The three different types of coatings used for this design study are shown in the Table 1 below:

Table 1: Swash-Plate Coatings used for this Design Study

Swash-Plate Coating Material Used

Coating 1 MoS2*

Coating 2 MoS2*+ PTFE **

Coating 3 MoS2*+ PTFE ** + Patented Binder***

* Molybdenum Disulfide

** Polytetrafluoroethylene or Teflon

*** Patented Binder [Includes nano-particles that show exceptional lubricant properties at high temperatures and pressures]

2. APPLICATION OF ANOVA FOR ONE-FACTOR DESIGN STUDY

Here I will present the test of hypothesis and use the ANOVA [with the help of Minitab] to make my recommendations and conclusions on the 3 coatings.

2.1 Test of Hypothesis

I have collected 20 data points from this durability test for each coating to analyze whether they produce the same results or not. The supplier company states that these 3 different coating are not much different when it comes to functionality with any kind of coating. I will analyze and prove that the supplier is wrong. Since we will be using coating properties as the only factor in the study, this is called the one-way or one-factor Analysis of Variance [ANOVA]. The numbers of population to be studied are 3, i.e., coating 1, coating 2 and coating 3. Lets denote the number of populations with k (=3). Let the population means of the seizure time be Ој1, Ој2, and Ој3. They are defined as:

Ој1 = Population mean of the swash-plate coating 1 seizure time (in seconds.)

Ој2 = Population mean of the swash-plate coating 2 seizure time (in seconds.)

Ој3 = Population mean of the swash-plate coating 3 seizure time (in seconds.)

The ANOVA will be used to determine if the true population means are all the same, or whether they are different from one another.

The null hypothesis can be written as:

H0: Ој1 = Ој2 = Ој3 [1]

The alternative hypothesis which is trying to prove that these 3 population means are not the same can be written as:

Ha: H0 not true [2]

2.2 Experimental Data Values

Table 2 shows the 20 set data for the 3 coatings that has been collected from the durability test done internally in our company lab. After each test reading the test set up and the component is cooled