·Home ·Table of Contents ·Reliability and Validation 1

Optimizing an NDT Method for Inspection of Synchronizing HUB in Automobiles B.N.Sankar Department of Physics, Anna University, Chennai - 600 025,India C.Srinivasan Engineer, Cooperheat Saudi Arabia Co. Ltd., Jubail - 31951, Kingdom of Saudi Arabia Consultant - Free lancer (Other than Saudi), 28/1, Northmada street, Nungambakkam, Chennai - 600 034, India M.V.Rajamani Electro-Magfield Controls & Services (EMCS), Ambattur, Chennai - 600 098, India Pankaj Kumar Jain Assistant Manager-Laboratory, Sundram Fasteners Limited, Hosur - 635 109, India Contact

ABSTRACT

Automobile industry is one among the various fields envisaging fast development and is having an impact on transportation and thereby the day-to-day life of mankind. The quality and reliability of a product is essential in a competitive world with respect to cost and high life time. The present work deals with a specific critical automobile component, namely, synchronizing hub, which requires stringent quality control. Various types of cracks are found to develop in the component, which may be harmful from application point of view. Hence an inspection methodology is derived so as to ensure 100% crack-free products. The most suitable nondestructive testing method for ensuring this is found to be the magnetic particle testing using central conductor technique with wet fluorescent medium. The component under study is manufactured by powder metallurgy technique and case hardened by the induction hardening process. A study is made to optimize the different machine and testing parameters that could yield the best results with respect to detection of cracks.

1. INTRODUCTION

Synchronizing hubs are used in manual transmission gear boxes for automobiles. They synchronize the process of switching between successive gears. One hub is present between every two gears and upto three hubs are generally required in a vehicle. The hubs are used for a dynamic application, failure of which can cause serious accidents. Thus requiring stringent checks to ensure crack free products. Wet Magnetic Particle Inspection (MT) method was found to be the most suitable method for detection of cracks and the process stage at which they originate. Among the different MT techniques central conductor technique with wet fluorescent medium was felt to be appropriate. When the inspection procedure was carried out with currents recommended by ASTM code of practice[1] there were difficulties in detecting the cracks which were originally present in hubs. This initiated an interest in probing the details of the technique and hence the present work..

2. SAMPLE PREPARATION DETAILS AND DEFINITION OF DEFECTS

2.1 Sample preparation
Synchronizing hubs used in automobiles have commonly been produced from wrought steel by forging followed by extensive machining. Industry has been looking at alternate processes to reduce extensive machining and raw material costs. Powder metallurgy (P/M) has its ability to produce near-net shape products thus drastically reducing machining operations. It is also a cost effective technique[2] particularly when huge production volume is involved. Hence P/M is chosen as the technology for manufacturing hubs for cars. The basic process typically consists of blending raw metal powders along with any necessary lubricants or additives; filling a die cavity with the blended powder; pressing the powder in a die to form a "green compact" and then sintering at a high temperature to bond the powder particles together and "strengthen" the product. Post-sintering operations are sizing to meet dimensional requirements, followed by machining and shot pinning. The final post machining operation is induction hardening. This is done to provide wear resistance to the hubs.

Fig 1: HUB and its cut view

Fig.1 shows both the full view and cut view of one of the three synchronizing hubs used in a car. The part has three levels or steps on the either face viz., the outer skirt with teeth, the middle web and the inner hub. It has keyways, multiple serration and grooves as shown in Fig.1.

2.2 Definition of Defects
Because of the complex shape, sharp corners, multiple levels in the product and in addition the large number of processes involved in its manufacture by powder metallurgy route, cracks in synchronizer hubs can occur at any stages from compaction to induction hardening. As the part is used in the transmission of the automobile, the part should be free from any cracks so that it may not fail in service. Hence it is important that 100% of the synchronizer hubs have to be checked for cracks. Keyway corners and splines on the inner diameter, which act as sharp notches, are the most susceptible areas where cracks could occur. Fig.2 shows various types of cracks, which may be present in synchronizer hubs.

Fig 2: Defects in synchronizing HUB

Following is the description of cracks:

1. Keyway cracks (KC): Emerging from keyway corners and extending over the surface or to the other face through keyway slot.
2. Groove cracks (GC): Emerging from splines on the inner diameter and extending over the oil groove.
3. Radial/Spline cracks (RS): Multiple cracks emerging from splines on the inner diameter and extending over the hub.

3. EXPERIMENTAL WORK

3.1 Selection of NDT Method
Widely used NDT techniques in industries are ultrasonic testing (UT), Radiography testing (RT), Liquid penetrant inspection (LPI), eddy current testing (ET) and magnetic particle inspection (MT).

Since the synchronizing hubs are porous materials manufactured by P/M, the UT method will result in a large number of echoes with different S/N ratio posing a problem in interpretation and requiring an expertise in it. UT also may not be suitable for large scale production of hubs. Since the defects that are to be detected are only on surface, application of RT is not opted. Besides it is also costlier and one has to be cautious about health hazard. When one thinks of LPI, the configuration of hub (lot of teeth, keyway and grooves) suggests a problem in employing it. The removal of excess penetrant completely from the surface is also extremely difficult. ET method involves a large number of parameters in its application. The variations in density and hardness at different locations make ET a still more complicated for its application. The three levels of hub make ET less reliable with respect to crack detection. MT requires the materials to be tested to be of ferromagnetic and the shape design does not restrict its application. It is faster in its application for mass production. Considering the wet method (fluorescent) with central conductor magnetisation it has the following advantages[3] over the dry particle method:

1. more sensitive for very fine and shallow discontinuity,
2. easy to cover all surfaces properly, especially the irregularly shaped parts,
3. faster than dry method, and
4. easy to adapt to a mechanised test system.
Hence, in the present work, central conductor technique with wet fluorescent medium has become the choice.

3.2 METHODOLOGY OF MT
Essentially the following steps were involved in carrying out MT:

• Cleaning the sample so that magnetic particles can migrate freely to defect locations and reduce contamination,
• Establishing the magnetic field,
• Application of liquid suspension,
• Viewing in black light, and
• Demagnetising the specimen.

In order that an optimal NDT procedure is identified for the hub samples the following parameters have been considered:

1. Current rating,
2. concentration of magnetic particle in carrier,
3. size and colour of magnetic particle,
4. flow rate and
5. black light intensity.

3.2.1 Current rating
As per ASTM code of practice, for parts with outer diameters up to 5 in. a direct (DC) or half-wave rectified (HWAC) or full-wave rectified (FWAC) current of 700 A/in.to 900A/in. of diameter for a single turn central conductor shall be used[1]. For alternating current (AC) 70% of these specified currents will be sufficient. Hence as per ASTM, current requirement may be obtained from the equations:

Current (in A) = (OD in inches) x 800	(for DC/HWAC/FWAC)	(3.1)
Current (in A) = (OD in inches) x 800 x 0.70	(for AC)	(3.2)

where OD represents the outer diameter of the part studied. For the three different hubs classified as A,B,C under study, the current requirements as per ASTM are calculated and shown in Table 1 along with their dimensions. In each of the three hubs A,B,C, 1000 samples were tested with AC as per ASTM. (In total 3000 samples were studied). Incidentally, none of the hubs under study revealed any of the three cracks defined in Section 2.2. Since this was not expected and this could not be possible, currents of different strengths have been passed through the central conductor by trial and error and the specimens studied. It is derived out of the experiments that current for the specific application may be obtained from the equation

Current in A = (WT in mm + d in mm) x 25		(3.3)
Current (in A) = (OD in inches) x 800 x 0.70	(ASTM formula)	(3.2)
Current in A = (WT in mm + d in mm) x 25	(New formula)	(3.3)

Sample ID	OD	ID	WT = (OD-ID)/2	Central core diameter	Current as per
					New Formula	ASTM Formula
	in.	in.	in.	in.	ampere	ampere
A	2.3906	0.9688	0.7109	0.75	927.7	1338.7
B	2.4375	0.7813	0.8281	0.75	1002.1	1365.0
C	2.6875	1.2500	0.7188	0.75	932.7	1505.0
Table 1: Current requirements as per Equation 3.2 and ASTM standard

where WT and d represent wall thickness of sample and diameter of central core conductor respectively. Currents have been varied upto ± 20% in steps of approximately 10% from the current value given by Equation 3.3. Currents were applied in two shots, approximately 1/2 s each. The appearance of defects is explained in Table 2, along with the details of other parameters, namely, flow rate, magnetic particle concentration and black light intensity used. For the present study, EMCS magnetic particle testing equipment (UFA 242) of Indian make has been used. The equipment has a provision to vary the current in steps of 25A. This equipment facilitates both longitudinal magnetisation and circular magnetisation. The magnetic particles used were iron oxide fluorescent powder (450 mg; 15 to 30m ; Medium ID: MG-OG; Batch No. TR/90 of Ferroflux products of India) in kerosene vehicle. An aperture type demagnetizer has been used for demagnetization of samples after the experiment. Since it may not be possible to present all data collected within the scope of the paper, in all Tables, data for seven different representative samples are shown. In general, the length of KC varied from 2 mm to 35 mm. The GC was of 2 to 3 mm length. Length of RS varied from 1 mm to 2mm.

Magnetic particle concentration	=	0.2 mL/100 mL
Flow rate	=	50 mL/s
Black light intensity	=	4500 mW/cm2 at distance of 400 mm from job

Sample ID	Defect	Current in ampere as per new formula					Current as per ASTM formula 1450 A for hubs A&B 1600 A for hubs C
		for A&C 700 for B 800	800 900	900 1000	1000 1100	1100 1200
A5	KC	LI	Sharp	Sharp	Sharp	PA	II, M, HPA
A6	KC	LI	Sharp	Sharp	Sharp	PA	II, M, HPA
B7 Side 1	KC	LI	Sharp	Sharp	Sharp	PA	II, M, HPA
B7 Side 1	RS	LI	Fuzzy	Sharp	PA	HPA	II, M, HPA
B7 Side 1	GC	LI	Fuzzy	Sharp	Sharp	PA	II, M, HPA
B7 Side 2	KC	LI	Sharp	Sharp	Sharp	PA	II, M, HPA
C1	KC	LI	Sharp	Sharp	Sharp	PA	II, M, HPA
C2	KC	LI	Sharp	Sharp	Sharp	PA	II, M, HPA
C3	GC	LI	Fuzzy	Sharp	Sharp	PA	II, M, HPA
C4	GC	LI	Fuzzy	Sharp	Sharp	PA	II, M, HPA
Table 2: Appearance of discontinuities for different currents

LI = Loss of Indication ;II = Irrelevant Indication ; PA = Powder Accumulations HPA = High Powder Accumulation ; M = Masking

3.2.2 Concentration of magnetic particle
The study was repeated with different concentration of magnetic particle in the carrier/vehicle, namely, 0.3 mL/100 mL and for the same currents of previous trial. The observations are shown in Table 3.

Magnetic particle concentration	=	0.3 mL/100 mL
Flow rate	=	50 mL/s
Black light intensity	=	4500 mW/cm2 at distance of 400 mm from job

Sample ID	Defect	Current in ampere as per new formula					Current as per ASTM formula 1450 A for hubs A&B 1600 A for hubs C
		for A&C 700 for B 800	800 900	900 1000	1000 1100	1100 1200
A5	KC	II, PA	II, PA	II,M	II,M,HPA	II,M,HPA	II, M, HPA
A6	KC	II, PA	II, PA	II,M	II,M,HPA	II,M,HPA	II, M, HPA
B7 Side 1	KC	II, PA	II, PA	II,M	II,M,HPA	II,M,HPA	II, M, HPA
B7 Side 1	RS	II, HPA	II, HPA	M,HPA	II,M,HPA	II,M,HPA	II, M, HPA
B7 Side 1	GC	II, HPA	II, HPA	HPA	II,M,HPA	II,M,HPA	II, M, HPA
B7 Side 2	KC	II, PA	II, PA	II,M	II,M,HPA	II,M,HPA	II, M, HPA
C1	KC	II, PA	II, PA	II,M	II,M,HPA	II,M,HPA	II, M, HPA
C2	KC	II, PA	II, PA	II,M	II,M,HPA	II,M,HPA	II, M, HPA
C3	GC	II,HPA	II,HPA	HPA	II,M,HPA	II,M,HPA	II, M, HPA
C4	GC	II,HPA	II,HPA	HPA	II,M,HPA	II,M,HPA	II, M, HPA
Table 3: Appearance of discontinuities for different current

LI= Loss of Indication ;II = Irrelevant Indication ;
PA = Powder Accumulations HPA = High Powder Accumulation; M = Masking

3.2.3 Size and colour of magnetic particle
The colour of the magnetic particle should be such that it contrasts most with the surface and in our experiment it glows as bright yellow - green when viewed under black light. This yellowish-green colour has better brilliance with the background and also is pleasant for eyes. Particles of smaller size than that used in our experiment may fill the crack and hence the discontinuities may become undetectable. Similarly particles of higher size may mask the finer cracks. Hence experiment was not repeated with other magnetic particles.

3.2.4 Flow rate
The entire experiment has been tried at two different flow rates, namely, 50mL/s and 100mL/s while magnetic particle concentration was kept constant at 0.2mL/100 mL. The higher flow rate simply resulted in washing away of all indications for all currents.

3.2.5 Black light intensity
As per the Standard Practice for Magnetic Particle Examination, SE 709 of ASTM the black light intensity at the examination surface (380 mm from the face of the light lens filter) should not be less than 800 m W/cm² when measured with a suitable black light meter. In the present study two sources of intensities 1650 m W/cm² and 4500 m W/cm² (at a distance of 400 mm from the sample) were used and the experiments repeated for distance range 300 mm to 450mm. The results are shown in Table 4.

Magnetic particle concentration	=	0.2 mL/100 mL
Flow rate	=	50 mL/s

Sample ID	Defect	Black light intensity: 4500 mW/cm2 (at a distance of 400 mm from job)Distance of sample from black light source				Black light intensity: 1650 mW/cm2(at a distance of 400 mm from job)Distance of sample from black light source
		300 mm	350 mm	400 mm	450 mm	300 mm	350 mm	400 mm	450 mm
A5	KC	Sharp	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible
A5	KC	Sharp	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible
A6	KC	Sharp	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible
B7 Side 1	KC	Sharp	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible
B7 Side 1	RS	Sharp	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible
B7 Side 1	GC	Sharp	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible
B7 Side 2	KC	Sharp	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible
C1	KC	Sharp	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible
C2	KC	Sharp	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible
C3	GC	Sharp	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible
C4	GC	Sharp	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible	Invisible
Table 4: Appearance of defects at different distances

4. RESULTS AND DISCUSSION

From Table 2 it is evident that the defects are revealed sharply for the current given by the Equation 3.3 for all hubs while the other parameters being kept constant.

Increased magnetic particle concentration results in high powder accumulation and irrelevant indications at change of sections as shown in Table 3. For higher currents complete masking of defects along with irrelevant indication and high powder accumulation is observed.

In the present study under the black light source of higher intensity the brilliance and sensitivity of defects were good only at a distance range of 300 mm to 350 mm while for lesser intense source none of the defects was visible for the entire range of distance in which experiment was conducted as indicated by Table 4.

5. CONCLUSION

From all the discussions based on the analysis of results obtained in the present work it may be concluded as follows:
1. At relatively low currents (than suggested by ASTM), at a magnetic particle concentration of 0.2 mL/100 mL, with a flow rate of 50 mL/s and with higher black light intensity the defects are revealed the best with better sensitivity.
2. It appears that separate MT standards may be thought of for synchronizing hubs manufactured by P/M for use in automobiles.

6. ACKNOWLEDGEMENT

The authors are grateful to Mr. K.Mani, President SFL-Hosur for according permission to carry out the work, the encouragement and support throughout the work. They are also thankful to Mr.M.V.Mohan and Mr.R.Vivek of EMCS for the support extended by them. They also thank Mr.V.Srinivasan and other staff of EMCS. One of the authors, B.N.Sankar is grateful to Dr.C.Mohana Doss, Professor and Head, Department of Physics, Anna University for the encouragement given.

REFERENCES

1. Appendix - B, Article 7, magnetic Particle Examination, Section V, T 745 1(b) and T 745.2(b), ASME, 1992.
2. Sidney H.Avner, "Introduction to Physical Metallurgy", Second Edition, Tata McGraw Hill Publishing Company, New Delhi, India, pp.605-632, 1997.
3. Article 25, Appendix C, Standard Practice for Magnetic Particle Examination, SE-709, Section V, ASME, 1992.

© AIPnD , created by NDT.net

|Home|    |Top|