A. SPINDLE ACCEPTANCE

R.A.S. has rigorous quality standards where all critical components are 100% inspected and documented for future reference (Figure 10 through 12). Acceptance requires that all criteria conform to the R.A.S. and customer requirements.

1. Shafts

100% inspected and documented for dimensional and geometric tolerances. See Appendix II.
100% dynamic two-plane balanced to G.O.4 speciﬁcation and documented.

2. Tube Spacers

100% inspected and documented for dimensional and geometric tolerances. See Appendix III.
100% dynamic two-plane balanced to G.O.4 speciﬁcation and documented.

3. Housing

100% inspected and documented for dimensional and geometric tolerances.
Housings with water cooling provision are pressure tested with air at 120 psi for l hour.

4. Detail Components

100% inspected with limited documentation.

5. Final Assembly

100% inspection and documentation.

i. Preload/axial spindle rate by means of an axial compliance check where 50 % of the bearing ampacity is utilized as an applied load and the deflection is measured.

ii. Temperature condition at steady state.

iii. Idle current load for motorized spindles.

iv. Flow rates and pressures for liquid cooled spindles.

v. Flow rates and pressures for oil lubrication

vi. Dimensional and geometric tolerances.

vii. Vibration analysis (displacement, velocity, acceleration and spike energy).

viii. Axis of rotation runout at static and diagnostic when applicable

B. RUNOFF

The R.A.S. runoff procedures entail thermal monitoring, vibration analysis and axis of rotation verification.

1. Thermal Monitoring

Thermal monitoring is required on all spindles during the runoff. The speeds must incrementally increase to prevent overheating of the bearings usually in 20% increments. The standard procedure is outlined below.

a. Run the spindle at 20% of speed for 1/2 hour or until a steady state temperature is reached. Temperature should not exceed 55°C on spindle housing and 65°C on spindle shaft. If the temperature exceeds the limits, shut the spindle down and allow it to cool 15°C. Restart spindle at 20% of speed and repeat (a.) until a steady state temperature is reached.

b. Repeat process (a.) at 40%, 60% and 80% speed.

c. At 100% of speed the spindle must be run for a minimum of 4 hours, even if the state conditions are reached earlier prior to taking vibration readings.

Note: To expedite difficult to “break in” spindles, shutting the spindle down during the first and second stages of the run in and using a fan to cool the spindle can be employed. Caution must be used carefully monitoring the shaft and housing temperature differentials. The temperature differential should never exceed 10°C.

2. Vibration Analysis

Vibration analysis is completed on all spindle units to speciﬁcation outline in VI. C or customer requirements.

3. Sputnik Check (Axis of Rotation)

Instantaneous axis of rotation is a direct function of component geometry and cannot be measured in the traditional fashion. That is, use of an indicator on the spindle face and pilot measuring run—out. The traditional test is very relevant but it should be realized that it measures the degree of concentricity of the local surfaces with respect to the rotational axis and which can be shown by surface deviations.

R.A.S. has utilized an advanced technique called a "sputnik check" which enables the detection of radial and axial changes in the instantaneous axis of rotation. It is the instantaneous axis of rotation which is the most important when we refer to ultra precision spindles because the roundness of a hole machined by a single point tool will follow the same excursion path as the axis of rotation of the spindle.

Enclosed in Appendix VI is a typical report supplied to the customer when ultra high precision ball bearing spindles were required for wrist pin boring pistons. Typically, an ultra precision spindle would be used in fine boring or grinding applications where Aerostatic, Hydrostatic or Ultra Precision Ball Bearing Spindles are employed.

C. PRECISION MULTI PLANE DYNAMIC BALANCING

R.A.S. two-plane dynamic balancing is performed on spindle shafts which should include all rotational components assembled on the shaft (i.e. electrical rotors, keys, sheaves, etc.) whenever feasible. Balance tolerances of assembled shafts with components should comply with ISO 1940 G0.4 or better. Additional considerations are outlined in section VI.C.

D. VIBRATION MEASUREMENT AND ANALYSIS.

R.A.S. has outlined vibration specifications with the intent to provide their customers with spindles having increased life, reliability, quality and productivity. Vibration analysis is a tool for the indication of machine system conditions and that these guidelines provide a target for machine acceptability ensuring integrity of our products.

Measuring vibration in rotating machinery is a function of the rotors condition that is, rigid, quasi – rigid or flexible. In these specifications only the rigid rotor vibration will be addressed. On machines with rigid rotor systems, vibration is measured on the main structural members such as bearing housings. The vibration levels measured are indicators of excitation forces generated by rotor imbalance, thermal stresses, bearing defects, electrical unbalance and other sources of excitation.

l. Specifications

There is no single vibration analysis technique nor one absolute standard in which every spindle can be evaluated. To establish standards requires physical data based on type and size of machine, type of application, mounting of system, and effects of spindle vibration. R.A.S. has established a speciﬁcation for it’s own purpose (see graph 15) and also recognizes several established vibration standards as its guidelines:

a. Standard Precision Spindles balanced in accordance with I.S.O. 1940, Quality Grade GO.4 or better.

b. Precision Spindles:

Velocity – in accordance with I.R.D. “General Machinery Vibration Severity Chart” not to exceed the VERY GOOD REGION <.040 in./sec. Peak Acceleration – in accordance with I.R.D. “Vibration Acceleration General Severity Chart” not to exceed the VERY GOOD REGION <l.0 in./sec.2 Peak.

c. Super Precision Spindles:

Velocity – in accordance with I.R.D. “General Machinery Severity Chart” not to exceed the SMOOTH REGION < .020 in./sec. Peak.

Acceleration – in accordance with I.R.D. “Vibration Acceleration General Severity Chart” not to exceed the SMOOTH REGION <0.5 in./sec.2 Peak.

2. Test Stands and Machine Frequency

Vibration tolerances used by R.A.S. and I.R.D. are based on measured vibrations on rigidity mounted spindles or bolted to a rigid foundation. Spindles that are mounted on resilient isolators, such as coil springs or rubber pads, will have higher amplitudes. Generally, the rule follows that one would see at least twice the vibration for a spindle mounted on isolators. If test stands are to be resilient these considerations should be addressed.

a. Test sub-bases which are lighter than the spindle and which are intended only to stiffen the spindle should have mass less than one-fourth of the machine.

b. Test sub-bases which are heavier than the spindle, which are intended to ﬁx the feet of the machine in place, should be at least ten times the mass of the machine.

For test purposes no major structural resonance should occur in the operating range of the spindle. The combined machine sub-base and soft mounts should be designed that all rigid body natural frequencies of the system are less than 25% that of the lowest natural frequency. It is important to note that spindles, machines and the support structures may each have their own degrees of freedom. This may be the case where the acceptable level of unbalance in the spindle may excite the systems, therefore careful analysis of the system must be accomplished to identify the damaging vibration.

E. MACHINERY DIAGNOSTICS AND ANALYSIS

The R.A.S. engineering, quality and service departments work closely with the customer to ensure a superior precision spindle for the design application. The responsibility of the engineering department is to analyse the requirements of the customer, and design an optimum spindle through the use of computer aided modelling. The quality department ensures adherence to specifications by utilizing rigorous testing techniques and the service department provides the feedback verifying conformance to requirements by analysis and observations.

l. Machinery Analysis

a.) Modelling

R.A.S. engineers perform rigorous modelling and analysis to new spindle designs and critical applications. All variables are considered during the analysis stage prior to finalization of the design. The most important variable which can affect the overall performance of the spindle is the tool configuration. R.A.S. models the tooling in conjunction with the spindle, working closely with the tooling vendor to optimize the spindle/tooling system. Listed below are the main programs that are employed to obtain an optimized integrated system.

TLEFF Analyse the tooling load effects on the spindle system.

RIGID Analysis of the static effects of loads and spring rate determinations on spindle systems.

Optimizes bearing spans and tooling overhang.

BSDRL Analyse dynamic and static variables on bearings yielding spring rates, power consumption, life and load distribution.

BTHRM Analyse the effects of ﬁts, temperature to determine working preload and stresses.

BRFRQ Calculates bearing components natural frequency.

TORSYM Dynamic analysis of torsional systems with several degrees of freedom, calculating natural frequencies of geared systems.

RHPDARB Dynamic and static analysis studying effects of speed and load on spindle systems.

CKYANS Boundary Element Analysis (B.E.A.)

Programmed for static analysis of elastic and thermal systems.

ANSYS Finite Element Analysis (F.E.A.)

Programmed for dynamic and static analysis of mechanical systems.

b.) Diagnostics

R.A.S. quality personnel tests and documents each spindle prior to shipment (outlined in section V.I.B.l through VI.B.3) for thermal, vibrational and axis of rotation performance a situation where R.A.S. has used temperature and vibration analysis is outlined below.

Prior to being used in production a spindle had undergone the standard vibration analysis by R.A.S., quality personnel and the customer, passing all vibration specifications. The spindle was mounted in the machine and put into production, after a short period of time, ovality in the parts was experienced. The spindle was monitored for vibration which indicated elevated Spike Energy (gSE) amplitudes and increasing acceleration noise. At this point, the spindle builder was notified of a failing bearing by the customer.

The spindle was removed from production and put onto a test base to monitor the apparent disturbances; however, no excessive noise was found to identify a failing bearing. It was then surmised that possibly the machine system may be indicating the high frequency vibration, but mechanical analysis did not reveal any disturbance. Consideration was then placed on the environment in which the spindle operated, where uncontrolled coolant splash was observed to be cascading over one side of the spindle. It was of interest to then simulate this coolant splash at our test facility by placing a cooling fan to one side of the spindle and observing its influence on the vibration (see Figure 11-14).

R.A.S. observed that the spindle Spike Energy (gSE) was .6 gSE, well within I.R.D. standards of 1.0 gSE when the cooling fan was not on the spindle. However when the fan was turned on, the Spike Energy began to rise to the amplitude of 4.5 gSE, at this point the spindle bearings began to emit a "hissing sound" so the fan was turned off. The Spike Energy amplitude began to fall off to the steady state amplitude of .6 gSE and the
“hissing noise” disappeared. Knowing that, broad band shock or bearing elements under stress can be caused by Spike Energy and acceleration.

The customer was notiﬁed of the ﬁndings and suggestions were made to protect the spindles from the environment. The R.A.S. service and engineering department were not satisﬁed with the degree of protection from errant coolant, the customer could provide and redesign the spindle housing to facilitate spindle liquid cooling jackets around the bearing pocket preventing localized thermal stressing. R.A.S. has seen an increase in life from the original design of spindles by 200% or 3 years.

The conclusions drawn from the diagnostic analysis are: unevenly distributed cooling creates ovality of the outer race of the bearings, increasing stress, reducing bearing life and reducing part quality and Spike Energy is an erratic parameter that should never be used to form a conclusion without additional investigation and measurement.

Field data accumulated and trended by our service department proves to be an invaluable diagnostic tool when original parameters begin to deteriorate. No single technique of diagnostics should be regarded as an absolute indicator of the spindle conditions without sufficient data or testing.

d. Standard Precision and Super Precision Spindles are to have Spike Energy or B.C.U. measurements for comparison trending only. It should be noted that these amplitudes have been good indicators of impending problems for spindles with taper bore roller bearings.

e. Special and Standard Spindles which are classiﬁed in accordance with I.S.O. 2372 Standard.

f. All new and rebuilt spindles require complete vibration signatures prior to shipment to ensure conformance to customer requirements.

g. Transducers/equipment should be capable of monitoring displacement, velocity, acceleration and high frequency energy (i.e. Spike Energy or B.C.U.'s).

h. Vibration measurements should be taken at every bearing location in the vertical, horizontal and axial (if possible) directions. Vibration measurements are to be taken at operating speeds including spectrum or windows illustrating amplitudes at running frequency and their order frequencies. See Appendix.

i. Bearing vibrations to be monitored at its component frequencies to identify the source of disturbances.

j. Test stands and spindles should be rigidly mounted to closer simulate their machines and the natural frequency should be at least 25% removed from the operating speeds.

k. Spindle shaft balancing should be performed with all rotational components assembled on the shaft (i.e. electrical rotors, keys, labyrinths, sheaves etc...) whenever feasible. Balance tolerances of assembled shafts with components should comply with I.S.O. 1940 GOA or better.

l. When possible, customer should make available to the spindle builder, prior to shaft balance, all spindle attachments (re. tooling, sheave, ﬂuid couplings etc...) to balance attachments and spindle in situ.

m. Balance weights added to motorized spindles should be of a non-conductive material (i.e. stainless steel). Balancing by metal removal may be better accomplished by pressing a collar onto the shaft/rotor assembly and removing metal from the collar. This will ensure structural integrity of the rotor.

n. Vibration measurements should be taken with regard to consistency of transducer mounts. Magnetic pickup holders are most commonly used because of convenience. When using stud mounted transducers, more vibration is transferred than from the magnetic holders yielding higher readings. Hand wands however, transfer less vibration yielding lower readings.

o. Stud mounted transducers require l/4 – 28 tap holes with a minimum diametrical pad of l.6 in. central to the tapped machine hole.