Data processing: measuring – calibrating – or testing – Testing system – Of mechanical system
Reexamination Certificate
2000-06-12
2003-03-11
Barlow, John (Department: 2863)
Data processing: measuring, calibrating, or testing
Testing system
Of mechanical system
C702S086000, C702S145000, C702S147000, C416S030000, C416S061000, C417S042000, C417S063000, C415S118000
Reexamination Certificate
active
06532430
ABSTRACT:
STATEMENT OF GOVERNMENT INTEREST
The invention described herein may be manufactured and used by or for the Government of the United States of America for governmental purposes without the payment of any royalties thereon or therefore.
BACKGROUND OF THE INVENTION
(1) Field of the Invention
The present invention relates generally to underwater projectile launch systems, and more particularly to a method for determining an rpm profile for a turbine pump based on a pulse train obtained during a pump test.
(2) Description of the Prior Art
Turbine pumps are typically used to launch a projectile, such as a torpedo, from a submerged tube. In order to test for the proper operation of the pump, measurements of the pump's rpm are taken during a test firing and compared to performance specifications. The rpm measurements are derived from a pulse train obtained by a Hall effect sensor or optical encoder mounted to the pump. The specific sensor used will depend on the type of pump being tested. As the turbine rotates, the sensor provides one or more pulses for each rotation of the turbine. The timing of the pulses, or pulse train provides the rpm indication. For example, a sensor providing four pulses per revolution, with a pulse rate of 16 pulses per second indicates an rpm of 240, i.e.,
(
16
pulses
second
/
4
pulses
revolution
)
*
60
seconds
minutes
=
240
revolutions
minutes
.
Currently, a frequency-to-voltage converter is used to convert pulse data to rpm data due to the time varying nature of the pulse frequency. Appropriate conversion factors are applied depending on the turbine pump type and sensor configuration. However, to ensure consistent and comparable results, the converter must be calibrated for each pump test. Additionally, it has been found that data acquisition systems used to obtain the pulse train should be a counter-timer that reads the pulses every 30 milliseconds in order to capture both low frequency pulses at the beginning of rotation and high frequency pulses when the pump rotates at maximum speed. This timing results in a relatively rough frequency verses time curve. Further, spikes tend to occur at the onset of ramp-up, i.e., when the pump first begins rotation and rpm's are increasing. Such spikes cause improper interpolation of the rpm data when included in the pulse data set. Thus any one or all of these factors may result in an erroneous performance evaluation of the turbine pump.
SUMMARY OF THE INVENTION
Accordingly, it is an object of the present invention to provide a method to determine a turbine pump rpm profile, which does not require equipment calibration for each test.
Another object of the present invention is to provide a method to determine a turbine pump rpm profile, which effectively eliminates spikes in the pulse train.
A further object of the present invention is to provide a method to determine a turbine pump rpm profile having a smooth frequency verses time curve.
A still further object of the present invention is to provide a method to determine a turbine pump rpm profile, which is highly repeatable, with a high accuracy level.
Other objects and advantages of the present invention will become more obvious hereinafter in the specification and drawings.
In accordance with the present invention, a method is provided to determine an rpm profile for a turbine pump based on a pulse train obtained during a pump test. The method obtains a pulse count over a fixed time period, preferably every 30 milliseconds, from a computer controlled data acquisition system reading the pulse data from a sensor at the pump. The 30-millisecond acquisition rate has been found to achieve an accurate rpm profile throughout the operating range of the pump, i.e., 0-1200 rpm. Once the pulse train data has been obtained, the method eliminates ramp-up spikes using a spike elimination technique, which incorporates previous test data. The method first compares the spike pulse count to the surrounding pulse counts and rejects data points responsible for abnormal (in relation to the previous data) pulse count increases between data points. The method further rejects data points that lie outside statistically acceptable pulse rate variations.
Once ramp-up spikes have been eliminated, a frequency, or rpm curve can be generated. Depending on the pump being tested, the sensor may be a Hall effect sensor or an optical encoder and the number of rotations per pulse may also vary. The method applies standard factors to the pulse train appropriate to the pump being tested in order to convert the pulse train to a rough rpm verses time curve. The rough curve is then smoothed by first infusing data points to achieve an acquisition rate of approximately 1000 points per second. The infusion is accomplished by interpolating between actual data points and equally spacing the infused data points along the interpolated curve. Preferable, a three-point interpolation scheme is used. Such a large number of data points is necessary to then apply one of many standards, 5
th
order, smoothing processes to the rough curve resulting in the final rpm curve.
Thus, the present invention provides a method to determine an rpm curve, or profile of a turbine pump system. The method uses the pulse train directly from the counter-timer of a data acquisition system in determining the profile. The method does not use a frequency-to-voltage converter and therefore calibration is not required for each test. Ramp-up spikes are eliminated from the data, in a manner which yields repeatable profiles. A rough curve is generated to which data points are infused, such that a smoothing process can be applied to the curve. The resultant rpm profiles based on this method have been found to have a 98% level of accuracy and be repeatable to within 2%.
REFERENCES:
patent: 5223207 (1993-06-01), Gross et al.
patent: 5804726 (1998-09-01), Geib et al.
patent: 5825657 (1998-10-01), Hernandez
Barlow John
Cherry Stephen J.
Kasischke James M.
Nasser Jean-Paul
Oglo Michael F.
LandOfFree
Method for determining a turbine pump RPM profile does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for determining a turbine pump RPM profile, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for determining a turbine pump RPM profile will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3007905