Doing a test soon on a fairly large test article and the control strategy for sine is average with 4 control. The question has been raised what happens if 1 control registers a low or poor signal will the other controls rise to make the average. my argument is that with an abort set for each channel to a level 25% higher than the max level needed the test will complete with out trouble.
Using more than 1 control channel is typically done for large shaker head expanders to take an average value over the large area. So if there is some physical distortion in the head expander, the average of all control sensors will be used for control, averaging out the distortion. Typically there should not be large differences in the amplitudes of these control sensors.
Another reason for multiple control channels is when low frequencies are used and a displacement type of sensor is used rather than an accelerometer, in the low frequency range. In that case the control will transition from displacement to acceleration, as defined in the “Contribution Profiles”.
For the typical case when using multiple control accelerometers on a large head expander, however there are provisions in the software to account for the case where there are low values measured on a control channel. This can happen if the signal is lost, like if a sensor comes loose and falls off. In that case the control will at first try and compensate the responses with an increase in drive, but only to a certain point, which is when the control sensor goes below a certain level. This lower level is defined by the “Openloop Threshold” parameter, and is set during the Selfcheck, and is calculated by:
This parameter is set automatically from the Selfcheck results, however it can be changed in the “Advanced” button of the Selfcheck menu. If the signal from any control transducer is measured below this value, then the test aborts with an “Open Control Channel” error.