The client scoured his factory, a considerably-sized one with dozens of production units, looking for even one that had a VFD with a read-out of both the frequency and the power. He found one – it was connected to a sweep - and so we connected it to the OmniBatch electronics to check it out.
I will be finicky with the math this time since what I want to show is that upgrading entails no compromise. An old factory with nothing there to measure, when (easily and inexpensively) upgraded, can provide measurement precision that is uncompromisingly excellent.
First we wanted to do a calibration of the frequency and OmniBatch voltages …
“Yuk”, you say, “keep your R-squared with all those 9’s. The intercept doesn’t go through zero!” Well, let’s see just how bad this is. We’ll compute the relative offset for all the selected frequencies by taking the intercept (volts) and dividing it by the value (volts) and then converting back to frequencies.
We have a fairly consistent 3 Hz offset throughout the entire range. Let’s live with it.
Now how about the power for a mixture that is stirred by this VFD?
Say, what you will, this is a great fit for a physicist. A nice, neat parabolic curve with a near-perfect R-squared after fitting three coefficients to six points? Only Mother Nature, not random chance, could make so many data points fit so precisely. This is awesomely precise.
That same tank has a homogenizer on it, but the readout is just for frequency.
I know expectations are not that high on accuracy, but, this one is pretty good:
The offset is consistently less than 1 Hz. That’s pretty good.
To get power, we wired another element to the VFD and behold:
The added element now provides data that is every bit as precise as had been found on the other VFD (on the same tank), the one I described above with both frequency and power outputs.
There is nothing to hold you back from having superb data and, with Electronic Batch Processing - perfect batches.