A while ago I made a post in which I discussed analog pressure gauges and how I still use them on very rare, but very real occasions. One such occasion presented itself last week, and since it happened in our shop, I was able to film it, and now I can present to you a concrete example of how a respectable, time-proven, and ubiquitous Parker ServiceJunior digital pressure gauge can give an unusable pressure reading when a pressure source is "noisy" - i.e. "spiced" with high-frequency pressure spikes, and how a simple mechanical (bourdon tube type, liquid-filled) pressure gauge, or a digital pressure measuring system that uses digital filtering (averaging) can successfully by-pass the noisy pressure issue.
Before I show you the video, let's examine the specs of the SCJN-600-01 I used in the test.
According to the catalog, the peak capture in this model is done every 10 ms - which is not bad at all, especially for a battery-powered device. I actually discovered that the gauge reads the transducer's resistive bridge twice every 10 ms in a pair of 1.3-ms bursts, employing AC excitation (more on this and the gauge's design here). Parker does not mention the refresh rate of the screen and the bar graph in the CAT-4054 brochure, but, luckily, they do mention it in the user manual - 300 ms for the digital reading and 50 ms for the bar graph.
These specs are more than adequate for a work-horse gauge, and I have been using this model for years, but apparently, these gauges, instead of averaging out the reading that they display every 300 ms, seem to provide you with a snapshot of the latest pressure capture, which results in values that seem to be "all over the place" when a pressure source is noisy. I am not sure if this is intentional or not, though. I, personally, would prefer an averaged-out reading because it makes setting "spiky" things easier. This is exactly why the pressure gauges that I built use an averaging algorithm in the app.
And now, let me show you what I mean: