Insane Hydraulics
Bold and audacious blog about fluid power
Today, my friends, I want to talk about how a graphic LCD with a well-thought-out interface can make a "classic" digital pressure gauge a thousand times better. It is going to be pure theory for now, but let me assure you that practice is on its way, and shortly, I will be adding a nice "show" to the "tell."
This is what I mean when I say "a classic digital pressure gauge screen":
A cookie-cutter segmented LCD that shows a pressure reading with an update rate of three times per second, a full-range analog bar graph at the top with a refresh rate of 20 HZ, and an additional line showing a record of either a pressure spike or a pressure dip. The gauge I just showed you is a Parker SCJN, but you will find it wearing many other "skins" - Stauff, Webtec, Minipress, Wika, Sika, Hydrotechnik, LR-Smart Tech, SKF, etc...
You can, of course, argue that these gauges are different and that models with higher precision or an extended range may carry improved circuitry or an upgraded firmware, but the LCDs are always the same, and if you work in this industry, I am sure that you've seen and used one of these gauges at least once.
Now, don't get me wrong - they are good gauges, super reliable too, and I have been using them as my "daily drivers" since forever, but it always feels like this display layout is overdue for an overhaul, and so I want to share my thoughts on how it could be done, and naturally, explain what exactly I don't (particularly) like about this classic screen we have all known and appreciated for so many years.
You can't see the max and min pressure records at the same time. The segmented LCD only has room for one of them, and in my opinion, this sucks big time. I appreciate the fact that the gauge has a 10 ms scan rate, but when I am hunting for spikes, I want to see the max and the min side by side, and having to press a button to cycle between the two feels cumbersome. A graphic display with a larger area could easily display both the max spike and the min dip at the same time.
It is there all right, and I can't say that it doesn't work, because it does, but... in certain situations it doesn't work that great, even with its adequately fast refresh rate.
You don't really need the analog bar when you are measuring static pressures, but (at least in theory) it should be very helpful when you are dealing with a rapidly changing pressure, like a chattering relief valve, or a system subjected to repeating pressure spikes/dips. Ideally, a glance at the "dancing" bar graph would instantly tell you the approximate range the reading is "dancing about", but in real life, you often can't see much at all. And I am not even talking about the ghosting effect of the LCD. I am talking about the two major drawbacks of any classical analog bar graph - the fixed range and the lack of persistence. Let me explain what I mean by that:
What if you need to see the dynamic behavior of, say, a 60-bar line? This is what 60 bar looks like on a 600 bar Service Junior:
Note the analog graph - it is exactly... three bars long! The graph's length is tied to the gauge's max range, which in this case is 600 bar, and of course, in this scenario, it is unusable. So, what do I suggest then? I suggest decoupling the analog bar graph from the gauge's max range by adding user-adjustable settings for the floor and the ceiling of the graph - in other words, make it zoom-able and thus much more usable in situations when you are interested in monitoring the dynamics of a pressure reading within a certain range. Obviously, it would be nice to have a clear indication of the floor and the ceiling values on the screen, and, once again, a graphic LCD would solve this by providing the necessary "real estate."
But, zoomable analog bar graph or not, in its classic form of a single line (alternatively, semi-circle) with alternating length, it still suffers from
So, let us talk about it. A single line on a segmented LCD can't have persistence, it's impossible, but a graphic LCD could totally change this, and here's how: the analog graph line would be flipped vertically and placed on the right side of the screen in a dedicated area below the main reading, and then it would not only change it's height at a high frame rate, but also leave a running trail, composed of pairs of lines corresponding to the max and min pressures, registered during every frame (50 ms periods for the 20 HZ frame rate) and displayed in the order of their appearance. Here's an example of what it would look like:
Yes - I actually already implemented this idea in the super pressure gauge project board, and from what little testing I've done so far, it is awesome even on this super-tiny screen! In this example, I set the ceiling and the floor of the analog graph to 30 bar and 0 bar (which you can conveniently see of the screen, along with the max. and min. pressure records displayed at the same time - take that, Parker!) , and I can easily monitor the pressure dynamics in this narrow range on a 600-bar instrument. How cool is that?! By the way, the "pressure waves" registered in the images above come from me yanking the crank of the Pressure Maker back and forth, and as you can see, my physical ability to create fast pressure spikes is definitely limited...
The screen of the development board is only 128 pixels wide, which means the trail, composed of 64 pairs of pressure records, is only three seconds long (at the rate of 20 HZ), but even so, it is much more informative than a single max and min pressure record. This could, for example, be useful for monitoring the charge pressure of a closed-loop transmission for pressure dips (see Closed Loop Failure Due to Line Elasticity for a good example of a bad example.) A single record of an abnormally low charge pressure will not tell you much - because it is what it is - a single record, and you'd need to be constantly resetting your gauge manually to confirm if these dangerous pressure dips are recurrent, but a trail like the one I just described, coupled to a zoomable graph range, would reveal this failure instantly. I can also see how easy it is to implement a function that would, for example, constantly store the last 60 seconds of the trail in memory, and you would have a pause button to stop the trail and roll it back for a closer look.
I already hear what some of you are saying: "Dude, we got an HMG 4000 for that kind of stuff!" I hear you, brother, we do too, and we love it as much as you do! But that is exactly my point - if our everyday throw-in-the-toolbox pressure gauges had this functionality, so easily achievable with a graphic LCD and an updated firmware - it would be like having a simplified version of a data logger always on hand, which would be expanding our diagnostic abilities without breaking the bank or having to reach for those pressure sensors and pesky cables (and don't even get me started on the wireless functionality of digital pressure gauges!)
I see all the reasons why the currently marketed digital pressure gauges are the way they are. People who make gauges are specialized in making gauges, and not in using gauges; therefore, their main priority is to manufacture a sellable product as cheaply as possible, and the segmented LCD is a good-enough component that cuts a lot of production costs. A quality low-power sun-readable graphic LCD, capable of sustaining refresh rates of 20 HZ, will be at least one order of magnitude costlier than its segmented sibling, so I get it, Parker, and I am not mad at you. But be warned - there's a mechanic/blogger "with ideas" whose hobby is designing digital pressure gauges that can beat your current design in every way!