Insane Hydraulics

Site theme image

The Cheapest Digital Pressure Gauge I Could Find

You know me - I like taking stuff apart, and naturally, when we got several of these "extremely economical" digital pressure gauges, I immediately decided to break into one. I think we paid more for 63mm analog Wika pressure gauges than for these, which makes this model the cheapest digital gauge I've ever seen:

In this post, I will open it to check out what's inside and how this thing works, and next week I'll be running some pressure tests to see how it fares as a pressure measuring device (and decide if it will be up for a long-term use review) So, let's get started!

The gauge comes in a simple cardboard box, without any instructions/calibration certificates, which is to be expected at this price range. You don't need instructions because there are only two buttons - one for turning the gauge on and off (long press) and switching between bar and Mpa (short press), and the other one is for zeroing. This model has a range of 1000 bar with the declared accuracy of 0.5% full scale - a standard figure for most mid-range digital pressure gauges.

The screen displays pressure and... that's it - no spikes, high-speed bar graphs, or anything else. To be honest - I don't know why they even bothered with MPa and bar.

And I simply can not omit to quote some of the marvelous phrases you find both on the OEM website and the Alibaba listing:

"...3305 series digital pressure gauge is suitable for calibration of the precision pressure gauge, general pressure gauge, pressure transmitters or as substitution of dead weight tester..."

"...with micropower components with its unique software processing technology..."

"high quality Intelligent digital pressure gauge"

Sounds like a bargain, doesn't it?

Here you can evaluate the gauge's size next to a "normal" digital pressure gauge (do note that one is like ten times as cheap as the other). I like the thick rubber boot with "fins" (or whatever these are called) - seems to be much more protective than the Parker's boot to me.

This is the battery compartment, which you can get to by unscrewing the three screws and removing the end cover (note: if something has "Super" on it than it must be of a very high quality):

The gauge is easy to disassemble - the front bezel comes off by turning it counter-clockwise, and inside you find the most basic (and practical) construction one can build - PCBs stacked with threaded spacers. There's a rubber gasket under the front panel, so the gauge seems to be well protected against "stuff ingress" :

This is the "heart" of this pressure-measuring apparatus - the pressure transducer cell. I didn't manage to pull it out (at least not yet) - it's pressed pretty hard in the body, but I did jam my camera inside and took these nice closeup shots:

Apparently, there's a stainless steel membrane with a load-cell pattern over it (2 radial + 2 axial cells), and a tiny PCB joining the 8 wires. There may be some passives (trim resistors) under it. The very thick and long wires are randomly jammed into the center of the round PCB, and you can zoom into these pictures and play the game "find where the wire goes" - very educative and entertaining at the same time!

The "business end" of the pressure transducer has a thread one could screw a snubber orifice in (no snubbers in there, though), and by the looks of it the hole goes straight down to the membrane:

The most basic design one can think of. This is where the "mechanical" part of the gauge ends. So far I can't state much. It's basic, and I don't mind basic. The quality of the welds does not matter at this price range. I will say more after I've tested it, and then even more after I've used it for a while (that is - if it passes any of my initial tests at all).

Let us move on to the electronic part of the gauge now. The PCBs are what I would call "DIY-grade". Two sides only, no ground planes or shields, and large SMD components. I like large SMD components because they are easy to solder by hand. I bet these long wires can pick up a lot of noise even with the stainless body around them, so by design alone this can never be a "fast" gauge - a lot of filtering must happen for this to produce an acceptable result. Once again - at this price range what this device should be able to do is (hopefully) read stable pressure with an acceptable accuracy - not more, not less.

The front end is based on a single AD8628 operational amplifier. I love it when the components are so big that you can figure out the schematic and actually read the values of the resistors just by looking at a PCB picture! This is a non-inverting amplifier configuration with a gain of 22. I also like the choice of this low-noise amp that has both auto-zeroing and chopping.

And finally - the "brains" of the gauge - the MSP430F4250IDL microcontroller (once again in a very DIY-friendly SSOP package). This is a logical choice for such a device. Plus this MCU features a 16-bit Sigma-Delta ADC - which beats the Parker's 12-bit ADC, so, theoretically, if the firmware is "solid", this gauge should be pretty accurate even with this cheap architecture - but we'll see about that when I get to the testing phase - this is a mere design evaluation report for now.

Luckily the PCB traces are large and apparent, and it is easy to see that the bridge excitation is straightforward - one of the bridge ends is tied to the ground (battery negative - and now that I mention it, the PCB that holds the batteries, has the positive terminal marked as "-" and the negative marked as "+" - I kid you not!) and the other end is fed from the positive terminal via a switching transistor. In fact - this switched bus feeds the bridge, the op-amp, and the analog voltage reference MCU pin (via a voltage divider) - which is a very simple yet perfectly functional design for radiometric measuring of a strain bridge cell.

Now, a device that boasts " micropower components" can not be feeding a 1 kΩ (yes - the cell is about one kilo-ohm across) bridge with constant excitation, and indeed it does not. Just like I thought, though - it is very, very slow. The cell is "shot" twice per second for about 30 ms:

Yep - you won't be picking up any of those fast pressure spikes with this gauge, by the looks of it!

The average current consumption is about 1 mA, and the standby current is about one microampere, so the operational/shelf life should be "golden".

My thoughts.

It's a very basic and, apparently, a very slow pressure gauge. Probably even slower than a mechanical gauge. Which makes it a contender for being an auxiliary (or emergency backup) pressure gauge at best. It should be suitable for simple measurements though - but only if it is accurate enough - and this is something that I will be testing for next week.

At this point, I can definitely see why it is so cheap, but I wouldn't use it before I confirmed that it's acceptably accurate and then reconfirmed it once more after a year of use.

I agree with the design choices - in the sense that if the components were coupled to a decent pressure cell, the gauge would be stable and precise.

I guess we'll see what happens next week then.

And - a Happy New Year to all!