Insane Hydraulics

Theme Image

Smart Industrial Monitor - Hardware Report from 26/05/2019

So, about a year ago I decided to build something with Bluetooth in it, without actually knowing anything about the Bluetooth and how it works, and then connect it to a smartphone App, without knowing a thing about how to build mobile Apps. Although I had enjoyed DIY electronics since I was a kid, this little project required the knowledge that I definitely didn't have...

Internet to the rescue! It's incredible how easily you can find guides on almost anything these days simply by consulting Google or YouTube! I still remember the times when having an internet connection was considered a luxury, and how I used to spend hours in the library searching for technical materials I would need.. And nowadays pretty much any information can be received instantly at your first request. Truly great times to be alive!

The primary "engineering tasks" that I established were:

Hardware - to create a small, portable, battery-powered device that can power an industrial sensor with an analogue current or voltage output and transmit its reading via Bluetooth.

Software - to create a mobile App capable of receiving this data, and then displaying it to the user in meaningful units.

And now, with goals defined, I had to acquire the knowledge that I was missing, and so I spent weeks doing nothing but educating myself about the Bluetooth communication and App development, and a couple of months later I could already name my "weapons of choice".

I decided to base the hardware part on the Silicon labs BLE121LR Bluetooth 4.0 module, mainly because of its long range capability and high sensitivity. As for the App - I chose the framework called React Native.

Now it was time to build something and code something, and so - one weekend later I already had my first working "prototype of prototype".

If you never saw a surface mount module hand soldered with small wires, or if you ever wondered if it was even possible - behold! This is the definition of ugly, yes, but I wanted to confirm if it worked - and, oh yes, it did! In this arrangement, naturally, the performance of the ceramic antenna was terrible, but I got to communicate a couple of bytes of data from the chip to my phone!

Here's another picture of the module after I cut it down from the "prototype of prototype" before soldering it to a proper PCB. Hideous? Yes! But a milestone nonetheless!

Ok, so now, with this "proof of concept" (so to speak) on my table, I knew it was time to get more "professional". The correct way, obviously, would be to order a proper development board, but hey - where's fun in that?

So I designed my own "development board", which, aside from testing Bluetooth data communication, also allowed me to test several other parts of the transmitter, namely the current channel protection circuit, the battery charging circuit and the voltage booster circuit (the part which is used to boost the Li-Ion accumulator voltage, which is 3.7V to the much higher 12-25V required by most industrial sensors).

In all - these boards turned out to be a good experience, but again - since I was learning stuff as I went - I discovered that the module radio performance when it's placed like that on the PCB, is very far from ideal. Be it as it may, these boards did allow me to conduct a series of useful tests and obtain lots of "know-how".

Many and many tests later I decided it was time to build something "field-worthy" - and thus the prototype that is presented in this video was born. I updated the transmitter with a multiplexer system, which would allow to swap the ADC input channels between input pins. And I also updated the booster so that I could change the supply voltage "on the fly" in the range between 5 and 25V from the App.

Then I coupled the transmitter to a Wika A10 400 bar 4...20 mA pressure sensor and (finally!) I had something tangible, something I could take with me in the field and amaze the clients with!

Well, after using the transmitter for some time, I realized that some things needed to be changed or improved.

For example - for a simple single sensor the multiplexer is not really needed. Most of the times I use the transmitter just as I would a normal digital pressure gauge - I take it out of the box, connect it to the hydraulic circuit, and press the ON button.

The connection between the sensors and the transmitter must definitely be upgraded to a more universal solution. I conducted dozens of tests with different pressure sensors and I came across the necessity to connect to four pin 43650 A-from connectors, C-from connectors, 12 mm circular connectors. So to make the connection easier I am redesigning the physical interface to a magnetic pogo-pin based custom snap-on connector, which will allow fast and easy swaps of sensors.

Also, I discovered my biggest blunder! Since I, obviously, had no understanding of how ADCs work, all of the prototypes that I built used only half of the available range, effectively halving the resolution. For example - in case of a 400 bar 4...20 mA sensor I was getting the resolution of about 0,3 bar, which is not that bad and, truth be told, never constituted that of a problem in the field, but if you consider the fact that you can have 0,15 bar resolution with the same ADC, the 0,3 bar resolution starts to look pretty lame. This, naturally, would be even worse in case of a 600 bar sensor.

This meant only one thing - back to the drawing board! So, I redesigned the transmitter, once again, tested it and right now I can say that I am pretty satisfied with the result! At this point a 400 bar 10V sensor gives me confirmed 0,1 bar resolution and a 600 bar 4...20 mA sensor - the 0.2 bar resolution.

This weekend, actually, me and my wife spent the whole Saturday in the shop testing and documenting different pressure sensors with the newest design transmitter on a makeshift pressure making test rig that I built (from scrap) specifically for this purpose (there definitely will be an article about these tests, and the "pressure maker" device itself).

In any case, at this point the latest and most precise Smart Industrial Monitor prototype board looks like this.

What's the current project status ? (Hardware-wise)

* The power control and the accumulator charging circuit is fully tested and functional.

* The regulated booster supply is tested and functional

* The current channel tested and functional (with improved resolution)

* The voltage channel is tested and functional (with improved resolution)

What's in the works?

After talking to several of our clients about this system, I decided that it wouldn't be a bad thing if I incorporated a temperature sensor. Several people asked for that. So, I am fixing a 1 wire interface to connect the transmitter to a DS18B20 temperature sensor - these sensors are sold everywhere and are just great! They have the operating range of -55 to 125C with 0.5C mid scale precision and 12 bit resolution - very decent parameters considering their dirt cheap price!

After the temperature sensor part is completed, I think I will start selling these boards for DIY enthusiasts. Simply because this board, combined with the App is a simple two channel wireless ADC, which, I believe, can be used to monitor about any analogue signal wirelessly, and also this is about the only thing that I can safely manufacture right now without making big investments or going through the hustle of certification.

Its always one step at a time...