IDT Wireless Flow Rate, Humidity&Temp Sensing Kit - Blog 2 - Kit Comparison

Post #2

Part of a series for the Road Test here:

IDT Wireless Flow Rate, Humidity&Temp Sensing Kit

Kit Comparison (for the flow sensor at least)

As I researched this kit, I became curious what else was out on the market. I had wanted to get a flow sensor (or flow switch really) for a while but hadn't spent much time researching what was on the market. Since starting this road test, I have found that there are quite a few options available. I had originally planned on doing a much broader comparison, but quickly fell down a rabbit hole and burned a few weeks building up a grid; but the grid started to include products that were further and further away from the original product. Ones that operate at 24 volts for industrial environments, ones that are flow switches as opposed to flow meters... and on an on. Newark has a bunch of options as does Sparkfun and other websites. I suppose it would be a proper summary to note that "there are options."

I have limited my chart here to show just the specs gathered for the liquid flow sensors that apply to this kit.

Here is a flow sensor from Turck for industrial environments for over $420 US. This would represent one end of the spectrum. And here is a page listing Sparkfun flow sensors (and some false hits for other products). There is a $10 version which contains just a hall effect sensor on an internal pinwheel requiring an external processor to count the pulses.

I also found the sensors themselves listed on Mouser (actually Sparkfun's datasheet links to Mouser). They list a lot of flow sensor options, including one flow switch with bubble detection within the liquid.

One other interesting resource that I found was this website: http://www.flowmeters.com/  which shows all sorts of different options for selecting the proper type of flow meter for any given application. There are nice animations and pictures to guide you through the selection process.

Screenshot taken from their website:

Also of note is that the configuration of the IDT flow sensor has an ic on board that talks to the sensor cube (or other devices) over I2C; so the heavy lifting of monitoring the device is put on the sensor board and not the main processor in a given application. This is as opposed to the Sparkfun paddle wheel mentioned above.

I found some sensors with data sheets that list them as "Food Grade" but others (like the IDT) don't have a mention of it. The datasheet for the FS2012 lists the following as potential uses:

So if used in a CPAP machine, I'm assuming it would be used to see how much volume is passed, which is air that goes directly into human lungs. And for liquid dispensing systems, the best use case I can think of is a automatic mixed drink maker (which would probably be considered a type of food).

Sensor type

The sensor for the flow meter uses the calometric principle

The gas flow meter is based on the calometric measurement principle so that the flow rate it assessed taking into account the known transferred amount of heat to the fluid flow and the temperature differences of the medium before and after electric coil of Thomas cylinder determined by thermocouples.

Source: http://www.cpo.hr/Paper%2015.pdf written by MARIÁN LÁZÁR, TOMÁŠ BRESTOVIČ, EVA SCHVARZBACHEROVÁ.

The easiest way to explain this sensing method is the following: Imagine a tube with a temperature sensor at each end, plus a heater in the middle. Fluid should always flow in the same direction (P1 to P2). The heater raises the temperature of the fluid and the delta can be found by comparing the two temperature sensors. The controller also monitors very closely the output current of the heater element to keep things consistent.

This image is taken from the article linked above.

The faster the fluid flows, the smaller the temperature delta. The controller can compensate by increasing the current to the heater element.

Since IDT advertise the FS2012 as containing "no cavity to cause clogging", my assumption is that they are comparing it to one of the other styles listed above with physical moving parts; although the article from Lazar seems to indicate that this style sensor does contain elements which would be directly in the fluid flow path (and thus could clog). Perhaps towards the end of the road test, I will physically open one of my two sensors to see what is inside.

I have looked at my sensor and surely the path is not straight through - it appears to have some turns in it. Hi-res images shown below. And even more, the actual orifice is quite a bit smaller than the port would seem to indicate.

Component side of the PCB

I imagine that Q1 is the transistor that runs the heater element; although it may drive the analog output which raises from 0-3V linearly based upon the percentage flow rate compared to the max. My sensors are 0-10 L/min for gas and 0-1 L/min liquid. U2 is the main processor but it has goop applied to keep the actual chip make/model a secret. U4 is a Microchip brand chip marked as "6L02E 746UK0" which appears to be a dual op-amp (Datasheet), which makes sense since this would have two temperature sensors. U1 in the bottom right, marked as "B47E", is (from the best of my googling) a flip phone from LG; although the form factor looks a little different than the Google Image Search.

Port 1 - Inlet.

Port 2 - Outlet.

These 'side shots' are pretty much orientated directly into the port, and the other side cannot be seen - the flow is not straight.

Let me know in the comments if you'd like to see one opened up and I'll do it later after testing the sensors out!

The next post will open up the kit and get it running.