I'll take a look at this. Thanks for sharing.
Is there any place along the stream where it passed through an area of fixed size, such as a large pipe underneath a major roadway, or an area where the banks of the river are retained by walls or other structures? That would make volume measurements much simpler.
Yes. I will look for a place in the stream where the transversion section is stable
I found a type of sensor that might be more appropriate for the stream conditions.
here it is at work:
Now, how to use it with an Arduino?
What information do you get from the spec sheets that indicates it will interface with a Arduino Uno?
If you do not read the spec sheets then you will not be able to figure out how to use the thingies, right? So, how do you expect us to know how to use those thingies if we do not look at the spec sheets. Frankly, when I search for thingies I look at spec sheets first. It is quite telling about a product when it does not have a spec sheet.
Stop swearing it's not nice.
The ultrasonic speed sensor for boats is a great option
Here it is at work:
And here's how to connect the interface it uses (NMEA 2000) with an arduino:
Not $6000, but quite expensive all the same.
And @prsoares hasn't yet stated the budget for this project. Maybe even a $300 sensor is too much.
$300 is much better than $6000
How are you going to get that to work?
It's for clamping onto pipes.
I've only ever seen them working on pipework associated with waste water treatment.
In any case, they don't come cheap.
As far as I remember, they are called Doppler flow meters or sensors.
I can only speak for waste water flow measurements.
By far the most commonly used method is to use an ultrasonic transmitter/receiver placed over some sort of flume, sometimes referred to as open channel.
The shape of the flume is designed to change the height of the liquid in the flume as the flow increases or decreases.
The big advantage is that it is non-contact and avoids the fouling you get in sewage.
I had some experience with river flow measurements and agencies like the UK Environment Agency liked to use a traditional float usually attached to a cable and pulley and a chart recorder.
The bottom line is that anything in contact with anything other than clean water is going to suffer from fouling, leading to inaccuracies and higher maintenance costs.
You could do worse than find out what your river authorities use.
I intend to use the UST800 SMART (https://www.boemarine.com/ust800-smart-nmea0183-bronze-hsg-st.html?qty=2) speed sensor and measure the cross section at the point where it will be installed. The biggest problem is communicating the NMEA 2000 interface with a microcontroller. It is based on the CAN network. I found some articles and posts that explain how to connect NMEA 2000 to an arduino, but I'm still reading it.
Some articles and posts:
https://www.electronicshub.org/arduino-mcp2515-can-bus-tutorial/ (tutorial for connecting CAN Network to an arduino)
NMEA 2000 Shield - #5 by system (Some comments on how to make an NMEA 2000 interface with arduino)
Sailboat Instruments: Building your own NMEA 2000 Device (Connecting a NMEA 2000 Compass to an Arduino )
...calibrating the system to determine flow rate from surface velocity only.
Even if you have the "stage" (depth) from another sensor, you will need to calibrate the system to obtain reliable flow rates.
And if you have "stage," you may as well just use that and discard the velocity sensor.
Apparently it can emit NMEA 0183 too. That's easier to deal with I suspect.
I think so too.
I found this project about connecting NMEA0183 with an arduino:
But apparently NMEA 0183 is being phased out. I asked this store ((https://www.boemarine.com) if they had a 0183 cable and they said they didn't.
Last I used 0183 (which was a long time ago) it was just RS232 (maybe officially RS422), so it should be possible to make a cable and interface it to a MAX232. Maybe talk to the manufacturer though - their site only mentions NMEA-2000. Maybe your vendor has an older model in inventory.
The OP means velocity sensor. Mid-stream velocity is used as a proxy in such measurements, often coupled with water depth and a channel-profile model.