Pressure test success!

A rainy Sunday provided an opportunity for some father/son #watergeeking. We rigged up some bungy cords and fired up the garden hose to test the pressure sensor for 3 m of head.
With Max on the hose, everything got thoroughly wet but we managed to fill the vertical riser pipe and record the sensor reading to inform calibration.

With no water in the pipe, I observed a reading of 101, while with 3 m of water I recorded 343. This confirms the sensitivity and gives as a linear adjustment factor for pressure calculations.
While I'm waiting for the second sensor to arrive I'm going to start putting together the prototype rig which will have a few configurations on the outlet to allow testing of submerged and open air discharge scenarios.

I'm also dusting off some high school maths to solve the head/pressure relationship for the check valve. This is proving to be a good mental challenge!

Sense and sensitivity

Monday evening served as a great opportunity to solder up my fancy new pressure sensor and see if it cut the mustard for measuring the low heads expected.

This week's laboratory was my mum's kitchen sink (renos are full steam ahead at our house so no bath).

Initial results from the Adruino serial plotter on my laptop were very promising, and I got a resolution of about 10 mm which should be plenty to work out flows passing through the WaStop. Much better than the 250 mm I was getting with my other sensor!

Next step is to calibrate it properly with a longer pipe then start putting together my testing/demo setup. I found a very nice plywood pallet on the side of the road last week which should make a perfect base.

Whizz bang sensor has arrived

A yellow santa (Courierpost) arrived with a new sensor and cable from Element14.
Out of the box it looks great, with a great finish quality and some nice sealing on the cable. Having a datasheet is awesome too!
I'll get this wired up for some testing in the next few days.

A solution to sensor resolution

With a pressure sensor mounted I could now start testing what pressure readings were returned for the low heads expected during operation of the Wastop.
I plugged in the sensor to the analogue input on my Ardunio UNO and ran a simple program which returned the direct output from the sensor (a value from 1 to 1050) via a laptop and set the serial output plotter running.
The photo above shows my son Max getting amongst the testing, being a big fan of water play there was plenty of splashing going on. We poured in water to the top of the 500 mm tall riser pipe to see what values were returned.
With the sensor just submerged I got a reading of 99, while when full to the brim I got only 102.
This range of only 3 for a water height difference of 500 mm indicates that the sensor I'm using has a pretty low resolution.
In a worst case, we may get up to 5 m of head behind the valve, and we need to be able to measure pressures of at least 20 mm difference in order to accurately determine the flows passing through the valve.
The sensors I have were a generic pressure transducer from AliExpress and after some further digging I found that they were actually rated for 1.2MPa which is about 120m of water head, way above what we needed.
It took some digging to find an alternative supplier, then even more to find a suitable sensor and connector which would work at the 5V that the Arduino operates at.
This fancily named Telemechanique sensor from Element 14 should fit the bill, and with a 1 bar rating (10m of head) it should have enough resolution for us to tell what is going on upstream and downstream of the check valve.
It also has a detailed data sheet which will be very useful as I calibrate the sensor.
At $260 all up it wasn't cheap, but it is a critical part of the development so I went ahead and ordered.
I can barely contain my inner geek waiting for it to arrive so I can plug it in and test the resolution!

Tapping up a storm

NZ is currently in a COVID-19 lockdown and I was missing a key piece of equipment to progress the testing and calibration of the pressure sensors.
In order to mount the pressure transducers to the uPVC pipe coupling, I needed a 1/4" BSPP tap. I had ordered one from AliExpress a month or so back, but with the delays and pause in non-essential shipping, it was unlikely it would arrive in the next few weeks.
As often happens, mentor Brett Ogilvie hooked me up with his Dad and we had a great chat on Zoom. Turns out he was formerly an instrumentation and control guru for NZ Steel and knew a thing or two about measuring pressure and flow. Heaps of ideas to ponder there, and he also had the tap I needed. Unfortunately his location at the other end of the country was a barrier in the current lockdown.
One option we had discussed was using the sensor itself to form the thread, a risky approach given the likelihood of cross threading the soft PVC, but with a combination of countersinking and careful alignment I managed to form a solid thread and mount the pressure sensor!
A small victory but a crucial one allowing me to progress to the next step of calibrating the sensors.

Wastop in the house!

Thanks to Brent at Hygrade Water, the NZ suppliers of Wastop, an awesome new PVC check valve turned up in the mail today.
At 110 mm, it'll be used for building a prototype to demonstrate and test the Smart Check Valve setup. Just waiting on some acrylic pipe to complete the setup, to which I'll add pressure gauges and eventually telemetry.