Are there any "cheap" flow meters for DIY metering with an Arduino/ESP32?

[gkoulomz]

I was looking into options for flow monitoring recently because I was interested in measuring the efficiency of my heatpump with Heat Pump Monitoring — OpenEnergyMonitor 0.0.1 documentation (Level 3).

I've not gone ahead but the best option I've found without getting a pump providing flow speeds is a non-invasive ultrasonic flow meter like "TUF-2000M" https://www.aliexpress.com/item/1005003995961461.html (various people sell them, no idea if this seller is any good) which you can find for ~$190. The ultrasonic sensors get clamped onto the outside of the pipe, the instructions do call for 30x the pipe diameter spacing from the pump to the sensors on a straight run which might be awkward (45" on 1.5" pipe)

A bit late for you, but others may be looking
I've been playing with the TUF-2000B. It works (after spending hours trying to understand it) with caveats.
First of all, the AliExpress shop where I got it offers NO SUPPORT!. If you want to get one, check out Good Price TUF-2000B Wall Mounted Ultrasonic Flowmeter Manufacturers and Suppliers China - Brands - Taijia Technology. I've been communicating with them via whatsapp, and although its slow, I do get some support. Prices seem good, but don't know about shipping.
Documentation is not very good. Looks like there are several clones / generations of this product out there, and you can search google and find various docs describing various versions. Between them, I figured it out.
The biggest problem I have encountered, it seems, is that I mounted the device about 60" downstream from my SWG. When the SWG is on, however, the device becomes quite erratic Took me a while to figure that out. I don't have a better location to attach the sensors, since the lines are buried, so using the totalizer functions will not work.
I performed a "bucket fill" test, where I plugged one return, put a short vac hose on the other end, and timed how long it took to fill a 10 gallon bucket. Figured out all the error bars. The Flow Meter reported 50.5 gal/min, while the bucket test gave me 48.95 gpm. About 3% error, which is well within my error bars!
The device has lots of features - you could easily go off the deep end on this. The most interesting for me are the ability (with extra sensors) is to measure BTU added to your pool by your heating system, interfaces to RS485 to capture data; interfaces to support current loop sensors.
The only other issues is that the readings are noisy, i.e. instantaneous values jitter quite a bit (+- 5%). There is a filter built in, but that means you always have to wait a while for the reading to settle down after a significant state change (i'm talking 30-60 sec).

 

[jakeblat]

@gkoulomz Sounds very interesting, makes me more tempted to buy one to mess with. I am interested in a more effective way to measure my heatpump output.

I currently just estimate the heating efficiency using a temperature sensor in the pool and use the rate of temperature change converted into effective heated watt hours for my pool volume. Then compare that to the realtime power consumption of the heapump.




Of course it's currently not really reflecting what the heatpump is doing because a lot of that temperature increase is actually coming from the sun and high air temperatures. But it gives a rough idea

​

 

[generessler]

@gkoulomz Sounds very interesting, makes me more tempted to buy one to mess with. I am interested in a more effective way to measure my heatpump output.

Yeah not just heat pump, but whole system flow, too. I wish flow meters were more recommended and included in automation by default. I added an "on the pipe" meter just before I switched to a VSP. The idea was to measure a baseline flow rate for comparison with the new pump. That worked great, but was just the beginning.

The meter is so nice as a general indicator of settings and health of everything on the pad: filter, skimmer, and pump basket clogging (far better than a pressure gauge), heater bypass position, pipes freezing in winter, etc. etc. It was surprising how much flow changes depending on valve settings and basket/filter cleanliness in addition to pump RPM.

In 3 years the only glitch was a seed that jammed the indicator bead one time. Five minutees to fix. And that turned out due to a missing screen on the filter's bleeder pipe.

So I intend to webify flow somehow. This device is an option. Or as a quick hack I could just point an outdoor camera at my existing meter.