Saving water with Flukso

Hello FLUKSO community!

We just installed a Flukso to monitor water and gas usage using reed switches on the pulse ports. Within a few days it was discovered showers were using around 130 Litres (!) each, when generally it is advertised in the media we should aim to use under 100 Litre/person/day.

The solution was to install flow restrictors (nominally 9L/min) and now showers are down to ~30L. The actual flow rate is 13L/min, calculated by measuring the time to fill a 10L bucket. I assume the little plastic flow restrictor disc depends on input pressure?

At this rate the FLUKSO should pay for itself. Further planned comparisons include: washing dishes by hand vs using the dishwasher, the effects of using different programs on the washing machine (small wash, large wash, water saving, extra rinse cycle etc).

We might consider changing the flukso daemon to export data to Pachube/COSM as well to overcome the difficulties of data analysis using the current Flukso website/graphs, eg:

------ cumulative counts of energy/volume used, with an option to specify a date range
------ cumulative count of energy used over logging lifetime, with the option to offset by the actual value on the meter, to get an idea of accuracy/to see if the Flukso is keeping up. (ref https://www.flukso.net/content/add-support-ading-real-meters-values)
----- live list of pulses or export raw pulse data to CSV (currently one has the ability to use either the Flukso website API or the FLM02 API but one would need a service regularly polling this as it seems the data is 'gracefully aged' in both the website and the FLM02 device)

Ref: https://www.flukso.net/content/pachube

If only PVoutput supported consumption other than kWh!! It currently supports auto-load of FLukso data via the API and it's server-side data management and graphing etc is very advanced.

icarus75's picture

I'm interested in the outcome of your water experiments. Keep us updated!

Features 1 and 2 should eventually come to the Flukso dash. As for the raw pulse data: we don't actually store the pulses. So there's currently no way to get historical pulse data out of the platform.