I'm playing with the idea to use an arduino (or esp8266) to control the water flow and temperature for the shower!
Now i'm fully aware that this would involve something to give input to control the water (like setting the temperature and water flow). This topic is not for that. Here i'm merely curious which components are best used to control it digitally. Controlling the controller (say via a display) it is something for another time
What i'm looking for is the following:
How to control the water flow? I think i found the solution in a motorized ball valve, is that the way to go?
How to control the flow based on temperature?
How to mix the cold + hot water?
Now i can find the individual components, but that means 1 valve for water flow, 1 for temperature. And i'd need two of both (for hot and cold). That is a bit too much valves i think. On top of that i'd also want to add a mixing valve to mix the hot and cold water. Just passive mixing, no knobs on there as the valves before it should already output the correct hot + cold water.
Ideally i'm looking for a valve that reports back the temperature + flow rate and allow me to adjust the flow rate with probably a motorized ball valve. I prefer to not have any additional stuff in the water flow directly for flow measurement so i think i prefer the ultrasonic based versions. The hall effect versions, while that effect itself is free of touching the water, the spinning magnetic wheel in the water is.. well.. in the water..
I think, to have full control from a software point of view, i would want 3 valves. One for hot water and one for cold and one for mixing the two outputs. But finding these valves seems to be rather challenging. To complicate matters slightly more, the resulting water flow must be "normally closed". So in other terms, when the valve(s) are not powered on, the flow must be off. This can obviously be done at the last valve in the chain (say the mixing valve) to only have a need for one normally-closed valve.
Just for a fun reference, this tech should exist! As you can buy this! It's waaaaaaay too expensive though.
I'm looking forward to your suggestions! I'm sure there are valves out there that can do this. I don't expect it to be cheap like $10 or so. Based on what I've found so far with valves is that they easily cost anywhere between $10 till $100 a piece.
Surely two proportional valves, one mixing-baffle then one temperature sensor (hopefully fast reacting).
Temperature feedback to set the relative valve levels, overall flow sets the collective valve levels.
Alternatively two temp sensors, two flow-rate meters, and you can predict the mixed temperature and feedback direct from the flowrates which is faster to respond (no latency due to transit-time, nor from thermal mass of temp sensor under fluctuating conditions).
When you write that something is way too expensive, consider it may function properly for 50 years or so. How long will your contraption operate with zero maintenance? How often will you be tearing into the shower wall to replace one of your cheap valves? How much will that cost? If you ever expect to sell your home, what will a prospective buyer think of your undocumented, cheap, no warranty devices? I would quickly walk back out the door.
There already exists a tempering valve for hot water systems ( in Aus at least) which are a mandatory fitting.
Note that untempered water not allowed in domestic situations.
Must be installed and adjusted by a registered plumber and in new installs, install is double checked by local council inspector.
Pretty foolproof zero maintenance device except for annual checks etc.
markg85:
I'm looking forward to your suggestions! I'm sure there are valves out there that can do this. I don't expect it to be cheap like $10 or so. Based on what I've found so far with valves is that they easily cost anywhere between $10 till $100 a piece.
I'm with @Paul_KD7HB on this. Building a DIY system that is reliable, and can be proven to be reliable, just would not be worth the effort when you consider the price at which you can buy a thermostatically controlled shower unit.
I think if (for whatever reason) I wanted remote control of the water temperature I would use a standard thermostatically controlled unit as the base and build a mechanism to adjust it.
I might not have been clear enough as to what i try to accomplish though. The net result should be the shower, yes. But at this point it's very much just an experiment to see if i can make it at all. I have no specific plumbing hardware for it, but i'd picture 3 buckets. 1 with hot water, one with cold water and an empty one. The empty one would be the output, hot and cold the input. The input would be fed to valves with actual plumbing but to create water pressure it would just be two aquarium pumps (one for hot and one for cold).
The idea of using a thermostatically controlled valve is great too! Why not use something that is specifically tailored for this
The thermostatically controlled valve only guarantees x temperature at any flow rate (heavily simplified). While it itself controls flow rate to keep a constant temperature, it has no mechanism for me to control flow rate. I can only control it's target temperature by turning it. (yes, i have this in my current shower, i know how it works)
So, to actually control it, i still need some form to manage the flow rate and adjust the temperature output of the thermostatic valve. But how would you go about that?
Reduced port , linear flow, motorized valves.
One reverse acting so as one opens the other closes.
Alternatively one motor and two linkages that control the mixing valve.
Flow mixer in the valve discharge stream .
Mixed flow temperature sensor as far downstream of the mixing device as possible.
Expect flow volume fluctuations unless you have a separate total volume control system.
Almost all of this can be done with springs in the flow stream moving ports on a valve.
