Really, I offer a tried and proven and very widely used scheme in post #18. Evidently nobody understands how it actually works. ALL of the other schemes depend on electrical power and quite a bit of circuitry and hardware.
Certainly a low pressure relief valve is a totally non-electronic method, and possibly it will be a challenge to come up with the right one. BUT it is vastly simpler and a lot more reliable.

Some pumps have their own relief valves.
Why the pump needs to stay on when tap is off is a question. Why does pump keep running when flow is not needed? In hydraulics the pump is usually connected directly to mechanical drive engine, always on. Maybe it's better to control an SSR for the pump, and, have a mechanical relief loop for when the tap goes off but pump is still spinning for short period. Blow-off valves for turbo setups work this way, to relieve surge.

Even simpler is two different hose sizes. Example, run the main line as 3/4, then just T off the pump 3/4-3/4-1/4, and just run sturdy 1/4 back to the tank. Yep, the loop always flows, but 3/4 will drop the psi a lot more when tap is open, so it takes most of the flow. Close the tap and the psi in 1/4 goes up, but it still allows loop flow to keep pump from stalling or cavitation. I gave example, but sized right you don't need electronics or valves.

 

The reason to keep the flow moving is because the integral flow-thru heater shuts down very quickly when the flow is stopped. (See post #1). Demand heaters do that sort of thing, some that have a very large heat input must shut down very fast. And if the pilot shuts off, then restarting is a big effort and can't happen rapidly. THAT IS WHY!!
But folks not familiar with bypass relief valves would not think of it.

 

The reason to keep the flow moving is because the integral flow-thru heater shuts down very quickly when the flow is stopped.

The Navien on-demand tankless water heaters do turn off when flow stops. They rarely have a problem with re-ignition.

With 1/2 line to tap how much heat input is needed?

 

Really, I offer a tried and proven and very widely used scheme in post #18. Evidently nobody understands how it actually works. ALL of the other schemes depend on electrical power and quite a bit of circuitry and hardware.
Certainly a low pressure relief valve is a totally non-electronic method, and possibly it will be a challenge to come up with the right one. BUT it is vastly simpler and a lot more reliable.

We would require more input from the TS, but I suspect the pump may be battery-powered. Continuous operation would deplete the battery. On the other hand, electrical power may be expensive, and maintaining a pump’s operation continuously could incur additional costs.

As to your suggestion, this works well in my pressure washer. When I squeeze the trigger high pressure water flows from the nozzle. When I release the trigger the back pressure forces open a valve that recirculates the water through the pump. I understand your suggestion, but it might not be what the TS wants.

 

We would require more input from the TS, but I suspect the pump may be battery-powered. Continuous operation would deplete the battery. On the other hand, electrical power may be expensive, and maintaining a pump’s operation continuously could incur additional costs.

Not only that, if the pump runs all the time w/o contributing to flow at the tap, then you get to MTBF w/o much flow at the tap, hence, ROI is bad, TCO goes up.

I have studied computer computer failure over the past 30yrs, and most failures stem from heat cycling the PC (turning on and off 1+ times daily). Starting motors falls into same realm, but it's not exactly the same thing as heat cycling failures. Inrush amps to motors is non-desired, and at the same time having a motor run constantly is bad for the bearings. The "however" here is, if you need the motor to run constantly, in this specific water case-scenario, probably better to run it at 1/2 or 1/4 speed when all it's doing is circulating back to the tank. Motor and-or pump longevity will go up, and, you save on power use.

That said, it's outside the scope of the actual ask about creating a mechanical loop fluid ckt, but those are the things I would think about.

 

MTBF

Hadn't even considered that aspect of it. Excellent point.

 

In my mind, this system appears to be a component of an RV water source. In such a scenario, it is not advisable to have the pump running continuously. Instead, it should only operate when water is required. Consequently, a drop in water pressure triggers the pump’s activation. This mechanism is similar to a well water system, where a tank is pressurized by the pump to approximately 60 PSI and then automatically shuts off. When the pressure drops to 40 PSI, the pump is reactivated to repressurize the tank.

Without further input from the TS - this is just guesswork.

 

In my mind, this system appears to be a component of an RV water source.

Maybe. The diagram posted shows the water tap on end of 100m hose reel. So maybe it's a hot water reel used to wash stuff down, up to 300ft?

I myself would perhaps think about adding a control switch where the reel is mounted. When you want hot water and pressure you flip the switch to 'on', instead of relying on a flow sensor. But still add in relief loop so that if the tap is closed while out 200ft, the pump can be relieved.

 

Maybe. The diagram posted shows the water tap on end of 100m hose reel. So maybe it's a hot water reel used to wash stuff down, up to 300ft?

If this is the case - the sensor could be mounted right at the hose reel. No need to run wires all the way to the hose end. IF we're talking about a hose at all. Until more details are available we're just flying in the dark, in a fog with no instrumentation. Only an idea of where the landing strip is.

 

The heat capacity of the recirc loop will play a huge role in design.

Hi, the heater sits after the pump with a separate heat exchanger and coolant loop

A 'coolant' loop. Please tell

 

Let me remind everyone about the complaint in post #1: " but it takes too long to start up and get hot water running again. " I am not aware of the brand or model of the heater, but certainly some are much better than others. And it seems to me that the TS is more familiar with that specific application than we are.

 

Let me remind everyone about the complaint in post #1: " but it takes too long to start up and get hot water running again. " I am not aware of the brand or model of the heater, but certainly some are much better than others. And it seems to me that the TS is more familiar with that specific application than we are.

So the return loop is looping hot water back to the tank? Wondering if it's always looping hot water when the tap is off. That seems like an odd way to go about it.

 

Battery power ? ? ? We haven't heard anything about that yet.

 

{editing your comment}
The pump is pumping to the hose reel.
When the hose reel is turned off the water flow stops.
The heater goes in to shutdown.
If (you) open the recirculation loop before the hose reel the heater will stay on.

So you shut off the hose reel and open the loop back to the heater. This maintains the hot water. You want to automate this.

OK, instead of ball valves you can use solenoid valves. With a DPDT switch you can turn one solenoid off and turn the other on. That being said you can further automate this by using a DPDT Relay. The relay can be controlled by your flow sensor. When the flow stops the relay drops out and closes the hose solenoid and at the same time it opens the heater loop solenoid.

In one comment you say the pump shuts off. In another comment you want the hot water to be available upon demand.

{editing your comment}
The pump is pumping to the hose reel.
When the hose reel is turned off the water flow stops.

How are you recirculating the water?

{editing your comment}
If (you) open the recirculation loop before the hose reel the heater will stay on.
I need to automate the recirculation loop opening when flow stops to the hose reel

So does the pump stop or does it keep running?

 

Like I suggested a bit earlier: When the hose reel shuts off, the pressure rises a bit and the relief valve opens and the flow is routed back to the tank. Then, when the hose reel flow starts again the pressure drops a bit and the valve closes. Totally automatic, no flow sensor and no solenoid valve.

 

Ok bear with me, lots of quote replies, apologies for keeping my original post simple, I felt less information in this case might've been advantageous as the crux is, water stops at end of 100m hose, solenoid/valve etc opens on recirculation loop then when water starts on 100m hose the recirculation loop closes.

OK, there is another way to monitor flow, which is a differential pressure switch shunting a check valve.
And now as I think about it, what sort of arrangement does the water heater have to sense when the flow stops?? If that switch is accessible, it could serve to trigger the re-circulation pump.

The big challenge is that the heater shut-down is a safety function, since probably there is enough heat to create steam.

The water heater goes in to standby mode whereby the OAT coolant continues to run through a heat exchanger to keep things from overheating, this happens when a flow controller for a 12v 100psi water pump senses the pressure switch in the pump has been activated, this causes the pump to DE (dead end).
When the tap at the end of the 100m hose reel is opened, the pressure reduces and the pump starts circulating and triggers the heater to fire up and make hot water again, but can take around 10 minutes to reach temperature.

The heater is webasto and has features in to to totally shut down when the coolant temp exceeds iirc 85c+ so you cannot make super heated water.

What is the volume of your recirculation loop? Could you sacrifice some pressure to dump over a pressure relief valve, inline with the recirc line.

The recirculation loop is less than 1m of 1/2" car heater hose.
I'm not sure there is anything to sacrifice but I honestly don't know

Suggestion using the Flow switch and Relay module.
The Relay has a delay timer that can be set up to 5 seconds before activation.
Wiring Diagram:
View attachment 341104

This idea seems to be what I had imagined, my only question is, my pump is calibrated to put out around 1.5lpm.
Would this flow meter-sensor work as low as this? as a lot i have seen are around 1-30lpm which doesn't seem "fine" enough

Really, I offer a tried and proven and very widely used scheme in post #18. Evidently nobody understands how it actually works. ALL of the other schemes depend on electrical power and quite a bit of circuitry and hardware.
Certainly a low pressure relief valve is a totally non-electronic method, and possibly it will be a challenge to come up with the right one. BUT it is vastly simpler and a lot more reliable.

This idea was my very first idea, it's what I wanted and what I would love to work above all other ideas as it's simple to run, fix and manage, except this issue, the 100m hose, when 100m out along a flat level surface requires the pump (which can operate up to 100psi) when calibrated to push 1.5lpm to the tap at the end of the hose, but it also needs to push 1.5lpm with only 15m of hose, or 50m hose along the floor and then 12m of hose vertical.
The issue I ran in to was, I used a ball valve before a pressure relief valve to reduce the flow to the valve, but it meant constant adjustment and even when removing the ball valve the PRV required constant fiddling with and in this scenario the system needs to work all day every day in many different and not always repeatable situations, I was unable to get the PRV to activate when the tap at the end of the hose was shut and pressure had built up in the 100m hose to trigger the PRV, it was constantly flicking open and closed which basically at 1.5lpm down 6mm bore hose means the water is stopping and starting the whole time.

The reason to keep the flow moving is because the integral flow-thru heater shuts down very quickly when the flow is stopped. (See post #1). Demand heaters do that sort of thing, some that have a very large heat input must shut down very fast. And if the pilot shuts off, then restarting is a big effort and can't happen rapidly. THAT IS WHY!!
But folks not familiar with bypass relief valves would not think of it.

This is correct, if the water flow stops for what I estimate more than 20 seconds the hot water goes in to shut down to a standby mode, but if you turn the water back on after that process has started you start to get cooler water until it has completed it's shut down, it then restarts and ignites and starts heating water up, this is a 12v system with 210a worth of batteries and it takes a big toll on them, but the main issue is, you can be without sufficient hot water for enough time to cause issues.

If this is the case - the sensor could be mounted right at the hose reel. No need to run wires all the way to the hose end. IF we're talking about a hose at all. Until more details are available we're just flying in the dark, in a fog with no instrumentation. Only an idea of where the landing strip is.

Hose is correct, the hose leading to the hose reel is 1/2" car heater hose. The sensor has to be mounted here or at least not on the hose reel or 100m hose.

The heat capacity of the recirc loop will play a huge role in design.



A 'coolant' loop. Please tell

I'm not sure what heat capacity of the recirc loop is, but it's 1m of 1/2" car heater hose with a ball valve splitting it to allow flow back to a 500L water tank (this tank capacity reduces throughout the day)
Now at the start of the day, this water can be between 2-15c and will rise to around 30c from 0800 to 1200 and roughly a 1/3rd of the water has been used, as the day progresses this tank water gets hotter and hotter. If the automated recirculation loop was to work and the tank water became hot enough to use without heating (or near as) the heater would be switched off and the tank used until cooled (i'd love to automate this also, but i think that would be running in the olympics but I'm just learning to walk currently)

The coolant loop is the loop of OAT that runs through a 20L expansion tank, through a heat exchanger and then through the 5.1kw webasto heater. It's completely closed and cannot be tapped in to.

So the return loop is looping hot water back to the tank? Wondering if it's always looping hot water when the tap is off. That seems like an odd way to go about it.

Yes, looping back to the tank to keep the hot water heater running, as when water flow stops the heater goes to standby with the flame extinguished and the water cools down very quickly, when flow is restored it has to finish it's shutdown and restart which can take a couple of minutes to do and then make hot water and push it along 100m of hose which has lost all it's heat, it can take a good ten minutes to get hot water back depending on ambient temp and length of hose in use.

Battery power ? ? ? We haven't heard anything about that yet.

12v battery, 210ah

So you shut off the hose reel and open the loop back to the heater. This maintains the hot water. You want to automate this.

OK, instead of ball valves you can use solenoid valves. With a DPDT switch you can turn one solenoid off and turn the other on. That being said you can further automate this by using a DPDT Relay. The relay can be controlled by your flow sensor. When the flow stops the relay drops out and closes the hose solenoid and at the same time it opens the heater loop solenoid.

In one comment you say the pump shuts off. In another comment you want the hot water to be available upon demand.

How are you recirculating the water?

So does the pump stop or does it keep running?

This idea sort of works except I can't put a solenoid on the hose reel, the tap at the hose reel is the "switch" as when this tap is opened it needs to trigger something, I'm thinking a flow meter/sensor would sense water flow to the hose reel and shut the recirculation loop.
If there is a solenoid, opening the tap would drop all pressure out of the hose which /might/ not trigger the flow sensor/meter as there is a solenoid preventing water making it's way from the push-fit tee to the hose reel, this is where i believe a check valve would be better as the pump would be pressing against the check valve but can't go anywhere as the tap on the end of the 100m hose is shut so the flow meter/sensor wouldn't rotate thus keeping the recirculation loop open, when the tap is opened, the pressure drops a bit, the check valve can open and hopefully with the right values on the meter/sensor and/or delay on the relay the solenoid would shut suitably quick enough pushing all the water down the hose reel.

 

In one comment you say the pump shuts off. In another comment you want the hot water to be available upon demand.

How are you recirculating the water?

So does the pump stop or does it keep running?

"In one comment you say the pump shuts off. In another comment you want the hot water to be available upon demand.
How are you recirculating the water?
So does the pump stop or does it keep running?"

The 12v water pump runs when water is flowing through it, when the tap at the end of the 100m hose is closed pressure increases in the hose back to the pump which has a pressure sensor, this causes the pump to go in to DE (dead end) and it sits there waiting for the pressure to drop and flow be restored.

I'm currently recirculating by doing all my work with the hose first which means the water can stop anywhere between 20-40 times in a 20 minute period (a lot of these times the system will have gone in to a standby shut down and restarted), the hot water is then recirculated between work anywhere from 5 to 50 minutes by closing a valve to the hose reel (thus keeping pressure in the hose which is important) and opening a valve all manually to the recirculation loop.

The pump does and doesn't stop, it stops running because it has a pressure sensor which tells it to wait until pressure is reduced. So if there was a flow meter on the hose reel that triggered a solenoid on the recirculation loop, opening that loop would drop the pressure and the pump would start up again, all within something like 3-5 seconds to prevent any chance of the hot water system shutting down

 

This idea seems to be what I had imagined, my only question is, my pump is calibrated to put out around 1.5lpm.
Would this flow meter-sensor work as low as this? as a lot i have seen are around 1-30lpm which doesn't seem "fine" enough

It says it will work at 1Lpm.
The purpose of the relay module which has a delayed turn ON is to help prevent switch "bounce".

 

OK, it sounds very much like you had the wrong pressure relief valve. The correct valve will open with a small increase, but not rapidly. And it will also close slow enough to avoid that oscillation that was described. It may also be required to add an accumulator so that when the flow is stopped, hot watter can flow into the accumulator for several seconds as the valve opens. That should avoid oscillation quite well.

 

The 12v water pump runs when water is flowing through it, when the tap at the end of the 100m hose is closed pressure increases in the hose back to the pump which has a pressure sensor, this causes the pump to go in to DE (dead end) and it sits there waiting for the pressure to drop and flow be restored.

Highlighting your words:
"this causes the pump to go in to DE (dead end) and it sits there waiting for the pressure to drop and flow be restored."
I believe the term is Dead Head. You can only Dead Head a positive displacement pump. And as far as that goes - you can't dead head a PD pump. Another pump, such as a centrifugal type which can flow a lot of water but not produce as much pressure as a positive displacement pump. The centrifugal pump won't have back flow preventer valves. When the centrifugal pump stops water will back-flow and drop pressure. Lower pressure starts the pump again. Pressure builds and shuts the pump off. When the centrifugal pump stops water will back-flow and drop pressure. Lower pressure starts the pump again.
Over and over. So your pump must be a positive displacement type.

So I have another question: What kind of pump are you running?

You said the pump runs on 12 volts. AC or DC? Battery or power supply?

Sorry, I guess I'm just trying to get a sense of what you're working with.