Off-grid well pump, lowest energy consumption: variable speed vs fixed speed?

[Hastings]

I have a big underground rainwater tank as the sole water supply for an off-grid house. The tank base is 3m below ground floor level and it is 25m from the house. All the house plumbing is ground floor only so total head max about 5m.

 

How do I decide between a variable speed drive (VSD) pump that only requires a very small expansion vessel or a fixed speed (FS) pump combined with a large pressure vessel (at least 300L)?

 

I understand that the VSD pump should last longer and they are promoted as being more efficient and using less energy but the comparison is generally made against a fixed speed pump without reference to the existence of any accompanying pressure/accumulator vessel.

 

The startup current of a fixed speed motor isn't really an issue as the power system can handle momentary big spikes in excess of 4kW and anyway 'soft start' fixed speed pumps are available.

 

The priority is lowest overall energy consumption due to winter power generation being low (a wind or water turbine is not an option).

[JohnMo]

Ours is a fixed speed, think it's a 100L accumulator. Works fine. Fixed are way cheaper than variable speed and for domestic use fixed speed is fine. If your using a 300L accumulator a VSD pump would run most it's life at full speed anyway. So why bother with the complications and additional expenses.

 

With those suction and discharge heads you are only going to pull 10s of watts anyway, with the right pump.

[dpmiller]

I put a Grundfos Scala on a job recently (ring main for a DI water system) and it's remarkably good at modulating it's speed and will shut off completely if dead-headed. Previously the customer had a single speed Grundfos boost pump and the reduction in waste energy shows in the holding tank (270l or so) being about 15 degrees cooler than previous...

[SteamyTea]

Can you make a water tower, then fill as fast as practical?

 

Thinking outside the box, how about a ram pump.

Pump 10 litres into it and the optimum flow rate, and get 1 litre at the head you are after.

(the key is the optimum flow rate)

[Hastings]

I already have a variable speed pump, specified and installed by others, but it's never worked well. Installed above ground at the house, it can barely achieve 1.6 bar and pulls its full 1200W at all flow rates. 2 plumbers have said it must be faulty, the supplier tested it at their workshops and deemed it not faulty. The suction piping resistance seems the obvious culprit - it is 32mm black (green stripe) MDPE but includes 3 tight 90 elbows (two of them very close together) and a mesh filter close to the pump intake which are all 'no-no's in suction line design - yet when I ran a test to eliminate the suction piping entirely by putting the pump in the supply tank as a submersible, it still couldn't achieve 2 bar pressure (hence the conclusion that it was faulty). The supplier suggests selling me another pump to put in the supply tank to help the existing one. I suppose they would, but I feel it's just wrong.

 

I can eliminate the two tight 90 elbows close to the pump by completely refitting out the entire pump house along with the pressure vessel, UV and string filters and can probably replace the remaining 90 elbow just outside the tank with two 45 degree bends (maybe even just a long gentle bend of pipe) - can do that myself I am fairly sure - then try the existing pump again. But would it make that much difference to be worth all the work?

 

Alternatively I am considering selling the existing pump and starting afresh. The obvious solution seems to be to go the submersible route. That would involve the expense of getting the groundworker and electrician back to get power the 25m to the tank. 

 

@JohnMo So all I need is a single 500W 'ish submersible fixed speed pump to run the entire system? I only need 2 bar pressure and 15L/min (20-25L/min would be better). Does yours have the pressure switch integrated into it and can the pressure be adjusted at all?

 

 

Edited June 6 by Hastings

[JohnMo]

There are several ways do things

First low flow (about 20L/min) and big storage (300L is quite large with 150L usable capacity), or high flow 40L/min and small storage.

 

Then there is pump location submerged or not. Not submerged requires care to ensure the net suction head available and required marry together. Get it wrong you cavitate the pump.

 

My pump is down hole 34m (vertical lift) and 25m to pump house and then upto 2.5 bar in accumulator, flow rate is 40L/min it pulls 1200W also.

 

Note: Looking at my central heating pump on speed 3, it flows 1m3 and pulls 33W, so there must be low wattage pumps out there.

[Hastings]

1 hour ago, JohnMo said:

My pump is down hole 34m (vertical lift) and 25m to pump house and then upto 2.5 bar in accumulator, flow rate is 40L/min it pulls 1200W also.

Is the water level 34m below ground? I'm interested to know the total head.

 

Thanks for your time. It's really helpful.

[JohnMo]

26 minutes ago, Hastings said:

Is the water level 34m below ground? I'm interested to know the total head.

Not exactly sure where the upper water level actually is. When they drilled they hit some water at 6m, but had to go down to 35m to get lots of it. 

 

Total head unsure, that would depend on the current water level. But with the amount of water we have had it could be quite high. Say 6m, plus 25 (0 to 2.5 bar) plus couple of meters pipe losses. So about 32 to 35m.

[Hastings]

@JohnMo Thank you. That sounds like our borehole too: 45m of drilling but the level is usually at 3-4m after a short rest. Ammonium in our new borehole killed it for household use.

[markocosic]

We run one of these:

 

https://www.electroil.it/en/p-23-inverter-archimede-immp11w.html

 

Approx €350 on fleabay

 

With a VS2/14 GRĘŽINIO SIURBLYS FRANKLIN VS 2_14 0.75kW 230V.pdf borehole pump ~ 40 metres down and providing 3 bar head at ground level for domestic use. Deffo doesn't run at 100% most of the time.

 

It doesn't however provide a residual 3 bar head if you're using a 32 mm pipe to fill a bucket. The pressure drop of the pipework between the pressure sensor and the bucket isn't enough. It will then run at 100%. (and most of the water will bounce off the bottom of the bucket and hit you in the face)

 

Note that a "2 bar" pump might only deliver 2 bar at zero flow; or full flow at zero bar; and half flow at half the rated pressure would be expected. All about the pump curve but marketers can be misleading...

[SteamyTea]

https://www.engineeringtoolbox.com/pumps-power-d_505.html

 

 

[Hastings]

@markocosic Interesting, thank you. Good advice about flow and rated pressure.

 

I've studied both your units on their web pages but am not entirely clear... Your combo is a single phase fixed speed pump converted to variable speed (constant pressure) by the Archimede sensor/control unit?

 

Seems more advanced than simple ON/OFF control, according to pressure, like the £20 https://www.ebay.co.uk/itm/272751286212 that @JSHarris describes in his system.

 

@SteamyTea Thanks. Using that calc it looks like a 400W submersible should be more than enough for me (confirming @JohnMo's advice).

 

[Hastings]

@markocosic Sorry, just found more info on that inverter and that answers my question. It does control motor speed, even single phase motors. Also has dry run protection.

[markocosic]

On 07/06/2024 at 12:02, Hastings said:

Your combo is a single phase fixed speed pump converted to variable speed (constant pressure) by the Archimede sensor/control unit?

 

Yes. 

 

If I had our time again I'd use a three phase pump to reduce any potential for reverse rotation; but the wide head ashtray bought a single phase pump before this and intended to use it equity a basic pressure switch.

[JohnMo]

5 minutes ago, markocosic said:

three phase pump to reduce any potential for reverse rotation

?

Just install a check valve on the discharge side of the pump, no reverse flow, no reverse rotation!

[markocosic]

On 09/06/2024 at 12:26, JohnMo said:

?

Just install a check valve on the discharge side of the pump, no reverse flow, no reverse rotation!

 

Check valve present. Still technically possible to reverse rotate a single phase pump even with this if you catch it at the right moment. Impossible to reverse rotate three phase pumps though; easier on wiring/currents/harmonic noise/capacitors in the inverter, no need for starting cap etc. Why wouldn't you?

 

Forced cooling doesn't appear to be neccessary on that inverter in a northern climate for what it's worth. Accidentally tripped on overheat - once - when testing with it laid horizontally on the floor and filling the sewage treatment tank. Else fine doesn't get warm enough to need fans that can make noise and fill with dust.

[sharpener]

On 06/06/2024 at 18:34, Hastings said:

when I ran a test to eliminate the suction piping entirely by putting the pump in the supply tank as a submersible, it still couldn't achieve 2 bar pressure (hence the conclusion that it was faulty)

 

We have a similar setup, with a 5 cu m rainwater tank just below ground level, with a Stuart Turner jet pump and pressure-sensitive control switch, conceptually the setup works fine for a two-story house even with a longish suction line. The response to a tap being turned on is pretty instantaneous as I think the elasticity of 25+ m of 20mm underground delivery pipe provides something of a buffer. It only runs when water is being drawn off so its consumption is not a problem even when we are running off our battery after dark.

 

The pump stopped achieving adequate pressure and after fruitlessly replacing the controller with one from Clark it took a lot of further effort to discover the reason was a hairline crack in the plastic impeller shroud which was I think causing a sufficient internal leak to negate the jet pump action. This was the second time it had failed, previously I had had to replace it because of a leaking shaft gland but at least I had enough bits left to make one working pump again.

 

They are only warranted for 1 year and no spares are available, Stuart Turner buy them in from the Far East and regard them as non-repairable (and their customer service was rubbish) so if it fails again I will replace it with a similar jet pump from DAB as it has a 2-yr warranty and you can get spares for it.

 

So I would think this DAB pump would suit your application pretty well, and installed above ground is more accessible than a submersible.

 

 

 

 

Edited June 18 by sharpener

[Hastings]

@sharpener Thank you, that's extremely helpful info.

I didn't know about jet type pumps and it sounds ideal. Steelpumps who make the pump I already have (and isn't working well as an above ground suction pump) don't seem to do them, or the UK agent doesn't import them.

I'm understanding now that submersibles use less power than pumps that have to suck water from below them, so that's probably the way I will go because of my winter power scarcity.

 

On 06/06/2024 at 18:00, SteamyTea said:

Can you make a water tower, then fill as fast as practical?

 

Thinking outside the box, how about a ram pump.

Pump 10 litres into it and the optimum flow rate, and get 1 litre at the head you are after.

(the key is the optimum flow rate)

 

Ram pumping is out of the question because I can't afford the water use. Already limited to maximum 2 person occupation of the house by private water supply regulations (new builds): have to provide 200L per person per day (167sqm of roof in a location getting an average 1,300mm/yr rainfall yields a daily average of 480L/day.)

Edited June 19 by Hastings

[sharpener]

10 hours ago, Hastings said:

I'm understanding now that submersibles use less power than pumps that have to suck water from below them, so that's probably the way I will go because of my winter power scarcity.

 

What is your control strategy/mechanism for the submersible? The focus on energy consumption and VSDs makes me think you will be running it all the time! (Submersibles often come with float switches but that is for use as a sump pump not a water supply.)

 

The pressure control unit on the jet pump (see this link) has both pressure and flow switches in it. It starts the pump on the pressure drop when you open a tap and it continues to run while there is a flow rate of > ?0.5 l/min. When you turn the tap off the system pressure rises until the pressure switch stops the pump. The non-return valve in the control unit then maintains the pressure in the pipe until the next time. There is a 10 sec max no flow run-on period to prevent dry running.

 

The user experience is like having mains supply. I chose the smaller of the two sizes of ST jet pump bc I was a bit worried about starting current causing the lights to flicker but it isn't in fact a problem. This pump is nevertheless powerful enough to allow two showers simultaneously. I have since installed a battery system with a 4.4kW Victron inverter and it copes fine with the starting current.

 

On 06/06/2024 at 11:11, Hastings said:

How do I decide between a variable speed drive (VSD) pump that only requires a very small expansion vessel or a fixed speed (FS) pump combined with a large pressure vessel (at least 300L)?

 

With a pressure/flow control switch as above you would not need any kind of pressure vessel so it will save you the expense and space. If you can install the pump near the tank it would be advantageous because it is better to have a short suction line and long delivery line that v-v but if not it may still work.

[sharpener]

52 minutes ago, sharpener said:

With a pressure/flow control switch as above you would not need any kind of pressure vessel so it will save you the expense and space. If you can install the pump near the tank it would be advantageous because it is better to have a short suction line and long delivery line that v-v but if not it may still work.

 

Edit to add:

 

I have no direct experience but would have thought the Clarke control box should work equally well on a submersible pump. If it has enough head to supply your house it should reach the required cutoff pressure which is IIRC 1.5 bar. You could install it anywhere in the delivery line e.g. where it comes into the house, much more compact than a pressure vessel.

 

I notice also that Clarke do jet pumps as well, the bigger of the two looks identical with the larger Stuart Turner one.

 

Edited June 20 by sharpener

[Hastings]

12 hours ago, sharpener said:

What is your control strategy/mechanism for the submersible? The focus on energy consumption and VSDs makes me think you will be running it all the time!

I would use a stop and start pressure switch - either a bullt-in one or separate one located at the house. I'm not particularly wanting VSD but just happen to have that at the moment because the supplier suggested it in order to reduce the risk of power spikes causing a problem with the house power.

At the moment my pump crashes the 5kVA Victron inverter with its 1200W demand unless the battery bank is fairly well charged, which is generally only in good summer weather. Any way I can reduce the total house peak power demand means I can put off needing to replace the battery bank.

If I get a jet pump and it turns out I have to put it nearer the tank it means I will have the added cost and trouble of housing it (and without spoiling the view).

 

What do you do for winter power?

[sharpener]

On 06/06/2024 at 18:34, Hastings said:

I can eliminate the two tight 90 elbows close to the pump by completely refitting out the entire pump house along with the pressure vessel, UV and string filters and can probably replace the remaining 90 elbow just outside the tank with two 45 degree bends (maybe even just a long gentle bend of pipe) - can do that myself I am fairly sure - then try the existing pump again. But would it make that much difference to be worth all the work?

 

Alternatively I am considering selling the existing pump and starting afresh. The obvious solution seems to be to go the submersible route. That would involve the expense of getting the groundworker and electrician back to get power the 25m to the tank. 

 

Having re-read the entire thread there were some points I had not taken in.

1. You are completely off-grid
2. You do not have electricity out to the tank location ATM
3. You already have a VSD pump but it does not work properly.

 

Your existing pump sounds as though it ought to do the job but I am not sure about the long suction line, if it is/can be used as a submersible then it is probably designed for flooded suction side rather than having a significant suction lift.

 

I would not have thought the bends etc would stop it developing enough pressure head though it might reduce the delivery flow. I have got at least 10m of suction line as well as a foot valve with integral strainer, several elbows and another fine mesh strainer before the pump.

 

If you need to run a cable out to the tank you could try it with an extension lead first. Have you tried the existing pump as a submersible to see if it works any better in this mode?

 

I am not off-grid so there is mains backup available to help with motor starts but I don't think it makes any use of it. Can't be completely sure as the Victron current graphs don't have the time resolution. Occasionally they catch the starting current of the freezer which produces a really big spike. If I remember I will test it tmrw without grid power.

 

IIRC Victron recommend an inverter 5x the nominal motor power for single-phase motos (and 3x for three-phase which are easier to start as upthread). Hence the 5kVA (which is only rated for 4kW continuously) is not really big enough for a 1200W 1-ph motor.

 

So the solution to this is probably a smaller pump, hence my suggestion of the small jet pump which IME is adequate for a similar house. They will handle a suction lift of 7m though the performance graphs show a significant drop-off in flow rate (from 40 to 20 l/min at 3 bar delivery pressure) but still enough for you. Depending on the relationship with your supplier he might lend or hire you the pump he thinks more suitable. Fitting a second pump to help the first sounds like a cop-out to me and no advantage.

 

It also occurs to me a standard centrifugal pump might be more suitable in your situation with significant resistance on the suction side, but I don't know enough about their characteristics to be confident. Maybe find a supplier (?Anglia Pumps?) who stock them all and so are in a better position to give dispassionate advice.

 

 

On 06/06/2024 at 18:34, Hastings said:

@JohnMo So all I need is a single 500W 'ish submersible fixed speed pump to run the entire system? I only need 2 bar pressure and 15L/min (20-25L/min would be better). Does yours have the pressure switch integrated into it and can the pressure be adjusted at all?

 

Yes, that should do it. If it does not have an integrated control (haven't seen any reply from @JohnMo) then with the Clarke control unit in the house feeding the pump circuit. But it would still mean a cable to the tank location, which you might be able to avoid with a jet or centrifugal.

 

 

 

 

Edited June 22 by sharpener

[JohnMo]

On 06/06/2024 at 18:34, Hastings said:

Does yours have the pressure switch integrated into it and can the pressure be adjusted at all?

Sorry missed the question. Mine has an adjustable pressure switch on the accumulator which is wired in to the pump control. Think it's currently set at 2.5bar. so pump starts and stops based on accumulator pressure.

[sharpener]

Thanks for adding that. Do you actually need an accumulator with your pump @JohnMo?, mine works OK with just the Clarke pressure/flow switch (which looks to be a copy of the original Stuart Turner one I think made in Italy), threshold 1.5 bar, not adjustable but works fine for us (assuming the pump does not have a cracked housing as upthread!). So can be used up to 15m delivery head which is a 3-storey house.

 

Have just done some tests starting the pump when off-grid. We have a Victron Multiplus II GX 5000 with 10.65 kWh of Pylontechs, currently at 34% SoC after cooking breakfast.

 

Pump is a ST Jet 55-45 rated at 55 l/min at open outlet and 4.3 bar closed. Curve is a straight line with a slight upwards kink so goes through (35 l/min, 2.2bar).

Nominal power quoted as 370W, maximum 650.

If I were buying new I would either go for the cheapo Clarke equivalent or the branded DAB as per link above.

 

Base house load is 350W

Pump adds another 550W.

Off-grid there is a brief dip on the lights when it starts but nothing serious.

I tested it also with the kettle boiling so a standing load of 2408W, the flicker is no worse when the pump starts. It then draws 2957W total so again about 550W for the pump which is well within the advice Victron give on motor sizes.

 

I would have run a test with a longer suction line but I do not have enough spare suction hose for this to be meaningful.

 

You do have proper spiral reinforced or rigid suction pipework don't you @Hastings? Any ordinary flexible hose at any point on the suction side will collapse under vacuum and seriously restrict the flow.

 

 

[JohnMo]

4 minutes ago, sharpener said:

Do you actually need an accumulator with your pump

Limits the number of start-stops, quite a bit.