Archived Forum

Sorry no experience of the Willis device that you ask for but to respond to your stratification comment. If you connect a pump from top to bottom of your tank and switch this on only when the stat opens, you can fill the entire tank with dumped power. The pump runs on PV power and only when the store is filling so no real extra running costs.

I have this working at the moment and experience a convection path through the pump when it is switched off which potentially loses heat so next time I drain down, I might slip a non return valve in to stop that happening.

Otherwise it works well and maximises the capacity to dump surplus PV.

Bob

Bob your idea with the pump is good. I have something similar but because in my system the fluid is 'radiator' water the pump is the main central heating pump and I use a thermostatically controlled three port valve to route the flow from tank top to bottom.

The increase in energy storage capacity is considerable and well worth the effort.

"no obvious reason why it cannot be used in same way as my version and heat 'radiator' water instead.

So instead of diverting a miserly ~2kWh you may be able to divert ~10kWh"

Apart from, why would you need to heat the radiators in the summer? and in the winter you won't have that kind of excess when you need it before sunrise.

A far simpler solution I found was to just fit a 36inch immersion replacing the 27inch one this now virtually reaches the bottom of the tank,

I'm looking at the new generation of battery charger/inverter systems that have appeared on the continent that "level out the supply" during times of non generation, they are getting cheaper and more accessible.

Rich, I think we had this conversation before and I suspect that my description was unclear then as well.

The water I refer to as 'radiator' water is the fluid that has the inhibitor in and flows around the boiler and the immersion heat exchanger coil. In the winter it also flows around the radiators if required. In summer only the immersion route is used.

A better term for the fluid is probably heating fluid.

Yes I agree that a long immersion heater is a good idea but even this results in a much reduced energy storage capacity than the 'stirring' technique provides.

On a sunny day I am succeeding in diverting 18kWh which for a small system is I think very impressive. I would be very interested to know how much energy your approach is diverting. In my case the DHW tank is 117 Litre and the tank with the heating fluid is 140 Litre.

I would also be very interested to know what battery technology you are considering to use in conjunction with the charger/inverters that you mention.

Brian

Unless my back of envelope calc are wrong, 18kWh  would raise about 450 litres through 35 degrees C ie 15 degree cold in, 50C DHW out (assuming 100% efficiency etc etc)

Put another way you could raise a store of 240 litres from 15 to 80C (when the immersion would max out) with that much energy

That would seem to mean you get through an awful lot of water - approaching 4 tank-fulls at normal consumption temperatures??

Bob

Bob your calculations are pretty good.

My two tanks hold a total of 257 litres and with our new strategy of hand washing dishes there is a small energy drain during the day plus a large drain in the  morning and at night for showers and baths.

The 140 litre tank containing heating fluid is allowed to reach about 75C before transferring energy to the 140 117 litre DHW tank.

I also did the back of envelope calculation some time ago and was skeptical so I knocked up this spreadsheet. You have full r/w access so you can change the numbers to suit.

The strategy I am using is that we don't get multiple consecutive sunny days very often so a large buffer store compensates for the cloudy days.

The tap temperature is about 60C which is too hot and we have to advise visitors - apart from that the system works well and balances nicely against a 4kW PV generator.

I want to question the 'long immersion heater route'.  This is fine in summer but in winter will result in a tank full of tepid water. A short immersion in winter might succeed in providing enough hot water for showers. I think the pump idea has merits.

Seatrout you are forgetting about vertical convection, heated water "floats" to the top of the tank along the convection current close to the immersion heater, the effect is the same but with the opportunity to heat more water if the solar is available. Question away if you wish but it works. I don't get a full tank of tepid water as its not being stirred, the gas thermostat that is half way up the tank will kick in in winter if needed, but that hasn'tt happened since Febuary. 

Brian I recall the conversation, I don't need that volume of hot water so haven't gone down that route, my tank holds 120L and by measurement along the tank I have all but the last 10L or so heated, with no losses due to a double heating system. Brian your system is ideal for a household that consumes a lot of hot water, but I'm looking at heating just the water that I need and then utilizing the rest of the excess solar in another way.

I will be posting up a new thread for discussion re solar power storage, although I'm by no means an expert there is enough European information to start things off.

Brian, you might want to think about adding a thermostatic mixer valve to the out put of your tank(s).  These come in a few flavours with some only designed to fit at the DHW outlet whereas others are often referred to as whole house versions and are suitable to fit at the source end.

60 degrees is really too hot and quite nasty to both young and old folk.

Bob

I was interested to read this. I am just about to install an external immersion.

This house is a lime brick construction and like Lathe & Plaster or clunch has limited insulation. As this house has a small Inglenook I am considering a Rayburn or boiler stove and considering a Willis or Dunsley Heat external immersion for DHW. Reason being that I presume it could be fitted onto the 'coil' fittings, which would allow for effective heating of the DHW. I could use a 36" immersion in the top of the tank as someone has said but they don't heat the water as well as the coil does and there is a case for more scale build up. At least with an external immersion replacement of the unit is easier wheras almost impossible from the hot water tank I gather (well... it is not a favourite job for a plumber... more often than not the cylinder has to be replaced).

This was just my humble deductions. I was interested to google Willis and the link below came up. The Willis is connected differently and I don't see why...

http://www.willis-renewables.com/immersion-how-it-works.htm

In the tank it says Indirect or Direct. So I don't follow the diagram as it would seem to be a more effective installation on the coil fittings of an Indirect cyclinder. Be grateful if someone could explain.

Emailed Willis today, there are no stockists in England, but are available direct from them, and the cost of a complete 3kW unit is £35.00 plus carriage of £11.00 plus VAT.

Not knowing that much about plumbing, does this device heat the tank via convection, starting at the top of the tank and gradually working downwards, until almost all of the tank is heated to temp?

Paul

I think it must work like that Paul, it looks very neat.

(not sure why I've ended up here!)

not sure why I've ended up here! - well it's very timely Robin!

Since April, I think that I've fired up the gas boiler for heating water on just 2 occasions, but as the sun get's weaker and the day's shorter..... I managed to divert a big fat zero watts to the immersion heater today!

So I need to maximize what solar energy we do get, and heat the entire tank, not just the top 2 thirds when we do get some sun!

The only thing stopping me fitting one of these is Limescale, as I have read that their efficiency decreases significantly with limescale buildup, especially as they work through convection.

But I do need to find a solution to heating the entire tank.........

Paul

But I do need to find a solution to heating the entire tank.........  well, there's no shortage of ideas here on that score!

I managed to divert a big fat zero watts to the immersion heater today! 

I've just added a similar facility to my own rig and yesterday's score was precisely zero.  At no stage did our panels produce enough power to exceed our baseload, which is around 200 Watts.  The triac did not fire once :-(

To heat the entire tank I fitted a 10" immersion to the bottom of the tank and have a 16A tank stat to switch from the top immersion to the bottom one when the top of the tank reaches 78 degC. This very simple solution ensures all the energy on low sun days goes into the top of the tank. On sunny days the second immersion kicks in and heats the remainder of  the tank. This is better than having a de-strat pump pumping cold water into the hot water you have already accumulated at the top of the tank.

I posted some pics of emoncms graphs showing tank temps with this system here.......   http://www.electriciansforums.co.uk/photovoltaic-solar-panels-green-ener...

I have also set the event module to notify me if the tank temp goes over 80 degC as an extra backup to the immersion stat fail-safes.

This sounds an excellent arrangement.  My Mk2i rev5b code, which supports multiple loads, can work in exactly this way.  When two output stages are fitted, the upper heater can have priority, and the lower one only kicks in when the upper stat trips out.  As soon as the upper stat closes, power returns to the upper heater because it has the higher priority.

My approach requires two separates output stages, each with triac and trigger (~£5). For this application, it seems reasonable to mount both triacs on the same heatsink as only one of them will ever be on at once.  But no other extra hardware is required.

My Mk2 PV diverter has been heating our hot tank since May. Still over 50 deg.C the following morning, this is enough for two showers and has only needed a boiler top up after a few really grey days.

(Many thanks to Robin for the development work, Paul Reed for showing me the way here from The Electrician's Forum and Robert Wall and many others for keeping me safe and entertained.

With the current thermostat setting, tank temperature reaches 68 deg.C over the top 1/3rd of the tank and I am concerned about the risk of scalding. The top mounted immersion heater reaches into the boiler calorifier coil, 500mm below the top of the tank. (Hercal tank 900 high by 450 dia. 120 litre capacity. Maximum immersion length specified: 700 top, 400 side, although there is no fitting evident).

I found a small 1/2" de-strat. circulator for around £80 and then "fell over" the Willis reference during an internet search. Having pondered the options, I am following the external  immersion route, as it should operate without any need for control other than a suitable thermostat setting. Also, I hope for a full tank of medium temperature hot water before the thermostat cuts out.

The heater is on order from Thermco Ltd who are in Martham, Norfolk (website: www.thermcouk.com). It is similar to the Willis and other externally mounted immersion heaters and is priced at £38 + £12 carriage plus Vat, a total of £60 delivered. Allen Brzeczek at Thermco was happy to provide 22mm fittings and an Incoloy immersion element at no extra cost.

I intend mounting this in the airing cupboard alongside the tank and  tee'd into the cold feed near the bottom of the tank and into the overflow pipe at least 600mm above the top of the unit to allow the thermosyphon to work.

No surgery is required on the tank, water connections will be made with two 22mm compression tees and the electrical connection will only need a longer cable - "simples", I hope, and importantly easily reversible if it doesn't work.

I will post again when I know more.

Mark.

Consider if you are fitting a boiler like a Rayburn use a Laddamat valve....as it allows for quicker heat up and less corrosion of the boiler. This site seems like a one stop shop for most things heating wise. Never used it but there's a lot on same. http://kotly.com/index.php?currency=GBP&cPath=8_250&sort=1a&language=en

Here are my notes following on from my July post.

With the aim of heating the whole hot tank to a lower overall temperature than was being achieved with the in-tank immersion heater, I fitted and evaluated an external immersion heater.

Having been through the experience outlined below, I believe that I could have achieved my original aim of a lower overall tank temperature by fitting a de-stratification pump with appropriate controls, although I am not clear as to the best way to provide control for the pump/heater combination. I have seen an unbranded mini-circulator on Ebay for around 80 pounds, and Grundfoss market one for almost twice this amount. Either of these could be suitable. Also for our system, a boiler circuit check valve would also be needed to prevent losses through the boiler and flue.

The external immersion heater is made from a piece of copper tube around 5cm diameter by 30cm long, necked-in at one end and with the immersion heater boss soft soldered to the other. The immersion heater is therefore integral to the assembly and the whole unit is not easily serviced other than for the replacement of the thermostat. Two 22mm compression fittings are brazed on, one to the necked end and one to the side of the tube near the heater boss. A good quality thermostat fills the safety cover and about 1 meter length of PVC cable is connected (subsequently this was replaced with heat resistant butyl).

We live in a bungalow with beam and block concrete floors. This, and my decision to make the integral thermostat removable, dictated that the unit was mounted 12 inches above the floor alongside the tank

Consequently the thermosyphon action stalled before the whole tank had been heated. However heating did extend beyond that achieved with the in-tank heater, reaching the boiler coil in the tank. On hot days, the boiler circuit started to thermosyphon, achieving over 50 degC at the boiler on a particularly good day, with 30 to 40 degC seen several times.

Water tank temperatures were around 60 to 70 degC with the 70 degC thermostat setting recommended by the maker of the external heater. To reduce this, I tried turning the uncalibrated thermostat in the external immersion down to 40 to 50 degC . This resulted in short cycling of the thermostat with the 3kW element heating the water faster than the thermosyphon action could remove it. Timings were around ten seconds on and 15 seconds off and I felt that the thermostat was unlikely to last long switching at this frequency, and therefore reset it to the higher temperature. There was more kettling sound from the external unit than had been heard before from the in-tank immersion.

I have spent around 100 pounds ( heater, pipe, tee's and isolation valves) and have decided not to fit a check valve which might restrict flow in the boiler circuit. I have therefore reconnected the in-tank immersion and will remove the external heater.

Although my efforts have not proven worthwhile, if you can meet some of the following conditions, you may achieve a better result.

heater.Although my efforts have not proven worthwhile, if you can meet some of the following conditions, you may achieve a better result.

You will need to live in a soft water region or frequent immersion replacement or cleaning will be inevitable.
Mounting the external immersion unit below the tank will ensure the whole tank is heated.
Fitting a check valve in the boiler hot water circuit would stop thermosyphon heat loss up the boiler flue.
Improving flow over the heater element and replacing the thermostat with an electronic controller will improve control. This is probably the most difficult hurdle as it will add cost and could lose energy to export.

For me, this last point was the deal breaker, as in order to achieve a lower tank temperature, both internal and external heaters would need a destratification pump.

Mark.

I've never worked out how a de-strat pump should be used for this application.  Should it be activated for a fixed amount of time, or should it work on surplus power only?  If the latter, and minimal SP is available, it could end up cycling on/off as per the dump-load which would probably not be a good idea.   

I'll probably never get around to it... but I was thinking of controlling a de-strat pump by using a number of strategically placed temperature nodes on/in the water tank, and then develop a logic to base decisions upon those temperatures.

Initially, when the topmost sensor reaches the desired temperature, and the sensors below are much cooler, then the de-strat pump could be activated for a short time, just long enough to shift a litre or so of water, then pause and re-check a couple of minutes later.
As the difference between the top & bottom water temperatures lessen, then the pump time could increase, to reflect the reducing differential in the temperatures, until eventually the whole tank is more or less at the same desired temperature, and then both the heating element & pump are switched off.

I was intending to develop the control logic in node-red (hence this post re MQTT) and then set the 'physical' thermostat to a few degrees above the 'desired temperature' in case it all goes very wrong!

Paul

Mark has given an excellent account of his experience, for what it's worth I thought I would chip in with mine! We have a 3.8 kW solar array with a Mk2 Diverter, our hot water tank is 160 L. It has a short immersion heater which cannot be removed without removing the tank as it is too near the ceiling!

The immersion would only sink about 2 kWh before cutting out with very little hot water! I also tried to source a Willis but ended up with a 3 kW Termco unit (http://www.thermcouk.com/index.php/in-line-heater.html) with 22 mm fittings - cost £60 delivered inc VAT. For servicing (without draining the hot water tank) I installed it with 22 mm (full bore) isolating ball valves.

This is by no means perfect (for all the reasons Mark has mentioned) but it was a dramatic improvement on the short immersion heater. It will sink ≈ 10 kWh & provides a reasonable amount of hot water. The thermostat is set quite high so the top of the tank reaches ≈ 70° with the lower temp (1 ft from the bottom of a 4 ft tank) ≈ 50° (I have not assessed the thermo syphon effect to the boiler but I will have a look at that!)

I too considered a de-strat pump but was not sure about the control...... Stratification is good, up to a point ....... So my conclusion (for the time being!) was that the external immersion was relatively cheap & easy to install & although not perfect does a reasonable job.....

PS Our system has a zone valve on the coil so thermo syphon through the boiler is not a problem...

Richard

I have a few new ideas to add (I also am limited by a short immersion heater because of a low ceiling in the loft above the hot water tank).

1. I have an external small tank (intended to go under a basin and to heat only about 10 liters. Cost about £90). This is connected to the top and bottom of the main tank. Once the small tank is hottish, natural thermal convection dumps hot water in at the top of the main tank, and draws in cold water from the bottom of the main tank.

2. de-stratification pump: this cost about £10, designed for a solar hot water system. I control mine by monitoring the small tank thermostat. When the small tank is hot (i.e natural convection is not enough to keep it cool), the pump comes on (through a relay) and empties the small tank until it is cooler. Hot water pours into the top of the main tank.

3. A thermostat on the inlet of the small tank monitors the temperature of the coldish water coming from the bottom of the main tank. When hot, it stops the small tank from being powered (heated), and no further heating takes place.

4. Using the technology of add-on thermostats (about £10 each) and 12v relays, all kinds of control is possible. For example, if the main tank immersion heater is being used, then a stat 2/3 way up could be used to switch on a de-stratification pump once the top of the tank is hot.

5. I have several cheap electrical thermometers (£10 each) monitoring temperature on the main tank at various heights. Thus the behavior of the hot water can be seen in detail.

6. Main tank cooling is a big issue. The tank originally had about 50 mm of fixed solid foam, and a jacket (about 100 mm). The tank would cool overnight by about 10 degrees C in 10 hours. Thus significant energy is needed just to keep the tank warm (even in the absence of any water consumption). Adding more insulation has helped. Insulating underneath the tank is important and difficult. There was also significant convective cooling through the cold water feed pipe, which was eliminated by inserting  a loop. Cooling is now about 4 degC/10 hours. Further improvements planned include replumbing tank connections so the pipes come straight out (horizontally) for at least 150 mm, and improved insulation.

Where did you get a destratification pump from for £10?
The ones I've been looking at are about £100+

Paul