Ditching gas, going electric immersion only, a wee project

Solarchaser

Solarchaser said:

but I am still loosing a few degrees to lack of good insulation, and so still some work to do. This year!! (I hope)

Was recently away from the house for a week, with no-one using the hot water, and in mid July I required an average of 10kwh every day to maintain the 84C tank heat I'd set.

Noting it was in a particularly warm week also, would suggest the heat loss would be far worse in the colder months, also noting that I was adding in 9Kwh each day last winter, so perhaps if I get the heatloss under control, I won't have to add anything??  and doing a quick calculation of how much this heat loss would cost me per year at 7p/kwh and it's over £250, so that was enough to give me the kick in the rear to get back under the house and start digging.

Kids have been off school so I've been employing a bit of slave labour, and I think we have moved around 1.5 cubic meters, which sounds not much, but when you consider its all been done on knees with a ceiling of less than a meter so no room to swing a spade and some pretty clumpy, clay and brick filled soil... yeah it still doesn't sound much, but it took a long time and alot of bruises.
*Side note, for anyone who has alot of knee pain and finds even the top brand knee pads don't really help, id highly recommend looking at Fento knee pads, they have a bit that comes down your shins and spreads the load and makes a huge difference to me, lasting 2-3 hours on knees instead of 20 mins max with my dewalt ones.*

I decided to go with 2 layers of damp proof membrane/tarp (soil was wet, so im close to the water table i guess) and then a layer of 50mm Celotex pir insulation. 
I've sat both tanks down on the pir and filled up they depress into the pir about 2-3mm I'd say, I expected a bit more crushing, but there you go.
My plan now is to put another layer between the bottom ofbthe tanks and the feet of the bra kets, giving 100mm on the floor and then line all the sides with 3 layers of the 50mm pir kind of interlocking with each other, and 1 layer on top.
I know most heat would try and escape out the top, but I've already got 200mm or so of insulation roll above (but compressed to around 80mm of void) and to be honest I couldn't bring myself to dig a further 100mm down to give the head height required for more insulation. I've a small get out that any heat lost vertically is escaping into the home envelope and so is contributing to heating the house.

As its been 2 years of use, I figured I'd give it some upgrades (only if they work), or refinements hopefully. 

1st upgrade is that at the moment (remembering they are horizontal tanks) I destratify by sucking water out of the mid point of both tanks on the right and pushing it along the top and into the top of the left side of both tanks, but I've decided that doesn't really need to be the case, and it's likely that I'm not necessarily destratifying the bottom of the tank, so I'm now going to pump from the drain spouts on both tanks which are probably 2/3rds to the left of both tanks, and pump it into the mid point of both tanks on the right.
So previously it was sucking from half way up on the right and pushing top left, now it will sucking from the bottom left and pushing into halfway up the right.
I figure it will be less hot pipework to lose heat and a cooler run for the grundfos pump.
It may be a failure, but I'll never know if I don't try.

2nd upgrade was suggested by another user who was going to do something similar with their tanks, but was going to feed the central heating through a mixer valve, to give consistent heat.
I didn't like that idea, I was concerned it may get blocked up or stop working when the CH got to a certain point, however I've convinced myself it will work, and if it does there are a couple of advantages to doing that.
First being that the radiators will always be at the same temperature.
I've been a bit concerned that in the dead of winter when the heating comes on from 84C tanks, that the radiators could potentially burn the kids, so I used a gate valve to mix in some of the cold all the time, but it's far from an exact science, and means the radiators change temperature through the day.
Second advantage would be that I'd theoretically use less heat from the initial morning heat as the tanks and pipes will get less of a thermal shock... well I think so.

Now I just need to make it happen before the months require some central heating in the morning.... wish me luck

Heedtheadvice

If I read correctly for destrat you are sucking out the cold and pushing into a hotter area.

I understood but could well be wrong that detrat usually takes our hot and pushes lower into the cold.

Are you doing it your way to circulate the cooler water through pupework and reduce losses? Are the pipes not in your insulated volume?

Solarchaser

You are indeed correct, normally destratification happens by pumping hot stuff from the top, into the cold stuff at the bottom to break up the boundary layer and basically mix the water.
However normally water tanks are cooler than my 84C and so I figured it would be better to pump from the middle to the top, the other thing is I'm using a (used) central heating pump, for destratification, rather than a specific destrat pump, and I figured it would be better for longevity to pump cooler water.
The piping is all lagged, but the pump isn't, as its control box is supposed to stay below 40C, so I lose heat through there. (Though I guess when i box it all in, that control box will be a bit higher than 40c)

Also I have 1 way valves in the piping to stop the heat leaching out of the tanks and into the piping, but this really means that any heat left in the pipes is realistically lost, so yeah if I'm circulating cooler water,  there is less heat loss.... well that's my thinking anyway.

So now to reduce that loss further ill take from the bottom and put into the middle and hope that at least works as well, if not better.

How much difference will it make?
I dunno honestly, I can't think it's gonna be kwh's a day, but as the marketing slogan goes... every little helps

Heedtheadvice

Certainly some good logic there.

I guess some things will depend upon where in your system the temperature sensors are ( for heating control ) and destrat.

First thought is that if your heating sensor is above( significantly) your destrat return then it will still cut out when the  'top' reaches cut out temp. If destrat continues to flow it will have little affect on the top layer. Destrat will mix lower part. Thus maybe reducing the overall heat capacity....?

That will not apply if heating elements are low and the sensors but then the top might get hot....too much.

( I cannot remember where your placements are! )

Do upu have a temperature sensor on the destrat pipe on low on the tank(s) that might help to determine if those thought have any value?

Sorry fluid dynamics are not my thing !!

Solarchaser

Good points that I hadn't totally considered to be honest, and good timing as my plan is to plumb this up today (no of course I'm not hanging about on forums wasting time to delay my descent under the house... honest guv').

The 500l tank has the heating elements on the end tanks pretty near the bottom, I have temperature sensors in middle top and middle bottom and due to the location of the heating elements, stratification isn't actually a problem with this tank, the maximum difference I've seen between the sensors is 2C.

The 300l tank however has the heating elements on the top which project roughly 2/3rds of the way into the tank and sensors I have in top and bottom can report a 10C difference.

Unfortunately the myenergi graphs don't let you click on them for a report of temperature at specific spots, but hopefully you can see here the difference made by the destratification pump coming on around 4am in January of this year (triggered by the 500l tank reaching its max temperature)

The larger tank is in green and smaller in white. The green block is the myenergi eddi powering the element, you will note they are both increasing though only 1 is powered, and thats because the other element in the tank, and one in the 300l tank are powered by PID's as recommended by yourself and QrizB.

Actually perhaps this is a better demonstration of both things in December. 


Here you can see the 1 element in the 300l tank is heating the water significantly quicker than the 2 elements in the 500l tank, why? Because it's only heating the top say 2/3rds of the tank.
Note that it plateau's for almost 2 hours due to its stratification, and only starts heating again after the destratification pump kicks on.

(You can also see both graphs move later in the day, but I'll explain that lower down)

The graphs give the idea of the 500l being stratified also, however this is due to the water being drawn from both and pumped around both as if they are one.
The eddi only has capacity for 2 pt1000 sensors, but I have a few others in various places that confirm the 500l doesn't have an issue with stratification, only the 300l, and there is one on the 300l destrat tube, which is oddly satisfying to watch when the pump kicks on and it starts from say 74C and over the course of around 10 mins going up 0.1C at a time until it gets up to about 84C.

The eddi sensor on the 500l is at the top, middle of the tank, and on the 300l is actually around the middle depth of the tank as its sitting in the tube of the heating element (the one not used during overnight heating) that would normally be occupied by the heating elements thermostat, (those are removed due to using the PID's and the PID sensor is in there also)

I know from the graphs that pumping from middle to top, definitely moved the water about and mixed (destratified) the tanks, and the pumps flow rate is 3.5 cubic litres an hour, so 15 mins should be enough to circulate 800l,  but I don't know if it's just going to circulate the same 100l, when I replumb it now to pull from bottom and push into middle, that's the concern.
I don't think so.
I think as the water heats from the bottom of the 500l tank it moves to the top causing a bit of circulation,  and so if I pump out from the bottom, into the middle, the different levels of heat will cause circulation outside of what's being specifically circulated. I may be wrong, we will find out.
Worst case scenario, I'll replumb again, but from bottom to top.

Now to quickly explain why later in the day, both tanks decrease. The CH coil is only in the 500l tank, the coil in the 300l tank is only for preheating tap water before going to the other coil on the 500l tank.

So why do both tanks decrease around 7:30am? Because to use the 300l tank to help with CH I kick in the destratification pump again to circulate all the water around the coil.
So my big concern here is with the CH coil being on the left side of the tank, and the destratification moving from being middle right to top left, to now being from bottom left to middle right,  am I going to get enough circulation in the left side of the tank to keep the CH topped up with heat from the 300l tank.

I'm concerned, but hopeful.
Fingers crossed

Heedtheadvice

Having read through and considered that fluid is not lumps that rise or sink  that has currents and eddies etc.  whilst heating and destrat it will be quite complex  ( according to my brain  cell ) you will maybe be fine. However you have enough sensors to see as you modify.

Solarchaser

Well after a few weeks of mucky, and sometimes painful work, it's all done.
Is it exactly how I wanted? No of course not.
My idea was to have full 2.4 x1.2 panels in there, but the "doorway" is only 800mm, so have I got 2.4 x0.8? No, because surprisingly enough when you try to take something that's pretty long and not flexible through a small doorway at a tight angle, it doesn't want to go.
Maximum I could get was 1.1 x 0.8, so every single "wall" was actually made up of between 2 and 7 panels.

I thought, that's OK, I'll glue them together in place with soudal plasterboarding foam, and literally seconds before doing this, I thought, hold on, how am I going to get them out again if I do that, and let's face it, at some point I'll need to replace a pump or heating element. So they are not actually attached to each other.
I cut the different layers at different sizes so there was no cold junction join, and decided since I couldn't glue them my only option was to pull them together with a ratchet strap and spread the loading of the strap with off cuts of wood wrapping round the corners.


She ain't pretty! But she seems fairly sturdy when squashed with the strap.
You can see there are 3 horizontal sheets, a vertical and then another 3 horizontal to make up the 2.3M which is the outer size of the box.
The pipe coming out is the over pressure/temperature overflow.
And does it work?


Seems to, it seems much better than before, which is a relief, I had visions of doing all that work and it making no difference. 
It's been fairly warm today, ive set the tanks at 70C rather than 85C for the moment and I'm sure it will weep some heat when we are in the hear of winter, but I'm definitely not losing 10kwh a day right now.

So that's the insulation,  what about the destratification? Well it doesnt look like much happened this morning as I only have the eddi coming on at 4:30 for an hour so that it can destrat if needed, and you can see a slight dip, but it's more noticeable in yesterday's heat cycle 


Eddi comes on for about 20 mins, hits max temp and destrat pump gets activated, and again just after 5am, and yeah it definitely does work.

So happy happy on those fronts.

I've had the heating on with the thermostatic valve and it seems to run fine at about 49C, which I think is about ideal, though u won't really know until we get into winter, but early signs are good.

For now, 2 out of 3 ain't bad.

Heedtheadvice

Congrats!

Another step towards reduced leaking of heat to the rest of us whilst keeping your tootsies warm.

Solarchaser

Realised I haven't updated in a while.
So I'll start with the good... ish news.
In the dead of winter, -2 outside



With tanks heating to 83C, lost 7.6kwh over 24 hours with no-one in the house.
Not as good as i was hoping, but definitely an improvement.

Turns out I was wise to not glue the pir together, as i came home one night to RCD having tripped,  which turned out to be the destratification pump had died. 
(Kept tripping after about 30 mins, was *fun* trying to figure it out)
At the outset of the project i bought two used pumps for £50 from ebay, so i suppose this was to be expected after 2+ years of use.
Handily I'd scavenged the 15 year old combi pump when scrapping the boiler, and grundfos pumps of that era have the same 4 bolt flange as I discovered.

It's a warning shot though, so I'll need to fit new pumps this year if I want reliability.
The project needs to come out of the project phase and just work without input.

I had used a hot water thermostatic valve set to 49C for the winter, and what this meant in real terms was 43C at the radiators, and that wasn't quite warm enough in the dead of winter, so I've got an underfloor heating thermostatic valve which goes up to 60C, and i figure I'll set it to 55C for the winter.

I haven't changed any of the radiators,  I am aware I could move to triple row radiators, but for the moment I'll go with just increasing the flow  temperature and see how it goes.

Heedtheadvice

What will you do for ' Entertainment ' when it runs all by itself?