Immersion Heater - PV electricity

[echase]

It is a solar water controller which uses a small PV panel to provide all the power to run the controller and pump so there is no mains needed. But it can be mains powered if you want. It’s mainly designed to switch the water pump on and off in response to the solar power and temperatures in the system. It also has 2 relay outputs to power the gas boiler/immersion heating top up function. These would do 1kW with an external relay needed for more than that.

It would not connect as is to a PV electric system, but with simple modifications:

1) You could add CT inputs. To mimic your unit paukdreed it would need a couple of input circuits to convert the CT output to a 0-2.5V aprox DC value.
2) It could log the power in the PV system too which is useful in system where the inverter is in the loft and so its power meter is not visible.


As you designed and soldered your own unit mine is not difficult. There are no surface mount components. But it’s outside the capability of a complete novice in electronic assembly.

June 09 covered the design and software with July for how to build, test and connect it.

 

[Gazzdid]

a simple thought...... dont forget to introduce a cylinder stat to your circuit so it wont continue to heat the cylinder to a dangerous temperature!

 

[pauldreed]

The input circuitry would add significantly to the overall circuit design. A simple passive diode bridge would not suffice, as I am sure you will be aware. This part of the circuit design was the steepest learning curve for me, and took the longest time to sort out.
You say that there are no surface mount components, but surely there must be components to provide the right environment for the PIC to operate? (sorry never seen a PIC or had anything to do with them).
Can you change variables in the PIC once it has been programmed (such as histeresys, switching thresholds ect) without actually re-programming it?

Edit - Yes I've got an immersion heater with an integral thermostat set about 5 degrees higher than my gas boiler water thermostat. So the central heating boiler does not heat the water (unless its been a dull day!)

 

[echase]

There are extra components on the analogue inputs but they are little more than resistor dividers and overvoltage protection resistors. But you would need to convert the CT ouput to DC with something like an op amp.


My standard device has >20 variables that can be altered by menu. If I changed the software to provide this function I’d add extra variables to alter the extent of input power averaging, switching hysteresis, etc.

 

[pauldreed]

There are extra components on the analogue inputs but they are little more than resistor dividers and overvoltage protection resistors. But you would need to convert the CT ouput to DC with something like an op amp.


My standard device has >20 variables that can be altered by menu. If I changed the software to provide this function I’d add extra variables to alter the extent of input power averaging, switching hysteresis, etc.

I use op amps to monitor the CT's which give me 3 volts at 5kw power flow (which can be changed via a shunt resistance).
If I can help, especially with the 'front end' circuit - let me know.
I'll PM you my email address..

 

[rondo]

Hi Millerman,
Just picked up this topic and very interested in your unit can we talk more on this.
Have been trying for some time to use my surplus PV into I/H which is a standard 3kw unit and not much joy as yet. Latest idea was to run the I/H thru. a site tranny (240/110) with a adequet continious rating of say 1500va,a trial gave me about 650w at the I/H. Auto control sounds a better way fwd.

 

[pauldreed]

Hi Millerman,
Just picked up this topic and very interested in your unit can we talk more on this.
Have been trying for some time to use my surplus PV into I/H which is a standard 3kw unit and not much joy as yet. Latest idea was to run the I/H thru. a site tranny (240/110) with a adequet continious rating of say 1500va,a trial gave me about 650w at the I/H. Auto control sounds a better way fwd.

Wouldn't it be more straightforward to simply fit a 1kw heater element and dispense with the transformer?

 

[Millerman]

There has been a lot of debate on the use of a 1kw immersion heater as a load dump. My first thought was to use a 2kw immersion heater then changed my mind to a 1Kw immersion. For my water tank the cost as about £75 (stainless steel) Then i had my solar system fitted that was constantly producing over 2.5kw, in theory wasting 1.5Kw. So i have purchased a unit that can be connected to a 3kw immersion heater (hence no cost "immersion heater") and can be run using 100 watts or 3000 watts. Spending pennies on a system to run the spare energy is fine. I have elected to splash out and use a plc and power controller, the cost is around £375 but over the period of the solar system life, it's peanuts. It works and is fully reliable.

 

[pauldreed]

....and can be run using 100 watts or 3000 watts

Can you expand on this?
Are you using a 1kw or 3kw element, and if 3kw - how are you limiting it's power dump?

 

[rondo]

Hi Paul,I looked at fitting a 1kw I/h at the bottom of the cyl. on an essex flange (popular years ago) but to-days thinking is that current copper cyl,s are too thin walled so it,s not recomended.A straight swop with the 3kw 27" top fitted is not good because they are only 11" and only heat the top of cyl.So my next trial was a site trany 230v/110v,I was given some duff info by an lecy friend who said I would get about 1.4kw at the I/h,but as you will probably know it,s more like 650w.The trany has apparently to be of suffient continious rating (va) I tried one rated at 2000va and seemed ok ,but you,ve got to switch it off/on manually as required and it takes some time to top up temp.Dont use much H/W so not worth investing much and I,m retired so have the time to monitor it.I would be pleased to hear any views on this.

 

[pauldreed]

rondo, you appear to be following the same path that I did!
But, if you give Howden electro heating company, electric heating manufacturer a ring they can provide you a 36" 1kw element which can be top mounted and replace your existing one.
I fitted a 26" element in mine and it works fine.

 

[mac1]

Hi
I noticed this on a website. It might be possible to use it to lower the power of the immersion to match the output of the PV system.
http://www.graham-laming.com/bd/Immersion_Heater_controller.pdf

 

[rondo]

Hi Paul,thanks very much,I thought the power and size were fixed,a 1 kw 26"from the top would be great.I was looking at the ideas on this thread regarding auto control and Andy appears to have a cracking bit of kit he has made but as I explained to him we use very little hot water, so as much as I would have liked to go that way I could,nt justify the costs, "Gotta use it to save"
Once again many thanks. Ron.

 

[pauldreed]

Hi
I noticed this on a website. It might be possible to use it to lower the power of the immersion to match the output of the PV system.
http://www.graham-laming.com/bd/Immersion_Heater_controller.pdf

This will not work as expected!
If you research the technology you will find that because it 'chops' the sine wave, it will use more than 1kw for a short period of time and then use no power at all for the remainder of the cycle (wasting generated power).
If you are not familiar with the technology principles - then please take my word - it won't work! (but nice try)

 

[PVFIT]

Hi Millerman,

Please can you forward me more information on your appliance switch you mentioned in your post on 06-01-2011.

Many Thanks

 

[mac1]

Ok thought it was too good to be true

 

[inie meanie]

Hi Paul
I am interested in your comment regarding the power control. Having just built an opto isolated triac control unit, I tested controling a 3kw kettle via the triac circuit as a test for my immersion heater control. Using a power meter at the plug of the kettle and whilst watching the output of my inverter and the mains input meter I performed vairious tests ranging form 300 watts to 2 kw taking it close to the total spare capacity that I was generating. At no time did my standard electricity consumption meter register any load. Can you provide a link to a paper / information that may cover your thoughts regarding the wave form cutting situation? Using a triac will provide a much more efficient way of heating your water as you will be able to drive the heater from a much lower level. It seems as though lots of people are catching on to the possibilities of using this spare power

 

[pauldreed]

Please see Light Dimming | Micro Basics which shows the principles of how a Triac works and which I refer to below.
You will see that the power is controlled by firing the Triac at a certain phase in it's cycle, thereby reducing the power 'average' through the full cycle.
Now, if we assume (for example) that you are trying to reduce the power of a 3kw heater down to 1.5kw, this will mean that the Triac will fire at 90 degrees at both the positive and negative cycle (as shown on the 3rd diagram down).
You will note that the current flowing as the Triac fires will be the full current of a 3kw load, of which only 50% will be supplied by the invertor, and the rest via the grid supply.

- The overall power reduction is only achieved by supplying full power for 50% of the time.

A further consideration is that in the 90 degrees before the Triac fires, the power flow is zero - which is not making efficient usage of the power provided by the invertor....

It is many years since I studied Triacs, so would be grateful for confirmation from other board members that this is correct.

 

[inie meanie]

Hi Paul
Yes I can see the logic of that. I wonder if the short duration of the extra spike is not seen by the input meter, almost like a transient spike. I willl carry out some more real life tests to see if this is what happens. I will get back to you.

 

[pauldreed]

I'm not sure how the digital input meters work, or how they deal with reactive/resistive loads, phase shifts ect, so can't really help there, but look forward to your results.
Maybe if instead of using a kettle for testing which will only run for a few minutes and be more difficult to measure as the input meters usually record in 1kwh units, you might consider an appliance which can be left running for an hour or so, such as a heater, tumble drier etc.