Managing and using your generated power - a whacky idea?

[ProDave]

Okay so I have my panels, MY FIT is secure at 45p and it's working well.

I've educated Mrs PD so that the Washing machine, tumble dryer and dishwasher are only ever used in the daytime when the panels are generating, to use up as much as possible of our generated electricity. Of course only one machine on at a time.

But that still leaves a large part of a sunny day when the panels generate a lot more than we use. So being a tight git, I want to find a way to use more of our generated power usefully, rather than just let it go back to the grid.

So I'm thinking, whenever it's generating surplus power that we are not using, to "dump" the power somewhere.

I thought about just dumping it to a panel heater, but we don't need heat in the summer, so that's no good.

So the most sensible place to dump unused power is into our hot water tank. Use the surplus power to heat the hot water in the day, so not only will my panels give me income, hopefully eliminate any cost for electricity, they could also cut down our oil consumption. (oil boiler heats hot water normally)

So, I'll rarely have much more than 1KW, often less of surplus power. So I'm thinking use the standard 3KW immersion heater, fed via a 110V transformer, which back of fag packet calculations should be about 687W of power.

So what's needed is some current monitoring and some electronics, so that whenever generated power exceeds used power by 687W or more, close a contactor to power up the immersion heater via said 110V transformer.

I'm not saying I'm going to build this straight away, but I have the rough design for the electronics in my head. Before I start, has anyone tried anything similar? or indeed is there an off the shelf unit that you can buy to do what I want?

Or am I just bonkers?

 

[Andy™]

at a lower voltage / power, will the element get hot enough?

 

[ProDave]

at a lower voltage / power' date=' will the element get hot enough?[/quote']Perhaps before I try making the electronics to do it automatically, I should just try the immersion heater on 110V and see if it does anything useful.

 

[steptoe]

good thinking dave,

I have set up timers to open close contactors for solar PV so that the folks that are at work have their various machines switching on and off one at a time,

your idea is deffo do-able, you need, I think, some sort of monitor on your tails that would switch the contactor when >xxW is flowing the 'wrong' way, that is going to be the tricky bit,

but Im sure canoey or SW have come across some sort of kit like that, ie, some sort of alarm monitoring system could Im sure be adapted, or even a digital ammeter.

 

[sprocketflup]

Worked on just such a system few weeks back with a homemade changeover switch with override for the not so sunny days.

Company called Chris Rudge Renewables in seaton, devon. Reckon hed sell the changeover boxes seperately

 

[ProDave]

The challenges I have yet to solve are power monitoring. A simple current transformer will tell me how much is flowing, but not which way. So monitoring grid current won't help me much.

I have a separate CU for the PV and a main CU, so I can independently monitor generated power and used power, without being troubled by which way (they can each only flow one way)

So when generated power exceeds used power by lets say 700W that will trigger the contactor to close.

As soon as the contactor closes, used power will increase by 700W, so it's no good just making the same comparison to determine when to release the contactor (otherwise it would release straight away). For that you need to compare generated power with used power, and when generated power is less than used power by say 100W, turn off the contactor.

Now this all could be designed with some basic analogue electronics. But I's rather do it with a microcontroller. That will allow switching thresholds to be easily tweaked, adjustable hysteresis, and probably some timers (to stop the contactor opening and closing every few seconds whenever the sun goes behind a cloud)

Having been out of the controls industry for nearly 10 years now, I don't know what's about in the market at the moment. Anyone care to suggest a simple programable device? must be cheap and be available in 1 off quantities as an evaluation board or OEM module, must have 2 analogue inputs, a couple of analogue outputs would be nice (but not essential) and some digital I/O. The programming software must be free or dirt cheap.

I saw a device the BBC are pushing mentioned on the tv recently but don't recall the details. They are trying to do for programming now, what the BBC micro did in the 1980's. This device will cost about

 

[steptoe]

there you go Dave,

FAQs | Raspberry Pi

is that what you need?

---------- Post Auto-Merged at 15:21 ---------- Previous post was made at 15:18 ----------

Raspberry Pi in action: our developer puts it through its paces - video | Technology | guardian.co.uk

 

[ProDave]

there you go Dave,FAQs | Raspberry Pi

is that what you need?

That's the one steps.

I'll have to look in more detail, but an initial glance suggests I'll need an I/O board to get the analogue inputs I need.

It also looks like a sledgehammer to crack a nut in terms of processing power. I could see this project growing arms and legs over time and getting more sophisticated.

 

[steptoe]

Im going to get myself one of those Dave to have a play with, looks like fun for 20quid!

 

[ProDave]

Experiment step 1 results:

I've found the "big" auto transformer I knew I had and hooked it up to my immersion heater. Here's the measurements:

Input: 247V @ 2.6A = 642W

Output: 110V @ 6.3A = 693W

So clearly one of those measurements was inaccurate but it shows my rough calculations of approximately 687W to be close to the mark.

The only "problem" is the "big" auto tranny I had turns out to only be rated at 500VA.

A short (half hour) test didn't show any problems with the tranny. But today is a dank overcast day and the panels are not generating much.

So next sunny day we get when I'm at home, I'll turn the heater on and give it a prolonged test, firstly to see if I can get away with that transformer, and secondly to see if it results in a useful heat input to the hot water tank.

I've also registered my interest with RS to get on the waiting list for a Raspberry Pi. (what's the fascination with calling electronic stuff after foods. Did anyone see the Ronnie Corbett sketch about that?)

 

[steptoe]

I've also registered my interest with RS to get on the waiting list for a Raspberry Pi

it seems that RS are being a bit awkward for some reason,

see here

 

[remustered]

Hi Dave, not sure I'm on the right track with this but...have you looked at arduino boards for diy I/O control and programming. There PC based usb I/O boards with digital /analog screw connectors and programmable chips.

there are numerous varieties for different uses and loads of ready made programs..also no queues to buy,pick them up maplins/online etc.

 

[steptoe]

any links remustered?

 

[remustered]

Arduino - HomePage

 

[steptoe]

ah, its a raspberry board!

 

[remustered]

got a usb experiment board I picked up, not used it yet.took piccie of back of packet ...basic spec.

 

[remustered]

don't believe it is raspberry pi board. these have been around years whereas RP's are brand new (hence waiting list for orders) also RP's are complete PCs, arduino is far more limited...no display output etc.

they do look similiar though.

 

[ProDave]

The Andruino looks to be perfect for the job.

Particularly as someone else PM'd me a link to someone who has done exactly what I am doing already using one (google "And yet another PV dump controller circuit")

So the design exists, the hardware is cheap, Roll on a sunny day to test that my 110V powered immersion heater works. Then I can order the parts and build one.

Just one dumb question. I assume after downloading a program from a PC, then the board just needs power and it boots and runs that program on power up? the PC connection is just used for program downloading.

It looks like this is programmed in C, or a C like language. I feel more comfortable with that having programmed in C before, rather than the Raspberries "python" language that I know nothing about.

 

[steptoe]

yea, sorry remustered,

they are based on the same technology, raspberry is just a bit more developed,

but dont let raspberry know you think they are a complete PC,

they are totally against the idea of anyone thinking that,

this is now why they have the hassle of CE marking with RS

bit of a thought though, how come those arduino boards are so expensive compared to the raspberry considering the limited ability of them?

someone somewhere is making money

 

[remustered]

Wow steptoe... your a sparkie and a geek...awesome.

my hero.:worship