Immersion Heater - PV electricity

[inie meanie]

EMMA uses AT20B10 for the PV sense and AT50B10 for the house sense. these sensors are only rated for continuous overload of 120%

Yes you should rate your coils for your own current range.20 amps is more than enough for me.

 

[bradburts]

very little protection and no filtering none of them would pass CE.

Please elaborate. Are you talking about the PFC chopping the AC? The EMMA seems to get by with a PFC and off the shelf filter. What filter would you use?
Otherwise the circuits I have seen have inputs protected with clamps, use opto isolators etc. I see no reasons why they could not meet CE so long as the build design is sensible.

My conclusion based on the prototypes and entry's on this forum is to be lazy and go for the AT20B10/ AT50B10 sensors, crydom phase controller, i/p filter, packaged power supply and an off the shelf PID controller ( not yet selected any suggestions welcome ). This is very expensive but I don't have the hassle of designing and building a pcb. The analogue cts were std stuff. I used precision rectifiers for the ct interface and then a simple type 2 op amp integrator

Agree, being a lazy engineer is usually best.
I would try and tempt you from the off the shelf PID controller though. There are plenty of ready made microcontroller solutions you can buy for £20 or so. Most have worked examples showing you how to make an ADC readings and drive a DAC outputs (or analogue out using driving resistors from your output ports).
If your PID controller has built in analogue and digital buffers then I guess its the best thing to use. Your system is then pretty much COTS and you have a £200 EMMA! Which PID controller are you looking at BTW?

 

[pmcalli]

Please elaborate. Are you talking about the PFC chopping the AC? The EMMA seems to get by with a PFC and off the shelf filter. What filter would you use?
Otherwise the circuits I have seen have inputs protected with clamps, use opto isolators etc. I see no reasons why they could not meet CE so long as the build design is sensible.


Agree, being a lazy engineer is usually best.
I would try and tempt you from the off the shelf PID controller though. There are plenty of ready made microcontroller solutions you can buy for £20 or so. Most have worked examples showing you how to make an ADC readings and drive a DAC outputs (or analogue out using driving resistors from your output ports).
If your PID controller has built in analogue and digital buffers then I guess its the best thing to use. Your system is then pretty much COTS and you have a £200 EMMA! Which PID controller are you looking at BTW?

To meet CE radiated and conducted emissions you need a filter designed for dirty circuits like thyristors. An off the shelf one is about £50. You also need to protect for failure mechanisms which could cause a fire so fuse is required for the thyristor or crydom unit. An MCB is required because the crydom unit and thyristors have a significant leakage current when off so regs demand a switch as its above 25VA

I have just been costing it up and it comes to more like £500. I am also having problems sourcing the PID controller to match the 10 V o/p cts that EMMA uses. they need a load > 2Mohms and the controller needs an external analogue set point. so I have been tempted to the dark side by my son and a colleague who have offered to lay out a pcb and do the micro controller programming . This should bring the cost to under £250

 

[bradburts]

To meet CE radiated and conducted emissions you need a filter designed for dirty circuits like thyristors. An off the shelf one is about £50.

Would you mind posting a link to a suitable unit?

I have just been costing it up and it comes to more like £500.

Thats a lot! I was expecting a lot less but I have not looked at the cost of the CTs.

I think that you are best going with a microcontroller. As pointed out the Arduino projects at openenergy monitor forums are well worth looking at.
There are examples showing how to log your systems performance to the cloud and then you can monitor your system from your smart phone or web browser. You would struggle to do that with a PID controller + the extra cost of the PID controller.

 

[pmcalli]

Would recommend a two stage filter Buy Power Line Filter 2 stage chassis mount filter,16A Schaffner FN 2060-16/06 online from RS for next day delivery. this one is under £30 you can get a single stage for half that but not sure it would pass CE. DIN rail mounted ones are available as well but horribly expensive £100 touch

 

[bradburts]

Thanks.

 

[suntrap]

I have been tempted to the dark side by my son and a colleague who have offered to lay out a pcb and do the micro controller programming . This should bring the cost to under £250

In that case, how about going all the way and using a triac (opto-isolated ones available for about £5) rather than the Crydom controller? Get a sniff of AC voltage and generate the trigger pulses yourself.

 

[suntrap]

Would recommend a two stage filter Buy Power Line Filter 2 stage chassis mount filter,16A Schaffner FN 2060-16/06 online from RS for next day delivery. this one is under £30 you can get a single stage for half that but not sure it would pass CE. DIN rail mounted ones are available as well but horribly expensive £100 touch

Or if you really wanted to economise you could buy a couple of washing machine filters (they seem to be single-stage) and put them in series. They can handle the current. I did wonder if the suppression would be less effective due to the size of a washing machine motor, but the component values seem to be comparable.

 

[bradburts]

In that case, how about going all the way and using a triac (opto-isolated ones available for about £5) rather than the Crydom controller? Get a sniff of AC voltage and generate the trigger pulses yourself.

Yes, I would go all the way on the TRIAC.
Generating the pulse yourself would be easier than interfacing to the Crydom.
To generate the pulse you just need to detect zero cross - a diode on a digital input would do (some braver sole may forget the diode and just use the digital input's ESD protection). Arm a counter counting down from the DIO trigger and you're done. OK some calibration needed. All microcontrollers will have a counter and/or PWM on board.
With the Crydom you have to generate and buffer an analogue signal using a DAC or resistor chain. Needing an onboard DAC is going to heavily cut down on your microcontroller choice. There are plenty available but perhaps not on a ready built evaluation boards.
I am digital & find analogue tricky, perhaps you're the other way. The timer / PWM arm software is just a few lines and there will be plenty of examples available.
If you use DIO then you just need an analogue circuit to protect the CT inputs. If you buy an microcontroller evaluation board then there should be space enough in the prototyping area to fit that circuit.
Without knowing your CTs cost I would say that your down to £100 or so with your line filter.
Evaluation controller £30 + £5 TRIAC + Case&Heatsink + £5 CT clipper + CTs + On/Off switch & fuse + £30 line filter.

 

[pmcalli]

Yes, I would go all the way on the TRIAC.
Generating the pulse yourself would be easier than interfacing to the Crydom.
To generate the pulse you just need to detect zero cross - a diode on a digital input would do (some braver sole may forget the diode and just use the digital input's ESD protection). Arm a counter counting down from the DIO trigger and you're done. OK some calibration needed. All microcontrollers will have a counter and/or PWM on board.
With the Crydom you have to generate and buffer an analogue signal using a DAC or resistor chain. Needing an onboard DAC is going to heavily cut down on your microcontroller choice. There are plenty available but perhaps not on a ready built evaluation boards.
I am digital & find analogue tricky, perhaps you're the other way. The timer / PWM arm software is just a few lines and there will be plenty of examples available.
If you use DIO then you just need an analogue circuit to protect the CT inputs. If you buy an microcontroller evaluation board then there should be space enough in the prototyping area to fit that circuit.
Without knowing your CTs cost I would say that your down to £100 or so with your line filter.
Evaluation controller £30 + £5 TRIAC + Case&Heatsink + £5 CT clipper + CTs + On/Off switch & fuse + £30 line filter.

Yes I find the analogue easy as I have experience in phase lock loops and analogue control loops. The crydom is attractive only as no interfacing at mains voltage is required and I don't have to design a snubber which is notoriously difficult to test without creating a bucket full of dead bits. I may also consider the option of an entirely analogue loop with a micro-controller to supervise,feed a display of power, logging and some load switching. I need to get on and build a prototype ( pressure from a queue of friends who followed my advice with the PV system and want the immersion controller ) and so will probably postpone the triac for a phase 2 cost reduction. Many thanks for the inputs very useful

 

[echase]

I have just been costing it up and it comes to more like £500.

Sacreblue! I can sell you one for under half that or as a kit for even less, but fine if you prefer the challenge of designing your own. Mine do work, as stated by early adopters such as:-

From a retired CEO of a housing authority:-

After some fine tuning the I2 worked well. I can tell from my web based power monitoring system that the unit cuts in and out at the right times. On a sunny day it provides all the hot water we need.

From a qualified electrician:-

The unit is working really well and keeping the thermal store topped up with heat on the days where I might not have usually produced quite enough from the solar thermal system.

I was monitoring the system the other afternoon and taking measurements with my clamp on wattmeter. This can differentiate between import and export power and shows a negative reading on export. Watching the meter keeping the power exported to around the 50-100W mark while the PV system was generating 1-1.2 kW, I observed the current to the immersion was gradually reduced down to 0.3 amps and then go off as the sun dropped in the sky.

I will probably be looking to buy another unit from you to install in my Mother's house as she has just had a PV system installed.

(To be fair not everyone is yet getting that low an export but it's an adjustment issue I think).

From an airline pilot:-

I have now the means to measure import/export accurately! Your device works perfectly and does exactly what you say! Very happy.

 

[trebormint]

we have developed a system which connects to a standard 3KW immersion heater which will reduce the power consumption to a more acceptable 1.2KW. It monitors the array output & household consumption & will switch the heater on when the PV is generating enough & switch the heater off when either the PV isn't producing enough or if the household consumption increases above the demand for the PV.

 

[bradburts]

we have developed a system which connects to a standard 3KW immersion heater which will reduce the power consumption to a more acceptable 1.2KW.

Sounds interesting.
Please share for others to build with/on. How did you go about it, which design did you use?

 

[pauldreed]

we have developed a system which connects to a standard 3KW immersion heater which will reduce the power consumption to a more acceptable 1.2KW. It monitors the array output & household consumption & will switch the heater on when the PV is generating enough & switch the heater off when either the PV isn't producing enough or if the household consumption increases above the demand for the PV.

Who is we?
What is your professional status, are you a registered company?

 

[pauldreed]

we have developed a system which connects to a standard 3KW immersion heater which will reduce the power consumption to a more acceptable 1.2KW. It monitors the array output & household consumption & will switch the heater on when the PV is generating enough & switch the heater off when either the PV isn't producing enough or if the household consumption increases above the demand for the PV.

Wot, 5 day's later and no response??
hmm.... is it me and my mate Fred who knows a lot about lectric stuff?