1. To continue with this build sequence, access to 230V mains voltage is required. Please take great
care, and do not undertake this activity unless you feel confident
to do so.
2. Before installing a MK2 control system, the associated immersion heater and thermostat should be checked to ensure that they are in good working order. After years of infrequent use, these peripheral components may need to be replaced.
When running in its "normal" mode, the Mk2 PV Router can reliably divert surplus power without need for any calibration. However, to provide an accurate indication of diverted power, and to operate correctly in anti-flicker mode, the system needs to be calibrated.
A simple way to do this is to clip each of the CTs around a test lead through which a known amount of mains power is flowing. If CT2 has already been built into the internal wiring, a test current can made to flow through that path by means of the override switch.
Calibration can be conveniently performed using the sketch
NB. For this sketch to run, the TimerOne library needs to be
installed within the Arduino IDE. A zipped copy of this folder is
available on the Downloads page.
As with most of the sketches for this board, the code needs to be
configured for whichever version of the hardware it is running on. This
involves a one-line change near the start of the code. If ICs 3 & 4
are in use, the line #define PIN_SAVING_HARDWARE needs to be included; otherwise, this line needs to be commented out (by adding "//" at the start of the line).
The sketch cal_bothDisplays.ino measures the real power that is flowing through the two current sensing circuits. CT1 and CT2 are intended to be used for the "grid power" and "diverted power" respectively. One set of results is presented at the 4-digit display and also at the Serial Monitor; the other set is discarded.
The "mode" connector is used to select which channel's data is to be displayed. If the two pins at the mode connector are shorted together (usually by means of the "operating mode" switch), the displayed values are from CT2; otherwise, they are from CT1.
The values for powerCal_grid and powerCal_diverted need to be adjusted until the displayed values for power are similar to the actual power that is known to be flowing. Those calibration values can then be transferred into the Mk2 Router sketch. The calibration values for each individual system will depend on the operating voltage and the burden resistor values. For a 3.3V system with 150R burden resistors which has an operating range of approx 4 kW, the calibration values will be around 0.045. For a less sensitive system with a greater operating range, the calfactors may be as high as 0.12.
This sensor channel would normally have a bobbin-type CT rather than a clip-on one.
Various points in the above photo are worthy of note:
1. To get the values for CT2 to be displayed, rather than those for CT1, I've plugged an external switch onto the 2-pin "mode" connector. This switch is in the closed state. If I were to open the switch, the displayed value would return to 0.000 because there is no sensor at the CT1 port.
2. The displayed value of 2.209 kW is rather high for my 2 kW heater. That's because I'm using the default value of 0.05 for powerCal_diverted. For this board, and all similar ones that I've assembled to date, the correct value is closer to 0.045. This change in cal value would cause the displayed value to be reduced to nine tenths of the value seen here, i.e. 1.988 kW.
3. The board that I'm testing has the extra pair of logic chips (ICs 3 & 4). The calibration sketch must therefore include the "#define PIN_SAVING_HARDWARE" statement. If those two ICs were absent, the #define statement would need to be commented out.
4. The 4-digit display will only appear the right way up if the the main board is orientated with the ribbon cable outlet at its top. In all other orientations, the ribbon cable will need to include an appropriate bend.
5. Not shown in the photo, but if the measured power is negative, alternate values are suppressed. This gives a flashing appearance at the display. Every value, however, can be seen at the Serial monitor along with a '-' sign when appropriate.
[I find it hard to believe that the 7-segment driver IC doesn't provide a '-' symbol!]
Calibration values are unique to each set of hardware, and can be usefully recorded within the unit.
For any pre-built unit that I supply, labels showing these values will be attached.
The pretty pink tape came as a freebie with the standard black & white tape that I normally use!
Having calibrated the system, the relevant powerCal values can be transferred into the appropriate Mk2 Router sketch. Several versions of the Mk2 PV Router sketch are now available via the Downloads page.
NB. For most of these sketches, the TimerOne library needs to be
installed within the Arduino IDE. For RF-equipped sketches, the JeeLib library will need to be installed too. For the latest RFdatalog sketch, which measures temperature, the OneWire library is also needed. Zipped copies of these
library folders are available on the Downloads page.
Having ensured that the system behaves correctly on the bench, it is functionally ready to be connected into the fixed wiring of the premises.
Providing that the Router has been assembled according to my wiring diagram, it should be capable of passing the standard electrical tests for Continuity and Insulation Resistance. While these verification activities are in progress, it is important that the two 240V AC switches are set correctly, i.e. mains switch 'off', override switch 'on'. There's more about this in the Assembly and Final Test section.
Section 8 of my Diverting Surplus PV Power article provides a wiring schematic for an installed Mk2 Router system. Although that article was written in the context of an Unduino Uno, the principle of operation has not changed. An updated version of that diagram is shown below, and a PDF is available here
There are just three connections to the outside world: - a permanently live supply from the Consumer Unit or fusebox; - a switched output to the load, which is usually an immersion heater; - the external CT which is clipped around either of the 'live' tails.
******************************************************************************************************************* After fitting the external cables, the tightness of all electrical connections should be rechecked. Loose connections on power wiring can lead to arcing and overheating. Rechecking the tightness of all power wiring after a few weeks of use would be a wise precaution. *******************************************************************************************************************
Here is a completed Mk2 PV Router of mine that's installed and working.
This photo was taken before PV generation had started for the day, so the display was still in its 'idle' mode.
The three external connections can be clearly seen here.
From left to right:
Switched output to the immersion
Permanently live 230V AC supply
current sensor (CT1)
For this installation, the existing cable from the immersion was rather short. I would normally have left more slack within the enclosure.
Each pair of Neutral and Earth wires share a single connector. These wires needs to be adjusted into the correct position (i.e. side-by-side) before the screw is tightened.
For a different installation, the existing cable from the immersion heater has been extended within the Consumer Unit.
Here, each pair of cores were joined using 2 -way Wago connectors, and placed inside some conduit for protection.
Two different sets of cable colours were in use.
With the extended wiring in place, plenty of slack was available for the power cables at the adjacent router.
When I first assembled this guide, my own system at home was still based on an emonTx.
From the outside, it appears and operates just as described in this Build Guide.
In the welcome Spring sunshine, our DHW tank reached its max temp today.
The external CT clips around either of the incoming tails at the supply point, whichever is most convenient.
The state of our Neutral cable suggests that this particular sensor has been moved around quite a few times!
The external CT has to be the right way around. If it's in the wrong orientation, the feedback in the system will be positive rather than negative. Once activitated, the load will remain firmly 'on' which would be no good for daytime use.
This incorrect configuration, however, can be very useful for after-hours checks. By reversing the direction of the CT, the load should soon come on and stay on. Then, when the CT is returned to its correct orientation, the load should go off. Having demonstrated this behaviour, you can feel confident that the system will operate correctly whenever surplus power next becomes available.
For general use, the anti-flicker mode (AF) is recommended. By setting the AF limits as wide as the supply meter will allow, any conducted disturbance to the mains supply will be minimised.
In conditions where only minimal surplus power is available, bursts of power will be supplied to the load very infrequently when running in AF mode. Visual activity can be increased by switching from "AF" to "normal" mode. Both modes are equally effective at diverting whatever surplus power is available.
On the Downloads page, there is an Installation Guide for pre-built systems that I have supplied. This document should be equally applicable for any system that has been built according to this Build Guide.
* * *
Please note that, to comply with the IET Wiring Regulations, 17th Edition, the installed system should be verified by a "competent person" (BS 7671, para 610.5).
Also, because the conditions of use are beyond my control, I am unable to accept any responsibility for the use of equipment that I have described on this website.
* * *
Thank you for sharing this journey. I wish you every success with your installation.