Archived Forum

Hi Canada man, you show the water heater just using phase 'A' 120v this will work fine as thats also where the inverter feed goes with the Mk2 as you will be using the 'A' line for the CT measuring the usage on this line. you can have a Mk2 chugging away on its own as it's independent of other devices.

Where I am confused is your need to limit the solar you produce, in the UK we are paid (a little) for any power we send back to the grid, in Spain they are penalized and are charged for what they feed in as its considered waste for disposal. What is the situation in Canada and why do you need to limit what is produced? After all if you don't need the power you just don't use it.

The Mk2 as you have probably read is a burst fire system, there have been others  who have experimented with phase control but this has limitations and problems of its own. It would demand quite a rewrite to even try PWM.

Rather than try to have the Arduino do two jobs (the Mk2 is basically a balancing system), I would suggest a Mk2 controlling the water heater and a separate measuring system somthinge like the emon or eco eye.

If you could explain your need to lose some of your generated power it would be useful.

b) limiting the amount of DC from the solar panels to GT inverter
by using a PWM regulator (based on a modified solar PWM charger).

That might be problematic.  The DC voltage in your typical domestic grid-attached PV system is very high....  many hundreds of volts.  You typically need a seriously heavy duty DC isolation switch.  The DC voltages and currents involved are pretty scarey.

And even if you could solve the switching problems, I think your inverter's maximum power point tracker would get very confused.

richmc: Hi Canada man, you show the water heater just using phase 'A' 120v this will work fine as thats also where the inverter feed goes with the Mk2 as you will be using the 'A' line for the CT measuring the usage on this line. you can have a Mk2 chugging away on its own as it's independent of other devices.

I suspect that Canadian metering works in a similar fashion to UK 3-phase metering in that it just measures the total power from both phases. If so you will only have excess PV power if it exceeds the total load on both phases so you need to measure both and only turn the triac on if the total load is less than the solar power.

The first point is Robin has done a Mk2 that uses phase control - the sketch is documented and listed here, so no rewrite is necessary. All that changes is the opto-isolator / driver i.c. Second, it would be wise to check early on whether the meter does in fact aggregate the two legs ("phases") for the purpose of calculating the nett energy flow, as that will define the strategy that you use. My suspicion too is it will, but evidence is needed.

I think MartinR might have a better feel for the capabilities of the Arduino than I have, but my first reaction would be to follow Rich's philosophy and have several devices each doing a specific task. I think it could be seriously slow trying to do all you require with one.

"The DC voltages and currents involved are pretty scarey." and they don't take prisoners.

Surely the answer is to control the MPPT, if the inverter allows it? And if not, get one that does. I can't believe it would operate anywhere near correctly when fed with chopped d.c. At best you'd need to heavily filter the output of the PWM switch.

"The first point is Robin has done a Mk2 that uses phase control "

Yes but the OP was asking for PWM, not quite the same and very inappropriate for AC.

And Rich mentioned phase control. I do know the difference between a.c and d.c, believe it or not.

I don't think I understand how PWM is intended to work here. 

With my 'Mk1' PV Diverter, I used a smoothed PWM output from the Arduino to create a pseudo-analogue signal which fed an external power controller.  The system did work, but was complex, expensive and slow to respond, and did nothing that a simple triac can't do.  So I went down the Mk2 route instead.  

Robin, I think the idea is to chop the d.c upstream of the inverter to limit the PV power (almost) at source. We don't yet know the reason for that - I think there is speculation that it might be down to (possibly ill-conceived) legislation, as in Spain. Whatever the reason, the next question is how will the inverter and the maximum power algorithm react. I fear it could get very messy.

I think the control scheme is that the Mk 2 is regulating the dump load to match the generated capacity, as the 'primary' load, and when that cuts out at maximum temperature, the PWM on the d.c. side throttles generation to match the house load - effectively taking the place of a second level dump load.

I know modern "smart inverters" allow remote control of their real and reactive power output.  I think this is now a requirement in Germany, but  I think the grid manager "owns" those controls.   Attached is an example of how one inverter manufacturer does it.   The master inverter is set up for direct connection to the 4 relays in a Radio Ripple Control Receiver and it then passes those settings along to the slave inverters via RS485.   The grid manager can then manage inverters in real time, and potentially even use them to inject reactive power deep into the network.

http://www.solaredge.com/files/pdfs/power-reduction-control-application-...

http://www.solaredge.com/files/pdfs/application_note_power_control_confi...

Hi everyone!

The first of all I would like to thank all for your fast and friendly responses!

My answers:
0) My small system made for increasing of my usage of power from local provider.
Not for commercial feeding....

Now, even in cloudy day, my PV supports my home appliances well.
Pls look at my "My usage" picture.

In sunny day I lost a lot of energy (marked by red text on my Hourly usage (from Hydro One web site))...

---------------------------------------------
1) My grid-tie inverter (from China): input 22-60(!)DCV, output 900-1000W @90-130ACV, 60Hz.
3 of 305W solar are connected in parallel.

For regulation I would like to make a modification of my broken Morning Star TS-60 charge controller to a PWM regulator.
12 of IRFB421 MOSFET transistors of TS-60 can work up to 150V, 60A (up to 3600W)/
For smoothing of "chopped" DC from PWM regulator I can install 2-3 big capacitors (and coil if it necessary).
Could help me to make this modification?

---------------------------------------
2) Reasons for limiting are very simple:

a) According my real local experience and real understanding of local Ontario "green" rules a procedure of legislation is very difficult (very long term bureaucrat way) and absolutely unprofitable (for my case).
As a result of this:

b) I don't want to feed lines of Hydro One (the biggest energy power in the World!) for free.

c) I want to increase a life of my solar panels.

d) For direct water heating I would like to have a independent solar water heating system (with vacuum tubes).

----------------------------------------

Thanks for your links and advises about using of other system.
Unfortunately there are not for my idea.

==================
Thanks ALL again!

"I do know the difference between a.c and d.c, believe it or not."

Yep, one tingles the other hurts. ;-)