Puppeteer

Author Topic: Solar Inverter - Load sensing  (Read 369 times)

veggie

  • Keep Calm and Start the Lister !
  • Sr. Member
  • ****
  • Posts: 453
    • View Profile
Solar Inverter - Load sensing
« on: July 18, 2019, 02:44:22 PM »
I would like to be able to feed power from ten X 220w panels into my cabin electrical system to help offset demand from the grid powered electric baseboard heaters in winter. I am not interested in "net metering" or selling power back to the utility company. When all heaters are on at the same time the total load is 4kw.
That creates a mess of paperwork and meter changes which in the end will not be worth my effort.
I am wondering if anyone is aware of a grid tie inverter that modulates it's contribution to the home circuit according to the current entering the house main breaker panel from the utility.
The idea being that my solar panels would contribute to the loads when the heaters ( or other loads) are on, and scale back (or divert the AC to a dump load like a water heater) instead of back-feeding to the grid when there is low demand.
These type of inverters are readily available from non-approved (and probably poor quality) suppliers on Ebay.
They have a donut current sensor which clamps to the incoming grid wire at the breaker box which is wired back to the inverter for monitoring the incoming grid amperage. The unit throttles accordingly.
They also claim to have "island protection" which will shut down the inverter if grid power goes out. Thus preventing back feed during outages.

I am wondering if any North American brands are capable of output modulation or load diversion based on mains current draw.
Anyone else using solar to help reduce electrical loads without backfeeding the grid?.

Summary:
I want to offset the power used by my baseboard heaters without back feeding the grid

« Last Edit: July 18, 2019, 05:01:54 PM by veggie »
- 6/1 GM90 Listeroid - Delco 33si Alternator
- Changfa R175 - Lease/Neville Alternator
- Kubota Z482 - 4kw
- JiangDong R165 Air cooled - 2 kw

veggie

  • Keep Calm and Start the Lister !
  • Sr. Member
  • ****
  • Posts: 453
    • View Profile
- 6/1 GM90 Listeroid - Delco 33si Alternator
- Changfa R175 - Lease/Neville Alternator
- Kubota Z482 - 4kw
- JiangDong R165 Air cooled - 2 kw

glort

  • Hero Member
  • *****
  • Posts: 2977
    • View Profile
Re: Solar Inverter - Load sensing
« Reply #2 on: July 18, 2019, 05:37:02 PM »
What you are after is an inverter with a " Zero Export" ability.  Most good inverters like Fronius, solar edge, Sunnyboy and others have that ability as well as load diversion which will send power not being used in the household circuits to a load like a water heater, Pool Pump etc.  I Imagine you should be able to get one or all of those brands in the states. They are all good brands used in OEM installs and will be approved for grid connected homes. As far as I';m aware, all the mainstream brands of inverters are anti islanding and have been for a long time.  I would think most if not all the Chinese ones are too. All the ones I have seen are.

I know the states is largely gone to smartarse meters but if you have an old analogue spiny meter, you may be able to feed back to the grid on the QT and rewind your meter. I have been doing that for 2 years and works real well. Just don't rewind past your last reading. If you are single phase (1 meter) I would allow say 100 Kwh to make it look like you are using something.

I don't know what your experience with solar is but as a heads up, it's pretty useless in winter and for heating Depending where you are.   If you are putting up 2.2 Kilo of panels and If you are roughly half way between the equator and north pole, I would be guessing you'll be seeing maybe 750W to a kilo  most of winter. Running the heaters and the hot water tank with excess isn't going to happen in winter.... unless you live on the equator.

I'm running about 22KW of panels atm and the last few days which have been perfect sunshine I have made 45Kwh of power.  In summer, I was making 70 Kwh from about half of that amount of panels.  45 Kwh may sound a lot but if you are like me and wanting to heat water and space electrically and live where it gets to freezing point at night ( or during the day)  You will eat that up real easy.  ATM I'm averaging about 60 Kwh day consumption. We have just come from weeks of overcast weather and some days I didn't even make 12 KWH.  that's from 22 Kwh.  if you had 2.2  Kwh of panels, it's possibly you may not generate a KWh for the whole day. many variables but more than a distinct possibility.

I guess it all helps but don't expect Miracles. In summer however you won't pay a cent for your hot water IF you set it up properly.  Probably cover all your daytime loads in the home as well.

If it's heating you are after, I would definitely set the panels up facing due south for you AND set them at the winter tilt which will be quite steep probably.  Forget about latitude or Ideal tilt, summer will take care of itself. Set everything up for best winter performance. You'll need all the help you can get.
If you can put on more panels even east or west, I'd do that. If you can't, I ordinarily say don't bother trying to make panels more efficient but in this case I'd say to have a hope of making them worthwhile You'll want to put them on tilt mounts if the roof is more than about 10o off which it's almost certain to be.

ATM I'm working on a couple of designs for PWM controllers to hoot to a hot water heater. If you hook panels direct they will always be pulled from their max power point for all but a small part of the day which looses efficiency rapidly.  With a controller that keeps them on or near that max efficiency level, they will produce Loads more power. this could go direct to you water heater via a suitable DPDT or SSR  or you could get another tank and connect to that as a pre heater which will save the main tank. May make more useful gains out of that than putting to the board heaters depending on how your tarrifs are set up too.


I would also suggest you crunch the Numbers at the PVwatts.site  https://pvwatts.nrel.gov/pvwatts.php and put in the data for you location, array size, orientation, tilt and don't miss down the bottom to put in your cost of power.  this will tell you how much power you will make and save a year and you can cross reference that to the cost of the install and see if it's worth while.  If you can run a meter backwards, then definitely worth it as you'll probably be able to offset your whole bill pretty much for 6=9 months of the year.
The PV watts site I have found for me to be surprisingly accurate in it's predictions so whatever the number comes back, I'd believe it.

Anyway what you want is a Zero export GTI by Fronious, sunnyboy or solar edge With load diversion.

Grit your teeth or get a bullet because these things are NOT going to be cheap..... Which is why I advise you crunch the numbers first. 

mike90045

  • Mendocino Metro
  • Hero Member
  • *****
  • Posts: 1378
  • Mmmm BBQ
    • View Profile
    • Mikes Solar PV page
Re: Solar Inverter - Load sensing
« Reply #3 on: July 19, 2019, 06:04:10 AM »
For sure, crunch the numbers before you buy anything.   

  There are several PV estimating sites on the internet, you enter your location, size, tilt and azimuth of your array, and you get a month by month output, adjusted for weather.  I'm north of San Francisco, and a sunny day in winter, I can easily harvest 18Kwh with 5kw of panels.  Cloudy days, I run the listeroid generator and top off my Edison Cells (48V 800ah bank)

https://pvwatts.nrel.gov/

Powerfab top of pole PV mount (2) | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | 3Kw 15, Evergreen 205w "12V" PV array on pole on a Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)|  Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
|| Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
|| VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

solar: http://tinyurl.com/LMR-Solar
gen: http://tinyurl.com/LMR-Lister

« Last Edit: July 19, 2019, 06:06:31 AM by mike90045 »

veggie

  • Keep Calm and Start the Lister !
  • Sr. Member
  • ****
  • Posts: 453
    • View Profile
Re: Solar Inverter - Load sensing
« Reply #4 on: July 19, 2019, 02:25:07 PM »

Thanks glort, I was not aware of the term "zero export".
I ran the numbers some time ago and the outcome is not very exciting.
However, I have these panels sitting on a shelf doing nothing at the moment.
This would be a "winter only" application with the panels almost in a vertical orientation and facing due south (for north American orientation).
The big problem is the cost of a "zero export" inverter rated at 3kw.
This system would NEVER pay for itself in 30 years when contributing only in winter.
(In summer the power usage is almost zero at this site so no need for solar).
I suppose the best application for spare panels is to heat a domestic hot water tank on a year round basis.
I can easily match the impedance of an element to the best efficiency point of my panels.
I also have an MPPT tracking converter which converts Panel power to a electric element sweet spot.
I bought one of these units a  year ago and plan to install it. I is designated for my hot water tank.
https://www.youtube.com/watch?v=YkPtPXfhiYk
I think it is meant to achieve what you are working on.

I think you are right, the amount of energy I would produce trying to offset my space heater consumption would not be worth the effort.
Keep us posted on your water heating project. This is something where PV panels can really do well.
I have several water heating applications that would benefit from solar so if you come up with a good DIY solution for tracking PV power points it would be very interesting to see it.

22kw of panels ... wow! that's a good sized array you have there mate! That's more than my house uses per day. Which brings up the burning question of excess energy storage. We really only have two options. Store in batteries or store as heat.
Charging a small electric vehicle would be very feaseable with your system.
- 6/1 GM90 Listeroid - Delco 33si Alternator
- Changfa R175 - Lease/Neville Alternator
- Kubota Z482 - 4kw
- JiangDong R165 Air cooled - 2 kw

glort

  • Hero Member
  • *****
  • Posts: 2977
    • View Profile
Re: Solar Inverter - Load sensing
« Reply #5 on: July 19, 2019, 03:57:13 PM »

Having looked at those Techluck units, I would STRONGLY suggest you uprate the caps in them. They are a marginal  rating for the job and I reckon in normal used one would be lucky to get 3 years out the things.

The way these work is by pulsing the power same as I am looking to do and storing the power from the panels in the off stage of the Mosfet. At this time the incoming power from the solar charges the cap which is then jumped to the element  when the fet switches back on. The pulsing keeps the panels at Power point.  The Duty Cycle and the cap rating is not really up to the job long term and fitting a larger, higher voltage cap would be a great advantage to the longevity of the unit.

The size of the cap isn't that important but if you are going to feed it with 2.2kilo of panels, I'd be going at least 1000Uf and a 450V rated cap.
On my unit I'll be doing 6x 1000UF Caps and If the  scrapped inverter I have up the back is what I think it is, I'll be running 12 1000UF caps.
When the unit is operating in midday I expect the fet to be almost constantly on.  When the sun is lower in the sky, the switching time will be greatly reduced but the power pulse will be much stronger.

You can run small caps but I'm the sort of guy that likes Low reving, High torque kind of setups.  :0)
With the cap on the teckluck I think you are trying to drive a Pickup truck with a small engine sitting on 8000RPM the whole time. It will work, for a while but then.... The ones on those board are not much bigger than you'd expect to find in a radio. Ones I am looking at are much fatter and higher voltage and that's only because I have them and the 4700Uf 450V ones I'd like are not cheap. 6 or 12K Uf will be enough though.

I know where you are coming from with the terminology.  I find that 90% of the frustration myself sometimes.  Look for the wrong term and you find nothing. Put the right term in and there is more choice and availability that what you could have imagined.

I knew the zero export units were exy which is why I said grit your teeth.  I have been looking for one for a while at an affordable price.
I have a 3 phase AC with one phase I can't feed back on.  I got a DPDT relay and put that on an inverter which matches the Units Draw  on that phase, Ironically, 2,2 kilo.  I have that inverter on what was 5 kw of panels so it produces late into the afternoon. When the compressor on the AC kicks in, the relay Switches and the inverter powers up again and syncs to that phase and supplys the AC.
When the compressor drops out, the inverter does too in the split second and then connects the power back to another phase so the power isn't wasted.
Knowing I am running on free power I tend to put the  thermo  on the lowest setting through the day to Cold sink the house much as I can and then just have the AC kick in minimally through the night if it needs to.

My zero export is just running the inverter to it's limit which nicely matches the load I want to drive and can't export any excess power.
It's not the most elegant of solutions but seeing it saved me at least $500 last summer and the relay was about $4, I think it's perfect!  :0)

I was thinking today and after my headache subsided, I ordered a DC PWM controller that is capable of up to 90V.  it is adjustable with a 0-5V signal. I'll Play with it manually then work out how to feed it the 0-5V range with a voltage divider circuit so it should be fully auto. I'll be able to hook  up to 3 panels in series and can then parallel them with 2 arrays but I might use the PWM to drive a larger fet setup so I can pump a couple of KW through it.
Not my idea atm, 15A@ 90V would be enough for the bit of heating I have in mind but if it could be easily expanded..... Be another way of addressing a solution.

I have a LOT more than 22KW of panels atm but I'm going to rationalise.  I'd need an impractical amount to do my winter needs unless I offload some of the heating demand which is the plan so even covering all the roof is not going to meet all the demand anyway. In summer I can switch half of what I had off and still be so far ahead it doesn't matter. Meeting winter power demands will not be a matter of throwing endless panels and inverters at the problem, it will simply be reducing the power demand by taking out a lot of the heating load that can be met by other sources much more easily.

I have had a lot of fun playing ( with very deadly Voltage )  panels but I have learned all I need to for practical purposes. Now it's time to Finalise my setup for the house, Package it all up nice and neat ( and safe) and let it do it's thing.  I'll keep one 2kw or so array on the shed roof for my experimentation and amusement and a couple more panels for the low voltage power stuff.

This week I have been Cutting down my 14Ft hedges up the back on the north side of the yard to 7 ft which is enough for privacy but I'm amazed how much more of the top yard is now in sunlight.  Just begging for a ground array  but I have to stop my solar obsession and find something else to get wrapped up in.

If I could store power effectively, I could easy make enough excess in summer to get me though the winter as well.
That's the problem now with the whole RE grid thing. There is NO way to store that amount of power even on a domestic let alone a grid level.

BruceM

  • Hero Member
  • *****
  • Posts: 2751
    • View Profile
Re: Solar Inverter - Load sensing
« Reply #6 on: July 19, 2019, 09:02:20 PM »
The Techluck is only a DC PWM unit, with no voltage shifting ala boost converter, so wattage to element depends entirely on the panel array and element voltage.  Open circuit max PV voltage is 250V, and max operating PV voltage is 200V.   Frankly MPPT is a waste of circuitry and complexity for this application, and a simple DC solid state relay driven from a non-load carry thermostat would perform similarly and without any EMI, at a small fraction of the price.

Your scheme will have the same limitation, Glort, in that 90V applied to a 230V, 4500 watt element (11.5 ohm) will only consume 704 watts.  You must either increase panel voltage or use a DC-DC boost converter (or other DC-DC transformer based design) to raise the panel voltage to match heating element, or find a lower voltage, high wattage element.  Ohms law lets you calculate it.

Most cheap hard switching and badly designed devices make me cringe; EMC is typically ignored, the cables to the panels will likely obliterate radio reception in the home.  Sadly it is now normal to have hobbyist design level switching power supplies in appliances of all sorts and this does not bode well for public health since conducted EMI goes back on all the home wiring.  Our home's wiring is typically totally unshielded, thus acts as a radio transmitter for high frequencies and is literally wrapped around us as we sleep.  Switching supplies were unheard of in appliances and HVAC just 50 years ago, so now we are in the midst of a huge public health experiment; both wireless and EMI levels  in homes are now typically 100,000 times higher than 50 years ago.  Neurodegenerative diseases, autism and other chronic illnesses are on the rise.  Hmmm.




glort

  • Hero Member
  • *****
  • Posts: 2977
    • View Profile
Re: Solar Inverter - Load sensing
« Reply #7 on: July 20, 2019, 12:59:25 AM »

 Frankly MPPT is a waste of circuitry and complexity for this application, and a simple DC solid state relay driven from a non-load carry thermostat would perform similarly.

Yep, that's what my first idea was based on. Use the thermo as a non load beading controller to drive a fet that does handle the load.  The thermo Board will be driven by a Little converter that will take high voltage DC and bring it back to 12V @ .5 a and that can drive the fet that is switching the the power from the panels and the cap. Divider to replace the thermo couple and run the power into a cap bank and switch that to the element with a fet.

Panels have some flexibility in them regards to the power point and I'm going to try 4v either side of their rated Pmax as I believe they will perform efficiently enough in that range and there will be variation anyway due to manufacturing tolerances. If I'm running 8 panels in series to get the voltage and power to where I want it, I'll get a fair voltage swing between  the on and off cycles to make  control easier and give the panels a decent but tight enough working range. The goal here is cheap and simple enough for a twit like me to make it work, not the sub watt efficiency an electronic genius like you have the knowledge to construct. I think this will work to a level where things like cloud and the amount of bird droppings will have more influence than the efficiency of the circuit design and if not, doesn't matter.
I'll be still getting a Hugely better efficiency out of something I can hash together for maybe $30 instead of the $300+ for a tekluck or similar.


Quote
Your scheme will have the same limitation, Glort, in that 90V applied to a 230V, 4500 watt element (11.5 ohm) will only consume 704 watts. 

Geez, I hadn't even thought to work that out! Sometimes the obvious escapes me.
That is going to be so much more power than I need for the job in mind. I'll have to incorporate a thermo into that or I'll have a kettle in no time with the amount of water I want to heat.  Could knock the panels back but I'll leave them as 3 to allow for crappy weather and power in reserve.  This will mainly be a winter thing so in reality I'll probably be getting 1/3rd of that output but still plenty more power than I was thinking for some reason.  :-[

For doing a domestic heater I'll just increase the panel array and add some mosfets to carry the higher load. That will only be for the exercise as I have no need for that in reality with my present setup but If I can make it work I can do a vid for others whom may find it helpful.
Even at 700W, a small 125L heater unit could go from 40 to 70oC in 6 hours. if you had a series of 3 panels in 2 parallel arrays, you could get that 6 hours of 700W for a lot of the year. The limitation of the resistance of the element would mean you could throw more power from the panels at the controller without cooking it because the voltage will stop the element sucking more power than the 15A the controller in mind can handle.


Quote
You must either increase panel voltage or use a DC-DC boost converter (or other DC-DC transformer based design) to raise the panel voltage to match heating element, or find a lower voltage, high wattage element.  Ohms law lets you calculate it.

Yep, don't want to add the converter so I'll use more panels and raise the voltage/ power and just use the controller to drive a fet on the high side.
There is a guy on YT that spouts a design he came up with that runs off 2 panels and has a voltage converter but the thing is only doing about 300W. He claims that is enough but then says hes only heating a 5 gallon tank.  Mrs would use that for the washing up! No way in hell I'd be living with 2 minute showers again.  I spend all day getting filthy and aching, I want a good long shower to get clean and take some of the ever increasing pains away.

Quote
  Sadly it is now normal to have hobbyist design level switching power supplies in appliances of all sorts and this does not bode well for public health since conducted EMI goes back on all the home wiring.

I was thinking of you the other day Bruce.
Something the daughter brought home and plugged in the other end of the house and I could hear the EMI through my computer speakers.  It turned out to be the AC power supply. Once she took that off the thing was fine ( well not producing interference my computer speakers could pick up anyway)  and I got her to charge it from a USB source. I must have rubbed something off on her. She has this large battery bank for her phone and other things and has been taking that with her to work and charging it in the car on the way and back home. She seems chuffed she can charge it in the car for nothing and then charge all her devices from that.  Spose it's a lot better than all those converters she had power board full of in her room.

BruceM

  • Hero Member
  • *****
  • Posts: 2751
    • View Profile
Re: Solar Inverter - Load sensing
« Reply #8 on: July 20, 2019, 02:20:08 AM »
Yep, little cheap switching power converters can be quite a radio transmitter.  One lady I was helping by phone had become newly electrically sensitive...she was getting headaches from her computer and other equipment.  I had her snoop the house wiring at the circuit panel with an AM radio...and in 20 minutes with some phone coaching she found the offending circuit at the panel, then traced it to the wall behind her desk (she was a writer) there was a POS battery charger plugged into an outlet inside that opposite room's closet.  She'd bought it and used it and then forgot that it was plugged in i that closet. Her problem got much better after that and fixing/unplugging some other things to dramatically reduce her daily exposure. 

In most homes now, there are so many sources it's downright discouraging.  The smart meters are a significant offender- cheapo switching power supply on the board, designed by an EMC illiterate moron, with not even a common mode choke between it and the line. Besides the switcher, the conducted radio frequency emissions exceed the insanely high FCC part 15 rules, but the power co is exempt.  The inverter type split systems are an EMI horror show, even when the room units are "off".  Crazy world, we have no idea what we're doing. 




glort

  • Hero Member
  • *****
  • Posts: 2977
    • View Profile
Re: Solar Inverter - Load sensing
« Reply #9 on: July 20, 2019, 03:37:44 AM »

  Crazy world, we have no idea what we're doing.

I disagree with that.
I think with 99% of things they know exactly what they are doing but for profit reasons, just deny it.
the list of things that people claimed and were made out to be nutters for 20 years until they could deny it no longer and they were proven to be right is long and painful.

I think in this regard the world is not as crazy as it is greedy and corrupt.

It's a lovely day here today. Just after midday and my panels are cranking out lots of welcome power.  3.2 Kw on one inverter, 2.7 on another which will peak around 1 Pm being all west facing and 3.4Kw on the other.  I broke even on one phase yesterday, 1Kwh down on another and 8 Kwh for the 3rd leg of the AC I'm not currently back feeding as I'm not heating through the day and only running that at night. Maybe today I'll make some progress winding back.

I'm quite looking forward to trying my experiments with the power point regulation on the panels.
I have asked about the ideas on several electronic websites and yet to get an answer to them. the boffins are quick to draw up better and far more complex solutions  which I don't seem to be able to get across to them I'm an idiot and incapable of  building them. I have offered to pay for them if the designers would like to build one for me but no joy there either.

I'm fairly confident I can make it work if not initially, that I can learn how to make it work and what I need.  I have learned a lot looking at the designs others have come up with even if they are well beyond my abilities.  I can see the principals they are using and think I can get to the same destination even if I have to take a more off the beaten track route to it.

The devices that do simple things like water heating IMHO are ridiculously over priced for the components involved as well as the R&D which would also be nothing more the learned couldn't come up with in a few hours if that. From what I have seen demonstrated, a lot of the electronically educated can come up with much better circuits than the commercial offerings in the 24 Hours they say " I'll see what I can come up with" till the time they show pictures of their prototype running.  If I can come up with something simple a moron like me can cobble together from available boards and show others that have some electrical rather than electronic knowledge I'll be happy with that.  :0)

Wish those slow boats from China would hurry up though!


veggie

  • Keep Calm and Start the Lister !
  • Sr. Member
  • ****
  • Posts: 453
    • View Profile
Re: Solar Inverter - Load sensing
« Reply #10 on: July 22, 2019, 01:32:22 AM »

Well Glort ... I'm a lot less proficient at electronics than you, so I will be watching intently on your progress.
Hopefully you are willing to share your findings  ;)
- 6/1 GM90 Listeroid - Delco 33si Alternator
- Changfa R175 - Lease/Neville Alternator
- Kubota Z482 - 4kw
- JiangDong R165 Air cooled - 2 kw

glort

  • Hero Member
  • *****
  • Posts: 2977
    • View Profile
Re: Solar Inverter - Load sensing
« Reply #11 on: July 22, 2019, 05:43:27 AM »

Well Glort ... I'm a lot less proficient at electronics than you, so I will be watching intently on your progress.

No, not possible, I don't know anything. I just fake it and tie wire stuff till it works.   :laugh:
I have to say though, I have had a lot of success with things that way. I have lumped off the shelf products together before and got the result I wanted very cheap if not elegantly. The specific gear for the job cost many hundreds or over over $1000 and I was able to kludge stuff together for next to nothing with old stuff people gave when they didn't want it anymore. Icing on the cake was people laughed at the concept let alone the method. More than a few have even called BS when I should it working and had been using it for months.
More satisfying than insulting.


Quote
Hopefully you are willing to share your findings  ;)

You can count on that!  I'll be very happy to show that my idea was correct on this ( given of course it does work as I hope!) that pre-made cheap parts can be wired together to achieve a result that would otherwise cost a LOT of money for job specific products.  I guarantee it won't be the most efficient or the most compact or elegant, but if it works at a fraction of the price of commercial things and pelicans like me wire together, then that's my goal achieved.
From there smarter people can modify it some more and give it better features.

I have put my ideas out on a couple of places where there are very expert people and none will tell me it's workable or laughable so I don't know what to make of that. I'm left having ordered the parts and having to put it into practice and if it does not fly, I'll learn a bit hopefully what I over looked.

It would be real easy to wire in thermo stats and voltage/ amperage meters, data loggers and even wi-fi monitors these days which a lot of designs have.
Lot different at my lack of skill level to work out where to attach a few wires than design it all into a neat electronic circuit like I see others do.  I couldn't even assemble the components let alone design it.

Hopefully a bit of Tight arse and cunning thinking may allow me to come up with an off the wall solution many other people can put together as well.