The inverters do sync themselves with the grid phase.
-
Categories
-
Platforms
-
Content
need help going green and willing to pay for help! solar / relay / timer control
- Thread starter henry barbera
- Start date
Fish. I've seen you post on alternative power quite a bit. Do you have any ideas that will work with the current setup?
It sounds as though, currently there is more than one array, and the inverters are connected to mains, either at or very near the compressors for the AC units.
The Solar arrays do not provide enough to fully supply the compressors, but supply almost all of the required power. Mains is relied on to take up this extra bit.
A UPS in reverse solution would not work, as the panels cannot support the AC compressors, but how can current from one of multiple AC sources be monitored to detect a fall in current caused by cloud cover and not a changing load?
*(Steve)*, do you have any ideas on a safe way to implement the clamp-on sensors? One at the output of the inverter, and one at the input of the compressor. You should be able to easily tell if the compressor begins to pull more from the grid rather than the panels by comparing the two values.
You should be able to do this from the mains input. A clamp on current meter and a voltage reading will enable you to determine the current load. This could be used to determine whether you should do some load shedding. And at the same time you could implement the 30 second delay for restarting the air conditioner.
these are absolutely grounded - per code.
yes i have multiple inverters. each inverter has three panels at 230w. three of these sets come together after the inverter outputs and go the AC unit. we're making this a bigger thing then it is. i just want to read from the combined inverter outputs (could be one or ten inverters at 100 or 300w - that doesn't matter) how many amps i am getting and to control a relay based on that number. the inverter will be at 120v 60hz. Gryd3 has confirmed we'll have to have the circuit under load to take a measurement - take a look at that thread.
Asking questions and thinking out loud, but I thought clamp-on current sensors only worked with AC. So using it on the Solar Panel side of the inverter is out of the question.
Depending on how the physical wiring is done, a clamp-on sensor could be placed on the line after the last inverter but before anything else. An additional clamp-on sensor would be placed at the input of the AC unit's compressor. My thought here is that a PIC, or other basic MCU could compare the values of the two sensors, and if the sensor on the AC unit read considerable higher than the sensor on the inverters, then it could be confirmed that the solar panel power has dropped. Without using two clamp-on current sensors, the voltage on the panels would need to be monitored in addition to the current output after the inverters.
I'm thinking two are required to prevent false positives if the AC unit's compressor cycles off momentarily causing the current draw in the circuit to drop. This would also help determine (again, without voltage sensing on the panels) if the cloud has dissipated without turning on the compressor for 3 minutes to 'test' for the current draw. (Cloud cover would present as higher current at the AC compressor than inverter. Clear would present as a higher current at the inverter than compressor, or as s slightly higher current at the compressor than the inverter)
This could be done with discrete components instead of a microcontroller, but a micro would allow for more features to be added much easier.
Ideas, concerns, points I missed? I wanna hear from you too Steve. These are only my thoughts out loud.
no that is not what i want. as we have it set up currently - the AC unit will pull from the grid & the array-inverters (some of each) to run. it's always using some of each because the array at it's best moment will produce just shy of what the compressor needs so any short fall is always covered by the grid. in addition, when the compressor starts there will be a spike on the draw to get it moving. so for that moment there is a bigger drawn which the array could not handle even if it was oversized. i could do battery back-ups but without going to into another tangent lets just say i have ruled that scenario out. in short that system would be costly and i would have to repeat it six times - once for each system we have. there is also hugh efficiency losses when batteries are added to the system. i would have to hook up to the grid as the primary source of power plus it would not in any way address what i want to accomplish. i have thoroughly investigated the battery backup system i would need and ruled that out long ago.
when i talk about the time delay i'm incorporating in the project i say that because 1) i know it's not good to constantly keep turning a compressor on and off. 2) many clouds are short and puffy - they come over and block the sun from shinning on the array for just a minute or so. to run the system for a minute or three is pretty cheap relative to running if for six hours with loads of serious cloud cover. 3) if the AC unit is turned off the handler which is still connected to the grid keeps pumping out cold air for another minute or so. 4) the AC handler unit has a label on the front which states 'do not turn the system back on for at least three minutes after it has been turned off. for these reasons the addition of a time delay is warranted.
you stated: if there is a "brief" interruption of power because of clouds, you would like to supply your air conditioners from the mains supply, but if the "cloud cover persists" you would like the system to simply shut-down.....does this accurately describe one possible way to view "solving" your problem?
YES, however the compressor as currently wired already is supplied from the mains (which completes the array circuit) and as less power is produced from the array the more the compressor would draw from the grid. yes, once again the compressor will consume all of the power from the array because the array is not large enough to produce 100% of the draw from the compressor. if the cloud cover persists i want to shut the system down. after three minutes we could then turn it back on if we have enough amps being produced from the array.
Fish4Fun
So long, and Thanks for all the Fish!
Henry,
Sorry for seeming obtuse, but in subsequent posts you have given more information about your solar array//inverters and this information is VERY important to achieving your goal...This little "tid-bit" fundamentally changes the problem:
This is the key-part most of the posts thus far have been dancing around and makes solving your problem//achieving your goal VASTLY EASIER. You don't need to monitor your Solar output @ all! You simply need to monitor your mains input.. If your AC Mains draws more than a predefined amount of power over a predefined period of time from the MAINS then you want to shut one or more AC units down in a predetermined manner until such a time as your mains consumption approaches zero or actually ""goes backwards" @ which time you would like to turn an AC unit back on. This is fairly easy to achieve and MUCH easier than attempting to monitor your solar input. This is very similar to "load management" systems where consumers allow the power company to turn their water heaters off during "peak hours" and reward them with a discount on their power bills. In your case your inverters are NEVER supplying all of the power required to run your house-hold, regardless of how things are connected, so what you would like to do is ensure that you consume 100% of the solar power 100% of the time, and you would like your primary "dump load" to be your AC units.
Before I go into detail about HOW you might achieve this goal, I want to make SURE I have re-phrased your objective correctly....
Fish
Sorry for seeming obtuse, but in subsequent posts you have given more information about your solar array//inverters and this information is VERY important to achieving your goal...This little "tid-bit" fundamentally changes the problem:
This is the key-part most of the posts thus far have been dancing around and makes solving your problem//achieving your goal VASTLY EASIER. You don't need to monitor your Solar output @ all! You simply need to monitor your mains input.. If your AC Mains draws more than a predefined amount of power over a predefined period of time from the MAINS then you want to shut one or more AC units down in a predetermined manner until such a time as your mains consumption approaches zero or actually ""goes backwards" @ which time you would like to turn an AC unit back on. This is fairly easy to achieve and MUCH easier than attempting to monitor your solar input. This is very similar to "load management" systems where consumers allow the power company to turn their water heaters off during "peak hours" and reward them with a discount on their power bills. In your case your inverters are NEVER supplying all of the power required to run your house-hold, regardless of how things are connected, so what you would like to do is ensure that you consume 100% of the solar power 100% of the time, and you would like your primary "dump load" to be your AC units.
Before I go into detail about HOW you might achieve this goal, I want to make SURE I have re-phrased your objective correctly....
Fish
let me chirp in if i may.
first i want to THANK EVERYONE for any and all comments. i know all of you typically deal with other electrically educated forum members. i know it's obvious i am not, none-the-less you have shown me patience and respect. i am grateful and i can't express enough how much i appreciate all of the time given by each of you just thinking about this design and build project. i know the world is a busy place and no one has extra time.
let me again try to explain what i think i need:
all along i have thought we need to read the power coming out of the inverter - how many amps (120v 60hz) so a clamp on meter sounds perfect. we don't have to read what the power draw of the compressor since that number never changes. when the AC unit is on the compressor is always running (thermostat temp is set to 60 degrees and the house temp will never ever get that low - the compressor will alway run!) the draw is always 9.2 amps at 120v +/- a very small amount. yes compressor start up will spike above the 9.2 but that has no relevance to us.
for a moment let's just forget about the time delay.
? so can we use the reading from that inverter clamp-on current sensor and say if the amps fall below 6 (my preset level) then make our relay go to position '1'. how?
lets say we can do that. now we want the relay to return to position '2' when the amps go above 6. Position '2' will complete the grid circuit coming into the compressor. we have one issue here with the reading - since the array / inverters are not under a draw from the compressor in relay position '1' there won't be any current to measure. so i will have to hook up to relay position '1' a string of lights, a pump, a heater (something) so that there is a load on the line which will give us the ability to get a reading. i understand that i should size that something to an amp draw greater then what the compressor draws - 9.2 amps.
so without taking into account a time delay we need this gadget that reads the amps coming from the outside of the inverter and based on the current reading will set the relay to the appropriate position. so from what i'm gathering we need a clamp on power sensor - and a relay which will go to position '1' or '2' based on the current reading. i have no idea on how to do this.
Position '1' breaks the grid current line to the compressor and completes the solar array / inverter circuit to something - say a string of lights.
Position '2' completes the grid current line to the compressor and breaks the solar array / inverter circuit to something - say a string of lights
i don't know if the settings, position 1 & 2 can be accomplished with one relay or if two will be needed. i have for discussion at this time ASSUMED it can be done with one. please note the solar array / inverter circuit will control (on/off) a separate distinct grid line. also the solar array / inverter circuit will always be under load - either the compressor or the string of lights.
in the example above i have talked about switching the relay at 6 amps. in the real world i may install more systems or replace my existing compressor and then will want the switching to take place at 4 amps or at 12 amps, who knows. so we need the ability to change that preset level. i do know that it will be 99% or the time in the range of 2 to 30 amps. in my stupid ass mind i think of a rheostat. so we need this gadget so i can adjust the preset level for the switching of the relay. what is this thing called? how can we add it to the clamp-on power sensor and the relay? i will for now call this the 'variable power setting switch'.
lets say we get our clamp-on power current sensor, relay and variable power setting switch functioning like we want. now we would want to incorporate the three minute delay before another relay switching could occur.
hopefully i have given everyone a more simplistic view of what i'm trying to accomplish. i really don't think we need to be concerned with the position of the solar panels, installing new micro panels and reading the power from them, checking on what the draw from the compressor is at any particular time, UPS's or any other gadgets. I apologize for not being able to adequately explain what i need and for soliciting those kind of responses.
henry
Henry, You have stated earlier that if, by odd chance, the solar panels+inverter supply more current than the AC compressor takes, that it will back-feed into the house and power other items.
From an electrical point of view, once the inverters are connected to you house they will always contribute unless the electrician who installed the system put in a relay or some other magic that disconnected the inverters completely when the AC compressors were off. This is why I really want to see a diagram, as I generally understand the setup, but there could be a small quirk that throws things off...
As Fish has said, you could always monitor the current... regardless of location, if the inverters were physically disconnected from the house when the AC units are off then this solution will not be complete. It will not be able to turn back on unless it turns the AC units on to 'test' how things function.
If I am correct in the assumption that they are always connected to the house, then when the AC compressors are off the house will benefit from a decreased grid usage.
You 'may' be able to test this by turning off all electronic devices in the house and looking at your electrical meter. If it rolls backwards, then we can determine that the inverters are always connected and can feed back to the grid. If it does not move at all... try turning on some devices (Not the AC units) to see if the meter changes behaviour.
YES the array-inverters could and would back-feed into the house if the power produced was not consumed by the running compressor. Because or the array size and the compressor draw this will never happen.
As previously stated the circuit from the solar array is not complete when the compressor is not running. the solar-inverter circuit is only complete when the compressor is running and the compressor must have grid power to make that happen. call it magic - a relay. what i was told is that the way the electrician hooked the solar-inverter power into the contactor (which i understand to be a switch - probably a relay) is that it will ONLY send power (complete the solar circuit) when the compressor is on. This is true and the inverters will not provide any power when the compressor is off. All the inverter lights stop flashing and fluking the inverter shows zeros even with several household appliances on. So yes i get no benefit from my panels when the compressor is off.
so you are not correct when you say 'i assume it is always connected to the house.' the solar-inverter circuit is only connected when the compressor is on. I know it seems illogical to be configured this way but in my world i have to deal with a utility company which is a monopoly, basically owns the island and can set any rules they want. thats a whole-nother topic and believe me you don't want to hear me screaming about that. in 2018 the rules here are scheduled to change, however, nothing happens as scheduled on this island - it;s the island time syndrome and i don't expect it to become less complicated for many years to come.
That is exactly why I wanted the drawing...
Fish, the current sensing idea will work for turning the AC unit off... but do you have any other ideas to turn the unit back on?
I am not certain if the DC voltage into the inverter would be a safe/reliable option to re-enable the circuit.
The electrician must have either added an additional relay to switch the inverter circuit on in addition to mains on for the compressor, or swapped out the relay for a double pole type to switch both at the same time... Would it be possible the relay switched a control line on/off for the inverter?
Can we get a picture of the inverter? You said the lights go out and you 'fluke' 0V. (but you have not said at what points)
We can ask more questions, but imagery will help.
*If we can't make absolute certain about how these things switch and connect we will resort to assumptions and guesses which could result in the proposed solution not working at all, or introducing more problems. There is still going to be a solution. worst case will require you to put sensors out for each of your panel arrays to measure the average light received on the monitored faces and switching that way. This will be isolated from what goes on in the house as opposed to using pre-existing equipment in the house to do the monitoring.
Fish, the current sensing idea will work for turning the AC unit off... but do you have any other ideas to turn the unit back on?
I am not certain if the DC voltage into the inverter would be a safe/reliable option to re-enable the circuit.
The electrician must have either added an additional relay to switch the inverter circuit on in addition to mains on for the compressor, or swapped out the relay for a double pole type to switch both at the same time... Would it be possible the relay switched a control line on/off for the inverter?
Can we get a picture of the inverter? You said the lights go out and you 'fluke' 0V. (but you have not said at what points)
We can ask more questions, but imagery will help.
*If we can't make absolute certain about how these things switch and connect we will resort to assumptions and guesses which could result in the proposed solution not working at all, or introducing more problems. There is still going to be a solution. worst case will require you to put sensors out for each of your panel arrays to measure the average light received on the monitored faces and switching that way. This will be isolated from what goes on in the house as opposed to using pre-existing equipment in the house to do the monitoring.
Last edited by a moderator:
there are three of these inverters. three 230w solar panels feed each of the three micro inverters. nine panels for the AC unit. the three micro inverters power the AC compressor. you can see the wires coming in to the front on this one inverter, they come from the panels. on the back of each inverter the power cord runs to a carlon junction box and inside the box the three inverter outs are tied together +/+/+ -/-/- there is a ground coming from the array frames into another carlon box where they are also tied together and then the ground goes back to the main where it grounded. the AC unit is also ground to same location. the + and the - from the carlon box go to the compressor where the circuit is completed through a relay when the compressor is running.
so the clamp-on power meter must be after the carlon box where the three positives are connected. i can read the total combined amps there. THATS THE IMPORTANT MEASUREMENT FOR ME.
i guess you didn't like the idea of sending the solar power to another circuit (a string of lights) when the amps fall below the 6 amp threshold and the circuit to the AC compressor is broken. my thought was to keep the array under load (actually it wasn't my idea but it is now) so that we could use the same clamp-on power meter results.
you are correct when you said 'Would it be possible the relay switched a control line on/off for the inverter? that is exactly what is happening.
you also said You said the lights go out and you 'fluke' 0V. (but you have not said at what points)'
When the compressor is off the circuit for the array is broken and i get zero power out from the inverters.
Yes.
It's too bad that the inverters were not always connected to mains... If the AC units turned off, the inverters would still contribute to the rest of the house saving you more money... but you run the risk of feeding back to the grid if the household requirement is too low while the AC is off... I'm sure your electric company would not like that, so in addition to the factory relay to cut power from the compressor, there is another relay (or the relay has been modified) to separately disconnect the inverter output.
I'd really like to pick @Fish4Fun 's brain on this one.
I want to help think up an easy to implement and reliable method for determining when the cloud is gone and the inverters can supply the appropriate amount to the compressor.
Fish, can the open circuit DC voltage of the panels be used for this?
Arouse1973
Adam
This must be getting irritating now for Henry. My first post mentioned current transformers which are these clamp on things others mentioned after this. Its time I think one of us drew out some wiring diagram for Henry to check that it is his current connections. I will do this next week as I am away at my brothers wedding this weekend. Then we can stop this tennis contest going on at the moment and look at a possible solution for him.
Adam
Adam
May I ask if this would not work...
Sun goes behind cloud...
Immediately have a relay turn the compressor off...
Wait 3 minutes...
Keep on waiting till the sun comes out again...
Switch compressor back on again.
All you would need is a PIC chip, a relay and a few LDR's or light dependent resistors.
Or a 555 timer and relay powered by a small 6V pv panel set to turn on after 3 minutes.
If there is no sun, the relay cannot turn on, and if there is no sun the 555 will not allow the relay to turn on before 3 minutes.
Sun goes behind cloud...
Immediately have a relay turn the compressor off...
Wait 3 minutes...
Keep on waiting till the sun comes out again...
Switch compressor back on again.
All you would need is a PIC chip, a relay and a few LDR's or light dependent resistors.
Or a 555 timer and relay powered by a small 6V pv panel set to turn on after 3 minutes.
If there is no sun, the relay cannot turn on, and if there is no sun the 555 will not allow the relay to turn on before 3 minutes.
This is the best drawing i could do and had to take an excel file and make it a pdf
Thank you very much Henry. This is perfect.
The anti-island feature may make this a little more complicated, as we can't just redirect the inverter output to a set of lights unless there is grid power on the set of lights as well... This also means that once connected to lights, or fans etc., you run the risk of back-feeding power from the inverter back to the grid which I assume is not allowed?
I think that attempting to use the reduced power from the panels with a cloud overhead will make this too complicated.
So.. my thoughts for a solution is:
-Clamp on meter for the 'grid' power-line into the air conditioner. When tripped will turn the AC unit OFF. (This will be read, and if the reading is too high, then it can be assumed that the panels are not receiving enough light. This will prevent false positives if for any reason the demand for the compressor dips or stops.)
-Voltage sense at the 12V input for the inverters. When tripped will turn the AC unit ON. (If the inverters are off, then the unloaded voltage from the panels should be able to provide ample information for the amount of light they are receiving. (Voltage from all 3 inverters will need to be looked at)
-Timer/Control board. This could be a 555 timer, a couple comparators and some misc parts. Or this could be a microcontroller. Microcontroller method will allow for more complex on/off programming methods for timing things, or comparing multiple panels. The trade-off is, if the circuit fails, a new microcontroller will need to be programmed. Building it with a 555 and other components will allow other technicians to rebuild/repair/understand the circuit better just by looking at it.
(Are you certain that the voltage from the panels is only 12V...
The anti-island feature may make this a little more complicated, as we can't just redirect the inverter output to a set of lights unless there is grid power on the set of lights as well... This also means that once connected to lights, or fans etc., you run the risk of back-feeding power from the inverter back to the grid which I assume is not allowed?
I think that attempting to use the reduced power from the panels with a cloud overhead will make this too complicated.
So.. my thoughts for a solution is:
-Clamp on meter for the 'grid' power-line into the air conditioner. When tripped will turn the AC unit OFF. (This will be read, and if the reading is too high, then it can be assumed that the panels are not receiving enough light. This will prevent false positives if for any reason the demand for the compressor dips or stops.)
-Voltage sense at the 12V input for the inverters. When tripped will turn the AC unit ON. (If the inverters are off, then the unloaded voltage from the panels should be able to provide ample information for the amount of light they are receiving. (Voltage from all 3 inverters will need to be looked at)
-Timer/Control board. This could be a 555 timer, a couple comparators and some misc parts. Or this could be a microcontroller. Microcontroller method will allow for more complex on/off programming methods for timing things, or comparing multiple panels. The trade-off is, if the circuit fails, a new microcontroller will need to be programmed. Building it with a 555 and other components will allow other technicians to rebuild/repair/understand the circuit better just by looking at it.
(Are you certain that the voltage from the panels is only 12V...
i need the ability to
Ah, but in my proposed layout i'm not redirecting the inverter Output (after anit-islanding) i am redirecting the inverter Input (before the anti-islanding), i used the green R1 R2 & R3 to show the location for (+) on the three inverter circuits and i proposed to run 12V DC motors when the compressor was no longer in the circuit. i also assumed we could hard wire the (--) to the fans and and let the (+) take care of completing the circuit. i did it this way to keep the circuit under load assuming it had to be to measure the power.
I see what your doing and i like it. i had no idea we could measure unloaded voltage. your method is a more idealistic approach for me as my decision to run the AC will be based directly what i'm paying (using from the grid). for others who may be following the thread this will also take care of turning on and off the AC unit every morning and evening at the optimal time (based on cost). currently i do it manually and if i'm not around then i have to decide to leave it on or off - that decision making goes away if we get this functioning.
you asked '(Are you certain that the voltage from the panels is only 12V..' No - it's not always 12v. i think it may be a range. i will fluke the the inverter input today and get several real world readings. the solar panels are wired in parallel so i've increased the amps not the volts. my best guess right at this moment (and i just started my first cup of coffee) is that it will come in at approx. 29.8 +/-
as to volts the panel stickers state: open circuit voltage Voc[v] 37.1 rated voltage Vmpp[V] 29.8
Max System Voltage Vmax[V] 600 V
i hope your idea whereby we trip the circuit back on when the open voltage is 'x' can work - that will simplify things considerably.
i never knew electric could be fun.
i just learned last night that the utility company is raising the rate by 50% thats going to $.75 per kilowatt hour. argg........ the cost of comfortable just went up.
The voltage across the solar panel will be at it's highest when it is open circuit (or not drawing any current)
If we redirect the lower voltage panels to fans or lights, this will keep them loaded, but you will need to contend with a voltage that is not constant. The more you draw, the lower the voltage will dip.
Earlier on, the idea to monitor the actual panel voltage was discouraged due to the potential high voltage. (Note the 600V system max voltage...) These systems can be connected in series or parallel, and if it's connected in series the voltages can be difficult / dangerous to work with. If most of your panels are in parallel, then the voltage should be at a low enough level to more safely work with.
When you fluke the panel input, please be careful. Try to avoid holding a probe in each hand, if the voltage arcs from the probe tip, or if the probe insulation is poor then the electricity will go across your chest which is a bad thing
If we redirect the lower voltage panels to fans or lights, this will keep them loaded, but you will need to contend with a voltage that is not constant. The more you draw, the lower the voltage will dip.
Earlier on, the idea to monitor the actual panel voltage was discouraged due to the potential high voltage. (Note the 600V system max voltage...) These systems can be connected in series or parallel, and if it's connected in series the voltages can be difficult / dangerous to work with. If most of your panels are in parallel, then the voltage should be at a low enough level to more safely work with.
When you fluke the panel input, please be careful. Try to avoid holding a probe in each hand, if the voltage arcs from the probe tip, or if the probe insulation is poor then the electricity will go across your chest which is a bad thing
Last edited by a moderator:
