Inverter with Limiter

[oldsalty]

Thanks for the reply, That makes sense. and YES, the inverter is wired DIRECT to the breaker box on its own dedicated breaker. The current clamps (2, 1 for each phase)) are wire in series, I am told to install the clamps with 1 arrow facing the breakers and the other facing away from the breakers. Is this correct?

 

[oldsalty]

Ampster, I had to put the CT clamps around L1 and L2 60 amp breaker that feeds the sub panel in my garage since there to small to fit around the ones coming from the pole, However, i'm not sure which way the clamps should go now. These 2 clamps are connected in series. If the clamps arrows are facing the same way I was getting a reading of 600 watts ( solar reading was 60 watts since the sun was going down) on the display. If one is facing in and one is facing out the display showed 1200 to 1300 watts. Which direction should they be facing?

 

[Ampster]

If one is facing in and one is facing out the display showed 1200 to 1300 watts. Which direction should they be facing?

You will need some way to verify. It doesn't matter as much which way they are facing since you can switch the wires at the input. I think they add the two shunts if you have them in series but the wires should have different colors so if they are black and white then one of each color to the inverter and one of each color connected together between the shunts, if that makes sense. They should be facing the same way.

The best way to be sure is to calibrate the reading from the inverter with a known reading from your meter. I am not trying to be vague but up could be down when it comes to the arrows on your shunts. If it is a sub panel then you may need to turn on the biggest load on that sub panel to be sure. As long as your inverter is feeding into that same sub panel that is all you care about.
If you have to leave them one way for a while to test them and a little power leaks back to the grid don't worry. It is all in the name of science.

 

[GXMnow]

I found a few listings on Ali Express for similar inverters. Probably all made in the same factory. They all look to be just single phase, either 120 or 230 volt. I do not see any marked as "Split Phase" for US 120/240 two leg power.

From the description of the CT clamps, I would expect they are basic current ratio transformers. This works fine for a single phase system, just putting one on the hot line. Running two of them in series will not add the two current values properly. You would basically get an average of the two currents, or even the lower of the currents, depending on the circuit reading the CT transformer. That may be good enough though. But I have to wonder how you plan to feed it to both legs. Even the 230 volt version is a neutral and a single hot line out. If you connect it across the 240 of our split phase power, will it tolerate the 120 volts on it's neutral wire? Be careful to make sure this is safe.

Once you do have it connected, if the inverter is in limiting all the time, then the CT's are wired backwards. Since the 2 phases are in opposite directions, the 2 clamps should be in opposite directions or wired opposite so the values add. If both legs are pulling 10 amps, and you wire them the wrong way, the reading would zero out. Wired the correct way, the current should add up and show 20 amps. As the inverter pushes power, the consumption shown on the CT's should drop as the inverter is powering the loads instead of the grid. If the inverter makes more power than the load, the CT's start to read negative, and the inverter reduces it's output to try and keep it at zero current. This does stop the back feed to the grid, but it is basically the same as shading your panels and wasting energy you could be making.

 

[Ampster]

This does stop the back feed to the grid, but it is basically the same as shading your panels and wasting energy you could be making.

I understand what you are saying but it is more convenient than shading the panels every time a load decreases. It is also less costly than a GT inverter and the process to get permission from his utility which to OP said he did not want to do.

 

[ianishomer]

Can I ask a related question.

I am currently looking components for setting up a system, but I need a grid tie inverter with a limiter, utility does not do net metering and will charge me for excess energy that I put into the grid , I have seen inverters with limiters advertised, but I have been offered an unused Hosola 3000 T smart inverter, which doesnt come with a limiter.

My question is, can you add a limiter to any inverter or do they have to be supplied with one?

Sorry if this a stupid question

 

[Ampster]

I do not know of any third party limiters per se. There are dump loads but i do not know of many that are variable. You would have to look at the specs of the manufacturer to see if it can be fitted with an external current sensor and if it has the software to measure that current sensor. The term used to describe that process of limiting current is "non export".

 

[GXMnow]

Unless the inverter has some form of current control in it's design, it would not be an easy task to add current limiting. Most "Grid tie" inverters are designed around converting as much solar energy as possible and pushing it into your breaker panel. Now that more places are setting up rules for back feed power, current limiting is being added in newer designs, and the prices on the older ones are falling.

Any external controller that would have the ability to add a limiting function to a grid tie inverter would likely cost more than the cheap limiting inverters that are showing up online now.

While thinking about this, I did come up with one relatively cheap way of doing it. First off you would need to have a bidirectional power meter at the main grid feed. I have found a few for about $120 US that can do 120/240 3 wire split phase. There are cheaper ones that can do a single phase 2 wire feed like Europe 230 volt. Then have a small logic controller read the power meter. If the power going in to the home falls below 20 watts or so, have it turn on a relay. That relay will connect a heating element to the DC from the solar panels which will pull down the voltage and the inverter won't be able to make as much power. If that heating element can be part of the hot water system, then you are not really wasting the power. With more programming, you could do it in steps. Have a 100 watt, 200 watt, 400 watt etc. If it turns on the first and the power starts to export again, turn on another. With just those 3, you can get any 100 watt watt step up to 700 watts. Figure out what your lowest draw might be near solar noon, subtract that from the maximum the solar panels can make, and you need to be able to push that power into heaters.

Of course, it still may be cheaper to just get an inverter with limiting built in.

 

[fafrd]

Ampster, I had to put the CT clamps around L1 and L2 60 amp breaker that feeds the sub panel in my garage since there to small to fit around the ones coming from the pole, However, i'm not sure which way the clamps should go now. These 2 clamps are connected in series. If the clamps arrows are facing the same way I was getting a reading of 600 watts ( solar reading was 60 watts since the sun was going down) on the display. If one is facing in and one is facing out the display showed 1200 to 1300 watts. Which direction should they be facing?

Oldsalty,

I’m thinking about following in your footsteps.
Are you happy with the quality / performance of the GTIL inverter you got?

Can you share the vendor you purchased from and were you happy with them? There are so many on AliExpress that a recommendation is worth a lot...

Thanks.

 

[fafrd]

I found a few listings on Ali Express for similar inverters. Probably all made in the same factory. They all look to be just single phase, either 120 or 230 volt. I do not see any marked as "Split Phase" for US 120/240 two leg power.

From the description of the CT clamps, I would expect they are basic current ratio transformers. This works fine for a single phase system, just putting one on the hot line. Running two of them in series will not add the two current values properly. You would basically get an average of the two currents, or even the lower of the currents, depending on the circuit reading the CT transformer. That may be good enough though. But I have to wonder how you plan to feed it to both legs. Even the 230 volt version is a neutral and a single hot line out. If you connect it across the 240 of our split phase power, will it tolerate the 120 volts on it's neutral wire? Be careful to make sure this is safe.

Once you do have it connected, if the inverter is in limiting all the time, then the CT's are wired backwards. Since the 2 phases are in opposite directions, the 2 clamps should be in opposite directions or wired opposite so the values add. If both legs are pulling 10 amps, and you wire them the wrong way, the reading would zero out. Wired the correct way, the current should add up and show 20 amps. As the inverter pushes power, the consumption shown on the CT's should drop as the inverter is powering the loads instead of the grid. If the inverter makes more power than the load, the CT's start to read negative, and the inverter reduces it's output to try and keep it at zero current. This does stop the back feed to the grid, but it is basically the same as shading your panels and wasting energy you could be making.

I’m considering using a couple of these for easily-wired time-shifting.

My thoughts are do not try to use a single inverter with two limit sensors, but rather, use two inverters wires in parallel, each with it’s own limited sensor (one on L1 and one on L2 from 240V utility feed).

While my 240V microinveryer-based Solar is producing more than consumption, both inverters should be off.

Once sun has gone down and consumption exceeds remaining production, battery-powered inverters should kick in to produce enough to compensate for net consumption so smart meter reads net of zero. And once battery is drained, inverters shut down and consumption reverts to grid.

Thought?

The only issue I am struggling with is whether I also need a pair of autoinvertets or not. With an autoinverter for each of 2 240V inverters, I can basically create a split-phase inverter which will convert the 240V L1 + L2 consumption to a split-phase 120V + 120V + Neutral feed that should offset all consumption.

These inverters claim to be 120V / 240V with ‘autodetect’ but I am skeptical and have seen others stare that they are 240V.

I’m hoping to find a way to avoid needing to feed into existing wiring (and avoid requiring a sub panel) while allowing a battery charged from few additional DC-coupled panels to supply enough juice to cover peak demand from 6pm to 11pm after the sun has gone down.

Any thoughts or advice appreciated.

 

[Ampster]

Do you mean autotransformer? I do not know what an autoinvertet is.
This seems complicated compared to a split phase inverter that can be programmed for no export.

 

[fafrd]

Do you mean autotransformer? I do not know what an autoinvertet is.
This seems complicated compared to a split phase inverter that can be programmed for no export.

Yeah, autitransformer, sorry for the typo / brainfreeze.

I have not yet seen a split-phase inverter that does not need to be wired in between main panel and critical loads subpanel (Conext, Radian, Magnum, etc...).

I am looking for a grid-tie inverter that supports an external limiter sensor and will run off of 24V or 48V battery power to feed offsetting split-phase power into the main panel through a pair of circuit breakers.

Any recommendations for how to do this as easily and inecoensively as possible greatly appreciated.

 

[Ampster]

I have not yet seen a split-phase inverter that does not need to be wired in between main panel and critical loads subpanel (Conext, Radian, Magnum, etc...).
I am looking for a grid-tie inverter that supports an external limiter sensor and will run off of 24V or 48V battery power to feed offsetting split-phase power into the main panel through a pair of circuit breakers.

The first part is easy using a GT inverter with a no export program.
The second part, to power your electrical panel from batteries is where it
gets cumbersome or expensive because you will need some way to isolate your electrical panel from the grid. This could be as simple and inexpensive as a manual interlock. That would be a switch that you would
change every evening when you want to go off grid and run on batteries. In the morning when the sun came up you would connect up to the grid again. You could put that switch on a timer but that is the least expensive solution. The only option that I am aware of is a hybrid inverter like the ones you mentioned that would require a separate loads panel to be wired through the automatic transfer switch of that inverter.

Have you done an analysis to understand the loads that you will need to support during the evening?

 

[fafrd]

The first part is easy using a GT inverter with a no export program.
The second part, to power your electrical panel from batteries is where it
gets cumbersome or expensive because you will need some way to isolate your electrical panel from the grid. This could be as simple and inexpensive as a manual interlock. That would be a switch that you would
change every evening when you want to go off grid and run on batteries. In the morning when the sun came up you would connect up to the grid again. You could put that switch on a timer but that is the least expensive solution. The only option that I am aware of is a hybrid inverter like the ones you mentioned that would require a separate loads panel to be wired through the automatic transfer switch of that inverter.

Have you done an analysis to understand the loads that you will need to support during the evening?

I’ve done all the analysis and have everything figured out using the ‘prime-time’ solution of inserting a hybrid inverter between the main panel and the critical loads subpanel. For home backup, that is the only alternative but I am primarily interested in time-shifting, not backup and am hoping to find a way to avoid any rewiring.

So I need a grid-tied inverter with an external clamp sensor that only generates any power when the net consumption from the grid is positive.

During the day when the 4kW mucroinverter-based PV is exceeding consumption, battery inverter stays shut off (and smaller DC-coupled PV array charges 24V LiFePO4 battery).

During ‘peak’ period in the evening once consumption exceeds AC solar production, battery inverter kicks-in and generates sufficient power to offset consumption so that net meter reads zero consumption.
Late in the night once fridges have drained the battery to cut-off, battery inverter shuts down and consumption reverts to grid power.

If I can make this work, it is easy-peazy because the incremental wiring is trivial (2 breakers to feed power into the main panel from the battery inverter) and the component cost is also reasonable ($100s rather than $1000s).

I just need to make sure that these limiter grid-ties inverters shut down when ‘consumption’ is negative (AC-coupled generation exceeds consumption) and figure out how to translate two legs of consumption into split-phase output suitable for being fed back into the mains.

If I can put a sensor on L1 for L1-Neutral 120V output and a second sensor on L2 for L2-Neutral 120V output and feed that L1-N-L2 feed directly into the main panel through a couple breakers, I’m good to go, but I’m not understanding whether neutrals can be joined in that way from two inverters and it will all work out and it just seems safer to couple two 240V inverters into a dual-pole Autotransformer like Victron makes (with L1-L2 240V input and L1-N-L2 output).

Any thoughts or ideas appreciated (especially any split-phase inverters with external rather than internal sensors).

 

[GXMnow]

What I think you are tying to do is certainly not cheap, but it is doable. But if the plan it so save money, I don't think it will pay off.

If I was going to build from scratch, I could connect my 16 x 300 watt panels with a solar charge controller to my battery bank. Then setup my Schneider XW-Pro as just an inverter with both "Load Shave" and "Grid Sell" modes turned on with the proper time ranges set. Even in my current system with my solar AC coupled, the XW-Pro is able to "Sell" power back to my main panel and power my entire home through my 5 hour peak rate evening time. The problem is making it then charge the batteries when the sun is up again. Using the DC coupled solar does fix that issue. I do not have true grid current limiting at this point, but it can be added with a WattNode module connected to the Schneider Conext Gateway. For now I just have it programmed to export 1,400 watts, and that ends up still pushing about 300-600 watts out to my grid, unless the A/C kicks on, then I am buying 1,500 watts or so. The Watt Node could be adjusted to ramp the current to keep the grid current zeroed with changing load, but it's not that big of a deal. The extra savings available is really not that much.

Before I bought my XW-Pro, I did look into trying to do it on the cheap with some chargers and grid tie inverters off Amazon or AliExpess etc. But between the shoddy quality, not having the large autotransformer internal, poor efficiency, and questionable programming interface, I figured the extra spent on the Schneider was well spent. The idea of pumping thousands of watts and KWH's through a cheap box was not worth it. But I did run into the stupid software issue on the XW-Pro which I have still not worked around. I am getting closer, and hope to have it working before I need to run A/C again. Since I am using all AC coupled microinverters for my solar, the XW-Pro can't automatically switch between grid tied discharging and battery charging. The "Grid Support" will shut down at 0.5 volts higher than the battery recharge voltage, so it just won't start charging again the next day, since the battery is above the recharge volts. It is such a dumb programming error, but Schneider does not see it as a problem. They just say to use DC coupled solar with their way over priced charge controllers.

When I was thinking about the cheap AliExpress solution, I knew I was going to have to program my own controller to handle the power time shifting. I did not think I was going to need that with the Schneider, but here I am programming a small PLC controller that can send the Modbus TCP commands to the Schneider system.

 

[Ampster]

For home backup, that is the only alternative but I am primarily interested in time-shifting, not backup and am hoping to find a way to avoid any rewiring.

Operationally backup and timeshifting are the same and both require a means to switch power. If you are trying to avoid a critical loads panel then you would need a hybrid inverter that can handle the rating of your main panel which is presumably 100 Amps. That is 24kW. Have you priced a 24kW hybrid inverter compared to a 6kW hybrid with a 60 Amp critical loads panel? There may be a way to derate your main panel. then you might be able to get by with a hybrid inverter smaller than 24kW. You seem more concerned with finding external CT sensors than understanding how to wire a hybrid inverter to allow it to load shift and operate automatically like you want it to. With a separate loads panel you don't need external CTs. The cost of materials for a separate panel is less than $200. Are you doing this yourself?

 

[fafrd]

What I think you are tying to do is certainly not cheap, but it is doable. But if the plan it so save money, I don't think it will pay off.

If I was going to build from scratch, I could connect my 16 x 300 watt panels with a solar charge controller to my battery bank. Then setup my Schneider XW-Pro as just an inverter with both "Load Shave" and "Grid Sell" modes turned on with the proper time ranges set. Even in my current system with my solar AC coupled, the XW-Pro is able to "Sell" power back to my main panel and power my entire home through my 5 hour peak rate evening time. The problem is making it then charge the batteries when the sun is up again. Using the DC coupled solar does fix that issue. I do not have true grid current limiting at this point, but it can be added with a WattNode module connected to the Schneider Conext Gateway. For now I just have it programmed to export 1,400 watts, and that ends up still pushing about 300-600 watts out to my grid, unless the A/C kicks on, then I am buying 1,500 watts or so. The Watt Node could be adjusted to ramp the current to keep the grid current zeroed with changing load, but it's not that big of a deal. The extra savings available is really not that much.

Before I bought my XW-Pro, I did look into trying to do it on the cheap with some chargers and grid tie inverters off Amazon or AliExpess etc. But between the shoddy quality, not having the large autotransformer internal, poor efficiency, and questionable programming interface, I figured the extra spent on the Schneider was well spent. The idea of pumping thousands of watts and KWH's through a cheap box was not worth it. But I did run into the stupid software issue on the XW-Pro which I have still not worked around. I am getting closer, and hope to have it working before I need to run A/C again. Since I am using all AC coupled microinverters for my solar, the XW-Pro can't automatically switch between grid tied discharging and battery charging. The "Grid Support" will shut down at 0.5 volts higher than the battery recharge voltage, so it just won't start charging again the next day, since the battery is above the recharge volts. It is such a dumb programming error, but Schneider does not see it as a problem. They just say to use DC coupled solar with their way over priced charge controllers.

When I was thinking about the cheap AliExpress solution, I knew I was going to have to program my own controller to handle the power time shifting. I did not think I was going to need that with the Schneider, but here I am programming a small PLC controller that can send the Modbus TCP commands to the Schneider system.

Our setups sound very similar and I have an entire setup architected around either the Conext or the Magnum PAE. The reason I am hesitating is not so much the cost as it is the hassle of requiring my panels.

Obviously if the cheap Chinese inverters crap out, that’s an option that will need to be abandoned, but it’s relatively easy and inexpensive to try.

2 240V 1000W inverters for $300 each and an Autotransformer costing another $100-150 (and UL, so low-risk).

I’m not trying to time-shift my primary solar output, just the 2kW output from the new smaller DC-coupled array I am adding.

With the change in peak window, my solar generation will be valued at ~67% of what it was for the most important 4 months a year and I’m going to be charged ~200% for most of the grid energy I consume. I’ve estimated that the value of covering my peak consumption from a 6kWh battery charged by a new DC-coupled 2kW array will save me close to $250/year once the new rates set in.

The 280Ah battery has already been purchased so that’s now a sunk cost.

The 2kW in new solar was planned anyway to cover the extra consumption associated with a planned-for EV.

So rather than expanding the AC-coupled array, adding a new DC-coupled array and finding a way for a battery-powered grid-tied inverter to cover peak consumption from 6pm to 11pm seems like the smarter thing to do (especially if I can find a way to keep the additional wiring as simple as it is with microinverters).

If you know of any UL-Listed battery-powered inverters supporting external clamp sensors for current-limiting, that would be interesting - so far, these Chinese inverters are all that I’ve found.

 

[fafrd]

Operationally backup and timeshifting are the same and both require a means to switch power. If you are trying to avoid a critical loads panel then you would need a hybrid inverter that can handle the rating of your main panel which is presumably 100 Amps. That is 24kW. Have you priced a 24kW hybrid inverter compared to a 6kW hybrid with a 60 Amp critical loads panel? There may be a way to derate your main panel. then you might be able to get by with a hybrid inverter smaller than 24kW. You seem more concerned with finding external CT sensors than understanding how to wire a hybrid inverter to allow it to load shift and operate automatically like you want it to. With a separate loads panel you don't need external CTs. The cost of materials for a separate panel is less than $200. Are you doing this yourself?

Undrrstand everything you have stated but don’t think it is completely correct.

Backup needs switching off of the grid but time-shifting does not.

In general, switches are needed to switch between grid power and off-grid power, but not in the case of these external sensor limited inverters. When consumption is zero (or hopefully also when it is negative as in the case of my AC-coupled PV generation exceeding consumption), battery-powered inverter output is zero, so from the grids point of view, those inverters have been switched off.

When net consumption is positive (such as will be the case when my PV output tapers off in the evening and consumption ramps up with preparation of dinner, the battery-powered inverter senses that consumption and generates an equal and offsetting amount of power. So from the grids point of view, consumpion by my house has changed from decreasingly negative to zero.

Once the battery is drained and the battery-powered inverter(s) shut(s) off, continued home consumption is supplied by the grid and becomes ‘visible’ to the smart meter.

When the sun comes up in the morning, DC-coupled array begjns to charge battery enabling battery-powered inverter to turn on again, but AC-coupled PV array exceeds consumption to battery-powered inverter stays off.

The only issue could be on cloudy days when consumption exceeds generation, battery energy could be consumed off-peak while it would be better to use cheap grid energy and conserve battery energy for peak window in the evening, but that’s easily solved on way of the other by a timer if it becomes frequent enough to be justified.

A backup system has to have sufficient power output to power the entire house, but a time-shift system does not. My fridges consume an average of 0.4kW/h and my peak-window consumption averages 2kW/h with occasional spikes to 2.5kWh/h.

Of all loads I have, the electric oven is the heaviest, consuming as much as 3kW itself.

If I only have 2kW of battery-powered time-shift generation, I will not be offsetting ALL of my peak co sumption, but merely most of it. When I use the oven, they grid may still need to supply ~1kW of the power, for example.

If it can be done, the wiring is much simpler than having to reconfigure panels (not to mention redundancy - an inverter failure just means no time-shift, but the grid is still there...),

 

[Ampster]

Undrrstand everything you have stated but don’t think it is completely correct.

Backup needs switching off of the grid but time-shifting does not.

As far as a UL certified Inverter is concerned my statement is correct. I was assuming a hybrid inverter that was capable of running off grid as well as in grid tie mode.

However, I looked into the GTIL inverters discussed and I presume they are GT inverters that just connect to one leg or phase of the 240 volt panel. They are not to code but if that is not a concern go for it. They need the grid signal to work, so I presume they would not work if the grid was down. If you connect one inverter to L1 and another to L2 I also assume they will sync up to their respective phases. I wouldn't even know why you would need an autotransformer but I am making assumptions based on a complex interrelationships and very little knowledge of those inverters. You mention AC coupling but I am not sure where AC coupling fits in here because these do not sound like they could provide an AC signal to run your micro inverters when the grid is down. Perhaps all you mean is that they sync to the grid. They can produce up to a 1000 Watts based on the CT feedback so there is no export. @GXMnow has done more research than I have so his feedback is more valuable than mine.

 

[fafrd]

As far as a UL certified Inverter is concerned my statement is correct. I was assuming a hybrid inverter that was capable of running off grid as well as in grid tie mode.

However, I looked into the GTIL inverters discussed and I presume they are GT inverters that just connect to one leg or phase of the 240 volt panel. They are not to code but if that is not a concern go for it. They need the grid signal to work, so I presume they would not work if the grid was down. If you connect one inverter to L1 and another to L2 I also assume they will sync up to their respective phases. I wouldn't even know why you would need an autotransformer but I am making assumptions based on a complex interrelationships and very little knowledge of those inverters. You mention AC coupling but I am not sure where AC coupling fits in here because these do not sound like they could provide an AC signal to run your micro inverters when the grid is down. Perhaps all you mean is that they sync to the grid. They can produce up to a 1000 Watts based on the CT feedback so there is no export. @GXMnow has done more research than I have so his feedback is more valuable than mine.

Yes, not to code, and sounds like you now understand as much as I do.

I only mentioned AC-coupling to distinguish my new planned-for 2kW DC-coupled PV array from my existing 4kW Microinverters-based PV array, which is to code, on net metering, and which I don’t want to touch.

These GTIL zero-export inverters need grid sync and will not function when the grid is down, so backup means using a separate battery-powered PSW inverter (not automatic, which is fine for my priorities).

The only thing I am unclear about is whether two can be connected with common-neutral so one is synced to L1 and the other is synced to L2 and they can each feed in through their own breaker or there is some issue with that (which is where an Autotransformer may be a solution).

I’ll reach out to GXMnow - thanks.