SRNE Split Phase Hybrid Solar lnverter ASF Series 10Kw

[natehawkv8]

I purchased the SRNE 10Kw ASF48100U200-H inverter from Alibaba and have been using it for 80 days.

Split Phase Residential Storage lnverter 8KW-10KW - SRNE Solar

ASF series All-in-one Solar Charger Inverter Rated Output Power 8,000W/10,000W Rated Output Voltage 120/240Vac Rated Battery Voltage 48Vdc

www.srnesolar.com

So far no problems. I made a quick video:

It was looking for a single all-in-one that was able to do split phase and 10kw or more. I also wanted MPPT that could accept high power arrays. This model was the best fit I could find. Each MPPT in this unit can accept up to 5500 watts of solar and up to 22amps. MPPT voltage range is 125-425 with a max VOC of 500 Vdc.

I wish there was an option to parallel it to a second unit in case I add more arrays, but there is no option for that.

 

[Wakintx]

Add more solar arrays? Add additional panels through a new mppt connected to the battery Bank...?

 

[Pale1290]

Hello, I have a problem srne 8kw I put the batteries in L16 (lithium) but setting #4 does not let me change. They are not connected to the communication cable. I can change the other parameters except #4. Has anyone of you had this problem and if so, how could you solve it? Thanks in advance.Hello, I have a problem srne 8kw I put the batteries in L16 (lithium) but setting #4 does not let me change. They are not connected to the communication cable. I can change the other parameters except #4. Has anyone of you had this problem and if so, how could you solve it? Thanks in advance.

 

[natehawkv8]

Hello, I have a problem srne 8kw I put the batteries in L16 (lithium) but setting #4 does not let me change. They are not connected to the communication cable. I can change the other parameters except #4. Has anyone of you had this problem and if so, how could you solve it? Thanks in advance.Hello, I have a problem srne 8kw I put the batteries in L16 (lithium) but setting #4 does not let me change. They are not connected to the communication cable. I can change the other parameters except #4. Has anyone of you had this problem and if so, how could you solve it? Thanks in advance.

Make sure setting 15 is lower then setting 12, and that setting 12 is lower than setting 4. Setting 4 must be lower than setting 14. If that's not the case, you may not be able to adjust setting 4.

 

[Pale1290]

Make sure setting 15 is lower then setting 12, and that setting 12 is lower than setting 4. Setting 4 must be lower than setting 14. If that's not the case, you may not be able to adjust setting 4.

Thanks

 

[Pale1290]

Make sure setting 15 is lower then setting 12, and that setting 12 is lower than setting 4. Setting 4 must be lower than setting 14. If that's not the case, you may not be able to adjust setting 4.

Thank you. i did it and it worked

 

[Cheap 4-life]

Can this inverter supply grid tie power and utilize external CTs for zero export?

 

[fafrd]

Can this inverter supply grid tie power and utilize external CTs for zero export?

I don’t believe so. Here is the manual: https://www.srnesolar.com/wp-conten...ase_solar-charger-inverter_usermanual_1.4.pdf

For the OP (natehawkv8) or other owners, a question:

The ‘bypass overload current’ is spec’d at 63A - does this mean that the inverter includes a split-phase transfer switch rated for 63A / 15kW?

‘Switch time’ is specified as ‘10ms (typical)’ which I assume means there is a transfer switch that disconnects from grid in 10ms when there is a grid outage, but the manual is sparse on details.

Since the Midnight Solar DIY series of HF inverters are manufactured by SRNE, I suspect it’s just a matter of time before we see split-phase 8kW and 10kW offerings under the Midnight Solar DIY brand (along with improved documentation), but in the meantime I’m interested in any feedback on the quality and performance of these SRNE split-phase AIOs…

 

[QBear]

I recently purchased this all-in-one system to replace my existing setup. It has a capacity of up to 10kW output (solar + battery), but I need to ensure that the output of each leg does not exceed 5kW, as it could cause the system to trip. Generally, I'm satisfied with this system considering its price range (I paid $1650). However, there is one drawback: the inverter consumes a significant amount of grid power (45-50W) when running on battery, regardless of the battery state of charge. On the other hand, when solar power is available, the grid consumption drops to only 5W, even if the solar power alone isn't sufficient to cover the entire inverter output.

Additionally, I discovered that this inverter lacks a manual switch for switching between utility and battery power. However, I later found out that I could use Modbus via Home Assistant to change the power supply priority. I successfully tested this method by switching a 3-4k load without encountering any issues.

Furthermore, I noticed a peculiar design aspect regarding the airflow in the system. Most systems utilize an upward airflow design, which allows the system to passively cool itself using natural thermal updraft when the cooling fan is not running. However, this particular machine has two air inlets on the sides and an exhaust vent at the bottom, which struck me as rather unusual.

 

[fafrd]

I recently purchased this all-in-one system to replace my existing setup. It has a capacity of up to 10kW output (solar + battery), but I need to ensure that the output of each leg does not exceed 5kW, as it could cause the system to trip.

OK, so it is a maximum of 5kW or 41.7A of inverter output per leg, but is it able to pass through as much as 63A or 7.5kW per leg of grid power?

Generally, I'm satisfied with this system considering its price range (I paid $1650). However, there is one drawback: the inverter consumes a significant amount of grid power (45-50W) when running on battery, regardless of the battery state of charge.

Meaning grid-consumption cannot be less than 50W or that that 50W is wasted / lost when running in battery priority while grid is present?

I’m considering to use an external transfer switch to disconnect the grid from the inverter AC input overnight to avoid any sprigs export when the sun is not shining.

I’m hoping your observation means battery consumption increases by ~50W when grid is disconnected (so it’s just a 50W minimum grid consumption level when grid is present).

On the other hand, when solar power is available, the grid consumption drops to only 5W, even if the solar power alone isn't sufficient to cover the entire inverter output.

OK, so 5W minimum grid consumption during daylight hours when grid is present and 50W minimum grid consumption at night…

Additionally, I discovered that this inverter lacks a manual switch for switching between utility and battery power. However, I later found out that I could use Modbus via Home Assistant to change the power supply priority. I successfully tested this method by switching a 3-4k load without encountering any issues.

Fantastic. If only there were a way to force it to disconnect from grid through Modbus commands…

Furthermore, I noticed a peculiar design aspect regarding the airflow in the system. Most systems utilize an upward airflow design, which allows the system to passively cool itself using natural thermal updraft when the cooling fan is not running. However, this particular machine has two air inlets on the sides and an exhaust vent at the bottom, which struck me as rather unusual.

What output levels generally cause the fans to turn on? Are they loud?

Are you generally happy with the build quality?

Do you believe the claimed efficiency numbers are roughly accurate?

 

[QBear]

OK, so it is a maximum of 5kW or 41.7A of inverter output per leg, but is it able to pass through as much as 63A or 7.5kW per leg of grid power?

I have never tested such a large load on this machine, even though it is designed to handle pass-through loads. I understand that the manufacturer labeled it as 63A because they installed a 63Amp input breaker. However, upon closer inspection, I noticed that the wire for the AC input and output is of a smaller gauge than 6 AWG. You can see from the pictures that the incoming wire in the AFC is 6 AWG, but the wire on the terminal block is only 8 or smaller. Based on my assessment, I would not recommend using it for a long period of time with a 63Amp pass-through load, as it would likely cause the wire to overheat.

Meaning grid-consumption cannot be less than 50W or that that 50W is wasted / lost when running in battery priority while grid is present?

I’m considering to use an external transfer switch to disconnect the grid from the inverter AC input overnight to avoid any sprigs export when the sun is not shining.

I’m hoping your observation means battery consumption increases by ~50W when grid is disconnected (so it’s just a 50W minimum grid consumption level when grid is present).

I am unsure whether the power drawn from the grid, approximately 45-50W, is being wasted or used for the inverter output. At times, I manually switch off the AC input breaker to completely eliminate grid consumption, resulting in 0W usage. In an interesting test, I utilized Home Automation (HA) to send a changeover Modbus command to switch the inverter to grid power while keeping the AC input breaker in the off position. However, the inverter did not change to grid power as expected. In HA, I have set up an automation that triggers this switch based on the State of Charge (SoC) of my Battery Management System (BMS). When the SoC falls below 15%, the Modbus command is sent to the inverter to switch over to grid power.
Nevertheless, if I turn off the AC input to avoid utilizing the 50W grid power, I will lose power entirely once the battery is fully drained. As for the actual utilization of the 40-50W grid consumption by the inverter, I have yet to investigate this matter. I plan to delve into it further tonight.

OK, so 5W minimum grid consumption during daylight hours when grid is present and 50W minimum grid consumption at night…

Yes.

Fantastic. If only there were a way to force it to disconnect from grid through Modbus commands…

To eliminate grid consumption and forfeit the grid backup, I can only manually switch off the AC input.

What output levels generally cause the fans to turn on? Are they loud?

There are four fans connected in a way that they all run simultaneously and at the same speed. Two fans are dedicated to the inverter compartment, while the other two are for the Solar Charger compartment. Each compartment is equipped with a temperature sensor. If either of these two sensors detects a temperature higher than 45 degrees Celsius, all four fans will start running at low speed. Once the temperature decreases to 41.8 degrees Celsius, all fans will switch off.

If the temperature exceeds 50 degrees Celsius, the fans will continue running at a higher speed. However, it is unclear whether these fans have two or three speed settings. Generally, when the inverter output or the PV (photovoltaic) system operates at high power levels, the fans can be quite loud.

Are you generally happy with the build quality?

This is a typical Chinese-made item with a corresponding quality level. It follows the principle of "you get what you pay for." To ensure a longer lifespan for the unit, it is advisable to operate it at around 80% of its maximum capacity. Based on the wire size and overall construction, it seems that the unit is somewhat undersized for its claimed capabilities. The wiring compartment and the wiring block are both quite small, which hampers airflow and may lead to increased heat. Additionally, the unit causes frequent LED light flickering. However, apart from these issues, it performs well overall. It can effectively charge my Tesla Model 3 at 240V with a 32Amp current, provided that I do not use high-power appliances simultaneously while charging at the maximum 32Amp rate.

Do you believe the claimed efficiency numbers are roughly accurate?

The efficiency claim of the product aligns closely with the manufacturer's specifications. Through my recordings, I have observed an efficiency range of 89% to 92% when converting power from the battery to AC output. However, I am unsure about the best method to test the efficiency of the MPPT

 

[natehawkv8]

OK, so it is a maximum of 5kW or 41.7A of inverter output per leg, but is it able to pass through as much as 63A or 7.5kW per leg of grid power?

I'm 95% sure even in bypass to grid power its still only 5kW/41.7A per leg. I've tripped it while on grid power before and pretty certain neither phase was using above 5kW although I wasn't watching it real-time. Also, I use Solar Assistant sometimes to switch between Utility output, and SBU and that works fine. As for quality and performance I can say it's been reliable so far. I have about 9,500 kWh of PV through it so far (6 months). I wish SRNE would add parallel function so I could add a second one and get more than 10kW output. There is a setting to set phase mode to 0 or 180 but I believe the software would have to support it and a communication cable would be needed.

 

[fafrd]

I have never tested such a large load on this machine, even though it is designed to handle pass-through loads. I understand that the manufacturer labeled it as 63A because they installed a 63Amp input breaker.

With US-based rules, a 63A breaker should only be used to protect wiring rated for 78.75A…

However, upon closer inspection, I noticed that the wire for the AC input and output is of a smaller gauge than 6 AWG. You can see from the pictures that the incoming wire in the AFC is 6 AWG, but the wire on the terminal block is only 8 or smaller.

8AWG is rated for 73A under chassis wiring but only 24A for power transmission…

Pretty sure AC input to or output from an inverter qualifies as power distribution, so unless there are multiple 8AWG wires in parallel, this sounds like a problem for anything close to 75A let alone 63A…

Based on my assessment, I would not recommend using it for a long period of time with a 63Amp pass-through load, as it would likely cause the wire to overheat.

Don’t see how they can get UL listing for 63A rating with a single 8AWG wire…

I am unsure whether the power drawn from the grid, approximately 45-50W, is being wasted or used for the inverter output. At times, I manually switch off the AC input breaker to completely eliminate grid consumption, resulting in 0W usage. In an interesting test, I utilized Home Automation (HA) to send a changeover Modbus command to switch the inverter to grid power while keeping the AC input breaker in the off position. However, the inverter did not change to grid power as expected. In HA, I have set up an automation that triggers this switch based on the State of Charge (SoC) of my Battery Management System (BMS). When the SoC falls below 15%, the Modbus command is sent to the inverter to switch over to grid power.
Nevertheless, if I turn off the AC input to avoid utilizing the 50W grid power, I will lose power entirely once the battery is fully drained. As for the actual utilization of the 40-50W grid consumption by the inverter, I have yet to investigate this matter. I plan to delve into it further tonight.

I’m guessing they maintain 50W consumption when there is no solar power to avoid any export at night (with some margin).

Yes.

To eliminate grid consumption and forfeit the grid backup, I can only manually switch off the AC input.

Understood - thanks.

There are four fans connected in a way that they all run simultaneously and at the same speed. Two fans are dedicated to the inverter compartment, while the other two are for the Solar Charger compartment. Each compartment is equipped with a temperature sensor. If either of these two sensors detects a temperature higher than 45 degrees Celsius, all four fans will start running at low speed. Once the temperature decreases to 41.8 degrees Celsius, all fans will switch off.

If the temperature exceeds 50 degrees Celsius, the fans will continue running at a higher speed. However, it is unclear whether these fans have two or three speed settings. Generally, when the inverter output or the PV (photovoltaic) system operates at high power levels, the fans can be quite loud.

Loud fans are a bummer, but I guess you get what you pay for.

This is a typical Chinese-made item with a corresponding quality level. It follows the principle of "you get what you pay for." To ensure a longer lifespan for the unit, it is advisable to operate it at around 80% of its maximum capacity. Based on the wire size and overall construction, it seems that the unit is somewhat undersized for its claimed capabilities. The wiring compartment and the wiring block are both quite small, which hampers airflow and may lead to increased heat. Additionally, the unit causes frequent LED light flickering. However, apart from these issues, it performs well overall. It can effectively charge my Tesla Model 3 at 240V with a 32Amp current, provided that I do not use high-power appliances simultaneously while charging at the maximum 32Amp rate.

The efficiency claim of the product aligns closely with the manufacturer's specifications. Through my recordings, I have observed an efficiency range of 89% to 92% when converting power from the battery to AC output. However, I am unsure about the best method to test the efficiency of the MPPT

OK. Thanks.

 

[fafrd]

I’ve been looking at the EG4 6000W split-phase hybrid and it is essentially 2 x 3000W 120VAC inverters 180-degrees out of phase.

This SRNE inverter seems to be based on a similar architecture but 2 x 5000W 120VAC inverters in this case.

5000W = 41.7A and should be carried by wires rated for 41.7A x 125% x 125% = 65A protected by a breaker of no more than 52A.

6AWG wiring throughout and 50A breakers on input and output would suffice but if those are single 8AWG wires carrying 41.7A, those would not be properly protected by a 63A breaker.

8AWG is rated for 73A under chasis wiring rules, so perhaps they are relying on that.

Using chasis wiring ratings would mean a wire rated for up to 73A could be protected by a 58.4A breaker and could safely carry 46.7A.

If the input and output breakers are reduced to 50A, 8AWG might suffice under chasis wiring rules but more conventional wiring in conduit would need to be 6AWG (to AC input and output).

And a 50A breaker should not carry sustained loads of over 40A to avoid nuisance trips.

So 40A or 4800W per leg = 9600W combined seems like it might be OK as far as internal wiring with 8AWG.

 

[fafrd]

I'm 95% sure even in bypass to grid power its still only 5kW/41.7A per leg. I've tripped it while on grid power before and pretty certain neither phase was using above 5kW although I wasn't watching it real-time. Also, I use Solar Assistant sometimes to switch between Utility output, and SBU and that works fine. As for quality and performance I can say it's been reliable so far. I have about 9,500 kWh of PV through it so far (6 months). I wish SRNE would add parallel function so I could add a second one and get more than 10kW output. There is a setting to set phase mode to 0 or 180 but I believe the software would have to support it and a communication cable would be needed.

Can I ask what you paid for your SPI-10K-U and where you bought it from (including shipping cost, if any)?

5KW per leg will be more than enough power for me so it is really just the possibility of spurious export when discontinuous loads such as fridges cycle off that is my greatest concern.

I’m considering operating the inverter with no grid connected to the AC input to avoid the possibility of any spurious export spikes visible to the utility meter.

A separate 240VAC 48VDC LiFePO4 charger would add a bit more cost but guarantees that no export spikes can be visible to the grid.

The new ‘dual-channel’ metering recently authorized in California will allow my smartmeter to detect and record export spikes lasting as little 0.58ms…

 

[Cheap 4-life]

Why are you worried about export when loads cycle? Wouldn’t the loads be powered via the load output so offgrid?

 

[fafrd]

Why are you worried about export when loads cycle? Wouldn’t the loads be powered via the load output so offgrid?

Yes, the loads are on the AC output, but if the AC input is connected to the grid through closed transfer switches, when a load such as a fridge on the AC output being powered by the inverter switches off, the inverter will take several ms or even as much as 10ms to sense the reduced load and throttle-back power being generated.

During that interval, there will be an inconsequential amount of spurious export sent out through the transfer switches to the grid.

For example, a running fridge load of 100W being offset for 10ms after the fridge has cycled off = 278uWh, so no big deal.

But dual-channel metering means the smart meter will be able to detect and accumulate that net export separately from the much greater net import that is occurring all night long.

If the inverter AC input is not connected to the grid, that spurious export cannot happen.

Charging a battery from the grid does not require any special permission or approvals, but connecting an inverter connected to the grid does, so having a true off-grid inverter with no connection to the grid and controlling a grid-tied AC charger to act as a generator when needed seems prudent until it becomes clear how the utilities make use of their new-found authorization to collect dual-channel data and what the final rules end up being.

 

[Cheap 4-life]

Yes, the loads are on the AC output, but if the AC input is connected to the grid through closed transfer switches, when a load such as a fridge on the AC output being powered by the inverter switches off, the inverter will take several ms or even as much as 10ms to sense the reduced load and throttle-back power being generated.

During that interval, there will be an inconsequential amount of spurious export sent out through the transfer switches to the grid.

For example, a running fridge load of 100W being offset for 10ms after the fridge has cycled off = 278uWh, so no big deal.

But dual-channel metering means the smart meter will be able to detect and accumulate that net export separately from the much greater net import that is occurring all night long.

If the inverter AC input is not connected to the grid, that spurious export cannot happen.

Charging a battery from the grid does not require any special permission or approvals, but connecting an inverter connected to the grid does, so having a true off-grid inverter with no connection to the grid and controlling a grid-tied AC charger to act as a generator when needed seems prudent until it becomes clear how the utilities make use of their new-found authorization to collect dual-channel data and what the final rules end up being.

Wouldn’t the AC input connection be one way. It’s not bidirectional with this inverter, the inverter can’t do grid tie output.. like with a regular offgrid inverter, where does the milliseconds of power go when the load turns off? I’d suppose unlike a grid tie inverter the power output would instantly stop therefore no backfeed even if it was possible to send power in opposite direction thru the AC input. If battery charging from AC is enabled I find it hard to believe power will go backwards like it’s a bidirectional AC input. Yes if grid assist was enabled I can see it being possible to unintentionally backfeed, but not when the input is merely setup as a charger. Guess I could be wrong.

 

[fafrd]

Wouldn’t the AC input connection be one way. It’s not bidirectional with this inverter, the inverter can’t do grid tie output..

If an inverter integrates ‘grid support’ and allows critical loads to be powered by grid through transfer switches, the connection is not ‘one way’.

like with a regular offgrid inverter, where does the milliseconds of power go when the load turns off?

A PSW will allow output voltage to increase for the few ms it requires to sense reduced load and respond by reducing output power.

With transfer switches closed to grid, reduced load does not translate to increased voltage but rather reverse current to grid sensed by CT sensors.

Response can be very fast but is unlikely to be less than 1ms and is likely closer to 10ms. So it is pretty much impossible to avoid spurious export of shutdown load x 1ms on any hybrid inverter with ‘grid support’ capability…

For a 100W load, we’re talking about ~28uWh of spurious export, so absolutely not an issue for the grid or safety.

But dual-channel metering will allow the utility to detect and record that minuscule amount of spurious export where before it would be lost when combined with import under single-channel / net metering rules…

I’d suppose unlike a grid tie inverter the power output would instantly stop therefore no backfeed even if it was possible to send power in opposite direction thru the AC input.

No such thing as ‘instantly’ - not possible. Whether using an analog feedback circuit or a digital control algorithm and sensors, output power only gets reduced after responding to increased voltage or increased export current (or decreased import current).

If battery charging from AC is enabled I find it hard to believe power will go backwards like it’s a bidirectional AC input.

True, an inverter that only connects AC input to an internal AC charger and does not allow loads to be directly powered by grid through a transfer switch could not ever export anything to grid, like a PSW.

If you know of any off-grid inverters designed that way, I’d appreciate a link…

Yes if grid assist was enabled I can see it being possible to unintentionally backfeed, but not when the input is merely setup as a charger. Guess I could be wrong.

If the grid is only connected to the battery charger or if transfer switches are opened to disconnect grid from inverter output, there is no issue. But I’ve not yet found any off grid inverters other than Victron that provide that capability…

 

[Cheap 4-life]

If an inverter integrates ‘grid support’ and allows critical loads to be powered by grid through transfer switches, the connection is not ‘one way’.

battery charging power from AC goes in the same direction as grid assist to AC Load out of the inverter. Nothing is leaving the inverters AC input like a grid tie capable hybrid would do

With transfer switches closed to grid, reduced load does not translate to increased voltage but rather reverse current to grid sensed by CT sensors.

couldn’t the transfer switches simply be opened stopping power from going to the grid, then how would there be current going to the grid with the path cut off..
I thought this inverter didn’t use CTs

Response can be very fast but is unlikely to be less than 1ms and is likely closer to 10ms. So it is pretty much impossible to avoid spurious export of shutdown load x 1ms on any hybrid inverter with ‘grid support’ capability

yes, with my grid tie zero export inverters I see slight backfeed everytime a large load shuts off. I just wasn’t thinking that would also happen with an offgrid inverter.. but I guess if it has a grid connection that is being used as grid assist then it could happen as you explain. Although if the grid connection was only used for AC charging would backfeed still be possible?

True, an inverter that only connects AC input to an internal AC charger and does not allow loads to be directly powered by grid through a transfer switch could not ever export anything to grid, like a PSW.

If you know of any off-grid inverters designed that way, I’d appreciate a link…

I’m fairly certain the old Aims offgrid inverters only had AC charging without grid assist. Or maybe they didn’t allow for any grid connection. I’d think there was offgrid inverters with AC charging before grid assist was a thing.

If the grid is only connected to the battery charger or if transfer switches are opened to disconnect grid from inverter output, there is no issue. But I’ve not yet found any off grid inverters other than Victron that provide that capability…

Grid assist is a useful function and more I think about it due to this conversation, It’s probably difficult to find many offgrid inverters that wouldn’t also allow the AC from grid or generator to passthru to the load if it already had the AC battery charging capability. However I’m sure there’s some that just don’t have a transfer switch and only have AC charging path to the battery.