Hybrids don't seem to be the best for PV production, there's a ramp up delay after a load cuts where it would normally push that power to the grid, but with our export limiter on it doesn't seem to be so fluid and little chunks of PV don't get harvested.

Has anyone com across convection cooled off grid inverters?

When I looked last month I only found the SMA stuff, which looks pricey.

13 minutes ago, rinners said:

Hybrids don't seem to be the best for PV production, there's a ramp up delay after a load cuts where it would normally push that power to the grid, but with our export limiter on it doesn't seem to be so fluid and little chunks of PV don't get harvested.

I think that's not uncommon. Certainly when grid drops and the comes back, or they diconnect and reconnect, they are designed to wait a bit and then ramp up - that's deliberate.

13 minutes ago, rinners said:

Has anyone com across convection cooled off grid inverters?

When I looked last month I only found the SMA stuff, which looks pricey.

Goodwe? I have a Goodwe hybrid and it has no fans. If you listen closely you will hear relays clicking.

For totally off-grid I think Victron is the way to go.

2 hours ago, Jacques Ester said:

For totally off-grid I think Victron is the way to go.

I don't know about pricing, and I believe Victron prices have come down a bit. My Goodwe can be run as an off-grid inverter and it does get a little bit more from the panels in that mode. But Goodwe tend to be on the lower end of the output power side.

23 hours ago, Jacques Ester said:

For totally off-grid I think Victron is the way to go.

Which one doesn't have fans?

On 2025/03/11 at 4:52 PM, Bobster. said:

I think that's not uncommon. Certainly when grid drops and the comes back, or they diconnect and reconnect, they are designed to wait a bit and then ramp up - that's deliberate.

Goodwe? I have a Goodwe hybrid and it has no fans. If you listen closely you will hear relays clicking.

I'm not talking about when the grid drops. I mean when load switches off (like a geyser) the normal course of action would be for a hybrid inverter to temporarily push the PV out to the grid (and maybe changes to DC to charge the battery if that setting is an option). With an export of 0W it instead drops the PV to the remaining load level and slowly ramps up the PV again. It's small but it adds up. With some inverters it's more noticeable and especially with loads like irons and cookers that are on and off all the time.

Off grid inverters don't have the export option, though most do have an AC input for generator which one could easily use the grid supply to charge the batteries.

1 hour ago, rinners said:

Which one doesn't have fans?

I'm not talking about when the grid drops. I mean when load switches off (like a geyser) the normal course of action would be for a hybrid inverter to temporarily push the PV out to the grid (and maybe changes to DC to charge the battery if that setting is an option). With an export of 0W it instead drops the PV to the remaining load level and slowly ramps up the PV again. It's small but it adds up. With some inverters it's more noticeable and especially with loads like irons and cookers that are on and off all the time.

Off grid inverters don't have the export option, though most do have an AC input for generator which one could easily use the grid supply to charge the batteries.

I'm no expert, which may make me part of the problem :-) But a hybrid inverter has to balance grid, PV and battery and make decisions about what it's going to demand from each. Now there must be an overhead to that.

So yesterday, for example, there was very changeable weather where I live, PV was up and down like a yo-yo, and I was getting frustrated because I could see the sun come out from behind the clouds. yet it took the inverter a while to react, and sometimes it would react so slowly that by the time it had decided it could increase the draw from PV, the sun was back behind a cloud again, They don't just accept all available PV, they demand what they need, and what they need is determined by some algorithm somewhere in the firmware.

If I put my inverter into off-grid mode it suddenly gets more responsive. I'm thinking that this is because it now doesn't have to do the multi-way balancing act that it does when it's being run as a hybrid.

So this is a long way of saying that maybe an inverter marketed as a hybrid (and it is, or it can be) might still work for you if, like mine, it can be switched into a mode where it emulates an off-grid inverter. Which would give you a lot more choices.

Edited March 13, 20251 yr by Bobster.
Korrrreckshuns

8 hours ago, Bobster. said:

I'm no expert, which may make me part of the problem :-) But a hybrid inverter has to balance grid, PV and battery and make decisions about what it's going to demand from each. Now there must be an overhead to that.

I for one would not agree that due to the balancing between different sources need to waste a lot of PV. I have used my Solis on grid DC to AC and yes the measured voltage and current from PV to AC via the MPPT has been between 96-98% I have also seen the speed of adjustment real quick. May be the difference comes in between different makes and not if it is on grid or off grid.

I have also checked the same panels when switched between my Axpert off grid and Solis on grid and I found both would behave at about the same speed to changing conditions. Part of my checking was different strings both facing north at 26 degrees at a time where 1 string was attached to the Axpert and the 2nd to the Solis and both behave the same even with cloud edge.

I know my checking was limited to these 2 makes/types and could indeed differ with other makes. Now with both these mentioned strings on the Deye 5kW I have had no need to check as I have ample power most times. As a low user of power I just had a look at my grid use and found my grid consumption is about 0.4 kWh per day for the last 7 days. I have to say the reason why my grid use is so low I have a number of days with no grid use as I tend to keep my grid just switched off via smart switch. This 0.4kWh is mostly when cooking or to ensure my battery does not discharge too deep I have my grid on from 05h to 07h in the morning. My SOC does vary a bit at the low at 06h between 20% and 40%. Some nights when I go to sleep at 23h I would not even activate the 05h-07h timer if my SOC is above 60%. Today even with rain around 17h30 my battery was still at 99% while normally it would be at about 96% even on sunshine days. Although my Deye is set to use 50W from grid always it is of theoretical importance as the grid is mostly OFF.

I would not be concerned to use an off grid instead of a hybrid. I might be wrong. This is if one wants to use grid when the need arises.

21 hours ago, rinners said:

Which one doesn't have fans?

I'm not talking about when the grid drops. I mean when load switches off (like a geyser) the normal course of action would be for a hybrid inverter to temporarily push the PV out to the grid (and maybe changes to DC to charge the battery if that setting is an option). With an export of 0W it instead drops the PV to the remaining load level and slowly ramps up the PV again. It's small but it adds up. With some inverters it's more noticeable and especially with loads like irons and cookers that are on and off all the time.

Off grid inverters don't have the export option, though most do have an AC input for generator which one could easily use the grid supply to charge the batteries.

These Victron Inverters are fan less.

https://www.sunstore.co.za/collections/solar-inverters-chargers/products/victron-energy-quattro-48v-inverter-charger?variant=33098227548194

I don't see the logical link, though. (As always I'm open to correction by anyone who knows better).

Whatever the speed is of the Maximum Power Point Tracker (MPPT) algorithm and processor on any given device, for me it doesn't seem linked to whether it is built into either a hybrid inverter, off-grid inverter, or standalone MPPT.

Could you find for argument's sake that hybrid inverter A (Sunsynk/Deye) has a different tracking than hybrid inverter B (Goodwe) or C (Solis), etc.etc. and that maybe one of them performs better under the conditions of rapidly changing loads to use more PV?

Does it necessarily follow that off-grid inverter A (Axpert/Kodak), or B (Growatt) or C (Luxpower), etc etc. will have superior performance in terms of tracking the load level, and return to PV sooner?

Would it make a difference if one would use an energy meter instead of a CT coil to give feedback to the inverter?

Don't have any answers, just thinking out loud and hoping someone else knows.

13 hours ago, Scorp007 said:

I for one would not agree that due to the balancing between different sources need to waste a lot of PV. I have used my Solis on grid DC to AC and yes the measured voltage and current from PV to AC via the MPPT has been between 96-98% I have also seen the speed of adjustment real quick. May be the difference comes in between different makes and not if it is on grid or off grid.

That could be so. I have heard from other Goodwe users that their inverter seems more responsive and efficient when run in off-grid mode.

But what I'm trying to achieve is to suggest that Rinners has options.

Greenfields raises an interesting point: Hybrids depend on a CT coil for one of the inputs. In the case of the Goodwe there is a proprietary device named "smart meter" (though it's not a meter that a person can look at and see numbers) between the CT and the inverter. But in any case, when you put a hybrid inverter into off-grid mode there is a whole layer of inputs that is removed, and the inverter can make quicker decisions.

My leads are taken from the inverter specs and it might be a small sample. Most if not all hybrids quote a higher efficiency than the off grid units a have looked at. Difference up to 6%.

On 2025/03/14 at 9:14 AM, Bobster. said:

Greenfields raises an interesting point: Hybrids depend on a CT coil for one of the inputs. In the case of the Goodwe there is a proprietary device named "smart meter" (though it's not a meter that a person can look at and see numbers) between the CT and the inverter. But in any case, when you put a hybrid inverter into off-grid mode there is a whole layer of inputs that is removed, and the inverter can make quicker decisions.

I think we are trying to think what inputs need to be measured.

If one had to actually do a test with grid on or off and switching say a 2 kW load on and off the actual PV generated remains the same. This is on my Deye and the Sunsynk should be the same.

Not sure how one would try and measure the response. I belief the CT is just a switch so it's reading will be affected only as far as what is used from the grid. This has the delay as is mentioned to get the grid to be stable after being off.

I used real time measurements on the inverter and not via any APP.