Hoymiles problem, or grid voltage too high?

[Krill]

I am lucky to have a 13kW export limit, in practice I don't get to use it. On a blowy late spring sunny day exporting 13kW will take me well over 260V. I have had a few things suspiciously die around the time I did the tests and I didn't let the voltage climb much above 256V. So I limit my voltage to 250V whilst exporting though might sneak it up a bit this year. The lure of more export money is a strong motivator.

It's not worth tapping us down much more as we have the opposite problem when I am drawing a lot of current in the winter (80A) the voltage drops quite a bit reaching around 220V at the Quattro and then 217-218 at the CU.

Out of interest, is there a rule of thumb for the "most efficient" way to export related to voltage, ie is it more efficient to test the high voltage limits, or does it not really matter for residential export? I imagine that system safety ie not frying any equipment is the more important consideration.

Context for asking: I've never seen a grid voltage higher than 247V whilst force exporting with some solar PV production, but usually it floats between 238V and 243V

[Andy]

I am lucky to have a 13kW export limit, in practice I don't get to use it. On a blowy late spring sunny day exporting 13kW will take me well over 260V. I have had a few things suspiciously die around the time I did the tests and I didn't let the voltage climb much above 256V. So I limit my voltage to 250V whilst exporting though might sneak it up a bit this year. The lure of more export money is a strong motivator.

It's not worth tapping us down much more as we have the opposite problem when I am drawing a lot of current in the winter (80A) the voltage drops quite a bit reaching around 220V at the Quattro and then 217-218 at the CU.

Out of interest, is there a rule of thumb for the "most efficient" way to export related to voltage, ie is it more efficient to test the high voltage limits, or does it not really matter for residential export? I imagine that system safety ie not frying any equipment is the more important consideration.

Context for asking: I've never seen a grid voltage higher than 247V whilst force exporting with some solar PV production, but usually it floats between 238V and 243V

When I first got my SolarEdge in 2019, it used to limit to 253V before cutting out . That's when I asked the DNO to retap as I was tripping quite often. The "Taps" were stuck (corrosion?) so they had to replace the transformer for free. As part of that they massively increased the one serving the three houses to future proof. So we essentially got an upgrade for free. Then as part of my home system upgrade I had to upgrade the Solaredge firmware. I missed that the change to the limits meant that the inverters would not cut out until 262V which Nowty mentioned in his post. I was actually monitoring my system to see if the Quattro would cut out at 253V correctly (based on my assumption of the G99 knowledge). I was confused when neither backed off.

I can see how much of a pain it must have been for the DNOs as more solar came online and assume the rise in allowable voltage was to counter the issues of local grid voltages cumulatively rising. But 262V seems a bit high for local equipment. I end up limiting at 250V exporting about 6-7kWh in the summer. So for a normal install I'd not worry about it. Especially as your base voltage seems pretty well placed. It's only much bigger systems that you need to keep an eye on.

I dynamically change the max export on the Quattro constantly as a result of monitoring the voltages and let the Quattro deal with the grid limits for the time to change power. 5 seconds I think? There are impedance issues with changing too quickly and often. It all seems to work well.

You can see the tracking in the graph below. The yellow line is the maximum allowed export. The green line is the voltage as measured at the Quattro.



.

[richbee]

Thanks for all the feedback - it does sound like I should monitor the voltage for a while & maybe get in touch with the DNO if it stays high & keeps tripping.
The sunsynk is, set to nominal 230v output, with max voltage limit at 260 I think.
If I reduce that to 250,will it just stop outputting at 250?

[richbee]

OK, so had the DNO - Northern Powergrid, out today, as the voltage on my Sunsynk was reading 253-254 for a couple of hours and the Hoymiles has gone off and sulked since 10am.
Nice guy, came out and measured at the meter at 251V (obviously the sun went in and the Sunsynk also dropped to 251 at the same time ).
He then removed the main fuse and measured the incoming without load at 248V.
He said basically they won't do anything as it is within limits and they are seeing a lot of problems on solar PV systems especially with EV chargers which are sensitive (& over read the voltage).
He went to talk to his boss and came back to say they would arrange to fit voltage monitoring equipment to see if it shows anything - not sure when as they are apparently in high demand - but better than nothing I guess.
As soon as he went, the Sunsynk went back up to 256V, obviously!!

It's a bit of a problem, the power grid people seem to want to keep everything simple and don't want to drop voltages if possible due to the risk of too low voltage in the winter.

So what to do next - doesn't seem a lot of point in fitting more panels, and a bigger inverter with higher export would definitely be a no-no.
Will the heatpump, when it is fitted, use enough to keep the voltage down? I guess it is a suck it and see thing

[Countrypaul]

I suspect the HP will make little difference during the sunniest part of the year due to the fact that when it is warm and sunny the HP probably won't be producing much heat. In those shoulder months you might get the HP running at the same time your PV is producing so there might be a benefit in keeping the voltage lower.

[ecogeorge]

OK, so had the DNO - Northern Powergrid, out today, as the voltage on my Sunsynk was reading 253-254 for a couple of hours and the Hoymiles has gone off and sulked since 10am.
Nice guy, came out and measured at the meter at 251V (obviously the sun went in and the Sunsynk also dropped to 251 at the same time ).
He then removed the main fuse and measured the incoming without load at 248V.
He said basically they won't do anything as it is within limits and they are seeing a lot of problems on solar PV systems especially with EV chargers which are sensitive (& over read the voltage).
He went to talk to his boss and came back to say they would arrange to fit voltage monitoring equipment to see if it shows anything - not sure when as they are apparently in high demand - but better than nothing I guess.
As soon as he went, the Sunsynk went back up to 256V, obviously!!

It's a bit of a problem, the power grid people seem to want to keep everything simple and don't want to drop voltages if possible due to the risk of too low voltage in the winter.

So what to do next - doesn't seem a lot of point in fitting more panels, and a bigger inverter with higher export would definitely be a no-no.
Will the heatpump, when it is fitted, use enough to keep the voltage down? I guess it is a suck it and see thing

Where are you measuring 256v at the Sunsynk or incomer ?? Bigger cables from Sunsynch to incomer ??? i had to fit 2 x 10mm in parrellel to keep volts down from inverter to fuseboard.
George.

[richbee]

OK, so had the DNO - Northern Powergrid, out today, as the voltage on my Sunsynk was reading 253-254 for a couple of hours and the Hoymiles has gone off and sulked since 10am.
Nice guy, came out and measured at the meter at 251V (obviously the sun went in and the Sunsynk also dropped to 251 at the same time ).
He then removed the main fuse and measured the incoming without load at 248V.
He said basically they won't do anything as it is within limits and they are seeing a lot of problems on solar PV systems especially with EV chargers which are sensitive (& over read the voltage).
He went to talk to his boss and came back to say they would arrange to fit voltage monitoring equipment to see if it shows anything - not sure when as they are apparently in high demand - but better than nothing I guess.
As soon as he went, the Sunsynk went back up to 256V, obviously!!

It's a bit of a problem, the power grid people seem to want to keep everything simple and don't want to drop voltages if possible due to the risk of too low voltage in the winter.

So what to do next - doesn't seem a lot of point in fitting more panels, and a bigger inverter with higher export would definitely be a no-no.
Will the heatpump, when it is fitted, use enough to keep the voltage down? I guess it is a suck it and see thing

Where are you measuring 256v at the Sunsynk or incomer ?? Bigger cables from Sunsynch to incomer ??? i had to fit 2 x 10mm in parrellel to keep volts down from inverter to fuseboard.
George.

256V is from the Sunsynk app - so not sure how accurate it is. But, the Hoymiles is the one which is tripping out once it sees 253V (the Hoymiles seems to consistently read a couple of V lower than the Sunsynk).
I have wondered before if the Sunsynk would benefit from larger cables from a temperature point of view, as it was wired (by the amazing solar experts!) in 2.5mm all the way from the consumer unit up to the loft and the inverter - pushing/pulling 3.8kW max. Would that help with the overall system voltage?

[Countrypaul]

OK, so had the DNO - Northern Powergrid, out today, as the voltage on my Sunsynk was reading 253-254 for a couple of hours and the Hoymiles has gone off and sulked since 10am.
Nice guy, came out and measured at the meter at 251V (obviously the sun went in and the Sunsynk also dropped to 251 at the same time ).
He then removed the main fuse and measured the incoming without load at 248V.
He said basically they won't do anything as it is within limits and they are seeing a lot of problems on solar PV systems especially with EV chargers which are sensitive (& over read the voltage).
He went to talk to his boss and came back to say they would arrange to fit voltage monitoring equipment to see if it shows anything - not sure when as they are apparently in high demand - but better than nothing I guess.
As soon as he went, the Sunsynk went back up to 256V, obviously!!

It's a bit of a problem, the power grid people seem to want to keep everything simple and don't want to drop voltages if possible due to the risk of too low voltage in the winter.

So what to do next - doesn't seem a lot of point in fitting more panels, and a bigger inverter with higher export would definitely be a no-no.
Will the heatpump, when it is fitted, use enough to keep the voltage down? I guess it is a suck it and see thing

Where are you measuring 256v at the Sunsynk or incomer ?? Bigger cables from Sunsynch to incomer ??? i had to fit 2 x 10mm in parrellel to keep volts down from inverter to fuseboard.
George.

256V is from the Sunsynk app - so not sure how accurate it is. But, the Hoymiles is the one which is tripping out once it sees 253V (the Hoymiles seems to consistently read a couple of V lower than the Sunsynk).
I have wondered before if the Sunsynk would benefit from larger cables from a temperature point of view, as it was wired (by the amazing solar experts!) in 2.5mm all the way from the consumer unit up to the loft and the inverter - pushing/pulling 3.8kW max. Would that help with the overall system voltage?

If the 2.5mm was the problem wouldn't you expect the voltage loss to be in the 2.5mm, so a lower voltage at the CU than at the inverter? Have you measured voltage at the inverter and at the CU when the sun is out to see if there is significant loss? - using the same meter!

A quick check on here: https://www.tlc-direct.co.uk/Technical/ ... eDrop.html assuming 6m between CU and inverter suggests youmight get 1.78V loss.

[Countrypaul]

Where are you measuring 256v at the Sunsynk or incomer ?? Bigger cables from Sunsynch to incomer ??? i had to fit 2 x 10mm in parrellel to keep volts down from inverter to fuseboard.
George.

256V is from the Sunsynk app - so not sure how accurate it is. But, the Hoymiles is the one which is tripping out once it sees 253V (the Hoymiles seems to consistently read a couple of V lower than the Sunsynk).
I have wondered before if the Sunsynk would benefit from larger cables from a temperature point of view, as it was wired (by the amazing solar experts!) in 2.5mm all the way from the consumer unit up to the loft and the inverter - pushing/pulling 3.8kW max. Would that help with the overall system voltage?

If the 2.5mm was the problem wouldn't you expect the voltage loss to be in the 2.5mm, so a lower voltage at the CU than at the inverter? Have you measured voltage at the inverter and at the CU when the sun is out to see if there is significant loss? - using the same meter!

A quick check on here: https://www.tlc-direct.co.uk/Technical/ ... eDrop.html assuming 6m between CU and inverter suggests youmight get 1.78V loss, but could be 2.87 if clipped directly to a wall.

[richbee]

Where are you measuring 256v at the Sunsynk or incomer ?? Bigger cables from Sunsynch to incomer ??? i had to fit 2 x 10mm in parrellel to keep volts down from inverter to fuseboard.
George.

256V is from the Sunsynk app - so not sure how accurate it is. But, the Hoymiles is the one which is tripping out once it sees 253V (the Hoymiles seems to consistently read a couple of V lower than the Sunsynk).
I have wondered before if the Sunsynk would benefit from larger cables from a temperature point of view, as it was wired (by the amazing solar experts!) in 2.5mm all the way from the consumer unit up to the loft and the inverter - pushing/pulling 3.8kW max. Would that help with the overall system voltage?

If the 2.5mm was the problem wouldn't you expect the voltage loss to be in the 2.5mm, so a lower voltage at the CU than at the inverter? Have you measured voltage at the inverter and at the CU when the sun is out to see if there is significant loss? - using the same meter!

A quick check on here: https://www.tlc-direct.co.uk/Technical/ ... eDrop.html assuming 6m between CU and inverter suggests you might get 1.78V loss.

I haven't measured like that ,no - would be a good idea.
Your method sounds sensible - that 256V at the Sunsynk - 1.8V from the cable losses equals 254.2 at the CU, which is still enough to trip the Hoymiles and hopefully read 253+ once the DNO guys set up their monitoring equipment.