H1-3.7-E, 4x Mira HV25, 10x Trina 425W, installed Apr 2023, G98,
firmware 1.63/1.02/1.70.
Not the fault in the "Error Code 28 / Vac 10min Avg" thread — my fault
register reports "Grid Voltage Fault" (instantaneous), never "Grid 10min
Voltage Fault". They're separate bits in the same word. I proved the
10-min rule isn't my trigger: over today's 05:36-14:07 window the
sliding 10-min average was above 253V in 243/243 windows (and above 255V
in 88%). If that rule were tripping me I'd be in permanent fault, not
intermittent. So raising Vac 10min Avg 253->255 does nothing for me.
What I actually see:
- Trips occur at 259-263V instantaneous. ~100 trips on 17 Sep, 131 min
off-grid, first trip 02:08 (pitch dark — no export).
- 74.9% of all 28,723 hours since Jun 2023 had a max above 253V.
7.0% above 258V. 29 hours above 262V. All-time max 264V (17 Sep 2026).
- The night voltage is WORSE than the solar-peak voltage: 00:00-04:59
mean 255.5V vs 10:00-15:59 mean 253.2V. Nothing to do with my export —
it's over-voltage at 3am with the panels producing nothing.
DNO (Northern Powergrid) visited, measured, agreed it spikes over 250V,
and said it's "within limits" at the meter so they'll do nothing.
Questions:
1. Does the H1 expose the INSTANTANEOUS over-voltage threshold anywhere
— front panel, installer tool, or modbus register? Or is it fixed at
the G98 stage-1 point (262.2V/1s) with no user access?
2. If the DNO refuses on the grounds that their own meter reads within
limits, what actually worked for anyone here? Substation tap change?
3. Anyone seen an H1 report the two faults separately, and does the
instantaneous one log anywhere useful?
I'm not looking to disable protection — I want to know whether the
threshold is reachable at all, and what the realistic route is when the
DNO says no.
Thank you in advance for any advice given.
It sounds like it's tripping on the Voltage High Limit 1 (often called 'Vac Upper 1' in the inverters front panel Settings under On-Grid, Grid Para) this would normally be set to 262.2V.
You could try lifting this value a few volts at a time until it stops tripping but note that Vac Upper 2 is usually next at 273.7V)
If you have access to the installer settings you can also adjust the VoltageHighLimit1ProtectTime to extend the period that a high voltage has to be sustained before a trip - this might the better option as it sounds like it's probably fast transients that are causing the trip.
Your DNO will always read the voltage at the incoming fuse with the inverter disconnected, this is not entirely helpful as if they are reading 250V, it will be much higher at certain times of day and in particular when the inverter is exporting. It used to be that if you kicked up a fuss the DNO would leave a recorder as instantaneous measurements are fairly pointless unless you take them just as the fault occurs, a data logger captures everything.
You could try lifting this value a few volts at a time until it stops tripping but note that Vac Upper 2 is usually next at 273.7V)
If you have access to the installer settings you can also adjust the VoltageHighLimit1ProtectTime to extend the period that a high voltage has to be sustained before a trip - this might the better option as it sounds like it's probably fast transients that are causing the trip.
Your DNO will always read the voltage at the incoming fuse with the inverter disconnected, this is not entirely helpful as if they are reading 250V, it will be much higher at certain times of day and in particular when the inverter is exporting. It used to be that if you kicked up a fuss the DNO would leave a recorder as instantaneous measurements are fairly pointless unless you take them just as the fault occurs, a data logger captures everything.
Hi Dave,
Thanks — genuinely useful, and you're the first person in this thread to point at the right setting. I'd been going round the houses on the 10-minute average before working out my trips are the instantaneous O/V, not the 10-min one. Your names line up exactly with the G98 table I'd been reading (stage 1 262.2V/1s, stage 2 273.7V/0.5s), so at least I know I'm looking at the right thing now.
One thing I'd gently push back on: I don't think it's fast transients. I log the grid voltage continuously, and on my worst day (17 Sep) I've got 91 separate excursions above 262V, over 13 minutes' worth in total, longest single one 2.2 minutes. Several happen at 3-4am with the panels producing nothing, so it isn't my export. And on that same day the inverter sat connected for half an hour with the voltage up at 262V and never tripped once — which is exactly what G98 says it should do, so the threshold isn't twitchy. It only drops out once it genuinely crosses 262.2. Stretching the protect time looks like it would paper over a supply that's sitting high rather than clipping spikes. My days with 258V+ sustained run to 21 hours out of 24, so it isn't fleeting.
On the DNO side: I haven't had them out yet — I've raised a formal complaint with Northern Powergrid (Ofgem GS5) and asked them to come and measure, so your point about them reading at the fuse with the inverter disconnected is going straight into that.
Two things I could still use your steer on:
1. Is the instantaneous threshold (or the ProtectTime) actually reachable from the user side on an H1, or does it need the installer login? I'm not looking to disable it — I want to establish whether it's reachable at all, because "just adjust the inverter" keeps being suggested to me and I don't think it's something I can properly do on my own anyway.
2. Have you ever got a DNO to actually leave a recorder behind, and did it take a formal complaint to get there?
Thanks again — this thread has been more useful than anything FoxESS or the DNO have told me.
Phill
Thanks — genuinely useful, and you're the first person in this thread to point at the right setting. I'd been going round the houses on the 10-minute average before working out my trips are the instantaneous O/V, not the 10-min one. Your names line up exactly with the G98 table I'd been reading (stage 1 262.2V/1s, stage 2 273.7V/0.5s), so at least I know I'm looking at the right thing now.
One thing I'd gently push back on: I don't think it's fast transients. I log the grid voltage continuously, and on my worst day (17 Sep) I've got 91 separate excursions above 262V, over 13 minutes' worth in total, longest single one 2.2 minutes. Several happen at 3-4am with the panels producing nothing, so it isn't my export. And on that same day the inverter sat connected for half an hour with the voltage up at 262V and never tripped once — which is exactly what G98 says it should do, so the threshold isn't twitchy. It only drops out once it genuinely crosses 262.2. Stretching the protect time looks like it would paper over a supply that's sitting high rather than clipping spikes. My days with 258V+ sustained run to 21 hours out of 24, so it isn't fleeting.
On the DNO side: I haven't had them out yet — I've raised a formal complaint with Northern Powergrid (Ofgem GS5) and asked them to come and measure, so your point about them reading at the fuse with the inverter disconnected is going straight into that.
Two things I could still use your steer on:
1. Is the instantaneous threshold (or the ProtectTime) actually reachable from the user side on an H1, or does it need the installer login? I'm not looking to disable it — I want to establish whether it's reachable at all, because "just adjust the inverter" keeps being suggested to me and I don't think it's something I can properly do on my own anyway.
2. Have you ever got a DNO to actually leave a recorder behind, and did it take a formal complaint to get there?
Thanks again — this thread has been more useful than anything FoxESS or the DNO have told me.
Phill
Hi,
To be honest it’s been a while since i’ve known a DNO engineer to leave a logger, and the grid reference voltage range is obviously getting worse as more and more renewables come on-line. There is some push from NESO to change the 230V +10%/-6% to 230V +10/-10% which will allow the DNO’s to come down another tap on the transformer.
Ok understood on the fast transients, yes i’d agree from what you have said it’s sustained high voltage and weirdly happening in the early hours, i’ve not seen that before - are you by chance near a power station, hydro electric energy source or similar?. But yes also agree with you raising a complaint on your DNO if you are seeing consistently high voltages like that - it would normally be a single tap change on the local transformer which would allow an approx 7V drop but obviously depends whether it’s already on it’s lowest tap; either way they should be doing a lot more than they have done for such high voltages.
On the protect time, no that isn’t accessible via the inverter panel, it’s only available to an installer/agent through the website remote configuration, and as it’s normally set just less than a second (I think from memory it can be increased to 100 seconds max) in your case it’s not going to help much, lifting the Vac1 upper limit a couple of volts will help… but for obvious it’s not a long term solution - most EV chargers will hard stop working if the voltage gets above 254V
To be honest it’s been a while since i’ve known a DNO engineer to leave a logger, and the grid reference voltage range is obviously getting worse as more and more renewables come on-line. There is some push from NESO to change the 230V +10%/-6% to 230V +10/-10% which will allow the DNO’s to come down another tap on the transformer.
Ok understood on the fast transients, yes i’d agree from what you have said it’s sustained high voltage and weirdly happening in the early hours, i’ve not seen that before - are you by chance near a power station, hydro electric energy source or similar?. But yes also agree with you raising a complaint on your DNO if you are seeing consistently high voltages like that - it would normally be a single tap change on the local transformer which would allow an approx 7V drop but obviously depends whether it’s already on it’s lowest tap; either way they should be doing a lot more than they have done for such high voltages.
On the protect time, no that isn’t accessible via the inverter panel, it’s only available to an installer/agent through the website remote configuration, and as it’s normally set just less than a second (I think from memory it can be increased to 100 seconds max) in your case it’s not going to help much, lifting the Vac1 upper limit a couple of volts will help… but for obvious it’s not a long term solution - most EV chargers will hard stop working if the voltage gets above 254V