Backup down under

[Northarmgazer]

Brrr.


I think we are pretty lucky relative to many of our American cousins on this forum when it comes to grid energy reliability. Gisspland does cop it a bit though, probably worse that us.

I recently got notice that we will have a 9AM-4PM planned outage on 21 June, so another good test of the system. On the Winter Solstice no less!


Yes, and the plan at the moment is to take most of our regular load off-grid, however removing grid connection is never going to be a realistic option for us. The power and energy demands for our heating and cooling would make that far too expensive. 100 kWh days can happen.

So I plan to leave our ducted aircon (heating/cooling), ovens and stoves grid connected only (hot water I'll get to separately below). During the day these will still be mostly powered by the grid-tied solar PV, so it's at night we will be importing energy for those.

But the rest of household loads can largely be taken off-grid, save for some pretty crummy solar days (which with all the rain this year has been pretty often although June has been much sunnier for us). I'm looking at 10-15 kWh/day in general consumption. Anything more usually means we have guests or we are doing some unusual activity. Just about all consumption above that level will be heating or cooling.


A big project. all the best with it.

I'm perhaps a bit less down on energy suppliers than you are, I get why our feed in tariff is what it is and have no real big issue with it. Most of the cost of energy supply is in the network and transmission. The energy generation part (which is what our solar PV is) is a significant but minor component of the cost (save for recent world events which hopefully will not go on forever).

But yes there is appeal in reducing reliance on the grid in rural areas where we have the space and deal with reliability issues.


I think there is a bit of general misunderstanding on what each of these proposed system changes are about and what they mean at household level (in reality it will mean bugger all difference to costs and in many cases it will mean much better outcomes for households) but even so it will be quite a long time before such things take effect, if ever.

You will never be required pay to export. The changes include a provision that residential customers must be given an option which is free of any export tariffs (which BTW would be levied by distributors to retailers, not to end users directly). There are other elements to this which are actually pretty positive for solar PV system owners but I'll leave that for now - more a topic for a different thread perhaps.

I am completely unconcerned about them as they will ultimately enable more people to have more solar PV connected to the grid.


We do have a fireplace but really only use it for visitors/ambiance reasons. Being a large open fireplace it is terribly inefficient and burns through wood like there's no tomorrow! I don't have a decent renewable supply of our own wood so it is also pretty expensive as a regular heating option. Last time I had a wood store pile all it did was encourage termites! If we can ever get to do renovations we plan to swap the fireplace out for a more efficient burning unit, and then I would consider using it more. Winter is not much more than 8 weeks long here.


They are fantastic options for reducing energy consumption.

I did look into one when I replaced our 30 year old HW tank a couple of years back but the layout of the house was not conducive to fitting the compressor and it was going to be terribly expensive, at least $3k more and closer to $4k more when all the extra plumbing and electrical were sorted.

Plus we just don't use enough hot water to justify it. With the regular electric resistance HW unit we average 4.9 kWh/day over the full year. So a heat pump is only going to save us about 3.5 kWh/day ~=20-25c/day. For something that would cost us ~$3.5k more to install it just wasn't worth it.

If we were a large family with a lot of hot water consumption, then absolutely I would be having a heat pump system installed. We are going to see increasing government support for them as well. They already qualify for federal STC credits (same credits as for solar PV) and the VIC govt has a program for helping to pay for swapping out gas HW system for heat pumps.

I did however have a smart solar PV diverter installed recently so the HW is mostly heated using the spare capacity from our grid-tied solar PV system. It's only early days with that, I'll know better over a full year what our grid consumption for hot water has been reduced by but I expect our hot water will become ~85-90% self powered, with the balance from the grid.

It's kind of fun to find ways to reduce consumption and shift loads so that you self consume more PV and import less from the grid. Our imports have been on a consistent downward trend over the past six years, and I'm looking to keep that trend line moving down.

The change to hot water should reduce grid imports by another 4 kWh/day, and if I do get to scale up my off-grid system in the way I plan to then I think I can drop it by another 5-7 kWh/day.

The last ~10 kWh/day, on average, is unlikely though as that's the heating and cooling demand. That will only be lowered through improvements to the thermal properties of the house. I have done some but there is much more work to do on that front.


That should be fun!


The off-grid system monitoring is via Solar Assistant. There are a couple of threads here on it. It works with specific types of inverters and some battery monitoring systems.

The grid-tied system is Fronius and I have a Fronius smart meter as well, so I get full visibility on grid-tied production, consumption, imports and exports.

I have both integrated into Home Assistant and the Energy monitoring for them both are combined.

Thanks for breaking down all those replies. I must say of everything you said, the one thing that sticks in my mind is your overall power consumption. 100kwh days is mind boggling for what I'm assuming is basically a residential rural property? (not that's its anyone's business) I ran a 300 cow dairy farm with solar on the house (not a lot due to grid limitations), 3 solar hot water services on the dairy roof and heat recovery from the vat refrigeration to help heat 5 different hot water services, one being for the actual milk plant (so up to 92C ), plus all the water pumps, lights, 2 phase milking equipment for 6hrs a day plus main farm irrigation pump water etc etc and in Jan we still only used 181 kwh/day and that gave me power bills of $1500 to $2300 a month varying.. no wonder you're motivated to do more...
The unreliability of the grid was my main driving force. Of course I've had no outages at our new place yet, but after those 2, one week outages, I swore that was enough.
Good on you for doing your bit. There are a lot of rude shocks coming for a lot of people I think.
I just googled power prices forecasts for us and the top results were all about prices dropping in 22/23 (AEMO etc), there were multiple reports on this.. funny how the exact opposite is happening

 

[wattmatters]

I must say of everything you said, the one thing that sticks in my mind is your overall power consumption. 100kwh days is mind boggling for what I'm assuming is basically a residential rural property?

They are not common but can happen. We did have one 100 kWh day this most recent Summer.

A warm Summer day is normally in the 60 kWh range. Our biggest ever day was back in 2018 at 144 kWh. One of those stinking hot humid days we can sometimes get with family staying in every room, multiple large aircon units blasting away. Old pool pump. You name it, it was a big day. Never had one like that since.

We have three buildings, the primary home, a granny flat (second dwelling) plus my large mancave which can be used for family overflow. On a hot Summer day with family/guests the aircon systems can be going in all three buildings. Only the granny flat, which was built in 2019, is decent thermally. The mancave isn't too bad but still it has a large garage door at one end so only so much you can do with that.

On a nice mild Spring / Autumn day we might use 12-15 kWh. That's thing with our home, the energy consumption is highly variable and makes off-grid quite a challenging prospect.

 

[WYtreasure]

They are not common but can happen. We did have one 100 kWh day this most recent Summer.

Was Mum making a barrel of tea and running the A/C at the same time?

 

[wattmatters]

Was Mum making a barrel of tea and running the A/C at the same time?

She was probably doing some holiday baking but it would have just been one of those multiple aircon system days. We had friends staying in the mancave and family staying in the house that day. When people are visiting in Summer or in Winter our energy consumption skyrockets.



In general though we have managed to reduce overall consumption since moving here. So far we have sliced 22 kWh/day off our daily average (that's excluding the granny flat which was added start of 2020 and runs on 5 kWh/day):

As mentioned before, we have more work to do to further reduce consumption. With some minor exceptions, we are an all electric property.

This is our electrical energy consumption breakdown:

Hot water energy consumption won't change much (4.9 kWh/day), it has been very steady with minor seasonal fluctuation - so the focus there was moving it from being 100% grid imported to using our excess solar PV capacity with the recent installation of a smart solar PV diverter:

I expect over the course of a year our hot water should become ~85% self powered, reducing hot water grid imports to under 1 kWh/day.

Last year I moved the pool pump operation over to the off-grid system described in this thread. For all intents and purposes it's now 100% self powered.

Granny flat is pretty thermally good (not perfect), there is not much to be done there. It could be taken off-grid one day but I don't see the need for that as for now it shares our existing solar PV and off-grid support.

Heating and cooling are the biggest consumers of energy, and they are quite seasonal in their operation. That is mostly about improving the thermal performance our our home.

For the next part of this off-grid project I am focussed on the general household consumption + granny flat. Daytime this is mostly covered by our grid-tied solar PV but overnight it's all grid imports. So the plan is to work on supplying general household loads "when the sun don't shine".

My small off-grid array (2.22 kW) alone won't be able to supply enough energy all the time (Summer half of year yes, Winter half not so much), so it will be a combination of strategic and tactical use of the grid-tied array to support recharging of the off-grid batteries. Our grid tied system still exports energy in Winter. E.g. yesterday we exported 12 kWh so there is definitely some spare capacity to be drawn upon.

In time I hope to increase the size of the off-grid array, say to ~5 kW, as that would mean it will manage a much greater proportion of days without need of the grid-tied system support.

I have in back of my head the need to support an EV in future.

 

[WYtreasure]

In general though we have managed to reduce overall consumption since moving here. So far we have sliced 22 kWh/day off our daily average (that's excluding the granny flat which was added start of 2020 and runs on 5 kWh/day):

I expect over the course of a year our hot water should become ~85% self powered, reducing hot water grid imports to under 1 kWh/day.


I have in back of my head the need to support an EV in future.

Reduced consumption is easier than building bigger. We were able to do some canning and consolidating recently which allowed us to shut down a fridge/freezer. ?

I picture a tracking solar carport spinning around in some folks yards in the near future. Not real practical but would be fun to watch.

How many more batteries will you be adding?

 

[wattmatters]

How many more batteries will you be adding?

I have 2 x 100 Ah LiFePO4 (16S configuration) server rack batteries on order, so 10 kWh.

Idea is the LiFePO4 will manage the daily cycling while the lead will mostly be there for backup and ballast, which is its main job now.

I'll start experimenting with the inverter I have now supplying overnight loads and if that goes well I'll probably upgrade the inverter to something which can pretty much be supplying general household loads full time.

Reduced consumption is easier than building bigger.

It's definitely a good strategy but sometimes it's not always easier. In our case getting the grid tied system up was easier than some aspects of consumption reduction. The biggest chunk of energy consumption we have is heating and cooling but many of the improvements for that are expensive.

Double glazing for instance (we have a LOT of single pane glass), and replacing the single pane skylights we have which leak heat like a sieve with at least double glazed and with blinds so we can deal with the blazing Summer sun which smashes in through these.



The insulation above those timer ceiling panels is sub-optimal. It would be a large PITA to fix that.

I just heard our builder is coming tomorrow to review plans we have for renovations. Costs have skyrocketed so we have to trim down the project. But improving thermal performance of the house is part of the plan.

And one part of the remodel will be to remove a redundant massive stone fireplace which is and will never be used. The best part about that is it frees up space on the roof for 10 solar panels facing NE. Perfect orientation for a morning recharge of batteries. I have acres of SW roof but that's not so good in Winter. It's great for Summer but it's Winter I'm more concerned with.

 

[wattmatters]

Two of these puppies arrived today:




100 Ah / 5.1 kWh x 2

So begins my hybrid battery experiment (and my off-grid expansion project*). Yet to unbox the other one.

Will start work on these tomorrow, have a couple of things I need to pickup for the installation.

The screen displayed showed pack voltage at 52.6 V, and that matched my multimeter reading. All the individual cells reported voltages within 1 mV..



* which is kinda of delayed because the inverter I want is not in stock. I can still perform tests and experiments with my existing inverter though.

 

[wattmatters]

Installed the new batteries today. Opened the second box and no issues, both batteries seem in good order.

Some basics on the order:

Order date: 22 May 2022
Arrived: 17 Aug 2022
87 days
Vendor: Guangzhou Tycorun Energy Co.,Ltd (via Alibaba)
Cost: USD 2,150 + USD 360 shipping + USD 215 customs import duty + USD 70 customs agent handling
Total: USD 2,795 / AUD 3,990

Two each of:
100 Ah 51.2 V Server rack Lithium Iron Phosphate batteries
16S configuration
Guaranteed capacity rating 98 Ah
100 A max charge/discharge current
450*442*178 mm
52 kg

I made up the all the cables and voila, all connected and running. Picked up a dolly from Bunnings to sit the batteries on. Fits nicely inside my housing.



They are now operating in parallel with the big bank of lead, so this is one part of the grand experiment I'll be reporting progress on.

10 kWh of LFP with 20 kWh of lead acid.

No issues when connecting, no in-rush currents. Had my DC clamp meter on to view but it was only a few hundred milliamps when I closed the breakers and connected them to the lead-acid bank.

Shortly afterwards I turned the solar PV back on and the new batteries were charging with ~ 5 A each:



Switched on a bit after 1:30pm today and turned the solar PV back on.

Solar PV ramped up production pretty much straight away and based on the shape of the PV output curve this afternoon it looks like it was pulling the max it could from the 2.22 kW off-grid array:

I'll let the system sit with the normal idle load overnight (at the moment the battery discharges at ~60 W overnight, which is a small load of office electronics plus the inverter's idle draw).

Will be interested to see how the combined bank's voltage goes overnight. Normally the lead acid bank voltage gradually drops during the night with this sort of daily pattern:



But now the LFP should maintain the voltage higher and discharge a little to the lead, keeping the whole bank closer to the lead's float charge voltage.

I'll monitor for a couple of days, see how much the LFP charges up after tomorrow. We have some sunny (Winter) days forecast.

The way I have connected them I can flip breakers and swap over to operate as follows without need to shut down the inverter or disconnect:
i. Lead + LFP (either or both LFP batteries)
ii. Lead only
iii. LFP only (either or both batteries)

I have a whole stack of experimentation to do.

Some early observations for reference:

• Delivery from order placement took just under three months, which in hindsight is pretty reasonable although I was initially under the impression it would be closer to two months.
  
  
• The sales rep was very responsive to queries, and she did seek to allay any concerns I had regarding the process and worked with me to ensure the units chosen complied with the specifications I wanted.
  
  
• The delivery status information was a bit confusing but eventually I was able to see tracking information after asking for help to see status.
  
  
• The batteries arrived in good order, although I wouldn't say the packaging was particulate robust but they were sitting on their own small wooden pallet. No marks, dents, scratches or any external signs of damage.
  
  
• Both batteries turned on OK, and the display of each operates as normal for these types of units.
  
  
• No user settable BMS parameters. Those are pre-set and I have all the settings in the manual (downloaded). Pretty standard stuff.
  
  
• No printed manual provided (I have that downloaded).
  
  
• There were two communications cables supplied with each battery. One for daisy-chaining to the next battery, one for connection to whatever you might connect it to (e.g an inverter). I'm not at this time planning to use the communications. I may explore that in future. They are supposed to be compatible with Growatt and a few other inverters.
  
  
• The terminals are OK but I feel like they could be a little more robust. M6 screws with hex head. I would like to fashion up some sort of protective covers for the battery terminals. No idea what at this time.
  
  
• This is an experimental crap shoot and I have multiple experiments to run in the coming days and weeks.
  
  
• So far, so good.

 

[BenCos18]

looks good

 

[wattmatters]

Will be interested to see how the combined bank's voltage goes overnight. Normally the lead acid bank voltage gradually drops during the night with this sort of daily pattern:

So here is what the entire battery bank's voltage looked like overnight, and compared with a couple of previous nights:



As I expected the LiFePO4 batteries are keeping the whole pack's voltage much higher (+2 V) overnight, while the voltage still slowly drags down with the parasitic loads (~60 W) plus the natural self discharge of the lead. There would be some current flowing from the LiFePO4 to the lead acid bank but I don't think it is a lot of energy.

The LiFePO4 batteries were no where near full charge at end of yesterday, the LED indicators were suggesting they were somewhere in the 50% SOC level (3 of 6 LEDs lit). One might have been a little higher than the others. This morning (just before dawn) this is the display for each:


Battery SOC indicator LEDs are each showing 3/6 lights.

I don't have individual monitoring to track how much current is flowing from one battery chemistry to the other. I can do spot checks with the DC clamp meter, which I didn't do at night and it's too late now with dawn underway. Given the SOC indicator LEDs of the LiFePO4 batteries are still showing 50% SOC, which is where they were at yesterday afternoon, it doesn't appear to be all that much energy being transferred.

That said, I also don't know how good the LiFePO4 battery BMS and LED SOC indicator is at reporting the true SOC.

Today's task is simple. See how much charge these batteries take and whether they will suck up everything my off-grid PV array has to offer. The day is forecast to be nice sunny Winter's day, a little cloud expected around 3pm.

 

[wattmatters]

I have multiple experiments to run in the coming days and weeks

Just on that, these will be my initial tests:

- monitor the charging/discharging behaviour
- see what's needed to get the bank fully charged (will my off-grid array be enough, or will I need supplemental energy from my grid-tied array?)
- test charging with the grid input (I've used it before, so I know the feature works, I just mean in the context of the hybrid battery)
- work on the Victron smart shunt settings to see what provides the best compromise of settings (especially how it calculates SOC)
- load tests the LFP units to see what capacity they have
- test overnight running of the regular household loads from the off-grid system
- test daytime running of the household from the off-grid system (while being careful about what is used during these tests to avoid overloading the inverter)

 

[wattmatters]

So far this morning it looks like the batteries are soaking up whatever the PV array can supply. So that's good.

Also I monitored the system while the pool pumped started up. It has a 5-min initial cycle where it draws ~900 W and the load is a little higher than the PV array supplies at that time of the morning. There were also a few wispy clouds which passed overhead during this period. After the initial 5-min it drops back to it's all day running mode which is closer to 300 W.

Naturally the battery supplies the balance of what the PV can't.

But which battery is supplying the current?

I monitored the total current draw from the entire hybrid battery with the Victron Smartshunt and watched the BMS data displays on the new LiFePO4 batteries and it was clear the LiFePO4 was supplying the entire demand supplied by the battery.

When the pump dropped back to its all day mode and the load drops back closer to 300 W, then the LiFePO4 batteries were taking all the charging current from the PV array. All through this the lead batteries are just sitting there at idle.

IOW the system is preferentially charging and discharging the LiFePO4 first.

Which is exactly what I had expected/hoped they would do.

For the rest of the day I'll let them charge away and see how far they get. One battery is a little lower SOC than the other based on the LED charge indicators.

I'm estimating the off-grid PV array is capable of supplying ~7-8 kWh (provided we don't get clouds), the loads will be a bit under 2 kWh so there should be about 5-6 kWh available for the batteries to charge. I don't think that will completely charge them today.

 

[wattmatters]

provided we don't get clouds

Effing clouds! Think I only got about 2 kWh into the batteries today. Hopefully tomorrow will be sunnier.

 

[Ozark Tinkering]

Effing clouds! Think I only got about 2 kWh into the batteries today. Hopefully tomorrow will be sunnier.

I know what you mean! If I'm not making power from the sun I expect to at least be collecting water from the rain! It rained me 750 gallons at the start of the week and now I'm catching photons! Next week we're supposed to be getting another 1200 gallons. Clouds only are useless in my operation.

 

[wattmatters]

I know what you mean! If I'm not making power from the sun I expect to at least be collecting water from the rain! It rained me 750 gallons at the start of the week and now I'm catching photons! Next week we're supposed to be getting another 1200 gallons. Clouds only are useless in my operation.

Today should be better. Supposed to be a cloud free day (as was yesterday so not sure what happened). Hoping the batteries get closer to fully charged today. If it looks like they won't make it with the off-grid PV I might try diverting some of my grid-tied PV output to give it a bit of a boost.

I just want to see a nice full production curve from the off-grid array. If that happens then I will count it as a success and know the batteries will soak up all the PV array has to offer.

We only have one small water collection tank attached to the secondary dwelling, it's 4000-5000 litres (~1200 gallons). Non potable, was meant for flushing toilet and clothes washing but I found the filter just wasn't good enough to remove the tannin stains in the loo so I switched it off as a supply for home use and it's really just there as backup supply for the secondary house garden. We have mains water supply. But of course the last two years we have had record rainfalls and so I haven't had to water anything for two years.

 

[Ozark Tinkering]

I'm only collecting water for my garden at this point. My well hasn't failed me in 30 years and I don't anticipate it going dry anytime soon, although my new neighbors have 750,000 thirsty turkeys and about 15 longhorns with no live water on the property. That does concern me just a tad. The longhorns busted out 2 weeks ago during the worst of our heat wave. I don't know if they want water from the creek down below or they were looking for green grass, but they were up and down the driveway several times.

 

[wattmatters]

Today I discovered my inverter's charge controller has the premature float bug. I suspected it might but the way my system has been used it's never really reared its head.

Worked my way through that with some manual charge parameter adjustments and now the new batteries are fully charged. I'll need to have a think about whether this is going to be a total PITA or be something manageable. I'll know more after tonight's test as to how the system responds tomorrow.

I think there will be a way to manage it in the not too distant future when Solar Assistant adds more inverter parameter control to its Home Assistant MQTT broker. That would enable me to set up some automations for adjusting float parameters, a sort of virtual partial BMS. I can use the Victron Smartshunt data for that.

Today's charts:


Tonight from 6:00 PM I am going to flip over to run the house from the off-grid system and see how that goes. See you on the other side.

 

[wattmatters]

After running the home from the off-grid system overnight the hybrid pack behaved just as expected with all the charge being supplied by the LiFePO4 batteries and the lead battery just sitting there doing nothing wondering what all the fuss was about. Perfect.

The shunt said 114 Ah were consumed, which if all that came from the LiFePO4 should = a 43% SOC. The charge status lights on the new units tell me the batteries were both in the 33%-50% SOC range, so that aligns. Voltage at 6AM was 52 V. I don't know how much current leaks from the LiFePO4 to the Lead but based on me quick checks with the clamp meter, it's bugger all.

But now it's the morning I am back to charging the batteries.

Today I discovered my inverter's charge controller has the premature float bug. I suspected it might but the way my system has been used it's never really reared its head.

Worked my way through that with some manual charge parameter adjustments and now the new batteries are fully charged. I'll need to have a think about whether this is going to be a total PITA or be something manageable. I'll know more after tonight's test as to how the system responds tomorrow.

So I've decided to test my theory about the float bug.

The float bug occurs when the when the charge controller sees a combination of batteries at/above float charge voltage while current is low. It interprets that to mean the batteries are full and so the charger should move into and stay in float mode. But there is more than one reason why current might be low while voltage is at/near float level - the solar PV array in the morning might not yet be producing enough power, or some clouds pass over head.

So to check whether this is the case I have added charging from the grid (supplied by my larger grid tied PV system, no grid imports) to supplement the off-grid array charging and so the LiFePO4 are sucking down the juice* currently drinking ~ 47 A total, half each.

With charge current nice and high when the batteries reach the float voltage level, because the current is high enough, it should not trigger the float bug and charging should continue on its merry way with voltage climbing to the bulk/absorb charge level.

And that's exactly what it did - it blasted through float voltage, so now the batteries will definitely get a full charge:



I now need to see (tomorrow) if it will manage to push past the float voltage with just the off-grid array supplying power.

* meanwhile the Lead battery is only sipping 0.5 A. Again, that's perfect.

 

[TGPB]

Do you have any wind power? This would be interesting to see how the system would react with wind added to the mix.

 

[wattmatters]

Do you have any wind power?

No. Domestic scale wind would be a monumental waste of money here. Totally unsuitable terrain for wind energy.