diy solar

diy solar

Incrementally Adding AC Batteries

This is an SMA AC coupled with a Radian using a DPST style ATS to have whole house backup rather than a critical circuit panel.
Essentially the whole house are critical circuits. In my neck of the woods and much of California, the meter and the main panel are integrated in such a way there is no room to intercept the feed from the meter to the main buss bar so a separate panel has to be installed.
If a bi-directional inverter is strange to you,
I think bidirectional or bimodal are terms often used to describe the modern crop of hybrid inverters. As you mention some of their ports can go in eitherbdirection.
 
so this idea still needs work.
The goals of the user need to be factored in to see if there is a market for such a device and the cost of such a device.
For example, I have plenty of inverter capacity but all I need is more battery capacity. I wanted a simple solution so it was a no brainer to buy more batteries. Now I can run my batteries at a lower SOC so there will never be any need to modulate any of my GT solar. To me that is better than having another device. Others may need more inverterv capacity and an AC battery may be just the thing they need if it is competitive with Encharge and others
 
It might be possible to "plug-in" more DC batteries to the DC side of an Encharge too to get more watt-hours.

But I suspect a lot of folks will want more of both from an Ensemble system.

During the day an Ensemble system has lots of watts as it's AC off the roof plus AC from the Encharge. But at night it's only the Encharge, which is only 1.28kW per 3 kWh battery. So, a 6kW array with 10 kWh battery gets you a max of 6+3.85=9.84 kW of inverter during the day, but only 3.84 at night. So, in my case I'll want both watts and watt hours.
 
Last edited:
Would a "standard" UL1741 compliant Grid-following inverter/charger work as the head for a "magic" AC Battery?

...

1593207511538-png.16236
Logic
  • No Voltage/frequency on L1/2, no output power, also satisfies anti-islanding requirements
  • No CT Tone received (Grid Active)​
    • Charge Mode​
  • CT Tone (Grid offline)​
    • Frequency Shifted - Charge Mode: max possible charge rate set by shifted amount (but of course might be less based on SoC)​
    • Frequency not shifted - Inverter timing source synchronizes (probably use the existing AC signal as transistor's base with cap offset for perfect phase alignment)​
    • Voltage > VMax (~247VRMS) - OverVoltage protection kicks in: charges battery if they can absorb the energy fast enough or if battery full dumped to waste heat. This needs to be fairly fast, it can occur when things like inductive motors stops suddenly (e.g., air conditioning turns off, see inductive kickback spikes). It only needs to sink as much as the inverter's maximum output and probably less than a second. Heat sink is sort of a worse case scenario, big caps and clamping voltage? Also, Jack Rickard said if the voltage on a standard inverter exceeded the output voltage, the current would back-flow into battery (e.g., collector/emitter reversed)...could probably make use of that with the right transistors (might be how existing charger/inverters work). This is a swag, don't know enough about this other than it's possible as existing inverters do have this built in (e.g., Outback, Enphase, Tesla).​
    • if Surge TSink > Thot and battery > 80% disconnect until Tsink < TLow
    • If Voltage <= VMax, Inverter mode, V set to match existing V or VMin, whichever is higher. If they have the same chemistry V should naturally stay in sink equalizing SoC for all the batteries in the system. But with VMin, the AC battery is guaranteed to exhaust before the Energy Storage controller - which is important since its the one providing the frequency reference.​
    • DC Voltage <= VLowDC disconnect​
The AC Batteries "power cord" is not a "suicide plug" because it won't output any current if it doesn't detect voltage, similar to any GT inverter.
Note: "grid-following" is a different inverter type than grid-forming (aka, off-grid inverter) or grid-tied.

If it's grid-following then:
  • it will stabilize voltage (e.g., voltage source) and frequency
  • If its UL1741 then it should know not to stabilize 61.5 to 63.17 Hz which is used to control renewable energy generation
  • Could know to recharge if frequency shifted (not available when frequency shifting isn't on the load center side of AC coupled systems)
  • Not grid forming, so if no grid is inactive (anti-islanding)
Seems to meet most the criteria with no need for CTs. Possibly missing surge/over-production protection and wouldn't know when on-grid to charge. AFAIK grid-following inverters aren't yet available for your average consumer's microgrid and not UL1741 compliant.
 
Last edited:
There is no AC battery. Any battery is DC. What you have is a system. Some systems can be combined, some can't.
 
Remember how those existing cheap GT inverters on Amazon had most everything needed to qualify for Svetz' magic AC battery with one pesky exception... that as charge sources they wouldn't hold back and drain the batteries as fast as they could once turned on?

Well, they're also made to accept a wide input of power (watts) ranges because solar panels offer a wide range of power.

So imagine a micro-controller in front of the batteries doing a PWM on the battery's output and adjusting the duty on the fly to control wattage output of the battery. Now the GT inverter is only outputting the amount of power the brain wants it to, from 0 to 100%. The microcontroller sets that based on frequency sensors and grid state.

Assuming a solar GT inverter is akin to the schematic @RCinFLA provided, then it might just be reprogramming the MCU inside, or hijacking the mosfet base via your own MCU. Use the battery side mosfets to control power input, and the output mosfets (not shown, but assuming they're somewhere) to synchronize frequency.

1597492362903.png
 
Last edited:
The moderator doesn't take well to criticism? My question is how do combined inverter generators, such as two Honda s, work. The same principal could be used to combine the DC - battery - with - inverter - system called "AC battery".
 
The moderator doesn't take well to criticism? ...
Nah, love criticism. Criticism is invaluable as it points out weaknesses/flaws; you can never have enough critics! Your post wasn't deleted for criticizing the moderator or the idea, in fact your post didn't criticize anything. It was just a repeat of post #45. Saying the same thing over in a thread without providing any additional thought/evidence/references is considered spam, it adds nothing to the conversation. That's why your post was deleted, apologies if the note explaining why it was deleted was insufficient explanation.

... the DC - battery - with - inverter - system called "AC battery".
It's not like I coined the term, don't blame me the term entered the vernacular. Even what you call a DC battery is a "system" comprised of cells. A battleborn battery for example is a series of LiFePO4 cells, temperature sensors, and a BMS. An AC battery just adds two more components, and inverter and a charger. But, by doing so removes the pains discussed in incrementally adding DC batteries (and adds a few of its own pains).

...l My question is how do combined inverter generators, such as two Honda s, work...
AFAIK, inverters and inverter gasoline generators that can synchronize the AC output use a "parallel" feature. Electrically behind the scenes a wire between them let's the inverters stay in phase.

This would be a great idea for someone building their own and not needing to interact with the grid or an existing AC battery as those devices already exist. The real question, as the op says, is how can we mix/match AC batteries from different vendors or build our own DIY version?

The grid doesn't have a "parallel" feature or a specific timing wire. But overall the concept is the same and obviously every GT inverter must match the frequency of the grid so must have a way to sense the existing frequency.

With a Tesla system you can add more Tesla AC batteries...they know how to talk to each other. Same for Enphase. Each of those companies has done the engineering and can handle the frequency shifting from CA Rule 21; but to add more battery capacity AFAIK you're locked into their product. That might not be true, any grid-tied AC battery should work together without solar. But I don't see how they can work well together without communicating with solar...someone has to be the master to set frequency curtailment and the other AC batteries need to follow.
 
Last edited:
  • Like
Reactions: Zil
Found a problem with the PWM approach.

PWM Recap
That approach used an off the shelf GT inverter as the base. Unlike a voltage source, a GT inverter is a charge source, so normally it would suck the battery dry as quick as it could dumping power whether you could use it or not (normally they convert all available power which goes into the grid acting as an infinite sink, or they shut down if the grid is off so as to not be a problem).

The PWM approach "fixed" that by putting a set of capacitors and power mosfets in-between the battery and GT inverter to control how quickly the battery would drain. That is the battery would look more like a constrained power source and the GT inverters output controlled. It might look something like this:

1613912115837.png
The Problem
The original thinking was that surges would travel back through a GT Inverter into the battery (Jack Rickard demonstrated this, although it might not work on all GT inverters). But, the mosfets in the PWM subsystem should be off due to a voltage spike/surge (because as the surge starts to spike the GT doesn't need to supply power and so would close them), leaving nowhere for the surge to go.

This puts us back to where we started, to expand an existing AC battery system with different technology needs something that operates as a microgrid inverter (grid following or grid supporting), that is the frequency is synchronized but it acts as voltage source. Ideally something that's UL listed and less expensive than a full blown AC Coupling solution. Right now the IQ8 fits the bill in terms of being small, but it's not available and its unclear if it'll play nicely with other technologies. Hopefully the V2H designers will implement this in their first pass solution.
 
Last edited:
Desired Goal
Expand the power (kW) and/or energy (kWh) of an existing PV/AC Battery system in an incremental way. In an AC Battery system (e.g., Tesla, Ensemble) you cannot easily add DC batteries; but the system provides power if the sun is shining or batteries have charge regardless of the grid state.

Revisiting the Problem
During the day the solar panels are providing a lot of energy, hopefully enough to power all devices and recharge the batteries. So, there shouldn't be a shortage of kW or kWh during daylight hours. Any addition is therefore really for night use.

In this use case, there might be a short-cut solution: Use a "controlled" Grid Tied Inverter (GTI) feeding from a battery to connect to the microgrid at night only. At night, the solar panels do not create energy and therefore you don't have to worry about timing associated with throttling. At night, in a microgrid environment, the environment is identical to a voltage source off-grid inverter setup.

Controlled GTI
GTIs still have one big problem, since they're current sources if you hook them to a battery they'll convert the power as fast as they can, and if there's nowhere for the power to go they'll jack up the voltage, potentially setting fire to something. So, any GTI used in that fashion would need to be "controlled". In the prior posts this has a been a flaw as the system to control the panels also has a brain and there was no way to coordinate in a mixed-vendor system. By restricting operation to only when the panels aren't producing, it should eliminate the need for the brains to communicate.

Most commercial AC Coupled solutions use a relay to disconnect the solar until the batteries are low enough that the inverter can back-feed those surges into the battery safely. Enphase's Ensemble (and probably the skybox) throttle their inverters using something quicker than UL1741. A few posts up it was suggested a GTI's power consumption could be throttled via some sort of capacitor/PWM duty cycle to mimic panel output, although variable resistance might be as effective. So, properly controlling one is still a design problem; but seems way easier than trying to eat the whole elephant.

Looking forward to folks poking holes in the logic!
 
Last edited:
Appended this to the OP:
WARNING/UPDATE
Enphase does use frequency shifting from the grid to throttle its microinverters. But as that's relatively slow the IQ8s also use high-speed IP over powerline for throttling when offgrid; so much of the "hypothetical" discussion in this thread is incorrect in regards to Enphase.
 
Last edited:
The diagram I posted is for a LF hybrid bidirectional inverter. HF hybrid inverters works a little different with a high frequency DC to HV DC converter that is capable of switching power flow direction. The high voltage DC is PWM switched with additional output MOSFET bridge to create AC sinewave. HF hybrid inverters are not as rugged because the HF ferrite transformer in the DC to HV DC converter is vunerable to hard saturation of its core during power surges. If the HF transformer saturates the DC input switching MOSFET's current shoots up to very high destructive levels. It also has to make MOSFET drive control changes to change power flow direction. This may not sound like a big deal but the processsor controlling the MOSFET switching must recognize the backfeed surge, by 60 Hz waveform sensing, allow DC to HV DC converter to flush any continuous transformer current built up from before the surge and reverse the switching coordination to reverse the power flow. Meanwhile the HV DC filter capacitor must absorb any backfeed surge.

A LF hybrid inverter can continue, initially with its previously established MOSFET switch cycling, and it will just inherently push any backflow power surge, without any immediate attention by controller, to batteries which can take the surge a lot better than the HF inverter's HV filter capacitors.

The original LF hybrid inverter designs used a heavy power transformer modified for higher than normal leakage inductance. This was nothing more than coating the 'E' transformer core laminations with slightly thicker varnish coating to gap them a little more. Normally a power transformer wants the lowest possible leakage inductance as it acts like an extra series inductor with power transformer that causes some loss in transformed voltage. However, for a PWM switching inverter the extra leakage inductance of the transformer becomes the series inductor for an L-C low pass filter to reject the HF PWM chopping component, without the need for an extra separate series inductor.

Some newer LF inverters decided to use a normal power transformer due to volume pricing advantages and put in a separate inductor for the PWM L-C filter. The extra series inductor must be able to take the 60 Hz component of current without saturating the inductor core. The extra 60 Hz current acts like a magnetic bias on the inductor core that can reduce its high frequency inductance if core is not large enough to handle the extra magnetic bias, resulting in insufficient PWM HF switching component rejection. This filter inductor 60 Hz magnetic bias also applies to L-C filter on a HF inverter output. On some cheap inverters you can see 60 Hz waveform get 'ratty' when more load is applied caused by inductor soft saturation allowing more HF PWM to get through to 60 Hz sinewave output.

A few years back, some folks started the term 'AC battery' for DC to AC bidirectional hybrid inverters to differentiate them from a one way DC to AC inverter. I personally don't like the name, but it sort of makes abstract sense from a functional point of view.
 
Last edited:

Synchronverter

Synchroinverters are a special kind of smart inverter that mimics a synchronous generators and are able to avoid grid voltage frequency fluctuations and faults by responding dynamically and autonomously to changes in the grid by software means, and providing an inertial response in an extremely short time. ref
 
Back
Top