The context of system is that was put together by my father, and I'd like to optimize this system.

There is 13.5kW of PV panels.

1200AH of 4 x 12V silver calcium sealed torque batteries ( Yes, I know wrong type, but due to a previously incorrect charging voltage seem to lasted nearly 3 years).

2 outback 60's & 2 outback 80's

One connected 48V 5KW Quattro purely fed by the battery.

This all supplies their granny flat on my sister's property rather nicely. The granny flat is also equipped with LED's, gas & EV geyser.

They can chop over to a an ESKOM single phase unmetered supply that tee's off my sister's metered 100 kVA 3 phase supply, only do during heavy usage in prolonged inclement weather.

They have no desire to ever parallel with the ESKOM supply. My sister has a 6kW  1ph generator that they would like to auto-start for her though. They would also like to pick up some of the sister's 200kWh/day usage to ease her electric bill, but rather by supplying certain specific loads.( Maybe an entire phase?). They are in the Drakensberg, on a rural supply so power cuts for every thunderstorm.

They have acquired a second 48V 5kW Quattro, but I have to commission it,  (my mother's stipulation).

I am in Ireland, but visit my parent's in SA annually.

I have been reading up on paralleling quattro's, and conditional on them having the correct models so that I can match the firmware, I am pretty confident I can do it.

Now for my question, should I do it?

I think that 1 quattro supplying the other quattro is a far more flexible arrangement.

It could allow for two standby generators ( granted only one at a time, but two independent gen starts).

It could allow for two separate battery banks, (which is important if one wants to gradually ease into the pylon tech route), but still use this battery bank until it starts to die.

Or alternatively, use pylontech batteries as the cycling batteries, with a really low discharge cycle of the lead acids, but their when you need them.

I  stand to correction but I also think it would allow for 3 independant stages of load prioritizing, as if there was an AC1, AC2 & AC3out.

Also think that because of the 2 Quattro's ability to pass through and back current there would still be 10kW available at any/all of the AC outs if they were on.

It seems such an obvious way of configuring two quattro's, that I think I must be missing something.

What shortcomings can anyone else see?

 

 

 

 

 

 

My head is a bit flat today to picture what you're aiming for. Daisy-chaining one unit after the other is not a very common configuration, that's about all I can say for now.

Lets say priority load AC out on Quattro 1 onto some load 1 the AC in of Quattro 2.

Quattro 2  set to export back to Quattro 1 as well. ( What country Grid setting now?)

Other AC out of Quattro 1 onto another load 2

Quattro 2 in turn has a two stage priority AC out. ( say load 3 and load 4)

Quattro 1&2 either fed of one battery or two different batteries with two different charging curves.

The idea being that load could be ditched or picked up at a more granular level during a power cut.

Say, Granny flat  high priority load,  sister high priority , gen start, granny flat low priority, sister low priority, something like that.

Normal usage: maybe based on different SOC's of a lithium bank and a LA bank, so as to deliver normal by light usage of the LA's but to use the lithiums deeper discharge and possibly avoid the downside of the lithium switching off.

Backup Gen if first Gen fails to start etc. Can more than 1 Generator be configured on an AC in of one Quattro.

Just an idea at the moment....

 

14 minutes ago, phil.g00 said:

Just an idea at the moment....

Don't see in theory why it would not work. Just one thing I am not sure of: If you put a grid-interactive inverter downstream of another one, it must reduce power if the frequency shifts up. I'm actually not sure how to do that with a Multi/Quattro.

"it must reduce power if the frequency shifts up."

I can understand an ac-coupled Fronius PV inverter having to to this, and if I added ac-coupled PV to the system it would still work as essentially all the AC would be synched to Quattro 1.

What I'd have to make sure of is that the 1:1 rule wasn't broken inadvertently, when the Quattro to Quattro AC was disconnected if I added PV inverters.

This would happen say when there was the disconnection of a heavy load.

However, I see the Fronius as a generator with inertia pushing power onto a grid as opposed to the Quattro supplying power that is being sucked out of a battery.

So on the face of it I don't see a problem that is specific to down-streaming an inverter.

 

 

 

 

 

I guess if there's no fundamental runaway or an some sort of unstable ringing ringing effect, this configuration will far more flexible.

Sure there will tricking around with settings and inputs and outputs, but that just detail tweaking.

Another advantage of doing this is the ability to mismatch equipment, like the stringent requirement of identical quattros etc.

Maybe a Multigrid and a quattro?

I don't know about this. You will normally install either the PV-inverter or the ESS assistant on an inverter that has another generator on its output, and this gives it the ability to shift the frequency and throttle the downstream generator. Now you have a Multi downstream from another Multi. That Multi has to know to reduce its own power when the frequency rises. It doesn't. There is no assistant to do this. A setup like this will work fine as long as the grid is there, but if there is an outage then one inverter will back into the other. The upstream one will generally push the frequency up to 54Hz, and if you configure the grid parameters correctly that should cause the downstream one to just shut down completely. Should. It is not designed for this. If you run into trouble there will be no support.

Ok, lets take this step by step.

Forget about PV inverters for the moment.

Given: In stand alone off grid mode, only fed from a battery a quattro matches its output to the load, at a stable frequency.

Even on grid, the grid frequency is not used to curtail ouput power. It is the load that pulls the current.

Not that I've personally tested, but from what I've researched, an inverter is pretty good at maintaining 50Hz.

The fundamental operation is that it doesn't push electricity at a load thereby raising the frequency, it maintains the frequency/voltage relationship whilst varying the current within its rated range.

 Why do you say the frequency will head to 54Hz?

In other words, why will the upstream inverter not act as the grid up to its rated output?

I am asking because although I have a thorough knowledge of electrical fundamentals,  inverters are not my area of expertise.

There may very well be some kind of attempt to raise the frequency of the grid by the second inverter to let the inverter know the grid is actually there.

Then again I believe there is a allowable configuration that will allow a PV inverter on the AC in side, and if a third party PV inverter is permissible I don't see how an own brand inverter would be unsupported.

Please don't think I am being testy, I have read this forum top to bottom and I know that you are an authority on this stuff, and thank you for taking the time to discuss this.

 

 

 

 

 

10 hours ago, phil.g00 said:

Given: In stand alone off grid mode, only fed from a battery a quattro matches its output to the load, at a stable frequency.

What I am thinking of is that you might run both inverters in ESS mode, grid-connected. If not, if you will be running them purely off-grid and the one chained from the other, then of course none of this applies. But you did mention pushing power back up the chain somewhere... or I think you did.

The reason why I bring up PV inverters is you have to use the lessons learned from that scenario. If you have a downstream Quattro running ESS, it is a potential generator on the output side of the upstream Quattro, and ideally it should behave like the other generators we know, aka PV inverters.

10 hours ago, phil.g00 said:

 Why do you say the frequency will head to 54Hz?

When you run either the PV-inverter assistant or the ESS assistant on a Multi/Quattro, and the grid is disconnected, and it has other generators (usually a PV inverter) tied to its output, it will automatically raise the frequency if there is too much power being injected by the downstream generator. This is an industry standard that operates between 50Hz and 54Hz. At 50Hz you're indicating that you want full power, and usually by 53.7Hz the power will be reduced fully to zero. 54Hz is an indication that you want all generators to shut down immediately.

Now my point is that while the Quattro/Multi can act as a controller for other generators, I don't think it can be controlled as such a generator. It will shut down at 54Hz (or whatever the grid code mandates), but there is no gradual control as with other generators.

10 hours ago, phil.g00 said:

Please don't think I am being testy

Oh not at all! I, on the other hand, worry that I'm dismissing a possibly perfectly-working example on the basis that "this is not how we normally do it", which is true of course :-) But I really do have concerns about this.

Even if it does work I am unsure if it will be very stable. I've tied a Ziehl UFR1001e (passive anti-islanding switch) to the output of my Multi once, and it was unhappy with the quality of the power. On the test bench where I have a Fronius tied to the output of a 3kva Multi, it is almost impossible to get it stable without the Fronius constantly disconnecting whenever there was a change in load which caused a slight disturbance in the output of the Multi. 3kva (in this example) is a rather weak signal compared to a grid, so it is understandable, but in similar vein I would not be surprised if one Multi tying with another one upstream is similarly unstable.

Yes, you're right I did say that.

And I can see your logic, that if the  downstream inverter is set to export into the grid, and if upstream there is insufficient load and it's battery bank is charged there is no mechanism to curtail the power backfeed.

Unless zero backfeed power is set on the downstream inverter. 

Which would mean that only 5 kva would be available at the ac out of the upstream inverter.  

10 kva should still be available  from the downstream inverter.

I don't like the idea of feeding back onto the ac input of the upstream inverter, that would seem to create the chance of a run away loop.

So maybe that's a genuine limitation of this configuration:

A.) The upstream inverter has to charge its own batteries and can only feed 5kVA.

And B.) The downstream battery bank still gets charged by both and 10kVA is available.

I would still consider this a more flexible configuration, as I would still get a two battery choice and two  ac outs and possibly two gen starts.

I still wouldn't get that from a straight parallel in an off-grid set up.

 

 

 

27 minutes ago, phil.g00 said:

10 kva should still be available  from the downstream inverter.

You can do that if you use PowerAssist (take the difference from the batteries). I would not do it using ESS, because even if you tell it to not feed into the grid, that is a catch-up game that it plays, adjusting the setpoint according to the load, and for short periods it could still feed in in this mode.

So your downstream inverter will be configured in a simple backup configuration, with PowerAssist for loads over 5kva.

Yes, that sounds great.

Will the downstream inverter still work from the DC if the AC is lost in this scenario?

I know I would lose  AC2 out if I lost the AC in on the downstream inverter, but could I set the AC2 out to switch off at a certain battery voltage with the AC in still on?

In other words do I now have the ability to load shed non-essential loads and keep 10kVA available?

And one more question please while I am picking your brains, am I correct in thinking that the 10kVA available on the downstream inverter is the sum of AC1 out and AC2 out?

in other words, the 10kVA could come entirely from either ACout, if nothing was being drawn on the other ACout of the downstream inverter.

 

15 hours ago, phil.g00 said:

but could I set the AC2 out to switch off at a certain battery voltage with the AC in still on?

I can't remember. It might be possible to do it with a relay assistant. Install veconfigure, tell it to fake a target with full options, and try it yourself. You could of course also just use an additional contactor and just switch it with one of the relays in either the Multi or the CCGX.

15 hours ago, phil.g00 said:

keep 10kVA available?

I have no idea how power-assist will work when the grid source is another Multi/Quattro. As I said, you're dabbling in this space of "this isn't normally how we do it".

Edited October 18, 20186 yr by plonkster

It looks like a virtual switch operating a contactor is the way to go to shed AC2 load for the time being.

Although the VS route excludes assistants, and that would exclude PV inverters assistants in the future. So I'll have to have a think about that.

Victron themselves have indicated the power assist will do the job.

Further to my last it would seem the programmable relay assistant is tailor-made for this without a contactor, that presumably will allow the use of the PV inverter assistant as well.

And some further reading leads me to believe that this assistant would allow me control AC2 out , while not even having the prerequisite of AC in, even in the conventional parallel configuration. The conventional behavior is forgotten when an AC2 assistant is used. 

Which solves load-shedding requirement, without having to re-invent the wheel.

Thanks Plonkster for your assistance.