Author Topic: Inverter droop?  (Read 489 times)

WillEert

  • Full Member
  • ***
  • Posts: 176
Inverter droop?
« on: May 18, 2020, 11:44:41 AM »
I have been following off grid system design and equipment coming out of Australia. For larger off grid systems AC coupling using suitable inverters looks to me like the way to go. Some off grid inverters like the Victron or Selectronics "droop" as the load changes. This means that their voltage changes slightly as the load changes - Higher voltage at low loads and lower voltage at high loads. A string inverter such as the Fronius senses this droop and regulates itself proportionally. If it senses voltage above a certain setpoint then it will have ramped power output down, following the power reduction curve and shuts down.

I am hoping that the new Midnite Inverters have the droop characteristic. Bob?

Thanks,
Will
(1) 24 ET 185 mono, Classic 150,8x Surrette 6CS25p, 3x Magnum 4448, homecooked diversion system;
(2) 15 Heliene 330 mono, Classic 150; (3) 2 X 6 Hansol 345 mono, 2 stand alone Hotspot ACDC12C Mini-split heat pumps.

boB

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 3385
Re: Inverter droop?
« Reply #1 on: May 18, 2020, 08:51:33 PM »
Hey Will !

When you say "droop" do you mean that when they are loaded continuously (meaning, not just a quick surge) that the AC output RMS voltage actually drops and stays low  OR  do you mean that during surges such as, motors starting, that you see the lights dim (voltage drop) and then it comes back to normal ?

If  you mean surge, this is VERY important to make it as fast as possible.  The Outback FX inverters for instance did a great job at this !

But if you mean poor voltage regulation, then this should just not be an issue and no reason for that unless some kind of reference voltage internally is changing, which should be fixed in the design.  OF course, plain old over-loading might also cause this.

OK, so was it one of those things ?  Which one ?  Or both ?
K7IQ

WillEert

  • Full Member
  • ***
  • Posts: 176
Re: Inverter droop?
« Reply #2 on: May 21, 2020, 10:21:48 AM »
Hi Bob,

By "droop" I mean that as the load on the off grid inverter increases the output voltage decreases incrementally. In Australia, and maybe other jurisdictions, due to a large uptake of grid tied systems the utilities may experience excessive voltage increase when all the grid tie systems are producing. As a result of this they have the requirement that when grid voltage exceeds a certain set point that the string or micro inverters start to reduce their output. Eventually a system may shut itself down entirely on a bright sunny day in order to protect the grid from excess voltage levels. Australians complain about this.

Some off grid inverter/charger/charge controller companies are using this feature of the string inverters to control an AC coupled off grid system. If a string inverter is coupled into an off grid system at times of low load the string inverter must reduce its output. If it does not then the string inverter will eventually carry all the load and the off grid inverter will shut down. When this happens the string inverter also shuts down as it has no frequency to monitor. The off grid inverter now restarts to carry the load and after the time delay the string inverter also restarts. Then the cycle repeats.

Fronius string inverters are made to the Australian standard. There are others also. They make the settings for adjusting the voltage output reduction available for tweaking.  I have attached two documents that you will no doubt understand much better than I can - in particular the second one. One Australian company sells pre-programmed string inverters which are compatible with their off grid inverters. The off grid inverter must however "droop" the voltage as the load increases. This allows the string inverter to sense when the load diminishes by the voltage increase of the system. It then reduces it's output to keep the off grid inverter on line.

I love this system topology as it should make a diversion system very easy to control. As the system voltage rises but before the string inverter cuts back additional load can be applied.

So I hope this feature is included or can be programmed into Midnites new equipment. AC coupling for larger off grid systems really makes sense to me and a diversion or energy management system that optimizes that system has to be the way to go. It would be so simple to PWN a suitable relay bases on a voltage set point to send the "excess" energy to a suitable load or loads. What I do is that at a certain PWM % I turn on loads that cannot be PWMed such as an EV charger.

This as all about money. The more KW a system can make on an annual basis the lower the overall cost is with an increased ROI. It bothers me to see people with systems that once they fill the batteries up just stop and Float while burning fossil fuel for stoves, hot water tanks and their car (my rant).

Stay safe
Will


(1) 24 ET 185 mono, Classic 150,8x Surrette 6CS25p, 3x Magnum 4448, homecooked diversion system;
(2) 15 Heliene 330 mono, Classic 150; (3) 2 X 6 Hansol 345 mono, 2 stand alone Hotspot ACDC12C Mini-split heat pumps.

WillEert

  • Full Member
  • ***
  • Posts: 176
Re: Inverter droop?
« Reply #3 on: May 24, 2020, 10:46:04 AM »
Bob?
(1) 24 ET 185 mono, Classic 150,8x Surrette 6CS25p, 3x Magnum 4448, homecooked diversion system;
(2) 15 Heliene 330 mono, Classic 150; (3) 2 X 6 Hansol 345 mono, 2 stand alone Hotspot ACDC12C Mini-split heat pumps.

boB

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 3385
Re: Inverter droop?
« Reply #4 on: May 24, 2020, 09:50:29 PM »
Hi Will.   I was trying to digest what you were saying.

EDIT:  I looked at the Fronius document again and it looks like this method is Rule 21 stuff.  Yes, we are doing this but at the request of the utility which requires communication from the utility to the GT system with and/or without batteries.  So looks like what you are talking about has the GT inverter sell back (or convert) reactive power rather than real power.  OK, for the moment, on with what I posted earlier.....

I will be repeating what you said but sounds kind of like Australia has a slightly different system than I am familiar with because as you say, the string inverter (Fronius and others) will reduce their output as the AC line voltage goes above a pre-programmed value to try and regulate that AC voltage from going any higher.   This is a good thing to make happen and is easy to do of course.  SMOP -- Simple Matter Of Programming.

Usually how this works is that the off-grid battery based inverter (the reference) shifts its AC frequency upward by up to 1 or so Hz so that the grid-tie inverter can follow by reducing its output power proportionally.  Then, if the battery based inverter is keeping track of time gained by the increase in frequency, it can later on, reduce the frequency slightly to make up for that time gain and the grid-tie inverter will not change its power level.  This is what SMA did with their Sunny Island and the Sunny Boy string inverters.  Then, some other companies followed their lead and shifted frequency or responded by changing power output.  Doing it this way keeps the battery charging Absorb and Float voltages correct without having them go too high when excess power is available.

I am not sure if just having the grid-tie inverter drop its power level is the best thing to do all by itself because it wouldn't be optimum.  BUT if adjustment of the GT inverter is good, it could help a lot.  Just not the best situation.

We will only be making a small-ish GT inverter or two and I don't know if it has this feature programmed in but will forward this on to let them know what you have said.

Now the battery based inverter(s) should be able to do the frequency shift thing to tell the inverter to reduce power as well.

There is also Rule 21 and Rule 14H  (Hawaii) that requires control of the GT utility connected inverters from the power companies if the system is connected to public power which we are also aware of and implementing on our new stuff...  Although as far as the battery based inverters go, that will come a wee bit later.

Also the time it takes for one of those GT inverters to come back on is the typical 5 minute wait that is required by UL and IEEE 1547 here in the US.

The fact though that Australia has  them reduce the power of the GT inverter when line voltage goes up is interesting.

You can also use dump loads to keep from going over-voltage as well but I would rather have the battery based inverter tell the GT inverters how to reduce their power level in some manner.  I would prefer a separate method than shifting line frequency such as a cable connection or something like that.
« Last Edit: May 24, 2020, 10:01:01 PM by boB »
K7IQ

WillEert

  • Full Member
  • ***
  • Posts: 176
Re: Inverter droop?
« Reply #5 on: May 25, 2020, 09:22:17 AM »
Hi Bob,

Thank you for your reply.

I will have to think about what you wrote to understand what you said. What I understand is that there are a few methods of having grid tie inverters cut back. My Magnum 4448's do some kind of frequency shift. It does not act to regulate the output of the AC coupled inverter incrementally however if I understand how it works correctly. What the Magnum seems to do is a step change to shift the frequency in order to get the AC Coupled inverter to disconnect fully, based on battery voltage. Then after the battery voltage drops it shifts the frequency back to allow the AC Coupled inverter to reconnect. On / Off control to me is far less than ideal.

I can see how a DC  inverter/charger which changes the frequency in proportion to the load would work similarly to the Australian voltage system, assuming the AC coupled inverter used the frequency shift to also vary it's output incrementally as the frequency shifted. I suppose if a suitable frequency sensing device is available then once again a diversion system could be controlled using frequency.

So the question remains but changes: Will the new Midnite inverters be capable of something like this to allow them to be used for AC coupled off grid systems that would cause the AC inverter to reduce it's output proportionately? If it is just a matter of programming then to me it would add value to this product. I appreciate that the design, fabrication and certification process has likely taken far longer than Midnite could ever have imagined. As we used to say in the pulp industry when estimating how long a project would take: Double it and add 4. Likely this would not be a good time to divert resources away from what is going on the bring the new equipment to market however maybe if the units do not have this capability now it could go on a list for later.

Thank you.
Will
(1) 24 ET 185 mono, Classic 150,8x Surrette 6CS25p, 3x Magnum 4448, homecooked diversion system;
(2) 15 Heliene 330 mono, Classic 150; (3) 2 X 6 Hansol 345 mono, 2 stand alone Hotspot ACDC12C Mini-split heat pumps.

boB

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 3385
Re: Inverter droop?
« Reply #6 on: May 25, 2020, 03:28:51 PM »
Hi Bob,

Thank you for your reply.

I will have to think about what you wrote to understand what you said. What I understand is that there are a few methods of having grid tie inverters cut back. My Magnum 4448's do some kind of frequency shift. It does not act to regulate the output of the AC coupled inverter incrementally however if I understand how it works correctly. What the Magnum seems to do is a step change to shift the frequency in order to get the AC Coupled inverter to disconnect fully, based on battery voltage. Then after the battery voltage drops it shifts the frequency back to allow the AC Coupled inverter to reconnect. On / Off control to me is far less than ideal.


You got that right !  Far from ideal !  BUT it is s quick fix for over-charging the batteries and also over-AC-voltaging connected appliances.

I can see how a DC  inverter/charger which changes the frequency in proportion to the load would work similarly to the Australian voltage system, assuming the AC coupled inverter used the frequency shift to also vary it's output incrementally as the frequency shifted. I suppose if a suitable frequency sensing device is available then once again a diversion system could be controlled using frequency.

I think I like that idea.  The GT inverters of course have that sensing.  They have to.

So the question remains but changes: Will the new Midnite inverters be capable of something like this to allow them to be used for AC coupled off grid systems that would cause the AC inverter to reduce it's output proportionately? If it is just a matter of programming then to me it would add value to this product. I appreciate that the design, fabrication and certification process has likely taken far longer than Midnite could ever have imagined. As we used to say in the pulp industry when estimating how long a project would take: Double it and add 4. Likely this would not be a good time to divert resources away from what is going on the bring the new equipment to market however maybe if the units do not have this capability now it could go on a list for later.

Thank you.
Will

Yes, of course we will make sure we design in any of these methods that make sense....  And what we have talked about there definitely make sense.

Also, you are absolutely correct about the design fabrication and especially certification takes time and money.

Certification itself for all of these new products we are working on will end up costing hundreds of thousands of dollars.

Some refer to these outfits as the mafia, LOL.

boB
K7IQ