Load Diversion advise needed

System

Using Magnum MS4448PAE 120/240 Inverter with 8 200 amp hour batteries in AC Coupled config with 15 Schuco 180's connected to 15 Enphase M190's feeding to the L1/L2 Hot Out on the Magnum. A second string of 3 180w with M190's feed to Main. Need to add Diversion Load in case the frequency shift setup in the "Custom" battery setup fails. Looking at using 3 HL-75 Heat to Air for load diversion tied to Morningstar Tristar 60. Any suggestions on proper way to wire the Morningstar to the battery side?

Cariboocoot

Re: Load Diversion advise needed

Off-grid inverter being back-fed with Enphase units? Who told you to do that?
Just my opinion but ... you are asking for disaster.

Photowhit

Re: Load Diversion advise needed

Just out of curiousity...

What are you charging your batteries with?

System2

Re: Load Diversion advise needed

Magnum folks advised me that PAE supports AC-Coupled with 4.1V firmware and not exceed 80% of inverter capacity. They signed off on the design. Several others have been running this way for a while. Batteries are charged off the same inverter.

BB.

Re: Load Diversion advise needed

"Back Driving" the inverter:

AC Coupled config with 15 Schuco 180's connected to 15 Enphase M190's feeding to the L1/L2 Hot Out on the Magnum

I have not read the Magnum manual/website--But I assume that you have confirmed that this is OK and they (apparently) do frequency shift when the batteries are fully charged?

Inverter has a transfer switch and battery charger--so you skip the charge controller with the Enphase units.

Is an interesting way to go.

The MorningStar TS 60 (not MPPT model--should be cheaper with PWM), setup as a diversion controller. Other than connect following the rules in the TS manual (PDF) I am not sure what your question is. Note that they appear to derate the controller from 60 amps--so double check the ratings meet your requirements.

You probably will have to watch how the various charge controllers "play" with the diversion controller...

With normal series charge controllers connected in parallel to a battery bank, there is not any real problems with power sharing... Which ever controller is set to the highest voltage at that moment, will be the "winner".

With a shunt controller, you could, for example, have the Magnum charger trying for 14.5 volts because it thinks the batteries need recharging (bulk/absorb/equalize mode). And the shunt controller is trying to maintain 13.8 volts for float charging... So now you have a potential waste of power on your hands--The Magnum is dumping current to charge the batteries and the TS diversion controller is dumping power to keep the batteries from over charging...

I assume this is a "Grid Tied" utility powered system with the Magnum as a backup if the grid goes down.

So, I would assume the off-grid power will be relatively rare and I would set the TS diversion to ~14.8 volts (no float, no equalize) and set the Magnum (and any other solar charge controller) to 14.5 volts maximum or so. (just a guess--I have not read the manuals in detail).

Or, forget the diversion controller, and get a solid state relay (or mechanical) that you put on the output of the GT array. Connect the relay to a voltage controlled switch set for ~14.8-15.0 volts... It will not be temperature compensated, but it will at least be your "kill switch" if something goes wrong.

I guess you could also get a small TS or other model diversion (smaller, less expensive) controller and drive the relay with (hopefully) a temperature compensated controller if your battery bank is not in a conditioned space.

My 2 cents worth of thought.

-Bill

System2

Re: Load Diversion advise needed

Bill, Thank you for all your comments. Yes, the system is grid-tie and the Magnum
PAE with V 4.1 will auto frequency shift via the "Custom" battery setup (referenced in their docs as "AC coupled Support Mode"), (Ref: P/N: AN0002, AC Coupling 05/2010). This mode, when activated, causes the inverter output frequency to change to 60.6Hz. To setup the inverter the "Custom" battery settings are set at - Inverter Voltage = 48VDC, Absorption = 57.0VDC, Float = 53.0VDC, Equalization = 55.0VDC via the remote display. The shift activates when the battery voltage increases 4-volts above the Absorb Voltage setting. The frequency returns to 60.0Hz when the battery voltage falls 4-volts below the Absorb Voltage setting. Magnum calls this setup their "Micro-Grid Application". I spent many days validating this setup with all of the Magnum technical folks and close to 100 phone calls locating others who have this setup and have been running for several months (all of which report no problems). Last week I sent the electrical schematic depicting all of the wiring and Magnum approved it. (We went with the enphase micro's due to shading issues during the winter).

Again, thanks for your feedback.

Monte

Photowhit

Re: Load Diversion advise needed

I'm sorry, I was/am completely clueless that you could charge running back through an inverter, Wow...

Way out of my league...

BB.

Re: Load Diversion advise needed

Yea, it is really weird... But many inverters (all, some, few--I don't know) will act like a synchronous rectifier when connected to a Grid Tied inverter.

Inside the inverter, we are used to the current flow from the higher voltage (inverted battery energy) out to the loads--but it turns out that if you feed current back (in phase) with the inverter's voltage, you can push energy backwards through the inverter and recharge the battery bank.

The big issue is that when first done in paper spec's/advertisements (I saw Southwest Wind was doing this with an off-the-shelf Outback TSW type inverter), the issue was there was no charge control and the battery bank could be overcharged (or at least until the DC input over voltage cutout circuit shut down the inverter).

Now, a couple years later, they are adding smarts to the inverter to allow it to shift its frequency >1% off standard frequency (my guess) which then causes the GT inverters to fault (turn off) and there is now no more charging done. When the OG inverter decides the batteries can be recharged, it brings the frequency back to specifications.

As I understand, this is also how the Sunny Island inverter/charger setups work too--Although, they may have some more range of control vs the On/Off of this type of setup.

Solar Guppy said that from his experience the Off Grid inverter needs to be TSW for the GT inverters to properly synchronize/qualify the AC Grid as "Good".

This setup allows you to use GT + OG inverters without having to pay for a separate DC Charge Controller for the solar panels and you also do not need a Hybrid inverter that is capable of both GT and OG operation (like an XW Hybrid inverter--Which, by the way, with new firmware (?) can also perform this "AC Coupled" function).

The way to brute force a charge controller is to use a voltage comparator on the battery bank that controls the GT inverter output and either use the "inhibit" GT inverter signal (if present) or use an AC relay in the GT Inverter's output to cut the "Grid Power" to the GT inverter (fault it). This can also be used as a backup charge limiter (instead of using a diversion controller on the battery bank as primary or redundant battery charge control, which can have its own issues as I talked about above).

-Bill

Photowhit

Re: Load Diversion advise needed

Thank you for the wealth of information, I still have too many questions, and since it appears I'll be off grid for life, I'll religate this to my curiously interested file, until curiousity over comes me...

stephendv

Re: Load Diversion advise needed

Monte,

One of the characteristics of a system like this is that it causes "micro-cycling" of the batteries when the frequency shift is activated. I.e. Batts hit absorb so all your PV all goes offline, then you draw from the battery even though there's full sun outside, until you hit the trigger value, then all the PV comes online again - and it recharges. Not ideal for the battery life. There's a good paper describing this configuration and the effects of micro-cycling here: http://www.studer-innotec.com/upload/temp/Partial%20AC-coupling%20in%20Minigrids.pdf

The other interesting point they make about AC coupling is that the system can end up in a lockedup state under specific circumstances. E.g. if for any reason the batteries are drawn down too far and the magnum shuts off because of low voltage; then even if there's full sun the next day the batteries won't charge because the magnum is still stuck in low voltage disconnect, so the enphase's won't come online. Might not be that much of a concern for you, depends on how you use the system and whether you'll always be there and/or have an automatically activated genset to charge the bats.

For these reasons I'd consider going with a partially DC coupled system with maybe 200-1000W coupled in a standard DC configuration through a charge controller. This could help reduce the micro-cycling and give you some security that the system won't end up in a lockedup state.

Do you know whether the frequency shift of the magnum is variable? i.e. does it varie between 60 and 60.7 depending on how much surplus power there is on the AC side?

EDIT: And are you using a backup generator through the magnum, and if so does it operate at 60Hz? AC coupling can sometimes cause problems if the GTIs backfeed into the generator.

Cariboocoot

Re: Load Diversion advise needed

There have been other threads here discussing this same question, and running in to the same problems. Add those to Stephen's remarks and you can see why I come to the conclusion "asking for disaster". For the average Joe, this one falls under the heading of "don't try this at home".