Adding a separate MPPT to an inverter system

Hi There,

In my setup, I two two strings of panels on east and west side of my roof. Each string has 6 approximately 400W panels serially connected. I would like to ensure that each string gets its own MPPT for optimal performance. I have previously found out that when I directly connected these strings in parallel, I ran into a malfunction in one of the strings (most likely bypass diodes in the strings). So I now want to do things right to avoid a similar future problem.

The issue is that my solar hybrid inverter has a single MPPT module. How can I best add the second MPPT to my system? It seems to be suggested that the second MPPT output should be directly connected to the battery. Doesn't it create a weird configuration where one string goes to an inverter and then to a battery and the other string directly goes to the battery (I mean after the MPPT). I guess if I make this connection, the inverter would be fully unaware of my second string which goes to the battery directly. Wouldn't it cause a problem?

Or would you suggest that using blocking diodes in a combiner box a better option for my system?

Thanks for any insights.

Hi Ahmet,

If I understand your description correctly, your first string doesn't actually go into an inverter,  it goes into an MPPT charger that happens to be packaged with an inverter. It's the same as if it were an independent charger.

You can,  therefore connect a second charger (I have two chargers on my system) but you want to make sure the charging parameter settings are identical to the charger in the inverter system.

Hi Nathanael,

Yes, the inverter is an all-in-one type inverter which includes MPPT. I do not see this MPPT but it must be inside the inverter box.  The description of the inverter states "Mppt 90-500v Pv Input 100a" , which I understand to contain an MPPT with the given specs.

If I now buy a second MPPT with the same voltage and current settings (I guess it doesn't have to be the same, but it must match my panel specs) AND program the MPPT to use the same charging parameters as the built-in MPPT that comes with the inverter, I should be good to go, right?

The only thing that I do not understand is that, the inverter would be unaware of this second MPPT. But I guess it is not a problem.

Thanks,
Oguz

You also need to figure out what the "bulk", and "float" settings are on your first charger so that the second charger can operate synchronously.

You are correct that the second charge does not effect the inverter; the inverter is "ouput" from the batteries (unless you charge with a generator or grid power) and the chargers are "input".

What's the brand/model of your All in One inverter?  With some brands, they make more than one model of either inverter or a stand alone MPPT charger.  Some brands are designed so that they can communicate with each other via a serial connection, and if your brand supports that, it might be able to slave a second controller to the AiO?  Some controllers, such as my Midnite 200, have this function, and controllers can be placed in parallel, but without knowing your brand/model, that's just guessing.

The second option, already suggested, is to just wire in a second MPPT controller in parallel, with the bulk/absorb/float setting identical to those of the AiO.  That is likely to be good enough.

Think about which string you want on the all in one, especially if you have lead batteries (not nearly as important with lithium). There is higher efficiency when the inverter can convert directly from solar instead of charging the batteries and then pulling the power out for the inverter. Choose the string that gets direct light when you use the most power—not sure when that is for you, is it in the morning doing laundry, making lunch, or in the afternoon making dinner or working in the shop or charging all your cordless tool batteries after working all day.

R Scott wrote:There is higher efficiency when the inverter can convert directly from solar instead of charging the batteries and then pulling the power out for the inverter. Choose the string that gets direct light when you use the most power

I don't think this statement is correct.  As far as my understanding goes, ALL solar power passes through the controller into the batteries, and then All power comes out of the batteries to fuel the inverter.

What you might be thinking about is system potential, instead of battery voltage?  When sitting idle at night time, with no power coming in, the system potential is the voltage of the battery.  In daylight however, the system potential is the battery voltage, plus the charging voltage.  So, for example, a fully charged 24V battery bank at 8pm might be 25.4V, but the system potential would be ~28-29V towards the end of the absorption phase during the day.

The system potential becomes very important when you're trying to run a very big load, such as a 240V well-pump.  A BIG load like the well-pump could cause so much voltage drop that the inverter shuts off from a low-voltage warning.  While charging, the higher system potential prevents inverter shutdown because the battery has to first drop from a higher potential, and because power is coming into the batteries, the voltage sag will not be as great.

Michael Qulek wrote:
The system potential becomes very important when you're trying to run a very big load, such as a 240V well-pump.  A BIG load like the well-pump could cause so much voltage drop that the inverter shuts off from a low-voltage warning.  While charging, the higher system potential prevents inverter shutdown because the battery has to first drop from a higher potential, and because power is coming into the batteries, the voltage sag will not be as great.

Very good to know as powering a well-pump is a top priority for our homestead and I suspect of that of many others!.....

It is not 100% necessary to have multiple MPPT charge controllers communicating with each other, or with the inverter or BMS, if you have a BMS.

The main thing to look out for is excessive charge current. If your battery, wiring, fuses are all large enough to handle the maximum input from all connected panels/chargers (on a cold sunny day with the thermal coefficient of the panels considered) simultaneously, then the MPPTs will not cause problems with excessive charge current, as they simply can't produce excessive charge current.

If this is not the case, you would have a problem to address.. you'd need some way to avoid excessive current.


It might also be a good idea, if the above is not an issue, to configure slightly more conservative charge parameters in all but one charge controller. The idea being to have the other controller/controllers end their charge first, so that around max voltage only one charger is operating, in order to avoid confusing behavior as multiple chargers try to manage the relatively low current/high voltage gain end of charge behavior. This may be completely unnecessary, but can probably be determined experimentally..

Michael Qulek wrote:

R Scott wrote:There is higher efficiency when the inverter can convert directly from solar instead of charging the batteries and then pulling the power out for the inverter. Choose the string that gets direct light when you use the most power

I don't think this statement is correct.  As far as my understanding goes, ALL solar power passes through the controller into the batteries, and then All power comes out of the batteries to fuel the inverter.

What you might be thinking about is system potential, instead of battery voltage?  When sitting idle at night time, with no power coming in, the system potential is the voltage of the battery.  In daylight however, the system potential is the battery voltage, plus the charging voltage.  So, for example, a fully charged 24V battery bank at 8pm might be 25.4V, but the system potential would be ~28-29V towards the end of the absorption phase during the day.

The system potential becomes very important when you're trying to run a very big load, such as a 240V well-pump.  A BIG load like the well-pump could cause so much voltage drop that the inverter shuts off from a low-voltage warning.  While charging, the higher system potential prevents inverter shutdown because the battery has to first drop from a higher potential, and because power is coming into the batteries, the voltage sag will not be as great.

That might depend on the model of all in one, but mine can be set up to run the inverter direct from solar if available and then divert excess power to the MPPT. I can’t find the video now but I think it was Will Prowse that explained it better.  You gain efficiency because you bypass the MPPT inefficiencies. For lead acid, there was an additional gain as well.

Ahmet Oguz Akyuz wrote:Hi There,

In my setup, I two two strings of panels on east and west side of my roof. Each string has 6 approximately 400W panels serially connected. I would like to ensure that each string gets its own MPPT for optimal performance. I have previously found out that when I directly connected these strings in parallel, I ran into a malfunction in one of the strings (most likely bypass diodes in the strings). So I now want to do things right to avoid a similar future problem.

The issue is that my solar hybrid inverter has a single MPPT module. How can I best add the second MPPT to my system? It seems to be suggested that the second MPPT output should be directly connected to the battery. Doesn't it create a weird configuration where one string goes to an inverter and then to a battery and the other string directly goes to the battery (I mean after the MPPT). I guess if I make this connection, the inverter would be fully unaware of my second string which goes to the battery directly. Wouldn't it cause a problem?

Or would you suggest that using blocking diodes in a combiner box a better option for my system?

Thanks for any insights.

Ahmet, what kind of batteries are you running on your system? Are they lithium and are they in communication with the inverter? I ask because it would affect how you connected the second charge controller. If I was doing it i would hook up the second charge controller not built into the inverter to the solar string that receives sun in the morning. That way as the afternoon sun hits your all in one charge controller which you can monitor better it is able to top off the batteries. If you are using lithium your settings will be very important since the second controller will not be communicating with the batteries and will be relying on battery voltage readings alone which is never a great thing with lithium. Charge at too high voltage and the bms will shut down the bank.  
Cheers, David

Hi David,

I am using Lithium (LiFePO4 to be specific) at 48V. The battery box has a BMS in it and I can monitor the individual cells (there are 16 of them) using an app. But the inverter has no communication with the BMS -- it is an hybrid inverter rated at 6.2 kW. So I think my inverter is already relying on the global battery voltage readings from the poles of the battery. Is this a problem?

If I add a second charge controller, which in the mean-time I realized to be quite an expensive option, it would also connect to the battery poles directly and would sample the battery voltage from there. But I will make sure to enter the same battery settings for the existing inverter and the new MPPT.

I do have a question regarding the MPPT choice. Given that my panel specs are 320W-410W 32V-40V 7.7A-10A and I will connect 6 of these in series, can you recommend me a budget-friendly MPPT that will work with them. Victron models are unfortunately very expensive.

Also, I noticed that inverters can consume non-negligible battery capacity at night. The aforementioned 6.2 kW inverter easily eats up around 15% of my 100 Ah battery. Do you know if charge controllers also consume such capacity?

Thanks.

Ahmet Oguz Akyuz wrote:Hi David,

I am using Lithium (LiFePO4 to be specific) at 48V. The battery box has a BMS in it and I can monitor the individual cells (there are 16 of them) using an app. But the inverter has no communication with the BMS -- it is an hybrid inverter rated at 6.2 kW. So I think my inverter is already relying on the global battery voltage readings from the poles of the battery. Is this a problem?

If I add a second charge controller, which in the mean-time I realized to be quite an expensive option, it would also connect to the battery poles directly and would sample the battery voltage from there. But I will make sure to enter the same battery settings for the existing inverter and the new MPPT.

I do have a question regarding the MPPT choice. Given that my panel specs are 320W-410W 32V-40V 7.7A-10A and I will connect 6 of these in series, can you recommend me a budget-friendly MPPT that will work with them. Victron models are unfortunately very expensive.

Also, I noticed that inverters can consume non-negligible battery capacity at night. The aforementioned 6.2 kW inverter easily eats up around 15% of my 100 Ah battery. Do you know if charge controllers also consume such capacity?

Thanks.

so, there are a lot of variables. One thing i would try to figure out is if you are better off getting a second charge controller or upgrading your all in one to a larger unit that has more mppt capacity. Sell off the old one to help offset the cost. If you upgrade try for one with closed loop communication to get the most out of your panels and your battery. As to brands; that changes depending on where you are in the world. Most of the higher voltage chargers are made by two or three chinese companies then are rebranded into dozens of different names. I stick to gear that has been certified to a certain standard but it adds cost. As to losses the ways the all in one units work means you will burn 50-100 watts per hour just keeping it running. The mppt adds very little to that. Its the inverter portion of the unit that uses most of the stand by losses. Good luck on the journey.
Cheers,  David

Ahmet Oguz Akyuz wrote:Hi David,

I am using Lithium (LiFePO4 to be specific) at 48V. The battery box has a BMS in it and I can monitor the individual cells (there are 16 of them) using an app. But the inverter has no communication with the BMS -- it is an hybrid inverter rated at 6.2 kW. So I think my inverter is already relying on the global battery voltage readings from the poles of the battery. Is this a problem?

If I add a second charge controller, which in the mean-time I realized to be quite an expensive option, it would also connect to the battery poles directly and would sample the battery voltage from there. But I will make sure to enter the same battery settings for the existing inverter and the new MPPT.

I do have a question regarding the MPPT choice. Given that my panel specs are 320W-410W 32V-40V 7.7A-10A and I will connect 6 of these in series, can you recommend me a budget-friendly MPPT that will work with them. Victron models are unfortunately very expensive.

Also, I noticed that inverters can consume non-negligible battery capacity at night. The aforementioned 6.2 kW inverter easily eats up around 15% of my 100 Ah battery. Do you know if charge controllers also consume such capacity?

Thanks.

Since your battery is LiFePO4, is there any risk of freezing temperatures where you are? If so, does your BMS have a functional low temp shutoff to prevent destruction of the battery by charging beow freezing?

If this is a possible issue, seeking out MPPT controllers with temp sensing/limiting options may be desirable. I am runnning cheap LiFePO4 batteries without temp protection, relying on the Victron MPPTs for this.. You may not need to worry about freezing temperatures, though?