Design Using DC Optimizers

How do you suggest we model a PV system which uses a single "string" inverter with DC optimizers on each module (such as SolarEdge technology)?

To clarify, I like to enter specifications for inverters directly using the "Inverter Datasheet" method rather than relying on the CEC database, so should I input voltage, power, and current data from the DC optimizer datasheets and use the same number of "inverters" as modules in the System Design with 1 module per string similar to a microinverter design?

Or, use the AC/DC inverter specification with very low (0) and very high (1000) MPPT tracking voltages?

I could also multiply the min/max MPPT values for each DC optimizer by the number of modules in each string to get a pseudo-MPPT range...

Curious what the developers recommend.


Thanks,
designerd

Dear designerd,

In the end, its is the number of modules per string, strings in parallel, and inverters that determines how SAM models the system. So, whether you use the DC-to-AC ratio and nameplate capacity to size the system, or explicitly set the numbers of modules and inverters, it is the numbers that are important. So, I would recommend setting the numbers yourself to have the most control over the system design.

Also, when you use the "Inverter Datasheet" option instead of the CEC database option, SAM ignores the "C" coefficients, which are for a set of polynomial equations that make small adjustments to the inverter output. In general, if you have the choice, I would recommend using the CEC database option when you can.

Best regards,
Paul.

Hi Paul,

Thanks for the reply. Though, my question was not regarding the size of a system, but how to model a system which uses a string inverter with DC/DC optimizers, specifically regarding how to handle the MPPT tracking.

Let me try to clarify:

• The manufacturer does not list an MPPT voltage range for the string inverter itself.
• The DC/DC optimizers do have a MPPT voltage range listed in their datasheet.
• The manufacturer states that the MPPT tracking is handled by the DC/DC optimizers at the module level, and losses should be minimal.

So, since I cannot use the specifications for both the string inverter AND the optimizers, there must be a "best practice" sort of middle ground for modeling such a system using SAM. I can think of 3 options:

1. Input the power, voltage, current, efficiency, and MPPT specifications for the DC/DC optimizers and model the system like a microinverter system ("strings" in parallel = # of "inverters" & modules per string = 1)
2. Use the power, voltage, current, and efficiency specifications for the string inverter, but set the MPPT voltage range very large (so MPPT tracking losses are minimized) and use the "proper" # of strings in parallel, # of inverters, and modules per string.
   Or,
3. Use the string inverter specs as in option 2, but multiply the Max/Min MPPT voltages from the optimizer spec by the number of modules (and therefore optimizers) in each string, thereby developing a "sort of" MPPT voltage range.

Regarding the use of CEC database vs Inverter Datasheet methods, my primary interest is in comparing the relative performances of multiple systems for the same installation site. So, if I have to use the Inverter Datasheet method for one system, it seems to me I should use the Inverter Datasheet method for them all in order to control my variables.

I appreciate your time.


Best,
Mike

Hello again,

So I did some playing around with the software and discovered that whether you go with option 1 (divide inverter AC output power by number of modules per inverter) or option 3 (multiply optimizer MPPT voltage by number of modules per string) you end up with the same kWh/kW efficiency.

Not sure if that's what you were trying to say Paul, maybe I misunderstood, but I figured I come back and update the thread with what I found out.

I would recommend using the "microinverter"-type design for designs involving strings of different lengths, the example system I was looking at was 50x 290W modules which didn't break down into anything even once you consider max string voltages require 4 strings.

• Divide inverter's AC output power by number of modules running through it (25), and input this into Max AC Output
• Use inverter's efficiency rating
• DC voltages all come from DC Optimizer spec
• System Design like I would for microinverters, string length= 1module, #strings= #inverters= total #modules

Drove me crazy since DC optimizers don't output AC power and the inverter doesn't do MPP Tracking, but these seem to be the only workarounds available. (Option 2 just seemed like cheating the more I thought about it)

As of Version 2018.11.11, SAM can model inverters with up to four MPPT inputs, and each subarray has its own "modules per string" input so the array can have up to four different string lengths.

See the release notes for a full list of new features.

Best regards,
Paul.