Tesla string sizing

I'm having 395w panels with a 45.18voc 36.99vmp. The designer put 14 of these in 3 strings, i'm calculating that to a 673v to an three Powerwall3s, Poewrwall3 lists 600v max, the pv dc mppt volt range is 150-480 and input range goes up to 550v. (amps is fine between limits). Questions I have are...

Does this mean the exceeding 480v is pointless since its the mppt limit? 14 peak at 632v nameplate (690 considering temp) and min around 599v
is the excess above 600 an issue? I'd more expected this in 4 strings of 10 rather than 3 of 14.

Working voltage, Vmp, is much different than VOC.

Not to downplay VOC limits and potential of damage to equipment.

But aren’t Tesla installers “certified” as to provide that sweet Tesla warranty?

I’d say you’re paying them for the install, no need to micro manage them.

Micromanaging is fine, but you should ask them next, and it’s easier to prevent it now than to get them back later. In theory.

If Vmpp exceeds working voltage you leave power on the table year round

If Voc exceeds maximum the MPPT is dead.

Doing my best not to micro, but I pretty much designed a system that I didn't have time to install based off of this formum and learned SO much great information. This concern triggered when reviewing the plan set I received, and when we met I was told they had fired the designer for too many errors. The new plan was for 8s2p that are within limits, but the two remaining strings had 12 on them and that still looks problematic at STC its close to the limit, when looking at temps it pushes it the parallel group pushes losses up on the hot end of the equation. This is a great brain teaser to balance all these. Fortunately the gentleman on here who shared his spreadsheet really makes it easy to tinker with. photonik is also pretty good, but those are string only so losses on the low end make it interesting.

Need the roof layout and MPPT input specs for folks here to be able to help with assigning panels into strings.

On a DIY system one might recommend lopping off a handful of awkward panels onto microinverters and AC coupling those or just gridtying those to grid and not bother about off grid operation for those. A turnkey installer would probably not want to do that.

Thanks Zany, it's a ground mount 40 panel array. PW3 has 6 MPPT(s) with a 480v operating range, 550 input limit and 600 withstand limit. After some digging I'm questioning if it could just have solar inputs on two inverters and the third not be used. Tesla's dang annoying with the limited information they allow us in the general public to have.

With 6 MPPTs I assume Tesla would break down the strings to get in the voltage range.
Spec sheet says PW3 can handle 20 kW so may only need a single PW3 and get the battery extensions when available.
Depends too on inverter capacity (11.5kW) to provide to all loads on the entire panel.

snewb said:

Thanks Zany, it's a ground mount 40 panel array. PW3 has 6 MPPT(s) with a 480v operating range, 550 input limit and 600 withstand limit. After some digging I'm questioning if it could just have solar inputs on two inverters and the third not be used. Tesla's dang annoying with the limited information they allow us in the general public to have.

Click to expand...

Ground mount keeps things more consistent from a insolation angle (everything facing same direction), I wonder why they don’t just divide it evenly. Maybe something to do with minimizing splicing between rows of panels.

You can most likely get specs from the Tesla solar subreddit. I’ve seen posts there for their string inverters (sourced from other vendors). So I’m sure someone knows the PW3 specs.

Keeping the strings balanced across the three PW3 allows the charging to be done DC coupled in a symmetric way. Otherwise they have to do AC coupling calculations. In theory the two would get full first and then send to the third (if operating off grid), which would be double the AC charging capacity of the third. Or, the third would not be able to charge when off grid, if PW3 only supports DC coupled charging from solar connected to the PWs.

Now if there was the right kind of stacking communications, potentially they don’t need to use frequency shift to throttle.

Also you would have extra losses from DC AC AC DC.