• Have you tried out dark mode?! Scroll to the bottom of any page to find a sun or moon icon to turn dark mode on or off!

diy solar

diy solar

NHX-10K / Adani 530 BiF Best String Config

DaveNSueM

New Member
Joined
Jun 25, 2024
Messages
41
Location
Beavercreek Ohio
Leaning towards the NHX (vs the Growatt SPH) mainly for known SA compatibility. Have developed the following hypothesis and would love additional eyes on calcs and confirmation I am thinking about this correctly; did I miss anything. TIA...this forum/information has been amazing so far and my daily read for past few weeks!

Hypothesis
To minimize clipping given parameters below, the best string config would be four 6S strings (as opposed to three 8S strings). With four 6S strings can use up to 10% bifacial gain (assuming it exists) with no clipping. There will be clipping overall; working to determine config to minimize clipping.

Panel/Wire/Location Info
Grid tied; no battery to start, interconnect agreement obtained
24 - Adani 530w bifacial (specs below)
Ground mount
Already purchased 8awg to connect 4 strings plus 8awg ground; 150’ in 1.5” buried pvc
Record low -36c
8S is max string @ -36c (8 panels VOC @ -36c = 461v; <500v)

Inverter MPPT info
4 mppt @14a max IMP; 22a max ISC
Voc range 120v-500v max
Max pv 15000w

Based on above, it seems implied each mppt is 3750w (?);
@14a max per MPPT, V calculates to be 268v
If watts per MPPT is 4000w (Watts247 website note) then
@14a max per MPPT, V would be 286v

VMP Calc
Panel vmp is 41.49v;
6 panels=249v;
8 panels=332v

Startup Calc
Min vmp @ 6 panels = 41.49v * .8 * 6 = 199v; >120v (saw the .8 derate suggested in a post)
Min vmp @ 8 panels = 266v; >120v

Imp Calc
6 panels=249v * 12.79a = 3185w
8 panels=332v * 12.79a = 4246w (over max w MPPT)

Add bifacial gain
10% : imp 13.97a * 249 (6S) = 3478w
10% : imp 13.97a * 332 (8S) = 4638w (over max watts/MPPT)
15% : imp 14.61a * 249 (6S) = 3638w (will clip @ 14a)
15% : imp 14.61a * 332 (8S) = 4851w (will clip @ 14a)


Panel Data (entire PDF sheet was too large to attach)
1722089546123.png
 
Haven’t looked at the numbers overall yet but what is the driving the system design requirement for minimal clipping? Usually you have more energy than needed in clipping parts of the year.

Do you have 1:1 net metering and therefore can leverage all energy generated above clipping points by banking it? (Even better, if on TOU and the otherwise clipped energy is force exported at peak rate hours and not in real time).
 
BTW with battery less grid tie operation all your anti clipping efforts are for naught because this only has a 10kW inverter to send it to grid. 🫗

Already purchased 8awg to connect 4 strings plus 8awg ground; 150’ in 1.5” buried pvc
This is a weird size but I guess you bought it already

#8 is only needed for 2p and up OR you are marginal on voltage drop for the start voltage. In the latter case I would argue you have picked flaky operating parameters and should have added 1 more series panel instead or picked higher voltage panels.

Where did you see the .8 derate? Does it get really hot in your area?

Did you upsize the EGC because you upsized CCC for voltage drop, and needed to apply the requisite size adjustment? The default chart, without adjustments, for EGC allows #10 to be used with up to #6 CCCs (this is a current based chart actually)

Did you use PVwatts to calculate clipping potential under their model? STC is very ambitiously optimistic
 
Haven’t looked at the numbers overall yet but what is the driving the system design requirement for minimal clipping? Usually you have more energy than needed in clipping parts of the year.

Do you have 1:1 net metering and therefore can leverage all energy generated above clipping points by banking it? (Even better, if on TOU and the otherwise clipped energy is force exported at peak rate hours and not in real time).
So still learning for sure, minimal clipping seems like a desired outcome but maybe not? We do have 1:1 net metering (for now anyways) with annual true up, no TOU yet.
 
So still learning for sure, minimal clipping seems like a desired outcome but maybe not? We do have 1:1 net metering (for now anyways) with annual true up, no TOU yet
There are multiple schools of thought and different objectively correct answers depending on the details of a particular site. I just like to clown on it bc some people are overly & arguable illogically precious about it
 
BTW with battery less grid tie operation all your anti clipping efforts are for naught because this only has a 10kW inverter to send it to grid. 🫗


This is a weird size but I guess you bought it already

#8 is only needed for 2p and up OR you are marginal on voltage drop for the start voltage. In the latter case I would argue you have picked flaky operating parameters and should have added 1 more series panel instead or picked higher voltage panels.

Where did you see the .8 derate? Does it get really hot in your area?

Did you upsize the EGC because you upsized CCC for voltage drop?

Did you use PVwatts to calculate clipping potential under their model? STC is very ambitiously optimistic
I typically like to upsize wire when affordable; lowes had 8awg on killer clearance in my area so it was cheaper than 10.

I would have to find the post for the .8, typical 80-90’s in Ohio summer…saw it was important to stay above the start voltage…I believe Will said in recent video to target 220v.

Yes to PVwatts.
 
I should add in my use case I have a NG Generac generator for most of my loads already so don’t really need (at least I think I don’t need) the added features/cost of say the EG4 18kpv. I do want to plan for battery though in future when utility changes up net meter rates
 
I would have to find the post for the .8, typical 80-90’s in Ohio summer…saw it was important to stay above the start voltage…I believe Will said in recent video to target 220v.
Hmm, I’m not really sure about this. You do want the Vmpp to be closer to nominal voltage for max conversion efficiency. Anyway I don’t think it matters for you because it’s not torquing your design in one direction or other
 
Yes to PVwatts.
I don’t know if PVwatts models both MPPT and inverter stage clipping. On this tier of hybrid the inverter is almost always smaller than the MPPT capacity. Sometimes the battery charger is also smaller than MPPT capacity

The design idea I guess is to fluff up the numbers, or, thinking that the 15kW from MPPT can be split across battery and inverting
 
I don’t know if PVwatts models both MPPT and inverter stage clipping. On this tier of hybrid the inverter is almost always smaller than the MPPT capacity. Sometimes the battery charger is also smaller than MPPT capacity

The design idea I guess is to fluff up the numbers, or, thinking that the 15kW from MPPT can be split across battery and inverting
I looked for MPPT model option in there but did not see one.

Agree on your 15 kW comment, that’s why part of me thought the 4 6S would be better as it seems I would lose more to clipping on each MPPT with the 8S config (?), in other words the 4 string would get me closer to 15kV overall where the 3 string would clip a lot more per MPPT

Thanks for your comments!
 
in other words the 4 string would get me closer to 15kV overall where the 3 string would clip a lot more per MPPT
With the 10kW AC limit though there is no difference until you get batteries

For your long string you might want to compare the voltage drop % loss with the #8. You will lose the same absolute amount but the relative amount will be less with higher string voltage

I would mainly go with 4 string for more future proofing. It maxes out your conduit anyway (if you go to 5 you also go from 70% current derate factor to 50%, which is rather punitive). 3 string is at 80% derate, 4 is 70%.

You can easily reconfigure the string later at the array by rejiggering panels, but pulling more wire will be pain.
 
6 panels=249v;
8 panels=332v
The MPPT is most efficient when you can get string voltages to around 360-370 volts. 250 volts is low. This comes about because the AC inverter section wants about 360 volts on the head rail. If the string input is 250 volts, then it has to boost that first. If the input is 450 volts, then it has to buck that. At 360 volts, there is high efficiency since there is no great effort to boost or buck.

Thus you definitely should prefer the 3 strings of 8 panels, IMO.

The MPPT will not overload itself, it will take the power it wants even if you provide more. Also, the STC conditions basically never happen in real life. The NMOT conditions are really what you should expect for performance.

If you do, somehow, make more than the MPPT wants, it will clip, but that clipping will be very small, occur for only a small period of time, and thus the energy lost will be minimal. I predict this loss will be less than the constant efficiency loss of a low voltage MPPT input from a string of 6 panels.

The NHX will make 10 KW AC and also charge the battery at the same time, so more than 10 KW of panel power does go to good use. If the battery is fully charged, then it will only use about ~10.7 KW of panel to make 10 KW AC, the difference being losses from the conversion.

Mike C.
 
The NHX will make 10 KW AC and also charge the battery at the same time, so more than 10 KW of panel power does go to good use.
The initial setup will not have a battery, per second paragraph.

For a grid tie system the ideal situation is for the battery to never get full, so the integral of the excess above 10kW needs to be less than the size of the battery for most of the year.

Perhaps one way to address this is to install 4 strings in the conduit (since the extra cost is only maybe $250 of THWN, and then reconfigure from 3 to 4 or vice versa based on operational data.
 
The MPPT is most efficient when you can get string voltages to around 360-370 volts. 250 volts is low. This comes about because the AC inverter section wants about 360 volts on the head rail. If the string input is 250 volts, then it has to boost that first. If the input is 450 volts, then it has to buck that. At 360 volts, there is high efficiency since there is no great effort to boost or buck.

Thus you definitely should prefer the 3 strings of 8 panels, IMO.

The MPPT will not overload itself, it will take the power it wants even if you provide more. Also, the STC conditions basically never happen in real life. The NMOT conditions are really what you should expect for performance.

If you do, somehow, make more than the MPPT wants, it will clip, but that clipping will be very small, occur for only a small period of time, and thus the energy lost will be minimal. I predict this loss will be less than the constant efficiency loss of a low voltage MPPT input from a string of 6 panels.

The NHX will make 10 KW AC and also charge the battery at the same time, so more than 10 KW of panel power does go to good use. If the battery is fully charged, then it will only use about ~10.7 KW of panel to make 10 KW AC, the difference being losses from the conversion.

Mike C.
Thank you. Great info on the MPPT voltage.
 
Perhaps one way to address this is to install 4 strings in the conduit (since the extra cost is only maybe $250 of THWN, and then reconfigure from 3 to 4 or vice versa based on operational data.
I do plan on including the spare 2 wires so could easily play around with the configuration.
 
I don’t know if PVwatts models both MPPT and inverter stage clipping. On this tier of hybrid the inverter is almost always smaller than the MPPT capacity. Sometimes the battery charger is also smaller than MPPT capacity

The design idea I guess is to fluff up the numbers, or, thinking that the 15kW from MPPT can be split across battery and inverting
Pvwatts does. You can change the dc to ac ratio
 
Leaning towards the NHX (vs the Growatt SPH) mainly for known SA compatibility. Have developed the following hypothesis and would love additional eyes on calcs and confirmation I am thinking about this correctly; did I miss anything. TIA...this forum/information has been amazing so far and my daily read for past few weeks!

]
I believe youre wrongly assuming that each mppt is limited to 3750w
 
I believe youre wrongly assuming that each mppt is limited to 3750w
Yes, I am taking the quoted 15kW and dividing by 4 MPPT’s - is there a better method of calculating it if it is not specifically called out (I honestly have no idea)?

On Wats247 site he states up to 4000 per MPPT so there’s that…

What is your opinion on the NHX vs the Growatt? You had originally suggested the Growatt to me in a different post and overall I like it but really want to use SolarAssistant and as you mentioned prior, no one has tried it yet. I did send an email to SA support but have not heard back yet.
 
Yes, I am taking the quoted 15kW and dividing by 4 MPPT’s - is there a better method of calculating it if it is not specifically called out (I honestly have no idea)?

On Wats247 site he states up to 4000 per MPPT so there’s that…

What is your opinion on the NHX vs the Growatt? You had originally suggested the Growatt to me in a different post and overall I like it but really want to use SolarAssistant and as you mentioned prior, no one has tried it yet. I did send an email to SA support but have not heard back yet.

Each mppt should be able to handle it's max current * voltage so 14a*500 or 7000w

You already know my opinion lol.

Sa already works with sph inverters but we don't know if it works with this one yet. I'm sure it will be added soon if it's not already compatible.

We will know a lot more once there are more units out in the field
 
Ah ok...

:love:

Ground mount started this past week - waiting for panel delivery now. Will decide on inverter within the next week...all the data I see folks post from SA looks amazing.


Awesome! Solar is fun.

I don't envy you having to make a decision at this point in time. Too much choice, too many unknowns, and victron threw it's name in the hat too with the price drop on the quattros.

Good luck with the decision making
 
@1201 you got me thinking overnight…since my near-term goal (1-3 years) is to simply offset power bill/feed grid coupled with so many new options (batteries and inverters), should I just look at Growatt 10k or 11.4k now and spend more later when I’m ready to add batteries.

I realize tech is always changing but maybe I wait on hybrid when I really need it…think I saw @timselectric make a similar comment in a different thread ( may be very good advice)

I do plan on installing unit outdoors on a wood structure where meter is installed but it does face north and is somewhat protected from elements with small wood overhang.

Thoughts on those units or in general?
 
@1201 you got me thinking overnight…since my near-term goal (1-3 years) is to simply offset power bill/feed grid coupled with so many new options (batteries and inverters), should I just look at Growatt 10k or 11.4k now and spend more later when I’m ready to add batteries.

I realize tech is always changing but maybe I wait on hybrid when I really need it…think I saw @timselectric make a similar comment in a different thread ( may be very good advice)

I do plan on installing unit outdoors on a wood structure where meter is installed but it does face north and is somewhat protected from elements with small wood overhang.

Thoughts on those units or in general?

It wouldn't be a lot more to get a 48v battery ready unit now, in case you decide to add batteries sooner.

You guys are brave putting your inverters outside in the weather. Not me. I keep my solar shed cooler than my house lol
 

diy solar

diy solar
Back
Top