combined negative but separate positives across MPPTs

[Zee]

What the title says.

I am replacing inverters on a legacy system.
New inverters are Sunny Tripower X-30-US with 3 MPPTs each.
The negatives for 3 strings are combined at the array into a single #8 AWG PV wire comig to the INV...........while (most of) the positives are individual (#10 AWG) for each string.
Can I wire 2 positives to one MPPT and the third to another MPPT, while "splitting" the common negative across both MPPTs?
(wire single common negative to both MPPTs with MC branch circuit "Y" connector AKA splitter or jumper)

I know this can be done with paralleled source circuits, but not sure about MPPTs.

OPTION 2:
More details:
Panel Imp is 8.68A (3 strings at 8.68 A = 26 A combined)
INV MPPT max DC input of 24 A operating.

So I guess I could wire it to a single MPPT but because there is ony on other neagtive lead.......I would only use 2 of 3 MPPTs on the INV. Not sure if that is a huge waste or none at all?

 

[ggunn]

However you wire your positives the negatives should be wired the same way, i.e. you can't wire a string's positive and negative conductors into different MPPT inputs. You need to check your modules' Isc (X 1.25 unless the inverter data sheet says not to) to make sure you don't exceed the MPPT's maximum input current if you want to combine strings.

 

[Zee]

However you wire your positives the negatives should be wired the same way, i.e. you can't wire a string's positive and negative conductors into different MPPT inputs. You need to check your modules' Isc (X 1.25 unless the inverter data sheet says not to) to make sure you don't exceed the MPPT's maximum input current if you want to combine strings.

Thank you. Yeah the combined mod ISC is under the max ISC allowable for the mppt input of inverter. So we're good there.

My concern is how the inverter works if the PV is not spread across all available MPPTs.
Specifically, if I take 28 KW of PV, but only wire it to 2 of the 3 MPPTs on this SMA 30 kW inverter , am I capping output or seriously ruining efficiency in any way?
(And leaving the 3rd mppt unused)

 

[ggunn]

Thank you. Yeah the combined mod ISC is under the max ISC allowable for the mppt input of inverter. So we're good there.

My concern is how the inverter works if the PV is not spread across all available MPPTs.
Specifically, if I take 28 KW of PV, but only wire it to 2 of the 3 MPPTs on this SMA 30 kW inverter , am I capping output or seriously ruining efficiency in any way?
(And leaving the 3rd mppt unused)

That won't be a problem.

 

[Zee]

In other words is it equivalent to wiring the complete array to 2/3 of the inverter's 30 kW rating?........ so equivalent to wiring it to say a 20 kW inverter? (2/3 x 30 kW)

 

[ggunn]

In other words is it equivalent to wiring the complete array to 2/3 of the inverter's 30 kW rating?........ so equivalent to wiring it to say a 20 kW inverter? (2/3 x 30 kW)

I'm not sure what you mean by that, but not using all the MPPT channels isn't a problem. BTW, when you are evaluating the allocation of current you need to multiply Isc by 1.25 unless the inverter document says that you do not need to, i.e., unless it says something like "maximum connected Isc". If it just says "maximum input current" you need to add the extra 25%.

 

[jaggedben]

I'd bet a nickle it would work fine because the negatives are common inside the inverter anyway. I'm basing this on SMA's single phase inverters from about 10 years ago where they had two MPPTs but the field assembled DC disco had four terminals for landing the field wiring and only three factory wires to connect inside the inverter: two positives and one negative. I realized looking at it that the MPPT electronics must only be on the positive side.

But, if SMA does not say that you can do it that way, then you're on your own. Hence my offering to bet only a nickle, not a sandwich.

 

[ggunn]

I'd bet a nickle it would work fine because the negatives are common inside the inverter anyway. I'm basing this on SMA's single phase inverters from about 10 years ago where they had two MPPTs but the field assembled DC disco had four terminals for landing the field wiring and only three factory wires to connect inside the inverter: two positives and one negative. I realized looking at it that the MPPT electronics must only be on the positive side.

But, if SMA does not say that you can do it that way, then you're on your own. Hence my offering to bet only a nickle, not a sandwich.

Don't most inverters check for ground faults by comparing the currents on the positive and negative inputs on the MPPT channels?

 

[jaggedben]

Don't most inverters check for ground faults by comparing the currents on the positive and negative inputs on the MPPT channels?

Interesting point. But those SMA inverters I mentioned must have done that somehow; they were transformerless inverters. Perhaps it's not necessary to measure that current on each MPPT individually.

 

[solarken]

What is the voltage of the string, and is this a ground mount system or rooftop? Was the old inverter limited to 600V and are the modules and wiring capable of 1000V? If you can restring and go from 3 strings to 2 longer strings at higher voltage, it seems that would give you an extra conductor to have separate negatives. I would ask SMA engineers before splitting a common ground into two separate MPPT inputs. It is not just the power conversion functionality to be considered, the Tripower Core 1 separately tests each MPPT input for insulation resistance and has ground fault and arc fault detection that individually serves different string groups, maybe the Tripower X has similar or even more granularity across individual MPPT inputs.

 

[Zee]

I'm not sure what you mean by that, but not using all the MPPT channels isn't a problem. BTW, when you are evaluating the allocation of current you need to multiply Isc by 1.25 unless the inverter document says that you do not need to, i.e., unless it says something like "maximum connected Isc". If it just says "maximum input current" you need to add the extra 25%.

I was writing my question before I saw your answer.
And my question was really just over explaining my same question again (wondering if using two MPPTs will limit power output versus using all three MPPTs....... which it sounds like it won't)
Thanks!

 

[Zee]

I'd bet a nickle it would work fine because the negatives are common inside the inverter anyway. I'm basing this on SMA's single phase inverters from about 10 years ago where they had two MPPTs but the field assembled DC disco had four terminals for landing the field wiring and only three factory wires to connect inside the inverter: two positives and one negative. I realized looking at it that the MPPT electronics must only be on the positive side.

But, if SMA does not say that you can do it that way, then you're on your own. Hence my offering to bet only a nickle, not a sandwich.

I remember those "udders"!
(Field assembled DC Disco)

 

[Zee]

What is the voltage of the string, and is this a ground mount system or rooftop? Was the old inverter limited to 600V and are the modules and wiring capable of 1000V? If you can restring and go from 3 strings to 2 longer strings at higher voltage, it seems that would give you an extra conductor to have separate negatives. I would ask SMA engineers before splitting a common ground into two separate MPPT inputs. It is not just the power conversion functionality to be considered, the Tripower Core 1 separately tests each MPPT input for insulation resistance and has ground fault and arc fault detection that individually serves different string groups, maybe the Tripower X has similar or even more granularity across individual MPPT inputs.

Max DC V of string: 959 V
Old inverter rated at least 1000 VDC
Module and PV wire rated 1000 VDC

Rooftop, no MLPE, 10 or more years old

Yeah definitely a question for the SMA engineers!

 

[solarken]

Max DC V of string: 959 V
Old inverter rated at least 1000 VDC
Module and PV wire rated 1000 VDC

Rooftop, no MLPE, 10 or more years old

Yeah definitely a question for the SMA engineers!

Ok so it is already at 1000V and restringing is not an option. You might want to consider combining the strings differently to free up one PV cable in the run for an extra negative. Remove the Y-connector harness from the negative of string 3 at the array, leaving strings 1 and 2 on the 8AWG negative. Combine the positives of strings 1 and 2 at the array using a Y-cable onto one of the 10AWG positive cables going to the inverter. Now you should have two 10AWG PV cables left, for the positive and negative from string 3. Then land combined str 1&2 on MPPT1 and str 3 on MPPT2, with individual negatives for each MPPT. You might have to make some jumpers or reterminate some MC4 connectors to achieve, but you can avoid any adverse impact of splitting the negatives.

 

[pv_n00b]

I can't find an internal SLD for this inverter so I'm just making an educated guess about what is going on. The negative input bus most likely parallels all the negative DC inputs. The positive inputs go in through 3 separate MPPT sections and are then paralleled before going into a single inverter section. This all means that having a common neutral from the array is probably not a problem. The more of the MPPT inputs you can use the better but if you can only use two it should work fine as long as you don't exceed the rating of an individual input. This is all an educated guess because the manual does not provide much detail.