Bypass, Blocking Diodes, and KW limit

[Grouch1980]

Hi all!
So I'm starting to delve into the field of Photovoltaics. I do have a few questions, some basic, any help would be great!:

1. I read that the purpose of the OCPD fuses on each string (where required) is to prevent fault current from the other strings from flowing into the faulted string. The fuse on the faulted string will then blow, and the currents from the other strings will continue to flow normally, to the load. Wouldn't the blocking diode located on each string accomplish that though?... doesn't that prevent fault current from flowing into the faulted string? Why need the fuse then?

2. The bypass diodes on the solar modules...when they bypass a certain segment of series connected solar cells, the solar module is still producing current from the other segments, minus the current from the segment that is shaded / damaged and bypassed by its diode. is this correct?

3. When you size a PV installation for a townhouse for example, what's the usual starting point? do you go by the maximum sized breaker you can place on the panel, not exceeding 120% of the panel bus between all sources (PV and utility)? I would assume that if you use that as a starting point, that's the max sized system KW wise that you can use. and then a further limit would be the size and shape of the roof, possibly shrinking the system even more. and then of course is what can the owner afford.

4. I've seen designs where the positive and negative wires from the strings each go into its own MPPT on the inverter. Is that typical for an MPPT, where you have one MPPT for the + wires and one MPPT for the - wires?

As always, thanks!

 

[jaggedben]

Hi all!
So I'm starting to delve into the field of Photovoltaics. I do have a few questions, some basic, any help would be great!:

Welcome to the field. From your questions it sounds like you are reading a book that may be somewhat out of date.

1. I read that the purpose of the OCPD fuses on each string (where required) is to prevent fault current from the other strings from flowing into the faulted string. The fuse on the faulted string will then blow, and the currents from the other strings will continue to flow normally, to the load. Wouldn't the blocking diode located on each string accomplish that though?... doesn't that prevent fault current from flowing into the faulted string? Why need the fuse then?

First of all I've been in this field for 12 years and I've never installed a blocking diode. For that matter, it's been about 8 years since I installed a fuse on a string (although I've been doing nearly all residential during that time, not commercial or groundmounts). You should research micro-inverters, optimizers, and rapid shutdown requirements before trying to design a rooftop system.

Second, diodes can fail if the reverse voltage is high enough, or just if they are defective, and shouldn't be relied on for overcurrent protection.

2. The bypass diodes on the solar modules...when they bypass a certain segment of series connected solar cells, the solar module is still producing current from the other segments, minus the current from the segment that is shaded / damaged and bypassed by its diode. is this correct?

That's more or less correct. If you're just looking to install the equipment and not design it, that's a good enough understanding. Note that the segments are usually in the long direction within modules, which means that when they might be shaded from the bottom, a 'landscape' orientation takes better advantage of the diodes. The purpose of the diodes is to prevent dirt or shade on one segment from affecting the others too much.

3. When you size a PV installation for a townhouse for example, what's the usual starting point? do you go by the maximum sized breaker you can place on the panel, not exceeding 120% of the panel bus between all sources (PV and utility)? I would assume that if you use that as a starting point, that's the max sized system KW wise that you can use. and then a further limit would be the size and shape of the roof, possibly shrinking the system even more. and then of course is what can the owner afford.

You start by calculating the panels required to produce a certain amount of energy over a target time. Usually for grid-tied that's one year. (Off-grid is different.) There's lots of software products out there that help generate residential PV production estimates and financial analysis. The quick and easy one for just production estimate is https://pvwatts.nrel.gov/ You need to account for panel orientation (tilt and azimuth) and shading.

The limitations of the customers existing electrical are an important secondary consideration. You might upgrade their equipment or downsize the PV system depending on goals and budget. Read Article 705 carefully as it's not as simple as just the 120% rule.

4. I've seen designs where the positive and negative wires from the strings each go into its own MPPT on the inverter. Is that typical for an MPPT, where you have one MPPT for the + wires and one MPPT for the - wires?

An MPPT is always one PV circuit, i.e. a positive and a negative. Multiple MPPTs has become typical for string inverters, but see the answer to question 1.

 

[Grouch1980]

An MPPT is always one PV circuit, i.e. a positive and a negative. Multiple MPPTs has become typical for string inverters, but see the answer to question 1.

Thanks for your responses and help! Understood about the blocking diodes (or lack of them i should say), and the bypass diodes. I have to look more into my 3rd question. Regarding your answer to my 4th question... so the positive and negative wires from a string both have to terminate into the SAME mppt? so what that other engineer showed on their drawings... might've been incorrect?... they show the positives going to one MPPT, and the negatives going to another MPPT.

 

[wwhitney]

The purpose of the diodes is to prevent dirt or shade on one segment from affecting the others too much.

Where "affecting" can mean "starting a fire", right?

Cheers, Wayne

 

[GoldDigger]

Hi all!
So I'm starting to delve into the field of Photovoltaics. I do have a few questions, some basic, any help would be great!:


2. The bypass diodes on the solar modules...when they bypass a certain segment of series connected solar cells, the solar module is still producing current from the other segments, minus the current from the segment that is shaded / damaged and bypassed by its diode. is this correct?

Instead of "minus the current from ...." I would say "minus the voltage of ......"
The bypass segments within a panel are in series.
If you have two or more panels or strings in parallel, the voltage reduction from activation of a bypass diode may mean that that panel or string will not contribute at all to the total power or may force the rest of the array panels to be used at a voltage below their MPPT.
Avoiding this is yet another advantage of optimizers or microinverters.

 

[Carultch]

Thanks for your responses and help! Understood about the blocking diodes (or lack of them i should say), and the bypass diodes. I have to look more into my 3rd question. Regarding your answer to my 4th question... so the positive and negative wires from a string both have to terminate into the SAME mppt? so what that other engineer showed on their drawings... might've been incorrect?... they show the positives going to one MPPT, and the negatives going to another MPPT.

The positive and negative of any given circuit have to go to the same MPPT zone. Otherwise the circuit would be incomplete. They do go to the separate sections of the same MPPT zone dedicated to connecting each polarity, but they do not go to different MPPT zones. As of NEC 2017, you are required to have a disconnect blade for both polarities of any system that isn't solidly grounded (which is most if not all PV systems). It used to be the standard to only have to disconnect the ungrounded polarity, when the other polarity was grounded through the GFCI fuse/breaker (now known as "functionally grounded"), but that is no longer the case. So any time you introduce 3rd party DC balance-of-systems equipment (e.g. combiners, disconnects) where NEC2017 and later applies, you need to make sure it is compliant with NEC2017 or later.

You may or may not be permitted to have multiple source circuits going to the same MPPT zone, depending on the constraints of the particular inverter. The inverter will have a maximum current per MPPT, which will govern the maximum quantity of circuits on that zone. Usually there is a maximum short circuit current and a maximum usable input current. You'll need to check with the inverter manufacturer for whether any safety factors (such as the 1.25 enhancement factor or the 1.25 continuous load factor) apply to module Isc, when determining what "maximum short circuit current" really means for them. Some inverters are specifically built with two source circuits per MPPT in mind, in order to strategically design the system so you normally don't use fuses. In any case, the strings have to be identical on any given MPPT zone, because they will output a common voltage. If you connect non-identical source circuits to the same MPPT zone, they will have to compromise on performance in order to force both to operate at a non-ideal voltage that will hinder the performance of both circuits.

What an MPPT zone really is, is a DC-to-DC converter as part of the inverter, that will seek the voltage closest to the maximum power point "sweetspot" of the source circuits on its input, and then buck or boost the voltage to the DC voltage that the inverter uses for producing the AC waveform. This is analogous to a transformer, where you can dynamically adjust the ratio of turns. It isn't the same thing as a transformer, as it is built specifically for the DC application. Multiple MPPT zones allow the same inverter to have diversity of string sizes, as well as localize losses within the strings on the same MPPT zone, provided the string sizes are uniform within any given MPPT zone. It can even allow for a diverse selection of modules, due to sectionalizing the power processing of the PV array inputs.

 

[Grouch1980]

Note that the segments are usually in the long direction within modules, which means that when they might be shaded from the bottom, a 'landscape' orientation takes better advantage of the diodes. The purpose of the diodes is to prevent dirt or shade on one segment from affecting the others too much.

That makes perfect sense. if oriented in portrait, the shading on the bottom could knock out all segments. In landscape, the shading would only knock out one segment... the long one towards the bottom.

 

[Grouch1980]

The positive and negative of any given circuit have to go to the same MPPT zone. Otherwise the circuit would be incomplete. They do go to the separate sections of the same MPPT zone dedicated to connecting each polarity, but they do not go to different MPPT zones.

Understood, thanks!

 

[GoldDigger]

Understood, thanks!

Depending on the design of the string inverter, one polarity connection or the other may in fact go to a common point inside. But to keep things readily understandable the + and - terminals associated with each zone should be used.

 

[Grouch1980]

Depending on the design of the string inverter, one polarity connection or the other may in fact go to a common point inside. But to keep things readily understandable the + and - terminals associated with each zone should be used.

got it. yeah, that's why i came here with my question #4... I kept looking at the wiring diagram on the design drawings and thought something was strange.

 

[Grouch1980]

You start by calculating the panels required to produce a certain amount of energy over a target time. Usually for grid-tied that's one year. (Off-grid is different.) There's lots of software products out there that help generate residential PV production estimates and financial analysis. The quick and easy one for just production estimate is https://pvwatts.nrel.gov/ You need to account for panel orientation (tilt and azimuth) and shading.

The limitations of the customers existing electrical are an important secondary consideration. You might upgrade their equipment or downsize the PV system depending on goals and budget. Read Article 705 carefully as it's not as simple as just the 120% rule.

That's a slick website, thanks! you just punch in the variables and it analyses everything. One thing I'm not seeing though... wouldn't one start by first seeing the limitation of the system, such as analysis of the 120% limit and other things? You calculate the max size of the circuit breaker that can go on your panelboard. and from there you work backwards, figuring out the KW of the PV system (the max that you can do). You then punch in the KW rating of your PV system on the website, which analyzes the kWh's and cost savings. Or do you first start with the website analysis?

 

[Carultch]

That's a slick website, thanks! you just punch in the variables and it analyses everything. One thing I'm not seeing though... wouldn't one start by first seeing the limitation of the system, such as analysis of the 120% limit and other things? You calculate the max size of the circuit breaker that can go on your panelboard. and from there you work backwards, figuring out the KW of the PV system (the max that you can do). You then punch in the KW rating of your PV system on the website, which analyzes the kWh's and cost savings. Or do you first start with the website analysis?

I would recommend starting with what can fit to fill the available rooftop area, after accounting for required clear space and obstructions. Even if it is far in excess of an AC constraint on the 120% rule, there are ways to get around that particular constraint. Such as a supply-side interconnection, or using interval data or a load calculation to justify reducing the main breaker rating to salvage compliance with the 120% rule. It is generally most economical to do a load-side interconnection, but when necessary, a supply-side interconnection is also a possibility. You are ultimately limited to the rating of the service as a whole, regardless of how you interconnect on distribution equipment.

 

[jaggedben]

Where "affecting" can mean "starting a fire", right?

Cheers, Wayne

I think you're educating me. I honestly thought it was mostly about performance but I also see how they would keep shaded portions from overheating.

 

[wwhitney]

I also see how they would keep shaded portions from overheating.

To spell it out for everyone, say a leaf drops on a single cell and shades it. The cell will not pass any current with a voltage rise (generation). If there are no bypass diodes, the other cells in series generate enough voltage to overcome the breakdown voltage of the shaded cell, allowing the string current to pass but with a large voltage drop. That means a lot of power dissipation in one cell, possibly enough to burn a cell or start a fire.

The bypass diodes are installed so they are reverse biased in normal operation and do not conduct. That means that the number of PV cells per bypass diode has to be low enough so that the generated voltage by the substring is less than the bypass diode's breakdown voltage. But when one cell is shaded, the associated bypass diode will conduct the string current with only the usual forward diode voltage drop and thus a safer level power dissipation.

Cheers, Wayne

 

[Grouch1980]

I would recommend starting with what can fit to fill the available rooftop area, after accounting for required clear space and obstructions. Even if it is far in excess of an AC constraint on the 120% rule, there are ways to get around that particular constraint. Such as a supply-side interconnection, or using interval data or a load calculation to justify reducing the main breaker rating to salvage compliance with the 120% rule. It is generally most economical to do a load-side interconnection, but when necessary, a supply-side interconnection is also a possibility. You are ultimately limited to the rating of the service as a whole, regardless of how you interconnect on distribution equipment.

I assume the only limitation to the supply-side interconnection is making sure the AC current (coming from the inverter) doesn't exceed the service panel's bus rating? as well as the service conductors that you're tapping to.

 

[Grouch1980]

You are ultimately limited to the rating of the service as a whole, regardless of how you interconnect on distribution equipment.

Actually i think that answered my question directly above.

 

[BillK-AZ]

Any separate blocking diodes would need to be Listed or Recognized for the application or contained in a DC combiner that is Listed. Correctly rated fuses are generally the best solution. Fuses have less voltage drop than diodes.