Tesla Solar inverters

[nwdiver]

The diodes on a panel are in series with the solar cells. They prevent backflow current. A diode will only let current flow in one direction. This keeps shorts from happening in each panel. That's the ONLY thing a diode does.

The diagram below has one blocking diode, however each panel can have its own - which is common.

BYPASS Diode; not BLOCKING Diode...

If the cell group becomes reverse biased as happens when there's shade the diode becomes a short. Bypassing current around the shaded section. If this wasn't there shade can easily cause a fire. I learned this the hard way using panels without bypass diodes...

 

[cwied]

I think this is the picture that illustrates bypass diodes:

(credit: https://www.electronics-tutorials.ws/diode/bypass-diodes.html)

As I understand it the advantage of optimizers over bypass diodes is that whatever is bypassed is completely unproductive, whereas with optimizers, the shaded portion still produces, just at a lower rate. Depending on the pattern of shade on the panels this can be a larger or smaller issue.

 

[Garlan Garner]

BYPASS Diode; not BLOCKING Diode...

OMG that's terribly inefficient.

 

[Garlan Garner]

I think this is the picture that illustrates bypass diodes:

(credit: https://www.electronics-tutorials.ws/diode/bypass-diodes.html)

As I understand it the advantage of optimizers over bypass diodes is that whatever is bypassed is completely unproductive, whereas with optimizers, the shaded portion still produces, just at a lower rate. Depending on the pattern of shade on the panels this can be a larger or smaller issue.

That is horribly inefficient.

 

[nwdiver]

OMG that's terribly inefficient.

How so? Every panel has them. Even Optimized panels work this way. No panel can use a cell group where 1 cell is shaded. Optimized or not...

I think this is the picture that illustrates bypass diodes:

Yep; Except every panel I've seen divides each panel into thirds with 3 bypass diodes. That way shading a cell only kills 1/3 instead of the entire panel...

 

[Garlan Garner]

How so? Every panel has them. Even Optimized panels work this way. No panel can use a cell group where 1 cell is shaded. Optimized or not...



Yep; Except every panel I've seen divides each panel into thirds with 3 bypass diodes. That way shading a cell only kills 1/3 instead of the entire panel...

Because of the voltage drop across the diode and the back current leakage.

In other words....the bypass diodes you guys are listing is changing the voltage of the panel. That is horrible in a string. You want to change the current of a panel and keep the voltage constant.

That's what optimizers do.

My optimizers ALL put out 425v DC....no matter what the panel is doing...so that the inverters don't have to provide back current.

 

[cwied]

Here's an example of my production with optimized panels:

As I understand it, without optimizers the panels at 216 and 232 watts would have to be at 0, 85, or 170 watts, depending on what the actual shading on the panel is.

 

[nwdiver]

Because of the voltage drop across the diode and the back current leakage.

??? Both are incredibly small with quality Schottky diodes...

 

[nwdiver]

Without optimizers, the panels at 216 and 232 watts would have to be at 0, 85, or 170 watts, depending on what the actual shading on the panel is.

Weird... what kind of shade is that? Solid like a chimney or speckled like leaves? I wonder how the optimizers are avoiding activating the bypass diodes...

 

[Garlan Garner]

Here's an example of my production with optimized panels:
View attachment 316188

Without optimizers, the panels at 216 and 232 watts would have to be at 0, 85, or 170 watts, depending on what the actual shading on the panel is.

I understand what you are saying, however your string array is terribly compromised. Sure it appears that it is working, however that chart is showing you what should be happening.

When a cell is shaded the voltage of that cell drops. That reduces the voltage of the panel. Now your panels are mis-matched. That's a terrible thing to do to the inverter.

I totally understand what the bypass diode is doing....however its reducing the voltage of the panel.

 

[cwied]

I understand what you are saying, however your string array is terribly compromised. Sure it appears that it is working, however that chart is showing you what should be happening.

When a cell is shaded the voltage of that cell drops. That reduces the voltage of the panel. Now your panels are mis-matched. That's a terrible thing to do to the inverter.

I totally understand what the bypass diode is doing....however its reducing the voltage of the panel.

I think you are missing the fact that I said "optimized." I'm not advocating for bypass diodes. I just posted the image to clear up the confusion between blocking and bypass diodes.

My array has SolarEdge optimizers.

 

[nwdiver]

That's a terrible thing to do to the inverter.

The inverter is designed for that. Do you have any idea how much voltage fluctuates hour to hour if it's partly cloudy or season to season? I've seen voltage spike from 340v to 440v and drop back to 340 in seconds when a cloud passes by...

 

[Garlan Garner]

Weird... what kind of shade is that? Solid like a chimney or speckled like leaves? I wonder how the optimizers are avoiding activating the bypass diodes...

Optimizers evaluate the voltage of the string between each optimizer - Which is set by the inverter.

Then the optimizer looks at what its getting from the Panel ( shaded or not ) and THEN matches the voltage of the string by adjusting the current its getting from the panel.

In other words....it says. I see 425 volts on the string....so I'm going to take the 12 or 24 or 36 volts I'm getting from the panel and adjust its current so that I can get to 425 volts. P=I*E

If your voltage ( E) of your panel goes down...then the optimizer adjusts the current received from the panel so that it can put 425 volts on the string. Each panel is isolated.

This optimizer 425 volt matching allows for you to also mix and match panels sizes and everything. Optimizing at 425 volts DC also drastically reduces the gauge of wire needed to get to the inverter.

 

[Garlan Garner]

The inverter is designed for that. Do you have any idea how much voltage fluctuates hour to hour if it's partly cloudy or season to season? I've seen voltage spike from 340v to 440v and drop back to 340 in seconds when a cloud passes by...

I totally get what you are saying.

Not with a power optimizer. That's my point. Power optimizers keep the same voltage on the string @ 425DC volts and each optimizer adjusts the current they offer the string. That's tremendously more efficient for the inverter and the panels themselves.

See - your Inverter has to adjust Voltage to Current because of your bypass diodes changing both your string voltage and current.

Why do you think String Inverter makers are now producing panel optimizers? They know what time it is.

 

[cwied]

Weird... what kind of shade is that? Solid like a chimney or speckled like leaves? I wonder how the optimizers are avoiding activating the bypass diodes...

It's most likely leaves from the trees next door. The shade probably is horizontal across the bottom of the panels, which would most likely be a very bad case for bypass diodes, assuming the substrings are vertical. Note that the installers weren't careful about the id numbers of the panels, so it's likely that the two shaded panels are actually beside each other, not as drawn.

 

[Garlan Garner]

It's most likely leaves from the trees next door. The shade probably is horizontal across the bottom of the panels, which would most likely be a very bad case for bypass diodes, assuming the substrings are vertical. Note that the installers weren't careful about the id numbers of the panels, so it's likely that the two shaded panels are actually beside each other, not as drawn.

With power optimizers....it wouldn't matter at all.

Lastly....how do know the output of each panel since there is no panel signature in a String inverter setup?

 

[nwdiver]

I totally get what you are saying.

Not with a power optimizer. That's my point. Power optimizers keep the same voltage on the string @ 425DC volts and each optimizer adjusts the current they offer the string. That's tremendously more efficient for the inverter and the panels themselves.

If your definition of 'tremendous' is <1% then 'yes'. There's a Solar Edge system that's of similar and orientation and size to a String system I designed. They're neck and neck in terms of production. If there's a big benefit to optimizers vs a well designed string system I've never seen it. Optimized systems might be sliiightly more effective but they almost always increase the cost / kWh of production which is why I don't use them in my systems.

Lastly....how do know the output of each panel since there is no panel signature in a String inverter setup?

The screenshot I posted is from a SMA system equipped with TS4-S rapid shutdown devices that collect module level current and voltage data.

One of these panels DOES have an optimizer and it's not the one you would think... which is why I don't use optimizers

 

[Garlan Garner]

Lastly....let me clarify again.

Power Optimizers at the panel are DC in DC out to a DC inverter approx. 24 parts inside each Power Optimizer ( SolarEdge)

Micro Inverters at the panel DC in AC out to an AC inverter. approx. 240 parts inside each Micro Inverter ( Emphase)

Its much more efficient to transfer 425 volts DC ( SolarEdge) than 125 volts AC ( Enphase ) from the roof.

AND tremendously fewer points of failure on the roof with so fewer parts that can fail.

 

[cwied]

With power optimizers....it wouldn't matter at all.

Lastly....how do know the output of each panel since there is no panel signature in a String inverter setup?

Read my previous response to your post: I have SolarEdge optimizers. I'm actually supporting your side of the argument.

 

[nwdiver]

Lastly....let me clarify again.

Power Optimizers at the panel are DC in DC out to a DC inverter approx. 24 parts inside each Power Optimizer ( SolarEdge)

Micro Inverters at the panel DC in AC out to an AC inverter. approx. 240 parts inside each Micro Inverter ( Emphase)

Its much more efficient to transfer 425 volts DC ( SolarEdge) than 125 volts AC ( Enphase ) from the roof.

AND tremendously fewer points of failure on the roof with so fewer parts that can fail.

Agreed. I generally dislike Optimizers but I HATE micros...