why do we need Diode ?

[caliman]

This is my first post but sorry for my bad english,that is why i can not search for the answer.
The solar panels produce DC volt but why still need diot for.thanks for answers.

[Naptown]

Welcome to the forum

To answer your question
To prevent a reverse current that would destroy cells in a module by overheating.

[peakbagger]

I remember seeing an article in Home power that some applications in the third world intentionally did not use a diode. The use was on remote village battery charging stations. If someone hooked up a weak battery backwards with a diode on the panels, the diode could heat up and eventually fail rendering the panel useless. Without a diode, there was a chance that the panels would survive. The report I linked to below stated that about 1/2 of the charging station in Thailand had failed this way.

Here is a link to a article that goes into more detail

http://www.palangthai.org/docs/BipassDiodes.pdf

[Mod note - just post the plain link, not a complex url]

http://www.google.com/url?sa=t&source=web&cd=3&ved=0CDcQFjAC&url=http%3A%2F%2Fwww.palangthai.org%2Fdocs%2FBipassDiodes.pdf&rct=j&q=diode%20failure%20solar%20panles&ei=s9d5TsnzEYHG0AHr1rGsAg&usg=AFQjCNENsGh9LcQol49SUYluspiJIb091Q&sig2=U7HxiudsaEYC_1BoXd-P5A

[RCinFLA]

There are two different uses for diodes in a PV panel. They are called blocking diode and bypass diode.

A blocking diode is placed in series with the panel in forward current flow direction. Its purpose is to prevent leakage current back into the panel. All PV cells have some defects in the cell that result in parallel resistance within the cell. On good quality cells this can range from 10 ohms to over 1000 ohms. So a 36 cell series panel may have less then 500 ohms of net shunt resistance. Without some means to prevent it, this resistance would be loaded across the battery during the night or periods of low illumination.

Blocking diodes are not normally needed if a charge controller is used as most charge controllers prevent the reverse flow from the battery to the panels at night or during little illumination of panels.

The second use of diodes are bypass diodes which are placed around every 12 to 24 series connected cells. When a single cell is shaded by an object it becomes reverse biased by all the other cells in the series that are still under illumination. This creates a higher voltage reverse bias across the shaded cells. The shunt defect resistance sites on the cell can now have a much higher voltage across them then the 0.5 to 0.6 volts across the cell in the forward illumination current situation. For 24 series cells every bypass diode the reverse bias voltage is limited by the bypass diode to about 10-13 vdc. Without the bypass diodes the reverse bias can be higher and the result is 'hot spots' at the defect shunt resistance sites that may damage the cell or even create a possible fire hazard within the panel.

A secondary benefit of bypass diodes is to reduce the overall voltage loss due to a shaded section of a panel.

[russ]

Hi RC - You provide excellent posts that are easy for electrical/electronic novices like myself to understand.

Thanks again,
Russ

[Mike90250]

I remember seeing an article in Home power that some applications in the third world intentionally did not use a diode. The use was on remote village battery charging stations. If someone hooked up a weak battery backwards with a diode on the panels, the diode could heat up and eventually fail rendering the panel useless. Without a diode, there was a chance that the panels would survive. The report I linked to below stated that about 1/2 of the charging station in Thailand had failed this way.

from the article:

The problem is interesting because photovoltaic
modules are generally the most robust component in any solar electric system.
Photovoltaic panels have no moving parts and are inherently immune from a variety
of electrical faults including short circuits and reverse polarity.

Not sure if all this is right, connecting a bare PV panel backwards, you forward bias the junctions, and cook the whole panel, including the bypass diodes. I've never done this, but from what I understand, connecting a panel backwards, kills it.

[RCinFLA]

from the article:


Not sure if all this is right, connecting a bare PV panel backwards, you forward bias the junctions, and cook the whole panel, including the bypass diodes. I've never done this, but from what I understand, connecting a panel backwards, kills it.

From the article, "The unnecessary inclusion of bypass diodes in these systems created an unexpected failure mode when
villagers misconnected their batteries with reverse polarity."

This is really dumb!

The bypass diodes would be forward biased by reverse connection and you would kill at least one of them. PV cells would be reverse biased and not conduct.

The PV cells are in forward bias direction under normal polarity. The series stack of cells with their near 0.6v conduction point per cell must be greater then the battery voltage or the panel cells would forward conduct due to battery voltage being too great for the panel. For a 36 cell panel, the battery voltage must not be more then about 17 vdc or there will be forward conduction of the PV cells. This is one of the reasons why 12v battery charging panels have 17-18 vdc Voc. When the panel is in the summer noon sun the heating of the panel drops the cell conduction point voltage lower, to about 0.5 vdc per cell. A PV cell, by itself, is not damaged by reverse battery connection.

There is a potential issue with bypassing diodes if their reverse breakdown voltage is not great enough. This is particularly important with typical GT inverter where there is high total stack voltage of several hundred volts DC. It is important to check the maximum stack voltage rating on the panel. For GT hook ups with high DC voltage, panels should have 600 volt rated breakdown panels.

Most commericial panels (at least outside of China manufactured panels) have bypass diodes. They will not likely pass U.L. approval without them. Typical configuration is three located in junction box. Two rows of cells are bypassed by one diode. By using the down and back rows, all the bypass diode connections are easily brought the junction box. There are some panels that put the bypass diode in the panel area. You will see small bumps in the back of panel, plastic laminar sheet.

For 12v (36 cell) panels, it is possible to not have bypass diodes built into panel if high quality cells are used. If these panels are used in a multi-panel series stack then external bypass diodes should be added.

With the price of panels dropping so much, manufacturers have to use as much raw cell stock as possible. This means cells with poor leakage resistance are more likely to be accepted by manf. to be used in panel assemblies. This makes bypass diodes even more important to prevent hot spots in these poorer leakage cells.

You can use higher leakage cells by decreasing the number of series cells covered by one bypass diode. The less series cells per bypass diode, the lower the maximum reverse bias on a single cell will be.

The article is not incorrect but it is misleading. If a single 12v panel is used (or a number of 12v panels in parallel is used), the maximum reverse voltage is limited on each cell by the low overall system voltage. Any bypass diodes needs to be within the panel, splitting up the 36 cells into, say, two groups of 18 cells to be of any benefit. Putting a diode across the total 12v panel output on a 12v battery system does little good. Do not misinterpret from the article that bypass diodes, in general, are unnecessary.

[jg6871]

Hi everyone - I'm hoping someone can offer advice regarding bypass diodes in solar panels.

Does shading, on modules with bypass diodes installed, lead to long term cell damage?

My understanding is that for a 72 cell module (with 1 bypass diode per 24 cells), where there is shading across at least one entire cell, the group will be short circuited by the bypass diode (i.e. it will produce no power). This is because the shade causes an inverse voltage across the shaded cell(s), which will increase until it hits the forward bias voltage of the bypass diode- when it does, the bypass skips the group from the module. Is it possible to get cell temperature rise between the diode being negative and forward bias? Furthermore is it possible to still get hotspots in a system with bypass diodes installed, and thus lead to long term degradation?

The reason I ask is because whilst sources state that bypass diodes protect cells from long term damage, I have completed infra red testing which suggests differently. I've set up a 4x175W array with a pipe standing vertically in front of the array (providing shade). I've attached three pictures showing a set of infra-red photos of the underside of the panels - taken at 12.10pm,13.10pm and 13.40pm. Given that each of these panels have three bipass diodes attached, I am surprised that there are hotspots (in yellow = >100deg c). Most of the cells under shading see a temperature drop (as no sunlight is hitting them), however surely the diodes would protect against the odd hotspot? You can see that the panel on the right, which had the shading on it earlier in the day, still has hotter cells even though the shade is not acting on the module- why would this be?

All advice will be much appreciated!

[RCinFLA]

For a panel greater then about 12vdc, shading on a single cell can result in high reverse bias on that shaded cell. If its overall shunt resistance is low or there is a defect spot that has low resistance then the high voltage will create heatng that would not normally occur in it normal illuminated forward bias condition of only 0.5 to 0.6 vdc across shunt resistance.

If you have a shaded cell with a low resistance shunt and it is forced into a high reverse bias condition it will get hot which will degrade over time. The primary manufacture / UL object with bypass diodes is to prevent it from getting so hot that it creates a fire hazard.

In some panels the bypass diodes are too small or insufficiently heat sunk to prevent them from early failures especially if panels are subjected to a shading cycle every day. There was a thread about one manufacturer putting three lower amperage axle leaded diodes in parallel to get current rating demand. This is a bad design.

As to your IR pictures. You did not include an unshade picture for reference. Are these panels in series or parallel?. The third picture looks like there is no shading on the right most panel that shows the hotter cells. The most likely reason for what you are seeing is the hotter cells are not making as tight a contact to the front glass of the panel which is the primary heat sinking source for the cells. Remember, with cell efficiency in the range of 15%, 85% of the sun's energy hitting the cell is just heating the cell.

[jg6871]

Very helpful & quick response RC, many thanks. To clarify a few points:

- It may be helpful if gave you some context on what I'm trying to design. I'm working on a University project where we are designing a PV system which will be deployed across Africa and India with a life span of 10 years. It will see shading every single day due to a mast attached to it - we are trying to integrate the panels onto the mast. Ultimately I am trying to find out whether daily shading across PV modules will cause long term damage to the solar panels (I can design for the drop in power by adding more panels/wiring etc). If there is long term damage, then I will have to completely rethink the design.

- You mentioned bypass diodes may be insuffiently heat sunk to prevent them from early failure especially if subject to cycle shading. Is there a certain module that is well designed for this, or is this affect applicable to all modules? Furthermore, does a bypass diode actually stop cell temperature rising, or does it just limit the rise? This would potentially explain why there are hotspots appearing on the IR image as the shade moves from right to left. Are these temperatures actually damaging to the cells (maximum 80-100 deg c) and are they expected? Note: the module type is a BP4175 175kWp with three bypass diodes 1 every 24 cells.

- IR Testing - The system is a 2 string (2 modules per string) array, the two panels on left are one string and those two on right are the other. I've attached the IR pictures from 12-3.30pm, the shading was removed completely at 2.30pm (should give you a reference). Interesting point regarding the heat sink on the front glass.

-Other than manufacture defects, what makes a cell shunt resistance low?

Once again, many thanks in advance for your help

[jg6871]

Further IR pictures - The name of the file corresponds to the time taken i.e. 13_05 = 1.00pm.

[RCinFLA]

Look at the spec on the diode and the heat sinking necessary to bypass worse case current. If your panel cells can produce 7 amps then the diode will have some bypass voltage drop, between 0.7 and 1.0 vdc, times the worse case current that must be bypassed. Diodes could be asked to have 5 watts or more of heat dissipation. This requires a fairly good heat sink.

Do not put lower amperage diodes in parallel trying to accomplish current rating needed. The hottest one will take most the current and self destruct.

The best bypass diodes are TO220 cased devices that can be attached to the junction box case for heat sinking. Usually the panel cells are arranged in a down and back row pair to allow easy connection to bypass diodes. Look for the lowest voltage drop diode at needed current. Diodes must have a PIV greater then the total series stack voltage just in case there are other failures that might put full series stack voltage across diodes. 600 PIV are usually sufficient.

[jg6871]

Hi RC,

Unfortunately I did not understand you last post (sadly not an electric expert), would you mind putting it in simple terms...

Ultimately I am trying to find out
1) whether cyclic partial shading across a solar array leads to long term degradation, even with bypass diodes installed (as the IR images clearly show hotspots forming on brand new panels with limiting reverse current of 5.58A and installed with 3x Schottky diodes).
2) If no to the above, then why would the IR images show hotspots (other than the heat sink on the right hand panel once the shading is removed)? Are these hotspots damaging to the PV panels?

Many thanks in advance,

[Mike90250]

Unfortunately , electronic engineering terms are not readilly explainable to the layman. Before you get too far along into this venture, you should get a couple electronics (not electric) classes. Night school, Jr College, maybe even some web based simple electronics classes, would suffice, to at least learn the language, otherwise, you will always "be in the dark". You have to meet the free help halfway.

Diodes = one way valves, that can blow out with too much (heat, voltage, or amps)

Heat is generally bad for all electronics

Local hot spots indicate something is wrong, heat on a PV panel should be fairly even. What is the degree graident on the photos - what color is what ??

Every thermal cycle introduces a small amount of damage, eventually the damage becomes so great, the part fails. Most well made, factory finished PV panels will withstand 20+ years of thermal cycles. Adding some extra daily shadows to that will reduce it, by causing extra cycles.

[jg6871]

Hi Mike,

Thanks for your reply, and yes you're spot on I do need to take some more electronics classes!

On the IR images, the white areas are hot, the black areas cold. The maximum temperate (white) is around 80-90deg c, the minimum is around 20 deg c (black). The panels are brand new with three diodes per module, and contain two strings connected in parallel, each with two modules.

I've used your posts on other threads to calculate the diode required for the cells in the IR image (each cell produce 0.6V/5.38A - BP4175T, 72 cells). The bypass diodes therefore should be sized to deal with around 66V/11A (diode voltage ~1.5x array Voc, ampage is 2x panel max amps). I know these panels contain Schottky diodes (with around 0.5V drop) - I'm unable to find the specification, however, and hence check their rating? Given that the diodes will forward bias at ~0.5V/5.38A, the heat dissipation is ~2.7W. Would you mind confirming my logic?

You mentioned that adding some extra daily shadows will reduce the panel lifespan even further than if the panels were only subject to the normal thermal cycle without shading. Would you be able to roughly estimate this figure of how many additional years it takes off the panels? I've heard that some manufacturers actually void their warranty if the panels have been installed in shade, as they are not sure about the long term degradation caused by shading on solar panels with bypass diodes installed.

Once again many thanks.