Jump to content

Halfcell Solar PV and Sunsynk MPPT


cvschalk

Recommended Posts

I'm having issues with my Sunsynk 8kW inverter and 550W Halfcell solar panels when partially shaded.  The Sunsynk MPPT algorithm does not seem to handle half-cell shading well (with their multiple power peaks).

When the top half of only one of my 550W panels are shaded (out of a string of 8 panels), I reach only half of the expected current.  If the MPPT was clever, it would have pushed the current to activate the bypass diode on the shaded panel and achieve the full production current.

As soon as the shading clears, the current jumps to double the value immediately:

image.thumb.png.bf7df68f5e3265ede937afd3491951ba.png

The nett result is pretty much 60% of the production I should achieve, for half the day - all only because half of one panel is shaded....

 

Does anyone know for a fact if the Sunsynk FW supports the periodic "current scanning" for half-cell panel peak detection?

Link to comment
Share on other sites

1 hour ago, cvschalk said:

I'm having issues with my Sunsynk 8kW inverter and 550W Halfcell solar panels when partially shaded.  The Sunsynk MPPT algorithm does not seem to handle half-cell shading well (with their multiple power peaks).

When the top half of only one of my 550W panels are shaded (out of a string of 8 panels), I reach only half of the expected current.  If the MPPT was clever, it would have pushed the current to activate the bypass diode on the shaded panel and achieve the full production current.

As soon as the shading clears, the current jumps to double the value immediately:

image.thumb.png.bf7df68f5e3265ede937afd3491951ba.png

The nett result is pretty much 60% of the production I should achieve, for half the day - all only because half of one panel is shaded....

 

Does anyone know for a fact if the Sunsynk FW supports the periodic "current scanning" for half-cell panel peak detection?

This is pretty normal and a lot of us will find the same result during shading. It is not always easy to quantify what the loss will be. 

In this case the drop in output happens to be 40% I feel the top panel is the real culprit. 

IMG_20230626_145753.thumb.jpg.45f44c8e83d7a4a96b657c140d3746e0.jpg

Link to comment
Share on other sites

@Scorp007 thank you for the feedback.  For traditional panels, shading in full sun is actually quite easy to quantify and handle, since the shaded cell current potential goes to very close to zero.  Then as soon as the MPPT starts to load the current, the bypass diode of the panel activates and that third of the panel is shorted out (each panel has 3 bypass diodes, allowing a third of the panel to be deactivated/bypassed if shaded).  The end result is that you still end up producing 2/3 of the power of that panel and if your String is 9 panels long, then you will still produce 96% of the power of the string (26/27 "thirds").

What makes things a little more complex to understand and to manage by the MPPT, is that half-cell panels are wired like two smaller full-cell panels in parallel.  You still have 3 bypass diodes, but the panel is broken into 6 sections.  These sections are wired in parallel, two at a time, each pair sharing the bypass diode. 

An Overview - Advantages of Half Cut Cells Photovoltaic Solar Panels -  Voltacon Solar Blog

So, given this configuration, if the panel now experience partial shading during full sun, the MPPT has two ways to handle this shading:

  1. it pushes the current of the string into the zone of expected current (assuming full sun) and thus forcing the bypass diodes in the affected area to activate.  This will result in 96% production for a limited partially shaded string.
  2. It only pushes the current a little bit.  The bypass diode of the panel remains inactivated, because the parallel complement of the shaded cells are still producing current - so current is flowing! But since its only half the current, the whole string are now forced to only produce half their current potential.  This results in exactly 50% production.

The problem with old school MPPTs are that they start to increase the current slowly.  Then they reach point 2 above.  Since this is a local power peak, their algorithm "gets stuck" here.  It does not know about and never reaches the second power peak, which is almost double the size!  The modern half-cell-compatible MPPTs knows that they need to sweep the full current range to identify where the true maximum peak is (the 96% one, not the 50% one).

I would have fully expected Sunsynk to be compatible with halfcell panels - given this is the newer technology panels and what everyone is installing.  But it seems they might not currently be.  I can only imagine how much wasted energy this produces every day because of partial shading on halfcell panels.

 

 

Link to comment
Share on other sites

22 minutes ago, cvschalk said:

@Scorp007 thank you for the feedback.  For traditional panels, shading in full sun is actually quite easy to quantify and handle, since the shaded cell current potential goes to very close to zero.  Then as soon as the MPPT starts to load the current, the bypass diode of the panel activates and that third of the panel is shorted out (each panel has 3 bypass diodes, allowing a third of the panel to be deactivated/bypassed if shaded).  The end result is that you still end up producing 2/3 of the power of that panel and if your String is 9 panels long, then you will still produce 96% of the power of the string (26/27 "thirds").

What makes things a little more complex to understand and to manage by the MPPT, is that half-cell panels are wired like two smaller full-cell panels in parallel.  You still have 3 bypass diodes, but the panel is broken into 6 sections.  These sections are wired in parallel, two at a time, each pair sharing the bypass diode. 

An Overview - Advantages of Half Cut Cells Photovoltaic Solar Panels -  Voltacon Solar Blog

So, given this configuration, if the panel now experience partial shading during full sun, the MPPT has two ways to handle this shading:

  1. it pushes the current of the string into the zone of expected current (assuming full sun) and thus forcing the bypass diodes in the affected area to activate.  This will result in 96% production for a limited partially shaded string.
  2. It only pushes the current a little bit.  The bypass diode of the panel remains inactivated, because the parallel complement of the shaded cells are still producing current - so current is flowing! But since its only half the current, the whole string are now forced to only produce half their current potential.  This results in exactly 50% production.

The problem with old school MPPTs are that they start to increase the current slowly.  Then they reach point 2 above.  Since this is a local power peak, their algorithm "gets stuck" here.  It does not know about and never reaches the second power peak, which is almost double the size!  The modern half-cell-compatible MPPTs knows that they need to sweep the full current range to identify where the true maximum peak is (the 96% one, not the 50% one).

I would have fully expected Sunsynk to be compatible with halfcell panels - given this is the newer technology panels and what everyone is installing.  But it seems they might not currently be.  I can only imagine how much wasted energy this produces every day because of partial shading on halfcell panels.

 

 

Thanks @cvschalk

I never worked out the 50% drop. Looking at the picture you will notice that my drop is actually more than 50% for only 2 of the 6 columns on the bottom left CORNER with shading. This relates to a different string of 6 panels in series. They all face north on a narrow wall but the gap between a certain 2 is too small and the top corner throws a shadow on the one bottom panel for about 6 weeks during this time of the year. 

As you can see the drop is from 750 to 314W for a short period and then shoots up back to about 750W.

This is on a string inverter. No train smash but a good practical outcome with a not too efficient MPPT. 

IMG_20230628_091325.thumb.jpg.899a2cad55f4d40943db3a86adff5083.jpg

 

Link to comment
Share on other sites

@Scorp007 I see you also have half cell.  What inverter do you have?  Imagine that you could have have more than 2kW production without this shading! (complete the graph to the left)

 

here are some simple formulas to calculate shading.  Assuming that panels are mounted in portrait orientation, so vertical shading is the blocking of the long edge cell and horizontal shading is the blocking of the short edge cells.

 

Full cell PV panels:

horizontal shading = loss of full panel for each shaded panel.  Rest of unshaded string at 100% production (so 12.5% loss for a 1/8 panel string, 25% for 2/8, etc)

vertical shading = 33% loss of full panel, per 1/3 of the panel shaded.  Rest of unshaded string at 100% production (so only 4% loss per 1/3 of panel shaded)

The above is true for both old and modern MPPTs

 

Half cell PV panels:

horizontal shading = same as above for modern MPPTs, but 50% loss of the full string for old MPPTs (because the bypass diodes never activate)

vertical shading = same as above for full cell.  Both for old and new MPPTs

 

You might wonder what the use of halfcell panels are then if they cause so much trouble?  If you have a modern MPPT, they actually perform very well if you have partial vertical and horizontal shading (a couple of cells shaded anywhere on the panel).  Then you can end up in the zone between 33-100% loss on a single panel, which can be very advantageous if you have a tree/chimney/antenna throwing shading multiple panels on a string (eg if you have horizontal edge shading on your full string, you loose you whole string for full cell, but only 50% for half cell).  But for this you need a MPPT with a half cell algorithm.

 

Link to comment
Share on other sites

1 hour ago, cvschalk said:

@Scorp007 I see you also have half cell.  What inverter do you have?  Imagine that you could have have more than 2kW production without this shading! (complete the graph to the left)

 

here are some simple formulas to calculate shading.  Assuming that panels are mounted in portrait orientation, so vertical shading is the blocking of the long edge cell and horizontal shading is the blocking of the short edge cells.

 

Full cell PV panels:

horizontal shading = loss of full panel for each shaded panel.  Rest of unshaded string at 100% production (so 12.5% loss for a 1/8 panel string, 25% for 2/8, etc)

vertical shading = 33% loss of full panel, per 1/3 of the panel shaded.  Rest of unshaded string at 100% production (so only 4% loss per 1/3 of panel shaded)

The above is true for both old and modern MPPTs

 

Half cell PV panels:

horizontal shading = same as above for modern MPPTs, but 50% loss of the full string for old MPPTs (because the bypass diodes never activate)

vertical shading = same as above for full cell.  Both for old and new MPPTs

 

You might wonder what the use of halfcell panels are then if they cause so much trouble?  If you have a modern MPPT, they actually perform very well if you have partial vertical and horizontal shading (a couple of cells shaded anywhere on the panel).  Then you can end up in the zone between 33-100% loss on a single panel, which can be very advantageous if you have a tree/chimney/antenna throwing shading multiple panels on a string (eg if you have horizontal edge shading on your full string, you loose you whole string for full cell, but only 50% for half cell).  But for this you need a MPPT with a half cell algorithm.

 

This string I posted the result is full cell with 60 cells. Only one of the 6 panels have this small patch of shade on 2 of the six columns on the one panel. 

This is on a Solis grid tied inverter. The string is 1.62kW in total. Currently on a full sun day the peak production varies from 1150 to 1250W. The graph is for the full day yesterday. Each day one gets the same drop and the time of the drop changes at about 6min per day. From now on the shading will reduce every day. Only 2 panels are too close to the next one so the shade is only a problem on 1 of the 6. Horisontal gap between these 2 panels are 800m. Panels are at 28deg and size is 1660x990mm. I have already drilled the holes to move this one panel but time is the problem to move it. 

As can be seen the problem only starts end of April with shadows getting longer. 

Don't laugh at my my small system 😀😀😀

IMG_20230628_130503.thumb.jpg.a67060e43d8e8f8db85183ca3a7f7d9b.jpg

Edited by Scorp007
Link to comment
Share on other sites

@Scorp007 Ah I see.  So the drop is only for a short period of time and only for <20% loss?  This will fits with my understanding of full cell panels.

 

BTW, on full cell panels the cell-columns are wired 2:2:2 between the three diodes.  So if you know which columns blocked, you will know how many bypass diodes activate (1 or 2 of the three)

Link to comment
Share on other sites

11 minutes ago, cvschalk said:

@Scorp007 Ah I see.  So the drop is only for a short period of time and only for <20% loss?  This will fits with my understanding of full cell panels.

 

BTW, on full cell panels the cell-columns are wired 2:2:2 between the three diodes.  So if you know which columns blocked, you will know how many bypass diodes activate (1 or 2 of the three)

Yes it would fit but the last graph is on 28 Apr. 

It then gets worse as per my graph of yesterday where the output drops from 750 down to 314W. Every day from now it will be a lower loss. By the 5th of Aug it is gone. 

Link to comment
Share on other sites

  • 2 weeks later...
On 2023/06/28 at 1:51 PM, Scorp007 said:

Yes it would fit but the last graph is on 28 Apr. 

It then gets worse as per my graph of yesterday where the output drops from 750 down to 314W. Every day from now it will be a lower loss. By the 5th of Aug it is gone. 

So just an update. While the shading drop was 430W on 28 June it is currently down to only 100W 😀

IMG_20230713_093550.thumb.jpg.b5924fcbc56e4aa25102b727748326cd.jpg

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...