question

nikbok avatar image
nikbok asked

What is best MPPT configuration in EAST-WEST orientation

Hi,

I would appreciate your comments on using MPPTs in east-west roof orientation.

I have a roof that has west(-70dgr) / east (+110dgr) position and I intend to place 16 panels on each side. The idea is to connect them in four 4S1P loops on each side. My original idea was to couple 2 loops to a 250/85 MPPT ie 4 loops to 2 MPPTs on each side.

What I'm wondering is how good it will be if I connect all 8 loops to 2 larger MPPTs (250/100) in such a way that two loops from the WEST and two loops from the EAST go to one MPPT.

screenshot-2022-10-21-at-104136.jpg

The voltage should be OK, because the four loops are connected in parallel. Max current can be a bit debatable but from what I know due to the position of the panels they will never produce the same amount of energy at the same time. My calculations show that the maximum current for two loops is approx. 75A at an extreme temperature of -20C and at 12-2pm in SOUTH position!

I think my panels in west-east orientation will never produce more than 60-65A as a max probably less and when one side is at its max, the other is at its minimum or very little. Therefore, I think it could work with two MPPTs without cut-offs. What do you think?

MPPT ControllersESSSolar Panel
screenshot-2022-10-21-at-104136.jpg (735.3 KiB)
3 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

kevgermany avatar image kevgermany ♦♦ commented ·

What's East/West, the angled part of the roof or the ridge?

Panels with equal irradiance should be connected to an MPPT.

Use the Victron calculator to decide on MPPT models.

0 Likes 0 ·
nikbok avatar image nikbok kevgermany ♦♦ commented ·

Yes, EAST side of roof is +110drg (from south) and WEST side of roof is -70dgr from south. Roof angle is about 30dgr.

All panels are same and yes I did use Victron calculator but it does not take in count panel orientation, It asumes that panels are facing south (0drg)

0 Likes 0 ·
Schunk avatar image Schunk commented ·

Hi freaks,

look for similar arrangement east - west with identic panal arrangement and exact 90 - 180 ° with roof angle of 45°.

Made all calculations with 1000W/m² and 800 W/m² - result same MPPT selction.

I contacted the dealer and via VICTRON the answer was : max PV power as per data sheet.

Data sheet shows 2 remarks: a) ~ possible increased current will be limmited by controler

b) PV voltage has to be higher than Vbat + 5 V etc

Your discussion on MPPT operating point might be reduced with new algorism in MPPT regulator with follows 3 tracking points intsted of nomal 1.

Optimaler Solarstromertrag Sorry for German Version, could find quik english one.

Das Solarladegerät verfügt über einen innovativen Tracking-Algorithmus. Dadurch wird der Energieertrag immer maximiert,

indem der optimale MPP (Maximum Power Point) angefahren wird. Im Falle einer Teilverschattung können auf der StromSpannungskurve zwei oder mehr Punkte maximaler Leistung (MPP) vorhanden sein. Herkömmliche MPPTs neigen dazu, sich auf

einen lokalen MPP einzustellen. Dieser ist jedoch womöglich nicht der optimale MPP.

wait for your comments

KR

Eckart


0 Likes 0 ·
8 Answers
gazza avatar image
gazza answered ·

Whatever you do I would keep the panels with the "same orientation" connected to a specific MPPT(s). So, NOT your option B since the mixed irradiance between east and west will be feeding in to the same MPPT and reducing the output to the lowest common denominator.

2 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

nikbok avatar image nikbok commented ·

Hi, thank You for the answer.

I was thinking like that basicaly from the beginning but starting to question that aproach of having "all same". You say:

"mixed irradiance between east and west will be feeding in to the same MPPT and reducing the output to the lowest common denominator."

Is that really so? I'm not 100% sure about that.

If you have 2 equal panels (eg 400W) in parallel facing same orientation and one of those panels is completely in shadow, lowest common denominator for this set is 0W (or very close to 0W, because of the second one in shadow) but the MPPT will still use energi from first panel, all 400W (just en example).

Am I wrong here?

I am not expecting to produce 800W in this setup, I do know it is physicaly impossible.

What I would like to know is how big degradation of panel ONE imput is realistic?

Is it couple of % or we are talking about 30,40,50%.

Anyone that have setup like this?

1 Like 1 ·
rikhf avatar image rikhf nikbok commented ·

...Is it couple of % or we are talking about 30,40,50%....

Around 5%

0 Likes 0 ·
gazza avatar image
gazza answered ·

You are correct that was a bad choice of words...! It will work.. although sub-optimally, since neither string will necessarily be at the max power point as the controller tries to find a compromise. I couldn't give a specific degradation % though? It's just not best practice.

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

rikhf avatar image
rikhf answered ·

Can you give the panels specs? And do you have shading?

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Schunk avatar image
Schunk answered ·

Hi, thanks for your comments;

I agree that it is not best practice but on the other hand you has to look for costs.

Use of one big MPPT for half a day is not economic; power from sun is free of charge.

MPPT input feel no difference if one string has different input indipendent if it is based on orientation or shadow.

My Calculation:

- Victron MPPT calculation is based on 1000W/m² - realistic is 800W/m² == factor 0,8

- orientation E - W is not prepared reduction == factor 0.8 - 0,85

- total power input than is approx. 0.66 of nominal MPPT capacity

- we will use this approach for our design and order ( for safty ond possible failure we will have separat cablung from roof to MPPT´s)

Hope for homicide arguments!

KR

Eckart

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

sharpener avatar image
sharpener answered ·

I think if there will only be meaningful output from one set of panels at a time (as here) you could combine the strings as in your scheme B by adding external isolating diodes for each string.

Then whichever string has the higher voltage will "win", and reverse-bias the diode to the other string so it produces no current at all.

This will ensure the MPPT will operate at the correct point for the "winning" string. Without the diodes it will not.

2 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

nikbok avatar image nikbok commented ·

I'm leaning towards option B myself but I would like to hear others opinion as well as real life experience with this setup.

I've heard from some people that "dark side" will not degrade input of "sunny side" and final result will be about 100%, but I would really like to confirm that with real life data.

Most of better panel have isolating diods if not doing it yourself is very easy.

0 Likes 0 ·
sharpener avatar image sharpener nikbok commented ·

Those are not isolating diodes they are bypass diodes. They are connected in reverse across the whole panel or more typically three across one-third of the cells each. They work by allowing current from cells that are in the sun to flow round cells that are shaded.

You could alternatively make use of them by connecting strings from opposite sides of the roof in series, then whichever string can generate more current will forward bias the diodes in the other string and it will effectively be shut down.

However I didn't/don't recommend this because

  1. they are not intended to be used in this way
  2. you will lose 3N x the voltage drop across the diodes because there are 3 per panel instead of 1 per string
  3. at some point in time you will have both arrays producing exactly equal current. Then both will be generating and the terminal voltages will sum so you risk over-volting the MPPT.


0 Likes 0 ·
gazza avatar image
gazza answered ·

It would be good if you do this to get some "data" and put this one to bed going forward?
There are definitely successful anecdotes even though the theory says it should be sub-optimal.(since for "most" of the day both arrays will not be at their MPPT). According to others you may find that in overcast conditions this actually works as well if not better since the irradiance is way more diffuse? So, if you are going to do it... it would be possible with a couple of isolator switches (which you should have anyway) to "switch" in the east/west arrays either on their own or together? This would give the best of both worlds. In fact you could even automate this with a contactor set up if the results deemed it worhtwhile?

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Schunk avatar image
Schunk answered ·

It took a time to install all panals, string cabeling to both roof sides at 45 ° in hot summer conditions.

Equipment: West 4x 3 x 410 WP ; East 3 x 3 x 410 WP (same Modules) with connection via double fuse. After installtion of DC curent meter per roof side we startet tests around mid time.

First East, followed by West and than both sides in parallel to MPPT 250 / 60.

This test were made with 2x 3x3 and with 1x 3x3 and 1x 3x4 modules, beside a short transient response, we found no differences in single current per side and summery of both sides.

From Diagramm you can see details of behaivier with parallel operation.

We will continiue investigation with higher radiation in spring time.

2023-08-14-g35-ess-mpii-mppt-250-60.jpg

2023-11-08 G35 PV O3 -W4 mit Su um 12.25.15_fec5cef3.jpg


2023-11-08-g35-pv-o3-w4-mit-su-um-122515-fec5cef3.jpg (42.6 KiB)
2023-08-14-g35-ess-mpii-mppt-250-60.jpg (2.6 MiB)
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Ibraheem H avatar image
Ibraheem H answered ·

Any update to your tests in spring time? I’m currently debating this exact decision. Thanks

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.