MPPT 450/200 apparent throttling

Question

question

b1ayd asked Nov 23, '23

MPPT 450/200 apparent throttling

Good day Ladies and Gentlemen

I seem to have a problem with my MPPT 450/200 throttling the solar yield when surpassing around 7kw.

System spesifications:

Victron Quattro 48/15000/200
Victron MPPT 450/200
Victron CERBO GX
2x Freedom Won Lite 15/12 LiFePO4 batteries
18x Trina Vertex Mono 550w solar panels

Please see below

All four strings are used and connected as follows:
1 - 5 x 550w = 2750w
2 - 5 x 550w = 2750w
3 - 4 x 550w = 2200w
4 - 4 x 550w = 2200w
Total = 9900w

The throttling starts when approaching 7kw and ends when the load drops below 6kw.

Please see overview below:

The max solar power was 6.77kw at 09:00 - 10:00. This is on a clear day with no cloud/haze/fog etc. The outside temperature was around 37 degrees Celsius. Surely 9.9kw worth of panels and a 200amp charger should be able to deliver more than 6.77kw in one hour?

Any advice or recommendations will be greatly appreciated.

I am a novice and have limited knowledge about the workings of solar.

MPPT Controllers

mppt-throttle-vrm.png (65.3 KiB)

overview.png (45.2 KiB)

2 |3000

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

Guy Stewart (Victron Community Manager) answered · Feb 01 at 12:42 AM

Hi @b1ayd,

Thanks for updating the firmware to the latest current version - it's always the first thing to do in cases like this.

I had a closer look at your system and the PV array.

What I see is that your PV array configuration with 4 panel strings is VERY close to the minimum permitted, with a maximum power voltage of only 128V (4 x 32V). This is a 450V MPPT, and operating it near the very minimum of its MPPT voltage range will mean it cannot operate efficiently.

I would suggest reconfiguring your PV array (by a professional installer as these are potentially fatal voltages).

A 6 panel minimum on 3 strings is one option, and just leave one tracker empty.

While still not optimal it will at least give your MPPT a chance to improve performance (max PV voltage 228V)

Two 9 panel strings MIGHT yield even better results (max PV voltage 342V), and is evenly divisible, but then you would be constrained by the 4000W limit on each of the two trackers (9*550W=4950W). Note that this is a 'soft limit' where the MPPT just won't track to exceed it, so it's permitted. As compared to a HARD limit like the PV input voltage limit of 450V.

It's difficult to say if 6 panels with limited (but improved) power point tracking, or 9 panels with limited tracker power would provide the best yield in your case. Both are still constrained from the maximum theoretical, BUT either would also be an improvement from your current performance and worth trying.

So to summarise, to get more power out of the MPPT I believe you need to reconfigure your PV array to increase the PV voltage.

This needs to be done in such a way as to stay within the other limits of the MPPTs but those aren't an issue here as far as I can see.

4

2 |3000

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

b1ayd commented · Feb 01 at 06:54 AM

HI Guy,

Thank you very much for the information. It does makes sense and we will test 3x 6 panel strings and also 2x 9 panel strings.

0 ·

Guy Stewart (Victron Community Manager) ♦♦ b1ayd commented · Feb 01 at 07:05 AM

Good luck, I hope it goes smoothly. Please report back once you have the results, I am curious if my suggestion resolves the issue (it might not), and also what the best performing configuration is.

0 ·

b1ayd Guy Stewart (Victron Community Manager) ♦♦ commented · Feb 01 at 09:04 AM

Will do

0 ·

b1ayd commented · Mar 06 at 01:11 PM

Hi Guy, I have to thank you for this information. The strings were changed from 4 strings of 5 5 4 4 to 2 strings of 9. This has eliminated the throttling appearance completely. So you were 100% correct that the voltage was too low for the MPPT to work with efficiently.

We are however hitting the tracker limits of 4000w for about 2 hours during the peak sun but I would say we are only loosing maybe 1kw to 1.5kw of power due to this.
We have since changed the strings to 3 of 6 6 6 but even though the results were better than the 4 strings of 5 5 4 4 it is not as good as the 2 strings of 9 9.

Thanks again for helping to get to the root of the problem.

0 ·

Modifications Space

Node Red Space

Answer 1 · 2023-11-23T06:49:10Z

nickdb answered · Nov 23, 2023 at 06:49 AM

Some things to consider, at 37C, which is hot outside, the panel temp is incredibly high and they will be derating, as will the chargers and inverter. So that amount does not look unusual for the temp.

You will never get the panel rated power, which is quoted for absolutely ideal conditions and orientation at temps of 25C.

2

2 |3000

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

Alexandra ♦ commented · Nov 23, 2023 at 06:55 AM

Just to add on 6700/9900 is about 67% so pretty close to expected yield percentage ( between 60 to 70% of installed) for the southern Africa region.

1 ·

b1ayd Alexandra ♦ commented · Nov 23, 2023 at 07:51 AM

Alexandra, I can understand the 67% yield. The problem for me is the apparent throttling. If the panels deliver less power it should not look like throttling?

0 ·

Answer 2 · 2023-11-23T08:01:32Z

nickdb answered · Nov 23, 2023 at 08:01 AM

By throttling are you referring to the dip?

We would need to see a lot more data to understand that:

What was the battery SOC?

What was its CCL and CVL at the time.

What loads were running?

Was that after grid returns?

How is ESS configured?

It would be guess work without that.

If an mppt appears to shutdown or dramatically drops PV production, it is because it is being instructed to do so.

The system is controlled by the batteries and unfortunately parallel FW battery combos have been known to have issues. Have had to help too many people with problems originating there.

The FW's also vary the CVL when they get full or hit an imbalance, when it drops the bus voltage the mppt's stop production and it begins to drain the batteries until it is happy.

Most likely your issue is something expectedly normal, or known to be abnormal.

11

2 |3000

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

b1ayd commented · Nov 23, 2023 at 08:18 AM

Hi Nick

Thank you for the information. The throttling I refer to is the part that starts around 09:30 and ends at around 14:00. The power delivery appears much more smooth before and after these times. The batteries charge at 9kw as set when the ESS is changed to charge batteries so I doubt that is where the problem originates from.

The batteries did not charge fully during this specific day.

I try to understand why the power delivery is smooth until it reaches about 6.5-7.0 kw after which it oscillates between 4kw and 7kw until the power goes below kw.

The oscillations is also not based on the SOC of the battery. If the battery is full and enough load is generated by other items (aircon, oven, kettle, etc) to test it does the same.

I will be switching off 2 strings and see if it appears to same to get some extra info.

0 ·

nickdb ♦♦ b1ayd commented · Nov 23, 2023 at 08:42 AM

You will need to zoom into the charts. If the sawtooth pattern has the dips spaced every 10 minutes, then that is the MPPT algo tracking the maximum power point, all MPPT's do that.

You will see it once the sun properly comes up and there are loads the MPPT is trying to cover.

1 ·

b1ayd nickdb ♦♦ commented · Nov 23, 2023 at 09:34 AM

It appears to be every minute.

0 ·

saw-tooth.png (30.1 KiB)

b1ayd nickdb ♦♦ commented · Nov 23, 2023 at 09:36 AM

And another one for 24 hours to highlight the dip when it starts throttling

0 ·

mppt-throttle-24hrs.png (25.0 KiB)

Alexandra ♦ b1ayd commented · Nov 23, 2023 at 09:52 AM

With the outside temps being above the 25° rating, (and these units run hot) you may need to look at moving the hot air away from it.

0 ·

nickdb ♦♦ Alexandra ♦ commented · Nov 23, 2023 at 10:23 AM

It is very hot at the moment, I am wondering if that RS isn't derating due to heat.

How/where is it installed?

0 ·

b1ayd nickdb ♦♦ commented · Nov 23, 2023 at 10:25 AM

This is not specific to today. It has been like this from the install date.

0 ·

nickdb ♦♦ b1ayd commented · Nov 23, 2023 at 10:29 AM

Humour me, send me the BMS limits chart for the Freedom won over the same period.

Can you also add the PV voltage/current chart for each tracker on the RS.

Can you also check that the battery voltage reported by the inverter is the same as that reported by the RS.

0 ·

b1ayd nickdb ♦♦ commented · Nov 23, 2023 at 10:37 AM

I think this should be everything.

VRM Portal.pdf

0 ·

vrm-portal.pdf (219.6 KiB)

nickdb ♦♦ b1ayd commented · Nov 23, 2023 at 12:17 PM

Thanks, that didn't convert well to PDF, the colours (legend) and scale/measurements are missing so it is a bit tough to read.

Weirdly, unless it is a conversion error, your CVL is 55.5V. Those batteries have a 55.8V max CVL that the BMS should send, so something seems to be limiting it.

0 ·

b1ayd nickdb ♦♦ commented · Nov 24, 2023 at 05:26 AM

I have printed it again

VRM Portal.pdf

0 ·

vrm-portal.pdf (329.0 KiB)

Answer 3 · 2023-11-24T06:48:39Z

Guy Stewart (Victron Community Manager) answered · Nov 24, 2023 at 06:48 AM

Hi @b1ayd,

It's really important that you update the firmware on your MPPT RS, there have been current control improvements added.

If you have two way communication enabled, this should be a simple one click update in the firmware update section of the device list on your VRM site.

10

2 |3000

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

b1ayd commented · Nov 24, 2023 at 06:55 AM

Thank you very much I will do the update ASAP to see if it makes a difference to the issue.

0 ·

b1ayd commented · Nov 27, 2023 at 07:24 AM

The software was updated but the same thing is still happening.

0 ·

b1ayd commented · Nov 27, 2023 at 11:11 AM

The update seems to have made the yield quite a bit worse.

0 ·

Guy Stewart (Victron Community Manager) ♦♦ b1ayd commented · Nov 27, 2023 at 11:23 AM

Hi @b1ayd

Thanks for that,

Now there is another update for the performance issue, please note there is a bug in this 1.16b1 that incorrectly reports an overcurrent condition that will be fixed shortly, but it’s no issue to use it now.

https://community.victronenergy.com/questions/241677/mppt-rs-shade-performance-testing-v116-beta.html

0 ·

b1ayd Guy Stewart (Victron Community Manager) ♦♦ commented · Dec 04, 2023 at 07:45 AM

Hi @Guy Stewart (Victron Community Manager)

We have done the update to 1.15 which has made the production even worse. We are now barely doing 6kw per hour in peak sun from 10kw of panels with a clear sky and no shade.

The interesting thing is it will settle at 5.7 kw and when switching off one string (2.5kw) will decrease to 5.2kw???

This is quite bizarre.

0 ·

nickdb ♦♦ b1ayd commented · Dec 04, 2023 at 08:06 AM

Did you load the 1.16 beta as requested, which addresses a known performance issue with 1.15?

0 ·

b1ayd nickdb ♦♦ commented · Jan 31 at 05:25 AM

HI Nick, the beta was loaded but issue is the same

0 ·

Alexandra ♦ b1ayd commented · Jan 31 at 06:36 AM

@b1ayd

The new firmware officially released (v1.16) has much faster tracking.

It is now a 5second scan every 5 minutes. The behaviour should have changed with tracking.

If it is throttling (or yeilds lower than expected) still then there must be some other interaction contributing to it.

0 ·

b1ayd Alexandra ♦ commented · Jan 31 at 06:45 AM

The behaviour has changed but it still appears as if it throttles and then settles on around 6kw with peak sun.

0 ·

Alexandra ♦ b1ayd commented · Jan 31 at 12:36 PM

Just FYI solar panels are rated at 25°C. If your temps are higher, then the panels will produce less power than installed at mpp.

Then you have irradiance. They are rated at 1000W/m² you hardly seen that in practice.

So probably related to the two other factors as well.

0 ·

Answer 4 · 2024-01-31T11:52:23Z

b1ayd answered · Jan 31 at 11:52 AM

Are there any experts or persons that can be recommended for a paid consult to try and resolve this issue?

2 |3000

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

Answer 5 · 2024-02-01T07:12:19Z

thanar answered · Feb 01 at 07:12 AM

Any change the MPPT is limiting production due to overheating? I would add a fan on the heatsink as a test.

1

2 |3000

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

b1ayd commented · Feb 01 at 09:06 AM

I doubt they are overheating but will add a fan to test.

0 ·

question