question

Ventura López Kalász avatar image
Ventura López Kalász asked

SmartSolar 150/100 not producing more than 2600W in a 48V system

Hi everybody,

I'm adressing this excellent community with the hope it'll help me solve a problem I've run into recently with my Victron off-grid solar installation.


First of all, the components:


12x 390W 144 half-cell solar panels, wired in 6 parallel groups of 2 in series panel.jpg

1x SmartSolar 150/100 MC4 MPPT (v. 1.59) mppt.jpg

1x Quattro 48/8000/110-2x100 (v. 4.30) inverter.jpg

1x ColorControl GX (v. 2.89) ccgx.jpg

3x PylonTech US3000C 48V Lithium Batteries battery.jpg


Up until now I'm incredibly happy with the setup, everything runs smoothly, I also have a small petrol backup generator I haven't had to use since 2020 wired to one of the AC inputs of the Quattro.


Now for the problem:

Recently I installed a pump for the well, which connects to a VFD regulating startup and some filters, everything wired to the AC output of the Quattro. The pump draws continuous 5000W of power. This has brought a problem to light which I had never noticed before:

The MPPT never draws more than 2600W from the panels! Or to be more specific, and this is what REALLY bugs me, it never draws more than a value that seems to be around the Charge Current Limit multiplied by the Maximum Battery Voltage; in terms of the solar array an interval centered around 33A at 75V. pumprun.jpg

Since the Pylontech batteries have their integrated BMS I can't really manipulate any values, all I can tell is that the CCL declared by the batteries is 44A, and the DCL is 111A (or 37A times 3 units, exactly the numbers from the manual). DVCC is activated with all sharings turned off, and the MPPT is running in networked mode as a Slave.

Now, I've checked and double checked, but at this point I think I'm going crazy. From what I understand, in 48V systems, shouldn't the SmartSolar 150/100 be able to produce up to 5800W (as per the datasheet)? Ideally, what I thought would happen, is that the MPPT would draw as much power as possible from the panels, and then "fill up" the remainder from battery power. What is actually happening is quite the opposite, the MPPT only draws up to the 2600W mentioned before and all of the remaining power is drawn from the batteries. During the day and with the panels in full sunlight and at a temperature of around 30ºC this seems quite stupid, as the battery capacity quickly becomes a bottleneck for the amount of time I can run the pump. What am I doing/understanding wrong?

The only thing I haven't tried yet is setting a limit for the charge current in the DVCC. Might this be the crucial point? How is point 4 in subject 8.4.1 in this link to be interpreted? And point 5?

https://www.victronenergy.com/media/pg/CCGX/en/dvcc---distributed-voltage-and-current-control.html


Any help is welcome! Thanks a lot in advance for any hints or ideas!

MPPT ControllersMultiPlus Quattro Inverter ChargerPylontechoffgrid
pumprun.jpg (3.4 MiB)
panel.jpg (1014.9 KiB)
mppt.jpg (3.7 MiB)
inverter.jpg (2.9 MiB)
ccgx.jpg (3.4 MiB)
battery.jpg (3.2 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.

3 Answers
Ventura López Kalász avatar image
Ventura López Kalász answered ·

Hi @nickdb! So I've finally been able to conduct some tests and measurements on-site, and I think the answer is a combination of stress, careless readings and unlikely coincidences.

I've been able to conclude that the Pylontech BMS is extremely reactive, adjusting CCL and DCL continuosly and on the fly depending on battery temperature and SOC. Now, the disorienting coincidence here was that with the short summer nights, the battery discharged to 80% at worst. At dawn, around 07:30, the panel production would start steadily rising, but by the time the irradiation levels would be high enough for the panels to produce more than 44 (Battery) Amps, the SOC was always already at a point were the BMS would cap the input at those 44A. Hence the limited production. The system is working exactly as I configured it. Facepalm for me.

The MPPT works perfectly well, and adjusting the CCL through VictronConnect over Bluetooth reacts instantaneously to the human eye.

Only one thing that bugs me remains. The panels producing well under their nominal factory values. I'm still seeing around 45A at 70V at best (or around 3000W), for a system that should nominally produce 390W*12 = 4680W, or (9.49A*6)*75V = 4270W. That means I'm only seeing about 70% of ideal performance. How impactful are DC cable runs? And temperature? It's true that it's well above the factory conditions of 25ºC... In fact, I'm consistently reading around 5V less Voc than advertised on each panel.

Again, thanks a lot for the patience and ideas! Victron's products AND community are what makes it awesome!

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.

nickdb avatar image nickdb ♦♦ commented ·
Hi, glad you're making some progress. Just to confirm, you did increase the mppt charge current to 100A?

Various factors affect panel performance, temperature is a big one, so if you're in a warm climate you will be losing power, how much will vary on your panel, it is a spec they publish for the panel so if you can find that you can work out roughly how much.

Panel age and the orientation of the panels also play apart. Is there any shading at all? How are they mounted? etc

Long cable runs will also cause loss, gauge contributes. Victron's wiring guide and toolkit can help calculate this.


0 Likes 0 ·
Ventura López Kalász avatar image Ventura López Kalász nickdb ♦♦ commented ·

Yes, I set it to 100A, and to a variety of other values just to check whether the modifications would register, which they do, almost instantly. But it was not the deciding factor; it only becomes one when set lower than the CCL set by the Pylontech BMS at that time; higher values do nothing as the MPPT is working as a slave over CAN. Everything just as described in the manuals.

Though early in the morning there is a bit of shade, usually I make my readings after 11:00, as then they're each and every one of them 100% irradiated, no shade at all.

The panels are 2 years old. Everything coming from the panels to the fuses and the general breaker is 6mm2 (should be enough for max 30A) from there to the MPPT it's 16mm2 (max 60A). And the total compounded run is about 50m.

That's why I believe temperature to be the main culprit. Ambient temperature is 34ºC. With the panels beign close to black... I guess they're at least at 45ºC. I'll check the 50ªC curves on the datasheet to see if it does match what I'm getting.

0 Likes 0 ·
nickdb avatar image
nickdb answered ·

Have you tested that all your strings are in fact operational?

Is the MPPT max charge current set high enough, though DVCC should overide this?

Any chance there is an inverter limit set in ESS settings?

Without any load, the MPPT should be supplying way more than 2600W to the pylons for charging, so if it isn't even doing that then you need to check settings and PV strings.

The pylons will be sending a CCL which DVCC should be honouring across the devices.

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.

Ventura López Kalász avatar image
Ventura López Kalász answered ·

Have you tested that all your strings are in fact operational?

Yes, all strings tested for voltage and correct. I even regrouped them and tested again.

Is the MPPT max charge current set high enough, though DVCC should overide this?

Currently it's set at 50A, though as you said, I didn't pay much attention to it because I was assuming the 44A imposed by the batteries would override this through the DVCC.

Any chance there is an inverter limit set in ESS settings?

To be honest, I didn't configure anything for the ESS. I thought this was unnecessary for off-grid systems.

Without any load, the MPPT should be supplying way more than 2600W to the pylons for charging, so if it isn't even doing that then you need to check settings and PV strings.

I don't understand why you say this. It's honouring the max 44A charge current the battery likes, isn't it? I do very much hope it's something with the settings, but I'm trying to pinpoint what exactly.

The pylons will be sending a CCL which DVCC should be honouring across the devices.

Yes, I think that's what it is doing, but somehow too strictly, as it doesn't allow any extra solar power beyond that to go directly to the AC consumers.

6 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.

nickdb avatar image nickdb ♦♦ commented ·
Set the mppt to 100A and see what happens.

A pack of 3x us3000c's should charge way above 44A, unless they are near fully charged or imbalanced - you should be able to saturate the mppt with a battery needing a good charge.

The CCL is only to the battery, it won't limit loads.


0 Likes 0 ·
Ventura López Kalász avatar image Ventura López Kalász nickdb ♦♦ commented ·

Ok, I'll try that!

How can I know if it's imbalanced? What do you mean by "saturate the MPPT with a battery needing a good charge"? There's definitely something weird going on. From what I understand from point 4 here https://www.victronenergy.com/live/battery_compatibility:pylontech_phantom the CCLt really should be 75A, and not 44A...

Thanks a lot for your help so far!

0 Likes 0 ·
nickdb avatar image nickdb ♦♦ Ventura López Kalász commented ·
The battery will lower the CCL if it is near fully charged, so at a higher SOC, or if the cells in the battery aren't evenly charged. Newer pylon firmware will report min/max cell voltages in the BMS section of the GX under details. You can also chart these in the advanced tab of VRM.

A bank of 3 pylons which need to be properly charged, should be able to accept around 100A or so, which is the limit of the MPPT. So if all is working properly you should get close to your 5000W just by charging.

Let us know how setting the current limit to 100A on the MPPT goes.

0 Likes 0 ·
Ventura López Kalász avatar image Ventura López Kalász nickdb ♦♦ commented ·
Ok... So if I understand you correctly, it would help the batteries rebalance if I let them discharge quite deeply? Now it the summer they hardly ever go below 80%. As for the firmware, my modules are from the first trimester of 2020, I don't believe they have the newer firmware. Should I dare and try updating them, if everything else fails?

I'll get back to you with the results of modifying the current limit setting on the MPPT on monday. It's an installation in a remote location and I won't be able to try it before that.

Thanks again for your help and patience.


0 Likes 0 ·
nickdb avatar image nickdb ♦♦ Ventura López Kalász commented ·
Your issue just sounds like an mppt setting. It will hopefully work when you raise the current.

There is no need to update firmware, please don't try without someone's assistance.

You can log a support call with pylontech, tell them you would like to update them so you can get access to the BMS details in the the Victron GX. Maybe they will help you.

If your remote location has internet access, you can make the change with the victron connect app, if not it will have to be onsite.

0 Likes 0 ·
Ventura López Kalász avatar image Ventura López Kalász nickdb ♦♦ commented ·
You're absolutely right! Accessed the MPPT through victron connect, and it is now set at 100A. But today it is cloudy -_- so there's 500W max going in anyway. While at it, I checked the remote console as well while the battery was reporting an 85% SOC, an now it reports a CCL of 111A (same as DCL). Maybe the lower CCL reported was just because of a 90+% SOC, will check again later. But this still wouldn't explain the key issue with the direct loads not going above 2600W...
0 Likes 0 ·