Victron MPPTs | Major Bug - Does not track Maximum Power Point running under load

[FurySolar]

I want to share my full experience with Victron MPPTs, because at this point, I feel completely let down. I’ve tested multiple units, logged data, compared against meters, even tried cooling experiments — and the results have been… disappointing.

---

The beginning — my first system​

I started Solar Simple, like every beginner.

• A 3kVA off-grid inverter (With Grid/Generator Bypass)
• A 60A 24V MPPT
• A 25.6V 150AH LFP Battery with 16mm2 cable and 0.5M from MPPT
• Two Bi-Facial 545Wp panels in series (42V Vmp, ~13A Imp), 35M 4mm2 PV Cables
  

I didn’t know much beyond the basics (VOC matching, panel sizing, etc.), but it worked.

---

Enter Victron SmartSolar 100|50 and SmartShunt 500A​

Wanting something more reliable and efficient, I invested in a Victron SmartSolar 100|50. This was supposed to be an upgrade - better tracking, higher efficiency (~98%+ advertised), and data logging.

All I knew at this point was that this is the best MPPT compared to others; looking at the real-time tracking was so satisfying. Whenever I was free, I usually kept that window open via Moto Smart Connect and streaming it to my PC via the mobile device.

After a couple of months, I bought a SmartShunt 500A, since my battery did not have a smart BMS, and I wanted to track the power discharge from the battery in and out.

After a couple of months, my dad picked up a used Rpi 3, and at this point of time, I already watched so many videos on victron, that I had plans. And this one was a surprise, so I decided to add it to the setup, I ordered the cables required to connect the units via the USB, and they both worked without any issues.

From here on, I did not need a mobile software to stream, I could just use the Victron Connect directly on PC, and see the stats. I even created the Node Red Dashboard, just for fun, not knowin, that it will be so handy in the future XD

Couple of months went on, and I noticed something off!. That one day, my batteries were full and I tried running loads, and I noticed in the MPPT window that solar power fluctuated like crazy instead of holding steady. I dug deeper, I checked my roof cam to see if anything was a cat on the solar panels or any bird movin around constantly, there was none. I checked the weather outside, and it was a pure sunny day, zero clouds floatin around.

It became obvious the MPPT was not tracking properly. Power was being wasted right at the MPPT. I made a couple of videos regarding this, and I reported this to my dealer.

Their response? “Normal behaviour.”

Detailed Report with video links: Victron SmartSolar 100 | 50

---

Cross-check — 100|15​

Not wanting to jump to conclusions, I borrowed a 100|15 from a friend with a small off-grid setup. Connected it to the same panels, same battery, same everything.

Surprise!!! The 100|15 tracked perfectly. Smooth power, proper MPPT operation. None of the “fluctuations” I was seeing with the 100|50.

So clearly it wasn’t my setup. I reported this difference to the dealer again. Completely ignored. I ended up selling the 100|50 in the second-hand market out of frustration.

Video link: Victron Smartsolar 100 | 15

---

​

Second chance — 150|45​

Still wanting to believe Victron could deliver, I gave them a second chance with a 150|45 (with future-proofing and expansion in mind).
And… disappointment again.

• Poor MPP tracking → panel power wasted.
• ~3% sensor drift between MPPT readings and actual power going in.
• Real-world efficiency stuck at ~95%, never hitting the advertised numbers.
  

Even under active cooling (unit temp 27–28 °C, ambient ~24 °C), the efficiency stayed flat. Remove cooling? The MPPT temp rises quickly (32 °C even at just 200W harvest).

I even tried firmware updates: v1.61 v1.64 v1.68
All of them were the same, losing the MPP tracking when running under load. So it’s not firmware. It’s the hardware, or could be something else.

Detailed report with video links: Victron SmartSolar 150 | 45

---

Real-world numbers​

Just to show what I mean:

• At 38V Vmp (single panel), efficiency was ~95%.
• At 1kW+ solar harvest, efficiency was still ~95%.
• Anywhere in between, efficiency was still ~95%.
• Compare that to the 100|15, which hit 98%+ even at 50W low power, or at its overheating temps of 50C, it was still converting at 97% - 98%.
  

This isn’t just “a little loss.” Over time, that 3%–5% difference adds up massively, especially in off-grid setups where every watt matters.

---

The cheap MPPT story​

Here’s the kicker: I also tested the first cheap MPPT (rated ~94% efficiency). And guess what? It ran like a champ. Stable tracking under any condition, always around its claimed 94%.

It didn’t overpromise. It didn’t underdeliver. It just worked.

Meanwhile, Victron, which I paid a premium for, promised 98%+ efficiency, perfect tracking, and delivered less than a budget unit.

---

Where I stand now​

I was actually planning to go all-in with Victron gear in a couple of years. The Multiplus inverter was on my roadmap. I wanted a fully Victron-based system.

But after this:

• Poor MPP tracking across multiple units (100|50 and now 150|45).
• Sensor drift leading to ~95% real efficiency.
• Extra heating because of wasted energy.
• Dealer/support brushing me off for months…
  

… Man! I’m done.

I’ve put the 150|45 up for sale in the second-hand market already, and it hasn't even been a month. I can’t recommend Victron anymore. I’d rather parallel another set of 4mm² PV wires and save more power that is lost in the cables.

---

My ask to the community (MPPT Experts preferred)​

Has anyone else seen this behaviour of bad tracking, sensor drift, and unexplained heat in their Victron MPPTs? Or am I just the unlucky one who got multiple bad units?

At this point, I’m sharing this so others don’t waste months (and a fortune) chasing performance that never comes.

 

[AH Trimble]

I've run 4 Victron MPPTs; 100/20, two 15/70, 250/70 and I've not experienced your issues. The only time got/get heat on the 150/70's is when they were high output/input (clear/cold/bright winter days), but then the heat wasn't that bad. Remember, they reduce output when heat approaches their rating.
My situation changed noticeably when i went with BMS control with the new Midnite Powerflo5's. There is a fair amount of charge "dancing" at 97% SOC and above. But, it gets done. More a result of the BMS control than the MPPTs.
Each of my 3 arrays is oriented a little differently and the MPPT charging shows it, but it appears that all are doing their job really well.

Do you have them "networked"?
Have you confirmed/verified correct settings?

FWIW...I did have one 150/70 lose its magic smoke earlier this month after recommissioning a refurbed/upgraded array. Victron is standing behind the warranty without any problems. I can't confirm if it was me or the MPPT at fault.

Have you tried contacting Victron directly for a warranty claim? I did, and entered the SN and got the distributor. Contacted them and they jumped right on it.

 

[BaylentheSky]

Thanks for the effort in your detailed write up. I haven't looked at any MPPT data's in all of my Victron's (3) because I assumed they all just work like they should. This is an interesting revelation, and even though I don't have the same units that you experienced issues with, I am now a little concerned. I'll take a look at them this weekend.

 

[sumpower]

Lot of missing information here! What is the measured irradiance per square meter? Do you live in the Bengaluru area in India, the
highly industrilized area with air polution?
Oh, didn't watch the video's. Figured someone hunting for more views.
Is thier a comparison of other equipment under the same conditions?

 

[740GLE]

Let see a 15a MPPT with over 1000w of PV behind it only able to generate ~400w

Vs the same 1000w trying to get each watt out of the array as possible?

I’d say connect 2000w to the 100/50 and then it’s an apples to apples comparison, over panel vs over panel.

Also I imagine a Ve smart network or a Venus linked MPPT to the smart shunt for all three MPPT tested?

For reference I also have a 100/50 on my 24v bank, I know they run warm when pushed but I haven’t seen any loading/generation issues with my array. Granted I peak at around 600w and have 824ah of storage.

 

[J.H.]

I have several of these units running..each with its own array…with more in the planing for fall..150/100 and 100/50’s.

Long story short, I always leave 25% headroom on the input voltage and amp output.

No maxing out anything. Very few thing like being maxed out in normal operations.

I use fan cooling on the heat sinks when they are on in summer and it drops the temp about 25F…
they are a lot more efficient when running cool.. almost all electronics are.

All mine seem to do exactly as advertised…and expected… for se real years now.

Perhaps you just got a wonky one…

I love their stuff…never had the first hiccup with one….and I love my 40 dollar grid power bill I have been getting since April …it used to be 150 to 200……

Sorry you had a problem with one … but I say Long live Big Blue……

J.

 

[outboard1196]

Almost sounds like a dodgy connection on PV wiring?
Smaller SCC didn't pull as much current to show the problem.

 

[Bluedog225]

Lot of missing information here! …

Oh, didn't watch the video's. Figured someone hunting for more views

Dude…

 

[740GLE]

Plotting battery voltage and wattage would be better to track.

The MPPT will adjust when if battery voltage dips and your floating.

One of the clips shows 200w around noon or 1pm, so is that battery voltage floating 200w going to load or cloudy?

Every cloudy day is different so no comparisons possible.

 

[scrubjaysnest]

I haven't seen this. Have 3 100/20's, 2 100/30's and 2 100/50's.
All are over paneled. Battery bank is 1060 Ah. 2 shunts 500 amp for the 400 Ah portion and 1000 amp for total.

 

[WorldwideDave]

just curious...2% difference over time could make a difference. But adding a used 200W solar panel for $25 would give you even more power and make up that difference. Not that you are trying to solve that problem, but used panels are very cheap.

I found that my non-victron MPPT did a good job outputting KwH each and every day, but it ran very hot compared to my victron, offered no connectivity options to my cerbo GX reporting and monitoring infrastructure, wasn't smart in any way - could not limit output current without a complex menu system, also could not change from 12V to 48V easily without restarts and complex menus, and stuck out from the wall about 3" further than my victron and was 5 inches taller than my victron that gave me the same results or better. I paid $200 used for my Outback FM80. I paid $250 for my used Victron 150/100.

While you are not getting the results you want, It sounds like you need to provide additional datapoints to arrive at your conclusion more accurately.

Or maybe you bought a fake victron product - heard they exist.

 

[Brucey]

• Poor MPP tracking → panel power wasted.
• ~3% sensor drift between MPPT readings and actual power going in.
• Real-world efficiency stuck at ~95%, never hitting the advertised numbers.

Try graphing PV Power and Battery watts for each scc using VRM advanced, should be able to see the efficiency at a glance over the day. Below can see where efficiency of 450/100 is much less than the 250/100.

 

[FurySolar]

Lot of missing information here! What is the measured irradiance per square meter? Do you live in the Bengaluru area in India, the
highly industrilized area with air polution?
Oh, didn't watch the video's. Figured someone hunting for more views.
Is thier a comparison of other equipment under the same conditions?

I live in the residential side of Bengaluru, there is not much pollution where I live in. So we can take out the smoky haze from the equation.

I got this old video, which you can refer to. One is the cheap MPPT, which I initially purchased, and the other is the SmartSolar 100 | 50. Both tested in similar conditions.

This is the data according to Global Solar Atlas for my area:
Direct normal irradiation 1420.1 kWh/m2
Global horizontal irradiation 1930.1 kWh/m2
Diffuse horizontal irradiation 903.1 kWh/m2

 

[FurySolar]

Let see a 15a MPPT with over 1000w of PV behind it only able to generate ~400w

Vs the same 1000w trying to get each watt out of the array as possible?

I’d say connect 2000w to the 100/50 and then it’s an apples to apples comparison, over panel vs over panel.

XD I only tested 15A unit, to check if it goes bad tracking after the 200W of incoming power, that's when these two units start to do abnormal MPP tracking. And for my surprise, the 15A unit ran perfectly; there were no abnormal higher PV tracking.

Also I imagine a Ve smart network or a Venus linked MPPT to the smart shunt for all three MPPT tested?

Yes, I have tested all three MPPT units to VE. Network via Bluetooth to the SmartShunt, and also tried turning on DVCC and allowing the SVS and SCS.

But none of them made any difference in the bigger units, and the smaller one (15A), ran perfectly with/without these things turned on

For reference I also have a 100/50 on my 24v bank, I know they run warm when pushed but I haven’t seen any loading/generation issues with my array. Granted I peak at around 600w and have 824ah of storage.

I do not know if your higher capacity battery bank can make the MPP tracking stable, unless there is around 2 kW of load during the peak power from your solar array.

 

[FurySolar]

Almost sounds like a dodgy connection on PV wiring?
Smaller SCC didn't pull as much current to show the problem.

I have checked the PV wire for any issues by shortcircuting, and it was showing the max amps for that time of the day, and since it is Bi-Facial panel, I have seen the amps shoot upto 17A - 18A in some occassions.

The smaller unit did not pull much current, since its max charging amps is limited to 15, and it will clip the power, however, you can check these two videos for comparison on a similar sunny day with almost no clouds to cause any fluctuations in the incoming power from the solar panels.

The Cheap MPPT and SmartSolar 100 | 50

 

[740GLE]

XD I only tested 15A unit, to check if it goes bad tracking after the 200W of incoming power, that's when these two units start to do abnormal MPP tracking. And for my surprise, the 15A unit ran perfectly; there were no abnormal higher PV tracking.

So you ideally want to see flat steady PV voltage which equates in your mind to peak MPPT efficiency?

It’d still be interesting to track battery voltage, you are saying when load is applied the MPPT behaves differently? Wouldn’t that mean your battery is close to floating voltage?

 

[FurySolar]

Try graphing PV Power and Battery watts for each scc using VRM advanced, should be able to see the efficiency at a glance over the day. Below can see where efficiency of 450/100 is much less than the 250/100.

I have added the graph as you requested. But here, to find the true efficiency. You will have to remove all the load that is running. In my case, when I turn on the relay on my Inverter, it stops inverting, and sends the incoming Grid / Generator power to the loads.

In this state, it consumes 0.15A of constant load on the DC side, which we can consider as 5W of load, but you can see that the Venus system shows that I have a 32 W of load connected. While the MPPT's sensors say the PV Power is 1085W, and converted to Battery Power is 1063W, which is 98% efficient.

But, look at the battery side, it says 1030W, we add this static load of 5W, so in total 1035W of actual output, which is 95.4% efficient.

Now, less efficiency = More heat, and it could also cause the MPPT algorithm to malfunction.

So you ideally want to see flat steady PV voltage which equates in your mind to peak MPPT efficiency?

It’d still be interesting to track battery voltage, you are saying when load is applied the MPPT behaves differently? Wouldn’t that mean your battery is close to floating voltage?

So, I am mainly concerned about the PV In > Battery Out efficency, which seems to be 95% on an average, while the MPPT's sensors report it being 98%.

Another issue is the MPPT tracking abnormality, if you just scroll forward in the video's I have attached, you can see that when I put load on the battery side, the MPPT's PV tracking starts to drift off from the MPP, and just keeps going up and never reach its MPP point, which means that the MPPT is tracking a lower power point wasting all the potential energy the solar panel can deliver, even after it does a sweep and finds the MPP, it looses it again.

I understand that the MPPTs throttle down when they hit their Absorption Voltage or to maintain the Float Voltage. But this behaviour is before the MPPT reaches this point, and when there is a load running on the battery side, in my case its an Off-Grid inverter.

 

[thanar]

I have also seen this on my 24V system. But ONLY under the following circumstances:
- Battery was full.
- 3 MPPTs capable of putting 2kW into the battery.
- Was trying to charge my EV using a cheap 24V inverter at 1.2kW.
- Inverter was connected CLOSE to the MPPTs (<0.5m) and AWAY from the battery bank (>3m).

The MPPTs would NOT keep up with the draw, although full sun was available. Battery draw was over 1kW, while the MPPTs were working very intermittently. I suspect some high-frequency sync issues between the internal of the MPPTs, the inverter, as well as the EV's charger. I guess the fact that the inverter was connected close to the MPPTs and away from the battery bank was magnifying the issue.

 

[Brucey]

I have added the graph as you requested. But here, to find the true efficiency. You will have to remove all the load that is running. In my case, when I turn on the relay on my Inverter, it stops inverting, and sends the incoming Grid / Generator power to the loads.

In this state, it consumes 0.15A of constant load on the DC side, which we can consider as 5W of load, but you can see that the Venus system shows that I have a 32 W of load connected. While the MPPT's sensors say the PV Power is 1085W, and converted to Battery Power is 1063W, which is 98% efficient.

But, look at the battery side, it says 1030W, we add this static load of 5W, so in total 1035W of actual output, which is 95.4% efficient.

Now, less efficiency = More heat, and it could also cause the MPPT algorithm to malfunction.

I would argue the scc is meeting its stated efficiency numbers (conversion efficiency). That you are adding other idle loads of presumably the pi etc doesn't change that.

In my system my cerbo-s takes about 5-6W, my 450/100 has an 11W idle crazy stuff.

Curious if you would still see this behavior with more battery capacity.

 

[FurySolar]

I would argue the scc is meeting its stated efficiency numbers (conversion efficiency). That you are adding other idle loads of presumably the pi etc doesn't change that.

In my system my cerbo-s takes about 5-6W, my 450/100 has an 11W idle crazy stuff.

Curious if you would still see this behavior with more battery capacity.

The only load connected to the battery terminals is the Off-Grid Inverter, and I know that it cannot utilize more than 5W on DC side in Bypass mode at any given time.

If the SCC is meeting its efficiency numbers, the DC load should hover around 4 - 8 watts, but not a consistent of 3% less power than what is shown by the MPPT's sensors. So, where is the missing 28W going?

 

[pollenface]

I have 6 victron mppts, all performing well after 4yrs. They replaced my agricultural Epever units. They produce steady power when a load is added, although I don't have a shunt. I only get voltage & temp from the Smart Battery Sense networks.

I have had a Victron Blue 48v PWM & Smart Battery Sense fail

 

[NC_hydro]

Do you have one of the pfangsun knockoff Victron’s by chance?

 

[740GLE]

You don’t have a shunt for your DC load do you? Measuring DC losses and efficiently in the cerbo are wild guesses.

So looking at the MPPT tracking when the inverter is on vs off, my only guess is there’s a small DC ripple when inverter is on that is impacting the MPPT. Why it does t do it for the 100/15? My only guess is that MPPT is keeping that PV voltage higher as it’s over paneled.

I’d love to see similar over paneling of the 100/50 and how it’d react in couldy situations.

I’ll try and grab some screen shots of my 100/50 as it charges today, so we can compare.

 

[outboard1196]

The only load connected to the battery terminals is the Off-Grid Inverter, and I know that it cannot utilize more than 5W on DC side in Bypass mode at any given time.

If the SCC is meeting its efficiency numbers, the DC load should hover around 4 - 8 watts, but not a consistent of 3% less power than what is shown by the MPPT's sensors. So, where is the missing 28W going?

Unless you have 2 shunts and one configured as a DC meter the DC power figure is a best guess figure.
I see my DC power almost double when charging at max 70A, but I know this number is a guess as I only have 1 shunt.

 

[Q-Dog]

What do you propose I replace my Victron gear with?