Continue to Site

Welcome to

Welcome to our site! is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

What's the deal with all these "MPPT" IC's with no current sense?

Not open for further replies.


Member level 2
Mar 8, 2013
Reaction score
Trophy points
Activity points

Can someone please explain to me what is the logic behind all these energy harvesting IC's on the market (i.e. ADP5091 or BQ25570) that supposedly come with "MPPT" (maximum power point tracking) built into them, but it aren't actually true MPPT's because they have no current sense?

Let me explain. The MPPT is normally used to track and extract the maximum power from a power source like a solar/PV module. The basic implementation is so that a controller monitors the voltage on the PV module and there is also an input current sensor (i.e. a current sense resistor). Then by multiplying the current and the voltage you get the power that the PV module is operating at. The next step is to slightly modify the switching duty cycle of the regulator and do the measurements again. If the calculated power is now greater than previously then continue modifying the duty cycle in this direction. If the new power is less then reverse the direction and continue backwards. Eventually this process will reach a point where the maximum power is being extracted from the PV module and will latch there. The algorithm is called "Perturb & Observe".

Now while looking for an energy harvesting IC I noticed that a lot of them that are advertised as MPPT aren't actually true MPPT's in a sense described above. They have no current sensor. Instead they assume that the maximum power point voltage of the PV module is simply some percentage (i.e. 80%) of the open circuit voltage. This percentage is normally configured by a resistor divider network and it is fixed. The IC will periodically stop switching to sample the Voc of the solar panel and then adjust the duty cycle accordingly. In a sense it is not a true MPPT because it cannot know the real power without knowing the current. Also, the resistors have to be properly selected according to the PV panel used. If the panel is changed then the resistors must also be changed and adjusted accordingly. The IC cannot by itself determine the true maximum power point like the P&O can.

So, the question here is: What's the deal with this type of "mppt" being so prevailant on the energy harvesting IC market? I've seen a dozen of chips use this approach, but only a few use the true P&O method (i.e. SPV1040). So why is this type of "mppt" without current sense so common? Are they afraid that the Rsense will dissipate so much power? Is this the new meta? What's the deal here?


The difference between "good enough" and "correct" probably would have added a penny to part and ten, to BOM cost. Location of MPP is estimated, how much is actually lost to being a bit off and by how much?

ADP5091 datasheet does well explain the implemented MPPT algorithm. To find out how's the efficiency compared to true P&O, you'd simulate the operation with PV characteristics at different irradiance levels.

Wiki has quite a good description of the various algorithms, these perform adequately or not depending upon the application.
For large multi-cell PV installations where dynamic shading occurs then constant voltage would be hopeless and is not used.
However for single or a few cells as implied by "energy harvesting" as mentioned by others the difference between "constant voltage" and theoretical perfection is probably not justifiable by extra complexity and in particular power consumption.
At the end of the day the choice is yours.

I see, thank you for the explanation. It would seem that I've been going at this from the wrong perspective. Having thought that P&O algorithm is the way to go I was looking for MPPT IC's with a current sensor and automatically disqualify those that didn't have one. That's why I ended up with very short list of IC's. I'll have to rethink my approach.

Thanks for all the answers!

Not open for further replies.

Part and Inventory Search

Welcome to