LT8490 MPPT Solar Charger

The Linear Technology LT8490 is a synchronous buck-boost battery charging controller for lead acid and Lithium batteries, featuring automatic maximum power point tracking (MPPT) and temperature compensation. The device operates from input voltages above, below or equal to the regulated battery float voltage. The LT8490’s full-featured battery charger offers many selectable constant-current constant-voltage (CC-CV) charging profiles, making it ideal for charging a variety of Lithium or lead acid chemistry types, including sealed lead acid, gel cells and flooded cells. All charge termination algorithms are provided onboard, eliminating the need for software or firmware and thus reducing design cycle time.

The LT8490 operates over a wide 6V to 80V input voltage range and can produce a 1.3V to 80V battery float voltage output using a single inductor with 4-switch synchronous rectification. The device is capable of charging currents as high as 10A depending on the choice of external FETs. The LT8490’s MPPT circuit enables a sweep of the full operating range of a solar panel, finding the true maximum power point, even in the presence of local maxima points caused by partial shading of the panel. Once the true maximum power point is found, the LT8490 will operate at that point while using a dithering technique to quickly track changes in the local maximum power point. With this methodology, the LT8490 fully utilizes the power generated by a solar panel even in non ideal operating environments.

The LT8490 performs automatic temperature compensation of the battery charge voltage by sensing an external thermistor on the battery. The STATUS and FAULT pins can be used to drive LED indicator lamps. Charging current limits can be adjusted by changing as few as 1 or 2 resistors, and a charging time scale can be selected with the appropriate resistor divider.

My LT8490 design follows closely the recommendation from Linear Technology. The schematic was designed in reference to the examples from datasheet and demo board. For the board layout I got help from a nice Linear guy, Tage Bjorklund, who previewed the layout a few times to ensure a proper design for high currents and low noise.

Down below are some design documents available for download. The BOM files including order numbers for all parts from Digikey. To design a charger for your own requirements a spread sheet (LibreOffice) is available to calculate all components that have an influence on the charging parameters.

CAD files for the layout will be not available any time soon. The board is a 4-layer design and requires a manufacturer who can handle the specifications for that, they are not cheap when ordered in a low count. I will probably sale the boards for a reasonable price when manufactured in a large number.

LT8490 Top Layout LT8490 Bottom Layout

LT8490 MPPT PPCB Bottom LT8490 MPPT PCB Top

The photos are showing the prototype design with a size of 30mm*60mm and a height of about 15mm. The series board will modified to provide mechanical provisions to attach a heat sink to the top side. The board will not be modified, wasting empty board space just to provide four mounting holes isn’t a cheap solution on a 4-layer design. Instead I have designed a frame that allows attachment to a 50x50mm heatsink. The frame is available at Shapeways.

Optional Heatsink

LT8490 Heatsink Mount ATS-CPX050050006-199-C2-R0

Heatsink ComponentsBoard in frame

Heat transfer and isolation sheetHeatsink attached to board

IMG_20140919_165417 IMG_20140919_165423

Required parts

  • Heatsink, ATS series, 50x50mm, fin height as required, Digikey part number ATS1590-ND
  • Heat transfer and isolation sheet, Digikey part number 1168-1978-ND, good for a few boards

Downloads

BOM Digikey 12.6V 5A
BOM Digikey 12.6V 5A
BOM_Digikey_12.6V_5A.ods
68.7 KB
596 Downloads
Details
BOM Digikey 14.2V 10A
BOM Digikey 14.2V 10A
BOM_Digikey_14.2V_10A.ods
69.2 KB
756 Downloads
Details
BOM Digikey 3.6V 10A
BOM Digikey 3.6V 10A
BOM_Digikey_3.6V_10A.ods
68.8 KB
425 Downloads
Details
DC2069A-4-SCH
DC2069A-4-SCH
DC2069A-4-SCH.PDF
64.1 KB
899 Downloads
Details
LT8490 Component Calculator
LT8490 Component Calculator
LT8490 Component Calculator.ods
54.3 KB
896 Downloads
Details
LT8490 Components Bottom
LT8490 Components Bottom
LT8490 components bottom.pdf
9.7 KB
674 Downloads
Details
LT8490 Components Top
LT8490 Components Top
LT8490 components top.pdf
26.0 KB
688 Downloads
Details
LT8490 Schematic
LT8490 Schematic
LT8490 schematic.pdf
22.0 KB
1340 Downloads
Details

58 thoughts on “LT8490 MPPT Solar Charger

  1. Hi Michael,
    Sounds very promising! I like the boost in this one, as it allows (high efficient) full 4″, 5″ or even 6″ cell integration in a wing of the sailplane with say only 11-12 cells in series!? Keep up the good work!
    Regards,
    Bob

  2. Depending on PCB and component costs, I might be down for a few units. Any idea when will you be arranging for the bulk order?

    • After a quick calculation for a batch of 30 units I can offer them for 50€/each. That would be a industrial assembled PCB with individual customization for the panel and charging parameters.

      • hello,

        is the 50 € for the whole assembled charge controller? Or is it only for the empty board?

        I wold take one or two controllers. please tell me if it is currently possible to get one.

        best regards, harald.

        • That is an rough estimation, only valid when at least hundred units are made. Expect I higher price, I unsure whether to have them manufactured or not. Don’t want to sit on that many devices if nobody has a need for them.

  3. Hi Michael,

    great job!!!
    I am interested in 24 V (lead batts) version…

    Please keep us posted 🙂

  4. Hey Michael

    Great work on the boards! I like how you kept the switching components so close together. It helps to keep the high current paths as short as possible.

    I have the DC2069 demo board charging some 200 ah lifepo4’s and it works a treat. No heat sinking and it only rises about 20 degrees above ambient at 180w output.

    Out of interest what net did you make the two inner layers? Are they both GND or GND and Vin?

    Cheers
    Anthony

    • Below switching components the inner layers are used only for control and return signals.

      Below the control circuit both inner layers are ground with some signal routing.

      • Great. Thanks!

        I’d love to know how the testing goes with the higher current version.

        I’m probably going to rout a PCB in a similar way to yours. The recommended layout in the LT8705 datasheet is OK but the high current paths in yours are a lot shorter.

        Good luck with the testing!

        • The company who asked for the 30A design is still testing. No negative feedback so far. Last comment was that it is working flawlessly under all conditions. I can ask for some numbers if desired, let me know. The high current paths were designed after rules given by an LT application engineer. He was a great help to get this right.

          • Hey Michael

            No need to ask for numbers. If it is a 30 amp version and it’s working under all conditions then that says it all. It’s quite amazing that you can push such a huge amount of power out of a tiny little circuit like that. Electronics have certainly come a long way in the last 10 years!

            I have used a LT application engineer that is based in Australia to help out before too. LT have very good product support.

            Do you do freelance work yourself or do you just do it as a hobby for enjoyment?

          • I mainly do the work for personal enjoyment, or hobby, but usually earn something for this.

  5. Hi Michael,

    I have a 12V (liquid acid) battery with a 12V, 1A solar panel. The open circuit voltage is about 15V, so it is charging the battery, but only in bright sunlight, and overall not very efficiently.

    Is this a possible configuration for this board? If so I would be very interested.

  6. I received one of the high current version boards and am evaluating the board. At this point I have not yet verified if the board can handle the maximum power without some adjustments, so I want to caution those of you who are interested in this high power version that there may be that the power components may need adjustments. And it will probably be necessary to adjust the layout so that component temperature is lower. Especially the input capacitor may have to be moved, and the partnumber changed to one with higher current rating.
    When the evaluation is done I will send a report to Michael with the results.

    • Hi Tage

      Id like to also know the results of the test if possible. Depending on the output power levels quite a bit of heat will be created and needs to be removed. Using a heat spreader like a copper plate on the SW1 and SW2 planes attached to a heatsink will help to withdraw the heat from the mosfets and inductor leads. I have a very similar layout and this is what I intend to do.

      Thanks

    • I am the one who built out and sent the 30A board that was designed by Michael to Tage for testing.

      I can verify that there are significant improvements that need to be made to the board to get it operate as smoothly as the 16A demo board that is offered by Linear Tech.

      I’m redesigning the board from scratch and while doing this I have found numerous issues with the current 30A design shown above and I believe that is the reason I was seeing poor MPPT tracking when compared to the demo board. At first it looked like it was working just fine but after testing it for a few days it was clear that there are design issues.

      For best results you need to design the board around your specific needs as far as battery voltage, and solar input voltage, and max current ratings considering all these affect how your need to design the board and how all the resistor settings need to change.

      The ceramic CAPs need to be XR7 rated. The large caps have to moved way from the mosfets if you don’t wand the board to fail in a short period of time if you plan on running any decent current through the unit.

      I have not built out a working 15A board although I have had the PCB’s produced so I can’t comment on how well it performs but I do know the IC layout is basically the same and I would expect to see the same poor MPPT tracking issues I was seeing on the 30A board considering I changed the sense resistors to the same as would be used on a 15 A design and it did not improve the MPPT tracking smoothness compared to the 16A demo board Linear provides which has superior design and performance. Tage has not tested the 15A board design either. I don’t think Michael has done any long term testing of the 15A board either.

      As I said at first the board looks to be operating fine but when you compare it to Linear’s 16A demo board you will see there are difference in how well they operate at high and low power levels.

      If you just want to try the design out then just buy the Linear 16A demo board. If you want to build the board out in quantity specifically for your needs then I would strongly advice that you start the design from scratch using the demo board as your reference point.

      Just my 2 Cents after playing with both designs for many months.

      • Hey Ryan

        Thanks for the feedback. I agree with all of your points in the previous post.

        Layout is extremely important when it comes to an application like this. LT went for a 6 layer design in their board which definitely helps with shielding and interference protection.

        I think that pushing a maximum of 20 amps may be the limit of a board like this. Heat removal would also play a massive part in the functionality of the board.

  7. HI Michael.

    I recived your design files a couple of month’s ago. (Thank you)
    Now i have 10 Pcs of PCB’s and have populated one board.
    Some components needed to be changed (Mosfet’s and capacitor’s rated for 100V), since it will be running at at around 48V.
    As some of the comments above state, i think a new design is needed,
    Your design looks to be designed to fit in a specific application / enclosure.
    I had some trouble sourcing 0805 capacitors, rated for 80/100V, ended up with 1206 standing up, or 5 pcs of 0805 on top of each other.
    The layout design is VERY “narrow”, especialy around mosfet’s, and the large ceramic capacitors.
    My initial testing was done with a 14.4V Lifepo4 battery, and a 0-30V 0-10A bench PSU, just to check if the board would start up.
    It sort of worked, effiency was in the 80-85%, but when the battery was fully charged, the MPPT board started to draw current from the battery, around 350mA. highest charge current i tried was 2A, and the mosfets got warm to touch, around 45 C.

    So do you think it can be a reliable MPPT for my needs. i DONT need the best mppt tracking, just a reliable Buck / Boost charger, and if possible MPPT tracking.

    My application is:
    40-58V input from solar cells, U/MP is specified to be 25.1V *2 @ 8A. 2 Poly panels used in series.
    44-54.2V battery voltage, depending on SOC. Li-ION battery made from 1560 pcs of Nokia 1100mAh cell’s. Approx. 6kW/h battery pack.

    If this board can run reliable at around 8A and 48V, then i would stick with it, else i will either try the Linear 16A demo board (if it can be modified to 48V), or design a new LT8490 board around TO220 Mosfets, larger 6 Layer PCB, throu hole electrolytic Capacitors / SMD capacitors combo layout, for reduced cost.

    Best Regards
    Rasmus Skjaer

    • Update to my first comment.

      Did some more testing after assy. Worked quite good at 14.4V, and 18-35V input, 10-15V Boost mode had some strange variations in voltage output to the battery.

      Now i have changed all the resistors reqiered for 48/54.6V use.
      Nothing works anymore. Status led flashes like before, will connect a RS232 logger for easy acces to status info later today.

      Will populate a second board later this week (if my wife behaves)

      BR
      Rasmus Skjaer

  8. Hi Michael,

    Is there an option for ordening already?
    Want to charge a 3s 18650 battery pack.

    Very intresting.
    Hope you can help me.

    Greets robert from Holland

  9. Like to have a complete bord too. For loading of 10s4p 16850cells ~600Wh from my solar panels. Hope you contact me 😉
    Thx

  10. HI Michael
    You have a very interesting web page and I appreciate the work that you have done so far on the LT8490 board
    SMPSU design is far from easy to get it right
    How is the project evolving as I can not see any more activity on this page since 8/9/14
    I have some serious interest in the LT8490 device and I would be potentially interested in several 10s of assembled boards per month on a regular basis.
    Perhaps you would be good enough to contact me at my email address.

    Kind regards

    Charlie

  11. There will be no progress and further updates as long as there is no feedback from evaluation.
    Stay tuned, the project isn’t dead.

  12. how much is the size of PCB? and how you stack the pcb?(for example most uses important signal/ground/power/signal you use this or others??)

    • 30mm*60mm
      Other layers are ground, inner signals. High current loops are separated from control block.

  13. Hi Michael!

    You did a really great work!

    I suppose you used CS EAGLE to design your charger. I also want to work with the LT8490 but I’m not sure if I have designed the LT8490 device correctly. Therefore I would be really thankful if you could send me your LT8490 device design for EAGLE.
    It would be also really great if you could send me your design files because you did a really great work on your boards and I think I can learn a lot from your design and your board layout to improve my own Projects.

    Your website is really interesting and I look forward to future stuff here.

    Hope you can help me.
    Kind regards

    Johann

  14. Hi Michael,

    nice work!

    Could you please send me the CAD files of your circuit components you used in your design?

    That would be really great!

    Best Regards

    Franz

  15. Hi Michael,

    just wanted to know what´s going on here.
    Basically i would be very interested in 1 or 2, the LT8490 is very cool, but i have no possiblities to do it on my own or to have pcbs done.
    Could you please provide an update on what´s going on here.
    I would be very happy and i´m sure i´m not the only one here if that project is going on and there would be pcbs for sale!

    That just looks too promising to ignore it!

    All the best for you and your project!

    Greetings from Regensburg!

    • See this comment. I have not yet seen any test results or got further info. At the moment the project is on hold and I’m using it only for personal purpose.

  16. Hello,
    I was just wanting some input. maybe im over thinking

    But I was thinking of using this chip to do the MPPT from each or my 32v 4.7a solar panels and feeding it to a say 28V DC bus. then i can use a dedicated Battery management solution to charge what ever batteries i want. That should cut down on heat production as each panel is doing its own Mppt. Do u see any problem with this? or maybe a better solution. this chip looks to be on the cutting edge and has a lot of potential.

    Thanks for all the information you have documented it has been a great help.
    Anyone interested in my setup so far feel free to contact me.

  17. Hi Michael, any updates on this project? Interested in having it charge a 6S 30ah custom built pack. I’d buy 2 if for sale soon.

  18. Hi Michael,

    Nice job!

    I´m also planning a project with the LT8490. I would however have to connect several LT8490s in paralell. Do you know if the LT8490 copes with parallel operation?

    Best regards
    Tom

  19. Hi again.
    Last year i tested the LT8490 on the 30A design from Michael.
    It worked sort of, and then nothing.

    I have just purchased another set of Lt8490 chips, and will populate another PCB.
    Just a little funny thing:
    http://www.aliexpress.com/item-img/500W-solar-controller-LT8490-single-chip-smart-battery-charge-control/32603725350.html?spm=2114.10010108.1000017.2.ZduD34

    211 USD as same as the Linear demo board, but with a couple more, and lager mosfets.
    If the PCB from Michael’s design, PCB #2 is not working either i will try this China design board.

    Br Rasmus Skjaer
    SP Denmark

  20. Hi, ich suche einen möglichst kleinen MPPT für einen 2S Lipo Akku. Hast du zufällig eine Quelle dafür?

  21. Hello, this would be a perfect MPPT and battery charger for a solar UAV which I happen to be working on. I know everyone keeps asking the status. But what is your results using the board for your purposes. I’m working with a friend using the SPV1020 MPPT chip but am having difficulty getting it to work properly. What you describe would match superbly to my project. Please post additional info if possible or if you have any recommendations for any other available solutions I’d much appreciate it. 12-36v solar in, 22v out, 6s Li-Ion batt, and max 7A load is our target.

    Thank you

  22. Hello!
    I am new to the concept of charging lithium ion batteries with the LT8490 IC. I am doing a project on it and i would like to know how the chip operates. I have saw the datasheet that mentioned perturb and observe and I would like to know how the electronics on the chip actually accomplishes this tracking method. Furthermore, how the electronics actually operate to charge the battery. I have been trying for sometime to understand and I’m not making any headway. Some assistance would be appreciated. Thanks in advance!

  23. iam using a 48v 12AH sealed lead acid battery. ineed a constant voltage of 56v 10AH to constantly full charge my battery. please advise whether lt8490 mppt solar charger suitable for my requirements, the power of my supply is 500watts. can i also see the power in to power out graph

  24. HI Michael.
    Your board looks great. Thank you for sharing your design.
    Not sure how is the project now after two years since this post. I noticed that Ryan and others were trying 8490 at 30A with your board and you seem waiting for feedback…But my gut feeling is that 8490 chip was designed not working at such high current, according to its current sensing resister selection R34 is supposed to be 0.0497/30=0.00166ohm. I doubt 8490 is designed to work with such extreme component. Besides, it will require special type and material of high precision resister and appropriate layout instead of design in your board….Well, I might be wrong since I have zero experience with 8490 comparing to you guys.
    One question out of curiosity, do you create 8490 footprint package yourself or get it from LT technology web? the footprint is available on LT’s web, but it seem a issue to get it due to Ultra librarian reader not supporting 8490’s odd pad. Do you know how to deal with this issue? Your help (or anyone else who may know) will be appreciated greatly.
    Thank you

    • Footprint created myself. I’m using Target3001 CAD where I can design footprints of any size and shape.

Leave a Reply

Your email address will not be published. Required fields are marked *

Comments will be moderated! Spam deleted immediately!
Before you submit form:
Human test by Not Captcha