Need High-Power Li-Ion Charging? How About 100 W

Ever want a seriously powerful PCB for charging a Li-Ion pack? Whatever you want it for, [Redherring32] has got it — it’s a board bearing the TPS25750D and BQ25713 chips, that lets you push up to 100 W into your 1S Li-Ion pack through the magic of USB Power Delivery (USB-PD).

Why do you need so much power? Well, when you put together a large amount of Li-Ion cells, this is how you charge it all at once – an average laptop might charge the internal battery at 30 W, and it’s not uncommon for laptop batteries to be dwarfed by hackers’-built packs.

A 4-layer creation peppered with vias, this board’s a hefty one — it’s not often that you see a Li-Ion charger designed to push as much current as possible into a cell, and the chips are smart enough for that. As far as the onboard chips’ capabilities go, the board could handle pack configurations from 1S to 4S, and even act as a USB-PD source — check the IC configuration before you expect to use it for any specific purpose.

Want a simpler charger, even if it’s less powerful? Remember, you can use PPS-capable PD chargers for topping up Li-Ion packs, with barely any extra hardware required.

21 thoughts on “Need High-Power Li-Ion Charging? How About 100 W

  1. Off topic, but still. Remember to work as close to your operational voltage as possible. Yes you can step-up/down anything to anything but voltage conversion ratios exist and you’re not getting any younger, so why complicate things?

    I recently made a flashlight from a single 3.7V 18650 and tried to drive a 20W 12V LED. Everything on the circuit was getting HOT! Then I put another 18650 in series, and increased the voltage to 7.2V. The heating disappeared, in fact I was barely able to feel it.

      1. No, if you’re going to get rid of the dc converter, and it’s a flashlight that needs to have multiple brightness modes, you want at least 4 cells. Say your minimum desired brightness mode is equivalent to feeding the LEDs 9 volts while the max takes 15. If you use only 1 or 2 cells, then even fully charged cells are slightly below the minimum voltage and you only need around a 2 amp boost converter. The efficiency at lower brightnesses will be just fine, and higher ones won’t be that bad either.

        If you have only pwm and/or linear regulators, then with 3 cells your output will be very low almost all the time except when the cells are very full. You’ll also waste energy. But with 4 cells, you’ll be able to maintain full brightness for more of the life of the cells, only now when they’re freshly charged you’re wasting a lot of energy and making a lot of heat. Much worse than a half decent 1s boost converter except for (often) the peak output, since you can get lots of current through a single FET when the cells are full if you want to go that route.

  2. Is it really “100 W into your 1S Li-Ion pack” though?

    It takes a 100 W (20 V 5 A) PD supply, and does not appear to do actual voltage conversion on the board. (sorry, no obvious inductors present, there’s no description of what it actually does, and no pdf of the schematic present, and I’m not going to download the effing repo and fire up kicad to check)

    Based on that, it looks like it will charge the battery or cell at the maximum current of the power supply, 5 amps, or 22-ish watts into a 1s pack, not 100 watts. A 4s pack will see close to 100 watts though.

    1. Ah, OK. L1 isn’t visible in the image at the top of the blog post here.

      Jeez, a viewable schematic and a basic description of operation and/or performance specs in the readme would go a long way.

      1. The datasheets of the two ,TPS and BQ, chips explane everything.

        I personal will go for a 4SxP (<20V while charging) and a BMS + active Balancer ISO a 1SxxP cos 120W / 3.7V = 23.4 Amps !

        I use a “6S Balance Li-ion Lifepo4 LTO Lithium Battery 2A Capacitive Active Balancer Board Equalizer 22AWG For Battery Diy” from AliExpress for balancing 20V (5S) powerpacks, works great.

        1. What efficiency are you hoping to achieve at 120W charging of a 1S battery ?

          95% is a reasonable number for inductor based DCDC, which equates to 6W of power loss…that is substantial. It may work if you have a big heatsink and stick it into your freezer

  3. Someone with electrical design experience needs to start up an open source module project. Something that creates tiny complete board segments that can be added to a design before creation or used as an add-on board of various types and handles common features that can then be fed into the main design.

    For example, a module that could handle USB-PD, and one that could handle different Li-on arrangements could be combined for a total power solution on a board.

    Could have modules that handles other special features as well. Ones that support display over USB-C with simple inputs, or ones that just provide an MCU with all the passives it needs.

    1. Actually working on something like that in my spare time, still have alot to go until it’s something presentable. It’s more than just USB-PD, it’s a modular system that includes MPPT, 120VAC charging, and an integrated BMS.

      1. If they are all able to be separated then that’s along the lines of what I mean.

        Ideally someone could fire up a design for a new board slap a couple locations for soldering on castellated premades and the chip they want to run, then be done.

  4. Neat! I’ve been working with a small startup developing our own 2s-4s battery system the size of a credit card that has USB-PD input up to 100 watts. We are still in prototyping stages but it is open source. Fully configurable voltages up to 4.3v per cell limited by a hardware monitor. Up to 9 amps output. Wide range of fault detection codes and automatic graceful shutdown procedures. Also does aux input for solar applications.

    You will know it when we release it.

    1. Cool. Will it have indicator LEDs / buzzers so the user knows what’s happening to their batteries? I guess this would require an onboard MCU, but would be a nice feature.

      1. Yes! gas gauge via four LEDs, and two simultaneously accessible UARTs to the microcontroller that allow for configuration, telemetry, etc. One of those UARTs is tied to an onboard USB to UART converter that’s connected to the USB-C connector.

        I just started a new job and am only about two paychecks in at the moment so kinda postponing the third prototype. It’s a dense board, and prototyping isn’t terribly cheap though our price goal is ~$80-$100 each.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.