Tools for unlocking Makita batteries

[tiendung166-sys]

Could you please help me write a program for an ESP32-S3 to unlock Makita batteries—specifically both the versions with and without the built-in LED charge indicator—using the ILI9341V display driver and the FT6336G touch controller? Thank you very much.

[zornfeld]

Enter the code in Claude or in ChatGPT and let it rewrite it yourself. It also worked excellently for me with a 32-Wroom. The pin layout will only change anyway.

[tiendung166-sys]

On the ESP32 board, make the following connections: GPIO 02 => Data pin GPIO 03 => pin_en Connect two 4.7k resistors from the 3.3V rail to GPIO 02 and GPIO 03, respectively. Board GND => Battery B- For older battery models, simply reading the data is sufficient.

[TheRepairforge]

One question: when you say it can't read the older batteries, which program and version did you try, and which pack (model, and the screen or serial output if you have it)? PocketOBI does read that generation, but those packs often answer only every second or third try, and older PocketOBI versions didn't retry. So I'd like to know if it's that or something else.

[TheRepairforge]

Here's a first test build for the ES3C28P: PocketOBI-v2.4.0-dev-es3c28p-test1.zip (attached). It's a dev build and I don't have the board, so expect a few rounds before everything is right.
PocketOBI-v2.4.0-dev-es3c28p-test1.zip

Battery wiring as you have it: DATA on GPIO2, ENABLE on GPIO3, 4.7k pull-ups to 3.3V, GND to B-. Never connect B+.

Flashing :
-in the browser, nothing to install :
Unzip, you get one .bin file.
Open https://espressif.github.io/esptool-js/ in Chrome or Edge, plug the board in over USB-C.
Click Connect and pick the board's port. If no port shows up: hold BOOT, press RESET, release BOOT, then try again.
Set the Flash Address to 0x0, choose the .bin, click Program.
When it's done, press RESET.
-Or with esptool: esptool --chip esp32s3 write_flash 0x0 PocketOBI-v2.4.0-dev-es3c28p-test1.bin

What I'd like to know:
-Does the screen come up, the right way round, with normal colours (not mirrored, red and blue not swapped)?
-Do taps land where you touch? If not, go to Tools > Settings > Calibrate touch, tap the three targets, and tell me if it's better after that.
-Plug in a pack and read it: what does it show? Try one of your older packs without the LED gauge too, and press read two or three times if the first one fails.

If anything goes wrong, a photo of the screen helps, and so does the serial log: open a serial monitor at 115200 (Arduino IDE works) and paste what it prints.

[tiendung166-sys]

After flashing the firmware, the screen turned on and the program is running fine. However, the touchscreen is not working as expected—it’s erratic and often selects the wrong items. I’ve recorded a video of this; please let me know if I’ve done something incorrectly or if you need any further information. Also, could you please add a feature to edit and overwrite the charge cycle count? Additionally, since I am Vietnamese, would it be possible to add Vietnamese language support? It would make it much easier for me to read and understand everything. My English isn’t very good—I’ve been using Google Translate—so please forgive any errors or misunderstandings. My apologies, and thank you very much!

[tiendung166-sys]

This is an older battery model without an LED indicator when plugged in, as shown in the image. The battery is still sealed and is not a repaired or replaced Chinese BMS battery.

[tiendung166-sys]

[tiendung166-sys]

IMG_1562.mov

[tiendung166-sys]

IMG_1557.mov

[tiendung166-sys]

IMG_1559.mov

[tiendung166-sys]

I'd like you to write it like this for easier and more convenient operation.

[tiendung166-sys]

[TheRepairforge]

The capacity is the value the pack's BMS stores for itself, PocketOBI just shows it. If it says 5.2, that is what this pack reports. I can't read the remaining charge in mAh reliably yet, so that is not planned for now.

[tiendung166-sys]

As I understand it, the manufacturer categorizes them by model, so BL1830B is an 18V battery with a capacity of 3A. If read correctly, it's 3A. For 18V systems, I've only seen batteries with capacities of 2A, 3A, 4A, 5A, and 6A. I've never seen any odd numbers before. I also tried disassembling a BL1830B battery, meaning it has a capacity of 3Ah, and I found it uses 10 Samsung U15R cells. I've never seen a 5.2Ah battery before. I compared three different programs, and only yours shows 5.2Ah; the other two read 5Ah.

[anhtaduongvtt]

@tiendung166-sys mình cũng là fans của anh ấy , pocket obi, mình đã tải về cài đặt đồng thời trải nghiệm với nhiều mẫu makita. So với kênh sáng tạo ( kst ) thì phiên bản này chuyên sâu và giám sát được rất nhiều cái, cũng là dự án mở để mình chỉnh sửa theo ý muốn, mình đã mất khá nhiều thời gian để chuyển qua màn cảm ứng điện dung cho đến khi thấy bản thảo này. Từ bản này mình có thể chỉnh sửa qua đọc mạch rời, có điều đối với f0513 vẫn là 1 mơ ước xa vời. Bạn liên hệ gmail mình để trao đổi thêm vài cái nha ( anhtaduongvtt@gmail.com

[tiendung166-sys]

F0513 THẤY AE ĐỌC ĐƯỢC NHIỀU RỒI .MÀ AE KHÔNG CHIA SẺ .NÊN CŨNG HƠI KHÓ NHA

[Alienbliep]

I have tried the latest test 4 firmware.
It works right after programming.

But after an restart it won't work. The program freezes when i click on the battery button.

The enable pin has no 3.3volts after boot except when i press the battery button.
Also data pin is 3.2volts.

This seems normal.

I used two 4.7kohm resistor like the instructions.

The only time i get it to work after flashing is to opem an serial monitor and then it reads out on the lcd and in the serial monitor.

Could this be because it is powered by an pc?

An powerbank seem to work fine. :/ XD

My question in claude:

That explains it. My initial suspicions regarding the hardware were incorrect. This isn’t a startup issue, but rather a case of serial output causing a blockage.

What is happening:
You are flashing a debug build (COMM_DEBUG=1). With every read operation, this build writes dozens of lines to the USB-CDC port—as seen in your log: cmd 0xCC len=2..., rx: ..., ==== DECODED ====. As long as a serial monitor is open, this data is read, and everything continues to function. Immediately after flashing, a monitor or flash tool is typically connected to the port, which is why it worked then. Once the power is cycled, however, nothing is reading that data anymore. The buffer fills up, and the firmware ends up waiting for a PC that never responds. Consequently, the screen boots up normally (since nothing is printed during that phase), but as soon as you press “read battery,” the system freezes. This happens even without a battery connected, because the print statements are triggered by the transaction itself.

Although there is a safeguard in the code (Serial.setTxTimeoutMs(0) in setup()), it evidently isn’t sufficient for your specific build.

Solution:
Build and flash a release version: set COMM_DEBUG to 0 and TOUCH_TEST_MENU to 0. These are set to 0 by default in the provided zip file, so check where they are being set to 1 in your configuration. This could be due to a -D COMM_DEBUG=1 flag in your build_flags, a #define in your own copy of the code, or the es3c28p PlatformIO environment, which sets TOUCH_TEST_MENU=1.
After flashing, verify that the battery-reading function works by performing a full power cycle—without having a serial monitor connected.

If you would still like to keep the debug logs, I can modify the print statements so that they only execute when a PC is actually connected (using if (Serial), which, in the case of hwcdc, only evaluates to true when a port is open). That way, it can no longer get stuck when there’s no monitor. Just say the word, and I’ll make that adjustment.

[TheRepairforge]

Thanks for testing. Test4 was a debug build that writes a detailed log to the USB port on every read. On a PC with no serial monitor open, nothing reads that log and the firmware waits for it, so it freezes. A powerbank has no PC on the other end, so nothing waits. Please flash test8 instead: the log is off there, and the PC bridge works again.
https://github.com/user-attachments/files/33551973/PocketOBI-v2.4.0-dev-es3c28p-test8.zip

Also, use 470 ohm pull-ups on DATA and ENABLE rather than 4.7k. That is the value I use on the PocketOBI board.

[TheRepairforge]

Glad it works now!

4.7k is the original value from the OBI project, and it can work, especially with short clean wiring like on a PCB. With flying leads and an adapter it was marginal on my bench. On 3.3 V logic the pack pulls the DATA line close to the threshold where the ESP32 can't tell a 0 from a 1, and reads failed now and then. 470 ohm holds the line up more firmly, and that's the value I validated and use on every board.

It's safe: at 3.3 V a 470 ohm resistor passes at most about 7 mA when the line is pulled low, which the ESP32 and the pack handle fine. On ENABLE the value doesn't really matter, since the ESP32 drives that pin. I use 470 there too so there's only one resistor value to buy.

So if 4.7k reads your packs reliably, you can keep it. If you get random NO RESPONSE or garbage values, switch to 470.

[Alienbliep]

Yes,

This seems solved now.

Thank you very much.

Why 470ohm? I saw some other guy use 4.7komh on this nboard and it seemed safer?

Thank you

[Alienbliep]

Thank you.
It works great.
I also made an case for it but is an remix of someone so i dont know if i am allowed to post the file.

[TheRepairforge]

Nice job !!!
You can post the remix, if original author allowed it ( which is usually the case).
Really well done, and this looks even more cooler with the new 2.5 interface :)
I will release it soon BTW

[tiendung166-sys]

CAN YOU HELP ME READ AND UNLOCK THE F0513 MCU?

[TheRepairforge]

I can't check here, because I don't have a pack like it. If you're up for one more, here's a build that only reads.

PocketOBI-v2.4.0-dev-es3c28p-test9.zip

It has no normal menus. It boots straight into one test screen.

Flash it like before: esptool-js, address 0x0. Your touch calibration is already the default, so there's nothing to redo.
Plug in the old BL1830 and tap the screen. It reads the pack three times in a row, which takes about 15 seconds, then shows the results.
Post a photo of the screen here. The pack answers on and off, so if you can, tap again and post a second photo.
It never writes to the pack. There's no unlock and no reset in this build, and you can flash the normal PocketOBI back afterwards.

One thing to check first. Last time every byte came back as 00, which usually means the DATA line is held low. The screen now shows "DATA idle" near the top. If it says LOW, check that the resistor on DATA (GPIO2) goes to 3.3V and not to GND. With the pack unplugged, DATA to GND should measure close to 3.3V. If it stays LOW, the rest of the screen won't tell us much, so that's the first thing to fix.

If your BL1850B still reads fine, a photo with that pack too would give me something to compare against.

If you happen to have a serial monitor open at 115200, the raw bytes are printed there as well. Paste them if it's easy, but the photo is enough.

[Alienbliep]

I am stupid, i thought the read-out test was for the BL1415n batteries. Haha

Anyway here is my read-out:

Here is the read-out

read-out serial.txt

[Alienbliep]

I don't know.
This is the license descriptionwith the original file:

This user content is licensed under a Standard Digital File License.

You shall not share, sub-license, sell, rent, host, transfer, or distribute in any way the digital or 3D printed versions of this object, nor any other derivative work of this object in its digital or physical format (including - but not limited to - remixes of this object, and hosting on other digital platforms). The objects may not be used without permission in any way whatsoever in which you charge money, or collect fees.

[TheRepairforge]

No sharing possible...
That is a shame, but the author can choose this !
I will eventually come up with a version of the case for CYD.