Ryobi Doesn't Want You To Know

Power Machine Tools
2 followers
I go on a journey to understand how Ryobi 18V batteries work and figure out what is causing good batteries to fail. It is a firmware lockout that is being incorrectly triggered. I also figure out how to reverse the lockout to revive the "dead" batteries.
Chapters
0:00 Intro
0:23 The Backstory
0:53 Initial Testing
2:38 Gathering Data
3:22 Verifying the Hypothesis
4:19 False Lockout Triggering
4:58 Concluding Thoughts
#machine #tools #Powertools
A software bug is bricking your IOBE power tool batteries, and in this video, I'm going
to show you what I learned and how I fixed all of these batteries.
It was a long journey, with a lot of reverse engineering and testing.
So let's get into it.
This whole saga started when three of my brand new batteries stopped working.
They would not charge or give an output.
They would just flash the four LED indicators to indicate an error.
I was able to get them replaced under warranty, but I was told to recycle the broken ones.
That did not sit right with me, and I was really curious what could cause brand new
batteries to just suddenly stop working like that.
I started by taking apart the bad batteries, and doing some basic tests to make sure the
cells were healthy and balanced.
They seemed fine, so I tried resetting the batteries by shorting the J1 jumper as some
people had suggested online.
When that also didn't work, I suspected some other hardware issue, but with no documentation
on the BMS circuit board, my only choice was to reverse engineer the circuit.
And after many many hours of work, I was able to construct a schematic that helped me understand
the design and what each component was doing.
I used this schematic to test the battery in more detail, comparing against a known
good unit.
But even that did not reveal anything obvious.
All the sensing inputs to the microcontroller seemed perfectly fine, but it was still not
enabling the output MOSFETs.
The analog front end chip was a mystery chip with no documentation, so I could not verify
if it was working correctly.
So I thought it could either be a bad microcontroller or the analog chip.
At this point, I was also able to put together a J-Link adapter and dump the memory from
the microcontroller, mainly just out of curiosity.
I swapped the analog sensing chip, but no dice.
But when I swapped the microcontroller, the battery started working again, and I thought
I cracked it.
I went on DigiKey to buy some new microcontrollers, and when they arrived, I soldered them on
and programmed them, and to my surprise, they didn't work.
They still gave the 4 flashing error.
I was completely stumped, and it made no sense.
I wanted to get more units to test, so I went on eBay and found many listings of seemingly
new batteries being sold for parts.
I picked up many of these and started testing them.
That is when patterns started to emerge, and I was able to bucket failure modes on these
batteries.
The biggest bucket resembled how my original batteries had failed.
I was knee deep in running all sorts of tests and experiments, until one day, the thought
came to me, what if the bad batteries had something in the software that could cause
them to lock out?
To test this theory, I got the memory dump from my known good battery, and programmed
it on a broken one, and amazingly, it started working again.
I could not believe it.
I had finally figured it out.
Battery after battery, I reprogrammed them all, and they all came back to life.
I was amazed, but also wanted to make sure that my hypothesis was correct.
So I tried to intentionally lock out a good battery.
I tried a lot of things, and it turns out the only way I could find to lock them up
was if I imbalanced the cells more than like 0.15V and put them on the charger.
I compared the memory dump, and was so happy to see my theory being proven correct.
Some bytes at the end of the memory had changed before and after the test.
After some trial and error, I found the exact byte causing these to lock out.
I was super excited to finally, after months of testing and hundreds of dollars spent on
dead batteries, figure it out.
So it seems like Ryobi has implemented a permanent lockout in these batteries, and if it gets
triggered, the battery will never work again.
It's arguable if that is really necessary, but the bigger question is why were these
healthy packs getting locked?
Most of the batteries I got were fully balanced, so that wasn't the reason for them to lock
out.
But even after lots of trials, I could not get these to lock out in any way that would
explain it.
My best guess is that something happens during long term storage where the software thinks
the battery is unsafe, and locks it.
Anyways, even though there are still some mysteries with this issue, the good thing
is that I was able to revive all the locked out batteries and get them back in the shop.
If you have batteries that have failed in a similar way, maybe you can fix them as well.
I will have a detailed guide on this and all the other failure modes I encountered in a
blog post that I'll link down below so you can check that out.
I hope Ryobi takes notice of this issue and implements a fix so that we don't have to
see good batteries going to e-waste.
Thank you for letting me share my journey with you.
I am so done working on these batteries and hate being locked into one brand of tools.
But there is not much I can do about that.
Not yet at least.
