Looks like the laptop can’t even stay up long enough for you to type framework_tool --smartbattery in terminal. In this case, remove the battery and try booting up using a 100W charger. If it can’t boot, the mainboard’s power circuit is probably at fault here. RMA or, if you are out of warranty, troubleshoot yourself of go to a local repair shop or buy another mainboard. If it can boot (it takes much longer if you don’t have an RTC coin cell) the battery has failed. Since the battery has one year extended warranty and the computer itself has at least one year of warranty, your battery is certainly within warranty. Contect support and have them send you an new battery.
1 Like
I unplugged the battery. Plugged the 100W charger and got a blinking red-blue pattern in the LED. It took about 40 seconds until I saw anything on the screen, but then the boot stages completed normally. But now he is running so that confirms your suspicion.
So ordering a new battery, I guess.
Should I also buy a RTC battery?
(The mystery remains, how did I manage to kill a battery of a laptop bought in Autumn '24? Also, upon inspection the battery is slender and not bulging/spicy at all.)For the interested, here’s the output of framework_tool --power -vv with the battery unplugged.
Charger Status
AC is: connected
Charger Voltage: 15400mV
Charger Current: 0mA
Chg Input Current:4500mA
Battery SoC: 0%
Battery Status
AC is: connected
Battery is: not connected
Error: “Fail”
Your battery is within warranty because you purchased the laptop less than 2 years ago.
You didn’t, the 61Wh battery is less reliable than the older 55Wh battery. Framework said that’s because of a higher energy density, some redditors think ATL is not reputable but I highly doubt it. I think the BMS is inaccurately calibrated which can leads to inaccurate battery health reading at best or shorten the battery longevity at worst. Nevertheless, it’s not your fault.
Without the RTC cell your computer is going to spend a long time doing “DDR5 memory training” each time booting. Other than that, normal operation.
Not all battery failures ended up in swelling. Some of them might be invisible such as unsafe imbalance (>200mV between cells). However it’s not easy to check because your computer can’t boot up with battery to type framework_tool --smartbattery. Then again, further identifying the problem is unnecessary as your battery is still within warranty and you can get a free replacement from support.
Many other laptops use “bypass charging” or “reverse boost” topology where the system voltage (VSYS) is equal to power adapter voltage (VADP), in this case, 20V. A buck or buck-boost converter is between VSYS and VBAT rail to charge the battery. These laptops will work happily without a battery.
The FL13 uses Narrow VDC topology, which uses a buck-boost converter to convert that 20V(if you use a 60W or 100W charger) into a VSYS “one notch above” VBAT to charge the battery while providing power to the system. If the battery is absent the VSYS and Charger Voltage are the same 15400mV. If the battery is limited to a setting, the VSYS will be a bit higher than VBAT. If the battery is charging normally the “Charger Voltage” will be 17800mV but the actual VSYS is still slightly higher that VBAT but will eventually be 17800mV if you decide to charge it to 100%. If the battery has failed, computers with Narrow VDC may be unable to boot as it doesn’t know what voltage to put into VSYS.
In your current case, the battery is absent so the Charger Current is 0.
1 Like
Surely it must be lower than that. 3.6V is the nominal (average during discharge cycle) voltage stamped on the side of many lithium cells. As I understand it, they are normally discharged to somewhere in the region of 2.8 to 3.2 volts before over-discharge protection kicks in. Perhaps you mean 2.6V ?
1 Like
2.6V certainly sounds a lot more plausible, that or they are just catastrophically out of balance.
It’s also possible that OP battery’s temperature probe went loose and made the BMS into thinking the battery is at -273C, preventing both charging and discharging
Or it just decided it didn’t like something else and decided to go into sudoku mode, then again they usually also stop talking to the host when they do that iirc.
Okay. But it’s weird that they would give a readout of some random point in the charging circuitry.
.. I like how all we did is theorize.
This is about a fairly representative Li-ion battery’s voltage-capacity curve. I want you to figure out which region the battery is the least stressed at. Note that this is not LiHV chem, since it stop high at 4.2V.
C means “capacity”. If you have a 5000mAh battery, 1C discharge would be 5000mA, or 5A. it’s “one capacity”. The faster it drains, the more steep the curve is. Variations of this two exist out there.
Most laptop cells are tuned for a balance of capacity and discharge performance – you can increase the thickness of conductors, which will keep the curve flatter, but this also mean you have less actual battery. I imagine the characteristics of laptop pack should be somewhere between 0.2C to 0.5C on the chart, for 1C behavior.
Oh and these are LiFePo4s. Note the similar shaped curve, but completely different voltages.
root@localhost:/mnt/Volume_A4/ectool-main# ./ectool battery
Battery 0 info:
OEM name: NVT
Model number: FRANDBAT01
Chemistry : LION
Serial number: 0375
Design capacity: 5491 mAh
Last full charge: 4954 mAh
Design output voltage 15480 mV
Cycle count 358
Present voltage 16775 mV
Present current 0 mA
Remaining capacity 4198 mAh
Desired voltage 17600 mV
Desired current 5491 mA
Flags 0x83 AC_PRESENT BATT_PRESENT
Its 4 cells, so do math. 15,48 / 4 is 3.87V (typical voltage) per cell.
Desired voltage is 17.6V (full charge), 4.4V per cell. 5491mA is exactly 1C.
We are at 16.775V, 4.199V, so. somewhere around the 85% mark.
This is typical LiHV (Lithium High Voltage) chemistry. Everything checks out.
@Wilson2 Well. hopefully we all learned something from this … minor incident.
Now, going beneath the “magical 3.6V” or whatever. That’s not immediately disastrous. How much damage depend on how uncontrolled the battery conditions are. If you very very very slowly charge/discharge, you can potentially nullify some damage. And some battery would let you trickle to bring it back. Doesn’t seem like its your case, though.
I don’t think so. OP’s battery is bricked at a higher voltage than I did the test (normal use to 3.3V). Discharge to lower than 3.6V, even routinely, doesn’t damage the battery aside normal wear and tear, especially when the current is higher since the battery’s “internal” voltage is higher but bought down by the internal resistance. In other words, OP’s battery is bricked due to uncooperative BMS. Either one of the cells fails (like voltage imbalance) or the circuit is broken (such as disconnected temperature probe). In either case, this is not caused by discharging below 3.6V
Not really, the shape of the discharge is almost identical because the high current curve is just open circuit curve moved down by internal resistance times current.
This is not a correct chemistry. Mainstream battery chemistries are lithium cobalt oxide (LCO), lithium magnesium oxide (LMO), lithium iron phosphate (LFP), nickel magnesium cobalt (NMC) and lithium cobalt aluminum (NCA). The 3.78V nominal 4.4V charge limit battery is LCO and the discharge cutoff voltage is typically 3.0V. Although the voltage will lower rapidly when below 3.3V, there’s no reason for it to damage at 3.6V
Characteristics
LCO has the highest voltage, high capacity but low longevity. Dendrite can build up if charged at high current on high voltages. In the Framework Laptop, the current is reduced from 1C to 0.7C when the voltage reaches approximately 4.15V/cell.
A typical LCO discharge curve source
NMC has the highest capacity, the voltage is between LCO and LFP, about 3.6V. As a result, the energy is about the same as LCO. NMC has better longevity than LCO but to get the same voltage the battery pack needs more cells so consumer electronics typically go with LCO. NMC is a jack-of-all-trades chemistry, used in power banks, electric vehicles, etc.
A typical NMC discharge curve source
LFP has the lowest capacity but highest longevity. Mostly used on energy buffers at renewable energy power plants.
A typical LFP discharge curve source
So given all of these findings, would any of you recommend anything else than replacing battery?
Sadly it appears that extended warranty does not apply to me, as such I will order one.
Why? What’s the reason?
Advanced: you can use a laboratory power supply, set voltage to 16V current to 0.2 to 0.5A. connect the output to the VBAT and GND pins. It’ll slowly charge the battery to about 60-70%. After the battery voltage has reached 16V, remove the battery and wait for a few hours or a day to let the BMS recalibrate, then put it back to the laptop.
Very advanced: if you know what you are doing you can try to use EV2400 or NLBA1 to reflash the BMS chip. However 99% of the time this way is more expensive than outright replacing the battery. Alternatively you can go to a repair specialist to do that for you. WARNING: Only works if the BMS has some sort of bug, if the battery cells arevery unbalanced or some of the components are fired it’s highly recommend to replace the battery as continue using can be dangerous
In the meantime they wrote back and said they will send a new battery.
Unsure what the mixup was.
Given all your suggestions it looks like the most reasonable choice is simply to replace battery with what they’ll send.
Thank you everyone!
2 Likes
Just to conclude this odyssey: I have received the replacement battery from Framework in an extremely short time (2 days after they confirmed they will send it). It is now installed and my machine is back! 
I also wanted to clarify the warranty as I finally understand it. My laptop was bought in autumn 2024, when FW had just introduced the 61W high-density battery. However, my laptop shipped with BIOS 03.05 from spring 2024. In the meantime, in was found that this battery requires more advanced power management or otherwise may get damaged, which FW introduced later in BIOS updates (specifically 03.09 for the AMD 7040). I wasn’t aware of the BIOS update which would improve this (FW may have announced it, however, I do not have access to the mailbox associated with my order anymore), so I have been sitting and cooking my battery for a year without the proper management (it seems it also took the SWIT Omni 99 power bank out with it when it died…).
Since many users got to use this 61W battery for a while before the update was available for all laptops, FW have kindly extended its warranty for 2 years. This is what the extended battery warranty is, see BIOS update required to prevent premature aging of 61w battery. As for my earlier post about not being in warranty, it likely was a misunderstanding until I confirmed to FW support via serial numbers the conditions for extended warranty: that my laptop has the 61W battery and was purchased before the BIOS update was available, after which they were extremely prompt in replacing it and providing all info I needed to do this smoothly.
As a final note, the guides say update BIOS and then install new battery. However, BIOS cannot be updated on charger/without a battery being plugged in, at least not in this situation (going from 03.05 to 03.20 in one hop!) So if you try, you may be welcome with a funky screen which says in red the update failed. Don’t panic! Install the battery according to the Battery Replacement Guide and then update to the right BIOS using Framework Laptop 13 BIOS and Driver Releases (AMD Ryzen™ 7040 Series). I completed this v smoothly using fwupdmgr.
To conclude, I want to give a last big shout to thank the community for all their help and support, this thread is so full of valuable information now!
And to FW for being honest and responsible about their products by providing this extended warranty to correct a limitation of the product.
3 Likes