CaffeineIsMyDrug
New Member
- Joined
- Dec 8, 2022
- Messages
- 2
Greetings folks!
Having some success reviving a used and abused Jackery Explorer 1000, I decided to acquire a Yeti 6000X that was in a flood. It was apparently in 11" of water, which was just enough to get the batteries wet. The upper electronics (front panel, DC to DC converter, 2000W inverter, and ~50V MPPT) remained dry and presumably still worked. The previous owner operated on it to fix some corroded cables, but was not able to get it to charge. I paid $320 for it, an eBay listing that was as-is, local pickup only, and I was fortunate enough to be within an hour of; taking a gamble that I could revive it, or at least make something fairly useful from the parts.
Upon disassembly, I got the 6000WHr battery pack (300-400 LG 18650 cells in a 3S, 10.8V config) and BMS separated from the rest of the electronics. Pre-BMS voltage was measuring near 0V. I, of course, living a little dangerously, tried to carefully charge it anyway..
With various combinations of my 5A bench supply, a 6V battery charger, and the 3.3V ~17A feed from an ATX supply pushing, I was able to push a few dozen watts into the pack, (and careful monitoring). Over the course of a few hours, I could only get each bank of cells up to about 2V. It would occasionally creep up higher, then drop ~0.2V, etc. I'm not entirely sure if this was cells awakening, but the LG cells listed do not seem to have their protection circuits. In the course of a day or so messing with it, I don't think I ever got the pack back up over 7V. Eventually I noticed a hot spot in the a corner of pack, so ceased my charging efforts. A day later, each of the 3 cell banks was down near 0V again. Oh well, this was kind of expected. I then removed the BMS from the top of the pack. Back down to the work bench...
At my work bench, I connected a ~120A SLA battery to the Lithium BMS. I made a resistor network to fake out the BMS cell balance 1/3 and 2/3 points, connected the thermister. At a little over 12V applied, the BMS would NOT activate the P+ terminals, and this BMS didn't have any kind of hard kill or obvious fuses. I then jumpered the B+ terminals to the P+ terminal and with the BMS comms connected to the front panel, the control unit came to life, showed 100%, and let me turn on the inverter/ DC outputs, etc. (With BMS comms disconnected, the front panel is not happy at all, reports 0%, and won't turn on anything)
Sadly, the Goal Zero has quirky controls in the app for controlling when the MPPT charger is 'allowed' on, configurable, like start under 90%, off at 95%, etc. That said, with the BMS reporting a 100% charged battery (~12.6V > 12.2V VMax) I couldn't quite test the ~50V MPPT (though the charge presence LED did come on).
So tl;dr:
-The 6000Wh of NMC lithium batteries are toast (fully expected); My local recycling center will take them, but I'll might want to save the racking first.
-The BMS powered by a 12V SLA still comms with the front panel, but wouldn't turn on the B+ to P+ (PFET?) network
-The front panel still works, including bluetooth/wifi comms, and DC to DC outputs, but needs a little love from the BMS to run
-The inverter works (limited testing up to ~60W; I didn't have it wired well enough to test w/ my ~1000W heat gun, LOL). I'm not entirely sure if the inverter enablement is a simple EN signal, or if comms to the front panel are required. The unit has 5 fans, and 2 are directly controlled by the inverter PCB it seems.
-I'd guess the MPPT probably works, but we all know they can be pretty fragile and unforgiving to certain overvoltage abuse.
Some questions for y'all:
A) Do any of the commercial all-in-one BMS's have a lockout/safety shutdown that require special tools/software to reset? (The Jackery has a physical switch to turn off the BMS, but it's happy-as-can-be once you get a low battery pack charged back up over the low-voltage cutout)
B) Are there any standards for communications between the BMS and front panel?
C) Is the reporting of voltage/current/remaining energy estimation over said BMS comms, or determined solely by the front panel microcontroller?
D) Has anyone tried to reverse engineer the communications between any of these BMSes inside of the commercial all-in-ones?
I'm thinking there's 3 ways out of this...
1) Build a new 6000WHr battery pack from scratch, reusing the same form factor and reuse the existing BMS, assuming I can verify it actually works completely and correctly
1b) Pick up a used 12V LiFeO4 or other 3S Lithium battery pack, (perhaps one with a bum BMS) and reuse existing BMS
1c) Keep trolling eBay and secondary markets for a 1500X/3000X/6000X with a good battery and bad top half
2) Fake the BMS comms with a microcontroller or something that lets me use an SLA battery or other battery just well enough for the rest of the unit (mppt charger, inverter, etc to run). i.e. Make a Franken-Yeti-1000X/2000X with a custom battery
3) Recycle the cells, and sell the Inverter/Front Panel/MPPT/questionable BMS on eBay where there currently AREN'T any 2nd hand parts.
Having some success reviving a used and abused Jackery Explorer 1000, I decided to acquire a Yeti 6000X that was in a flood. It was apparently in 11" of water, which was just enough to get the batteries wet. The upper electronics (front panel, DC to DC converter, 2000W inverter, and ~50V MPPT) remained dry and presumably still worked. The previous owner operated on it to fix some corroded cables, but was not able to get it to charge. I paid $320 for it, an eBay listing that was as-is, local pickup only, and I was fortunate enough to be within an hour of; taking a gamble that I could revive it, or at least make something fairly useful from the parts.
Upon disassembly, I got the 6000WHr battery pack (300-400 LG 18650 cells in a 3S, 10.8V config) and BMS separated from the rest of the electronics. Pre-BMS voltage was measuring near 0V. I, of course, living a little dangerously, tried to carefully charge it anyway..
With various combinations of my 5A bench supply, a 6V battery charger, and the 3.3V ~17A feed from an ATX supply pushing, I was able to push a few dozen watts into the pack, (and careful monitoring). Over the course of a few hours, I could only get each bank of cells up to about 2V. It would occasionally creep up higher, then drop ~0.2V, etc. I'm not entirely sure if this was cells awakening, but the LG cells listed do not seem to have their protection circuits. In the course of a day or so messing with it, I don't think I ever got the pack back up over 7V. Eventually I noticed a hot spot in the a corner of pack, so ceased my charging efforts. A day later, each of the 3 cell banks was down near 0V again. Oh well, this was kind of expected. I then removed the BMS from the top of the pack. Back down to the work bench...
At my work bench, I connected a ~120A SLA battery to the Lithium BMS. I made a resistor network to fake out the BMS cell balance 1/3 and 2/3 points, connected the thermister. At a little over 12V applied, the BMS would NOT activate the P+ terminals, and this BMS didn't have any kind of hard kill or obvious fuses. I then jumpered the B+ terminals to the P+ terminal and with the BMS comms connected to the front panel, the control unit came to life, showed 100%, and let me turn on the inverter/ DC outputs, etc. (With BMS comms disconnected, the front panel is not happy at all, reports 0%, and won't turn on anything)
Sadly, the Goal Zero has quirky controls in the app for controlling when the MPPT charger is 'allowed' on, configurable, like start under 90%, off at 95%, etc. That said, with the BMS reporting a 100% charged battery (~12.6V > 12.2V VMax) I couldn't quite test the ~50V MPPT (though the charge presence LED did come on).
So tl;dr:
-The 6000Wh of NMC lithium batteries are toast (fully expected); My local recycling center will take them, but I'll might want to save the racking first.
-The BMS powered by a 12V SLA still comms with the front panel, but wouldn't turn on the B+ to P+ (PFET?) network
-The front panel still works, including bluetooth/wifi comms, and DC to DC outputs, but needs a little love from the BMS to run
-The inverter works (limited testing up to ~60W; I didn't have it wired well enough to test w/ my ~1000W heat gun, LOL). I'm not entirely sure if the inverter enablement is a simple EN signal, or if comms to the front panel are required. The unit has 5 fans, and 2 are directly controlled by the inverter PCB it seems.
-I'd guess the MPPT probably works, but we all know they can be pretty fragile and unforgiving to certain overvoltage abuse.
Some questions for y'all:
A) Do any of the commercial all-in-one BMS's have a lockout/safety shutdown that require special tools/software to reset? (The Jackery has a physical switch to turn off the BMS, but it's happy-as-can-be once you get a low battery pack charged back up over the low-voltage cutout)
B) Are there any standards for communications between the BMS and front panel?
C) Is the reporting of voltage/current/remaining energy estimation over said BMS comms, or determined solely by the front panel microcontroller?
D) Has anyone tried to reverse engineer the communications between any of these BMSes inside of the commercial all-in-ones?
I'm thinking there's 3 ways out of this...
1) Build a new 6000WHr battery pack from scratch, reusing the same form factor and reuse the existing BMS, assuming I can verify it actually works completely and correctly
1b) Pick up a used 12V LiFeO4 or other 3S Lithium battery pack, (perhaps one with a bum BMS) and reuse existing BMS
1c) Keep trolling eBay and secondary markets for a 1500X/3000X/6000X with a good battery and bad top half
2) Fake the BMS comms with a microcontroller or something that lets me use an SLA battery or other battery just well enough for the rest of the unit (mppt charger, inverter, etc to run). i.e. Make a Franken-Yeti-1000X/2000X with a custom battery
3) Recycle the cells, and sell the Inverter/Front Panel/MPPT/questionable BMS on eBay where there currently AREN'T any 2nd hand parts.
