Is anyone using Additional Temperature sensors for DIY LifePO4 battery banks ?

Most of the BMS's seems to have low number of Temperature sensors (e.g. JK BMS 1 to 3 temp seonsors).
Since the cell count is usually high 8s-16s, if a cell that is NOT in touch with the temp. sensor heats up, BMS will NOT detect it and could lead to a catastrophic failure.

Is there anyone who use additional temp. sensors with their LifePO4 setup ?
Something like, temp. sensor per each cell and (using a micro-controller e.g. arduino) if at least one cell's temp goes above a threshold, trigger a alarm AND/OR disconnect the battery (e.g. using A High current Relay connected in series with the battery)

akumd said:

if a cell that is NOT in touch with the temp. sensor heats up,

Click to expand...

Is this a scenario that you have seen or heard about?

MisterSandals said:

Is this a scenario that you have seen or heard about?

Click to expand...

sorry if I confused you, what I meant was cell heats up due to some reason (e.g. a fault in the cell, faulty BMS) and NOT detecting it. I did NOT mean cell heats up due to NOT being in touch with a sensor.

I saw one YT video where a one Lifepo4 cell fumed and the owner said that BMS temp. sensor was NOT connected to the culprit cell (otherwise BMS would have cut-off the battery earlier).
Also in this forum too, in "Up in smoke" section saw few scenarios with LifePO4 cells fuming or catching fire.

In my experience, the BMSs cell level over volt and cell level under volt conditions would
enact a disconnect long before cell temperature became an issue.

Are there scenarios where a cell is within safe operating voltage (2.50V-3.65V) starts "fuming or catching fire"? I will add the battery over current scenario but the BMS would handle this too, unless the BMS mosfetts failed in a closed state (in which case a temp sensor telling the BMS to open is futile).

I'm not saying its impossible but i do not think i have seen a case of this type of failure where a BMS is still alive and functioning enough to act on a temp sensor reading.

MisterSandals said:

In my experience, the BMSs cell level over volt and cell level under volt conditions would
enact a disconnect long before cell temperature became an issue.

Are there scenarios where a cell is within safe operating voltage (2.50V-3.65V) starts "fuming or catching fire"? I will add the battery over current scenario but the BMS would handle this too, unless the BMS mosfetts failed in a closed state (in which case a temp sensor telling the BMS to open is futile).

I'm not saying its impossible but i do not think i have seen a case of this type of failure where a BMS is still alive and functioning enough to act on a temp sensor reading.

Click to expand...

one example of BMS NOT detecting the heatup,

YT video I mentioned earlier,

My Orion BMS has electrical provisions for 3x 10k thermistors so I ordered a few different ones to try.

I have a ring terminal one on a bus bar post near the middle of the pack, with the idea that the post is quite effectively conducting pack temperature up from within that cell in the middle, so it's my "internal pack temperature" reading.

Then I have a smaller free hanging one, that's on the end of a dual maybe 20 awg wire, and it is sort of lying gently on top of all the balance wires, so that it is reading the air temperature inside the case. You have to watch out with ordering 10k's cause some of them come on wires thin as hair.

And the last one is a smaller ring terminal that I put up on my positive charge bus, that's outside the case where charge sources connect. So if the bus bar ever overheats it will go off. That's only one out of a few different bus bars, but I was running out of ideas for temp sensors.

For that youtube report, my answer is that I have added a vent from my battery case to the outside. Which is a better place for smoke than indoors.

If I were to design a more thorough temperature monitoring system, it would be a thermistor on every connection, ever post, every lug, every bus bar, etc. They don't hurt anything being there on top of the lug under the nut so why not.

akumd said:

one example of BMS NOT detecting the heatup,

Click to expand...

The thread and YT video both showed what looked like a single cell having some issue. Its not clear to me or any of the commenters whether it was shorted against a frame or internal short because it was mounted on end.

I am also not sure if either of these scenarios would have benefitted from a BMS disconnect.

My takeaway is that there are 2 datapoints where cells mounted on end had issues.

I’m if you’re using a temp sensor to detecte high cell temps for all cells, the damage is already done.

Most of the safety features prevent damage. Under temp charging protection, over /under voltage protection.

I have a thermal fuse taped on to each cell (4S , 12 volt) which will stop charging in the event a cell goes over 65C. It's independent of the BMS, so a charge cut-off of last resort. Not intended to protect the cells, intended to avoid a fire or off gassing.

Marinepower said:

I have a thermal fuse taped on to each cell (4S , 12 volt) which will stop charging in the event a cell goes over 65C. It's independent of the BMS, so a charge cut-off of last resort. Not intended to protect the cells, intended to avoid a fire or off gassing.

Click to expand...

interesting ! (may be cell manufacturers should consider this )
What is the current rating of your fuse ? Also what was the approximate cost of a single fuse ?

I don't have any familiarity with thermal fuses, are they re-usable OR once "blown"(due to temp rise) has to be replaced ?

740GLE said:

I’m if you’re using a temp sensor to detecte high cell temps for all cells, the damage is already done.

Most of the safety features prevent damage. Under temp charging protection, over /under voltage protection.

Click to expand...

purpose is to prevent charging/discharging when at least one cell's temperature has risen. It will reduce the possibility of further escalation.

is good idea. have had a lead acid pack boil over with only one thermometer at center. monitoring each cell temperature regardless of chemistry makes good sense if the budget allows

akumd said:

interesting ! (may be cell manufacturers should consider this )
What is the current rating of your fuse ? Also what was the approximate cost of a single fuse ?

I don't have any familiarity with thermal fuses, are they re-usable OR once "blown"(due to temp rise) has to be replaced ?

Click to expand...

there are all kind of thermal fuses with difrent ratings. Some reset, some are one time use.

One time use is better for this application as you don't want charging to restart once a cell has gotten that hot.

This works for me as all my charge sources (alternator, SCC, shore charger) are controled by a voltage sense wire (like the remote inputs on the Victron) or i have them routed through a relay that is controled volt sense wire.