My Project Lithium Battery Caught Fire

there is a drastic difference between PHEV and NON PHEV....

for me, not seeing anyone mention is, unless I overlooked someone mentions anything about the terminals and all focus on the bms or signal doohickey... the move obvious reason is false positive signals from the terminals but I'm no battery engineer.

 

I'd love to respond, but I'm having trouble understanding what question you're asking. Can you please clarify the following:
1: From a BMS perspective, which "drastic difference" are you suggesting exists between PHEV and non-PHEV?
2: Which "false positive signals" are you referring to? The output signals from the Signal Soother?

 

Lithium batteries, especially hobbyist quality Chinese lithium cells light on fire if you try to manage them like NiMH without full monitoring. They are completely and utterly incompatible with NiMH management systems, so you can’t just hook them up as no battery chemistry behaves identically to another.

These cells as I worried are similar to the Chinese Lifepo4 cells I bought in 2011, they age rapidly and eventually can’t be balanced effectively.

The best Lifepo4 come fully matched from the factory with minimal self discharge
so you can charge and discharge thousands of cycles and many years before balancing them, but still need a full bms to stop the charge / disable the car if any imbalance or fault occurs. Otherwise fault = fire

 

lets back up a second... who didnt know the oe batteries in a PHEV are not NiMH?
this is what one of the four rows of modules looks like, who did not know this?

 

Did anyone know that the chemistry in the bz4x batteries by my research will work in the Prius PHEV models? If not mistaken the non AWD version is the meal ticket. This is how I plan to obtain 50kWh with hopes all is contained in OE case with some lifters under the top frame of the battery case if needed.

Not concerned about the repercussions of anything going wrong as it hasn't happened and I might as well keep on experimenting as so far its gotten this lil Prius pretty far!

 

I'm understanding that @mudder has already done a teardown of a nexcell module and knows more about the nexcell module embedded bms.

The kicker is if the module embedded bms is sufficient for it's purpose. This threads episode exposes a failure scenario where the BMS in each module is not capable of handling the issue that caused the thermal meltdown.

Which begs the question, is there / are there other scenarios that can cause the BMS to miss critical fluctuations within the pack resulting in an unsafe event.

 

That might break down into a few smaller questions:

• Is the in-module balancing circuit sufficient for its purpose when its cable is plugged in?
• Is the in-module balancing circuit sufficient for its purpose when its cable is not plugged in?
• Are fourteen in-module balancing circuits sufficient for serial charging/discharging of fourteen modules?
• How does the car's BMS try to respond to imbalance among the modules?
• Does the car's BMS get accurate information about imbalance among the modules?

 

and also

• Is there any corrosion on any of the connectors on the vehicle's OE battery components?

 

I think there is a lot here that folks with no formal battery training, are assuming based on how something they read works.

the thought that my limited battery knowledge is questioned as to say I don't know anything about what I speak, to find out, most didn't know PHEV doesn't have NiMH. that makes me think more, okay, if they didn't know that, and it doesn't appear they know about the Prius ECUs operating effectively when a non-vehicle specific engine is connected to the ECUs... let alone what happens on a bored-out 1.8L and how the ECU reacts... then why all the noise and hypothetical questions/comments about all this nextcell mumbo jumbo?

If it is to learn about how things work, maybe posts should include "I believe" or "to my knowledge, I could be wrong" versus written as 100% valid facts.

this internet is a wild place!

 

From my reading of this thread in relation to LFP vs NiMH, has always been in the context of replacing stock Gen 2 NiMH with experimental LFP. What you are seeking to do the your PHEV is really another discussion.

I'm willing to bet the programming of a stock battery ECU that controls NiMH modules in a Gen 2 Prius is quite different from the battery ECU controlling Lithium modules in a PHEV (and other lithium-based Prius). I'd also go out on a limb and say the two battery ECUs are not interchangeable.

 

well said, and I can agree that your assessment is a valid one. I guess to say, I can now see folks confused and questioning my words here while not realizing I have a 09(with LFP), 13 Plugin (with larger capacity LFP) and a 17 Prime.

so then in reality, it sounds like the issue is with the Gen 2 only... or maybe even the Gen 3 with NiMH from the factory. Id be curious now to know the differences between the components of the lithium and nimh Gen 3 batteries. if the Gen 3 has the same issue then the fix would rest in the differences then separately, one might be able to implement the differences from the Gen 2 into the Gen2 to eliminate the need for any battery cell add-on circuitry. I bet there could be something as simple as a resistor that should be added somewhere that would make all this work.

ECUs from Gen 4 Prime are interchangeable with 13 Plugin... and based on my findings, Toyota does a lot of mix and match which might prove beneficial to our mission.

the interchanagle from gen 2 to 3 not without some sort of battery harness converter as the connectors are different just by my looking and counting pins/connectors.

 

@TheLastMojojomo

You are wrong.

Search "project lithium soother test"

All crickets when you drop the pudding on their lap.

 

After reading that thread, glad to see there are some conditions that appear to cause check engine lights. However, without reviewing the Signal Soother schematic, I remain unconvinced that what it's doing is safe. For example, it could just be an LPF opamp that smooths out the signal, which might be ok (depending on the time constant). Or it could be an averaging circuit. Hard to tell without a schematic.

Please send me your Signal Soother for review.

 

So far I've reviewed V1 hardware. All it is is a comparator that enables a discharge voltage above a fixed voltage setpoint. That's it. There is zero intercell communication, zero supervisory control, and zero way to flag unsafe operating conditions (unless one of the OEM model-level tap voltage readings is grossly unbalanced.

I will review the V2 and V2.5 hardware just as soon as I get my hands on it.

So far I can only definitively say that the V1 BMS is insufficient. However, based on the images and videos I've seen of V2 and V2.5, I highly doubt that they are up to par, either.

Absolutely. Without a central CPU that can make decisions based on per-cell voltage readings, there is a high risk that a failing cell can cause a thermal event. For example, if a failing cell with high ESR is easily detected with per-cell voltage monitoring (what should exist, but doesn't), but is much more difficult to detect when you're only measuring every N cells (which is how the OEM BMS works).

 

Just to make it easier to find here is a direct link: Project Lithium Soother Test

Here is my reply to that thread:

 

Are you referring to me? If so, please re-read my first post in this thread for my qualifications.

I'm having a hard time following how your last few posts pertain to the discussion at hand.
Can you more clearly explain the point you're trying to make?

Which specific "100% valid facts" are you referring to?

 

come on man, I have been uplifting and trying to rally the troops together this whole time... not once have I resorted to calling you or anyone out, why all of a sudden would I fire shots at one? This is a group effort I thought...

basically, yalls argument on not feasible or logical to use lithium as a replacement for nimh when gen 3 comes with both. the battery cells them selves have no PCB hardware. that said, the system in gen 3 doesn't have a complete ecu and hardware difference between the two battery types to my understanding. now, if gen2 batteries can swap with 3's. one would think whatever is different between the two battery types would be our point of issue when swapping in lithium. plus, i think no one here taking into the fact, you can swap in non OEM cells long as the voltage matches. issa how now I have the larger capacity in the PHEV. this then points back to the topic of the major differences between the PHEV and non with relationship to how the vehicles ECU/BMS or whatever from factory adjusts itself to the new specs of hardware attached. its not just the batteries either! no ones been up for the adventure of testing to find what works and what doesn't, well in the past ill say.

anything someone posts with the intent of making a fact about the discussion at hand.
I'm not here to get into the specific use cases, wasting time.
just pointing out, I know I dont know all and might articulate what point I am trying to make in a way some dont understand. I'm human. therefore I do my best to preface the words I use, particularly when it's something I know I dont know the majority on. case in point obd2 devices, even tho I am working towards a wireless techstream obd2 implementation that also works real time with the rear motor controller using the pedal and other data points to propel me where Im going. I still can't write word as 100% factual information unless I had some sort of formal training, particularly if I am still not sure how a resolution will come to fruition.

 

A bit rich, considering that you've seen the cells are connected via solder joints.

More like the last 1176. There is a method to the madness, though, and this is how I imagine a jmyntrn post is composed:

1. Start on topic, and quickly go off on a tangent.
2. Imply that there are no facts and everything is just somebody's opinion.
3. Allege that Toyota is hiding the secret to perpetual motion somewhere within the Prius.
4. Mention AI, 3D scanning, Bluetooth, lift kits.
   
5. Include links to random products on personal website if feeling particularly salty.

 

Reading through this thread it strikes me that there are few here who have just enough knowledge to be dangerous.

The only swapping that is acceptable is like-for-like. Even if people are swapping Lithium for NiMH does mean that it is a good thing to do or a good idea.

I don't know where you got this idea, but if it were that simple everybody would be doing it. It doesn't sound like a good idea.

If a person really wants to do these retrofits well they actually need to school themselves completely on how the current system functions in its entirety. If that at least was done, I'm sure there wouldn't be so many cockamamie proposals on the table.

Just my positive encouragement for people to do it better.

 

I read the thread... I would still like to see the results realtime doing a test similar to the mutimeter voltage and battery ecu voltage comparison test I did above... I will link it again below for good measure:

I will also link my reply to that thread below:

Those pics are in no way perfect to verify that in no circumstance does the Signal Soother circumvent safety mechanisms due to errors that are occuring. Please do something similar to the video test I did. I would even be willing to edit it for you if that's what scares you. My video editing has improved alot in the last few months as you can see from watching the 1st minute or so of this video.

If the results of that test show no issues I will be willing to concede to some degree... but again... anything that alters voltage readings from the Hybrid Battery to something that is false is a horrible oversight in safety and logic in my opinion. It adds more complexity than is suppose to be necessary and only increases the likelihood of errors and safety issues long term.