-2
\$\begingroup\$

Purpose of this question:

In the interest of making my own BMS circuit designs safe and robust, i seek to understand causes of BMS failures in known systems. I'm not seeking consumer advice or general tips.

History:

In 2015, as reported in the press, many hoverboards (mostly from Chinese manufacturers) caught fire. Some of the hoverboards exploded while charging, others while riding and one while it was simply sitting. Consumers from 24 states reported to the CPSC of 52 hoverboard fires resulting in more than $2 million in property damage over an 11-week period. Then they recalled 1/2 a million hoverboards. https://phys.org/news/2016-02-hoverboards.html

Possible Causes:

Were the infamous hoverboard fires caused by over-voltage charging, over-current charging, over-current discharging, short-circuits due to faulty cells, short-circuits due to faulty wiring or poor mechanical design, poorly-manufactured cells, or something else?

This article speculates problems with switches, cells, chargers... but all speculation.

K.M. Abraham, research professor at Northeastern's Center for Renewable Energy Technologies, stated that the cause is a combination of poor lithium ion battery manufacturing, combined with over-current discharge. He also mentions cell punctures. However, I've not been able to find a copy of his essay-- it's unclear whether his conclusions are based on investigative, forensic research, or just speculation.

This article speculates people might be using the wrong charger (which the USCP echoed that in their public statement). However, the article says the charger jack on the hoverboards is dissimilar to phone and computer charge-jacks, so that's unlikely-- there's no hard evidence that people are using the wrong charger. The also speculate people are overcharging the hoverboards, but again, no data.

Multiple Causes:

While it's possible that each fire had a different cause, the fact that fires were happening repeatedly across the entire product category, including products from different manufacturers, in the space of just a few weeks, strongly suggests a common cause. The fact that UL and CPCS both conducted research to find "the cause" indicates that they too believe there's a common underlying cause.

Forensic Research:

CPSC devoted itself to finding the cause, but i've not been able to find their conclusions. https://incompliancemag.com/cpsc-investigates-safety-of-hoverboards/#respond

UL labs made hoverboard cells explode by pushing a nail into the center of the cell, and by applying indirect heat near the cell. But, we could make the cells from MANY products (which are not exploding during normal usage) explode in the same way.

UL reported: "some cells explode right away, others are built with backup measures designed to release gasses without bursting into flames."

That's approaching a possible solution, but then they'd have to show that gas build-up is the cause in actual hoverboard fires.

UL also conducted drop tests-- the hoverboard droped developed a serious crack, but it did not explode. That suggests puncture to the cells might not be the cause.

https://www.cnet.com/news/new-safety-standards-might-save-hoverboards-from-extinction/

Certified Hoverboard:

In May 2016, UL labs issued their first safety certification of a hoverboard, made by parent company of the Segway: https://incompliancemag.com/ul-certifies-the-first-hoverboard/

Official Cause:

In July 2016, the CPSC announced the official cause: overheating lithium-ion batteries. But many things can cause a cell to overheat-- they don't say what caused the hoverboard fires. I've not been able to find their final report.

Robson Forensic states:

"outcome of their tests found that the lithium-ion battery packs can overheat, posing a risk of fire and explosion."

But that's not news-- every electronics hobbyist and professional who uses lithium ion batteries already knew that lithium-ion battery packs can overheat. But exactly what caused the hoverboard packs to overheat?

Please share data about relevant electronics design of actual hoverboards, not general safety tips. I'm not seeking consumer advice.

\$\endgroup\$

closed as off-topic by Eugene Sh., PlasmaHH, winny, Dave Tweed Dec 6 '17 at 23:33

This question appears to be off-topic. The users who voted to close gave this specific reason:

  • "Questions on the use of electronic devices are off-topic as this site is intended specifically for questions on electronics design." – Eugene Sh., PlasmaHH, winny, Dave Tweed
If this question can be reworded to fit the rules in the help center, please edit the question.

  • \$\begingroup\$ This is pure speculation, unless someone can tell about a single isolated incident which doesn't tell anything about the general mechanisms. All mentioned and possibly many more things will be able to cause fires, and will likely have caused at least one. \$\endgroup\$ – PlasmaHH Dec 6 '17 at 22:12
  • \$\begingroup\$ @EugeneSh I've edited my question to indicate relevance to electronics design. Please un-hold the question. \$\endgroup\$ – johny why Dec 7 '17 at 1:14
  • \$\begingroup\$ @PlasmaHH You're stating that we can't assume that all hoverboard fires had a common cause. You're engaging with the content of the question. That indicates that you accept the legitimacy of the question on this website. Please unhold the question. \$\endgroup\$ – johny why Dec 7 '17 at 1:32
  • 1
    \$\begingroup\$ @johnywhy: with that reasoning every plane crash had a common cause. Also I am not obliged to provide evidence for anything, especially not if come here with your impudent attitude of demanding an answer. The community decides what is ontopic and what not, not you because you happen to want an answer. If you want to discuss that go to meta. \$\endgroup\$ – PlasmaHH Dec 7 '17 at 8:47
  • 1
    \$\begingroup\$ @johnywhy maybe it would be better to phrase your question from a circuit design perspective and post some designs and then ask how they could be improved. \$\endgroup\$ – Voltage Spike Dec 7 '17 at 16:19
2
\$\begingroup\$

So far just advertising from forensic experts.

Case Forensics: - cutting manufacturing corners or rushing products to market without rigorous testing can create catastrophic results. Details are not public, but you can call them.

Robson Forensics - lack of product verification safety tests, overcharge test, crush test and partial immersion tests, (OEM and counterfeits included) - resulted in UL 2272 test standards

The core requirements of UL 2272 focus on:

Ref

Wiring and connectors, with guidance on voltage, current and temperature ratings; averting reverse polarity connections and inadvertent short circuits in charging connectors; abrasion prevention; and instructions for user-serviceable parts and hazardous voltage spacing, per UL 60950-1.

Battery chargers, as integral to the hoverboard or a separate accessory, which includes guidance to follow UL1310 for Class 2 output circuits; UL 1012 for higher output circuits and UL 60950-1 when utilizing existing power supplies from other applications. Additionally, charging connectors must be keyed or otherwise prevent reverse polarity connections and inadvertent short circuits.

Cells must be evaluated or certified in accordance with UL 2580; additionally, based on UL 1642, allowances for larger size cells, hard short circuit tests and increased heating test time for larger cells are considered. An alternate cell test program, IEC 62660-2 is also provided.

Battery requirements are based on UL 2271. Considerations for the charging system within UL 2272 as well as guidance on overcharging, short circuits, over discharge and imbalance charge testing are also listed.

Motors must be evaluated based on the UL 1004 series, including locked rotor testing.

System level requirements include a temperature test, dielectric withstand/isolation resistance requirements and consideration of location and voltage levels of power throughout the system.

Protection circuits and safety analysis require a risk assessment/functional safety approach and safety Failure Modes and Effects Analysis (FMEA) on battery maintenance, shock/fire hazards, redundancy of safety controls/features and active/passive protection.

Other items for consideration include: handling, serviceability, regenerative braking and fully charged/partial charged conditions.

Other

According to this analyst, avoid anything <= $300 on Amazon or Ebay and buy direct from OEM.

Fact

The only reason LiPo's catch fire is they reach a critical temperature where thermal runaway causes unavoidable fire. But there are many causes of overtemp, some of which are unstoppable such as membrane leaks.

A Thermal sensor on each cell is good insurance unless the short is internal before the microwire fuse. ( even if there is one on each cell)

Leaking flammable electrolyte
Unprotected overcharge
LiPo Safety Issues from many stress sources

Examples of counterfeit batteries (substandard)
Which ones don't have them? Probably the cheap ones that went up in flames.

Why are the root causes not publicized?

Even if you were Samsung, would you want everyone to know why? Their mechanical tolerance stackup was too large that caused cell laminate to puncture and caused a cell self-short. Others under internal pressure with thermal duress from charging and talking at the same time or long after being dropped. (face explosion)

It just isn't good for your reputation.

Another reason is poor matched voltage cells result in mismatched ESR which under load means mismatched temp. rise of the battery. The other is aggressive mechanical stress to the batteries. ( I remember my 1st cordless drill that my old g/f threw down carpeted stairs and never worked properly again )

  • Each cell must have a good active balancer if you want extended life on the battery and this adds a SMPS to each cell to bypass when under/overcharged and adds to the cost.

  • Passive balancers are inefficient and rarely handle more than 5~10% Ah capacity mismatch and no cell balancers or poor quality cells are cheap and a good reason to follow the analyst's advice.

It also requires much tighter capacity and ESR balance on battery cells and this takes a lot of know-how from the supplier. Considering the imbalance accelerates to 100% when dead and << 1% when new, the rate of discharge on each cell if not actively balanced during charge AND discharge using SMPS bypass switched inductors which cost a lot, I think of a lot of design reasons why they can burst into fire. Overweight-user, over accelerated, unprotected from mechanical shock that can deform electrode foil and short out etc.

Reminds me of the cheap 5000 mAh rated LiPo cells on Bangood site, that were only 1000mAh. Don't go cheap on new unknown products.

p.s.

Having been in R&D, Mfg DVT Test Engineering (& Mgr), Mfg Eng Mgr of Electronic products and design Services Mgr over a 40 yr span, I can say with high confidence that;

A design is only good as its specs with verified MTBF to those specs (Design Validation Test) with Process Verification Test plans (PVT) on all components, sub-systems and final product. Any missing or overlooked specs and verification tests makes them all high risk.

I have done ever test imaginable (DVT) on the reliability of HDD's from dry ice to high altitude, vibration with Radar pulses and although not a health risk can be a financial risk if bank records are corrupted. ( Circa 80's.) I've done the same on many other technologies, since then.

If the same tests are not done to hoverboards with ORT ( On going reliabilty test plans) there is high risk of repeat failures. Just ask if you need a consultant for better specs than UL's.

\$\endgroup\$
  • \$\begingroup\$ Doesn't address design of the specific devices in the question. As i mentioned in the question, offer details about the specific devices, not "general tips". It's ok to say "i don't know" :D \$\endgroup\$ – johny why Dec 7 '17 at 1:19
  • \$\begingroup\$ You speculate about what MIGHT have caused the problem in hoverboards, but that's just speculation. Comments like "don't do such and such" is consumer advice. StackExchange Electronics is not a consumer advice resource. \$\endgroup\$ – johny why Dec 7 '17 at 1:25
  • \$\begingroup\$ "mechanical tolerance stackup was too large... Others under internal pressure with thermal duress from charging and talking at the same time" Do you have any data or documentation to back up that claim? Or are you just speculating? \$\endgroup\$ – johny why Dec 7 '17 at 1:35
  • 1
    \$\begingroup\$ A possible cause is confused sensor performance, such as the temperature measurement sensors, or current sensors or voltage sensors. In interference-laden situations, what can anyone trust from an important sensor, if the magnetic and electric interferences are not well controlled. \$\endgroup\$ – analogsystemsrf Dec 7 '17 at 3:33
  • 3
    \$\begingroup\$ There is a difference between speculation without experience and that with. \$\endgroup\$ – Sunnyskyguy EE75 Dec 7 '17 at 4:39

Not the answer you're looking for? Browse other questions tagged or ask your own question.