Rivian Patent - Charging Algorithm - May explain why owners have different charge rate experiences

[Zoidz]

There has been a fair amount of discussion regarding L3 charging where different users see different charge rates with no clear explanation. This patent application was filed in July 2023, so it's possible (likely?) that this algorithm is used in our vehicles. It uses temperature, voltage and current sensors via a mathematical derivation to predict an overheating situation and cut charge rate preemptively. It specifically says that the algorithm can adjust based on new data collected.

. It's possible/likely that owners will see different charging experiences due to this.

Sponsored

 

[DayTripping]

Great find and very interesting.

 

[COdogman]

 

[Zoidz]

For those interested in the extensive details, it's patent 20250042285 at https://ppubs.uspto.gov/pubwebapp/static/pages/ppubsbasic.html

You'll have to search it, it appears direct links to the search document don't work duw to how they handle session tokens.

 

[Nixapatfan]

Anyone that didn't know this is what all good BMS does probably isn't really going to understand this. They'll still stop by to complain.

 

[DayTripping]

Here is another way to access it.

https://patents.justia.com/patent/20250042285

I asked Chat GPT for a summary of the page I linked.

This patent describes an advanced electric vehicle (EV) charging system designed to make charging more efficient, cost-effective, and environmentally friendly. Unlike traditional chargers that simply supply power, this system can intelligently adjust charging based on various factors such as electricity demand, pricing, and renewable energy availability. By optimizing when and how an EV is charged, the system helps drivers save money, reduces strain on the power grid, and ensures that charging happens in the most sustainable way possible.

A major feature of this system is its ability to communicate with both the EV and the power grid. It can receive real-time data about electricity prices and demand, allowing it to charge the vehicle when rates are lowest. For example, if electricity is cheaper at night when overall demand is low, the system can delay charging until then. Additionally, it can adjust the charging speed based on grid conditions, slowing down or pausing charging during peak hours to help prevent overloading the electrical system.

Another important aspect is the system’s integration with renewable energy sources like solar and wind power. If clean energy is available, the charger can prioritize using it instead of drawing power from fossil fuel-based electricity sources. This reduces the carbon footprint of EV charging and makes better use of sustainable energy. The system can also store energy information and predict when renewable sources will be most available, further improving efficiency.

For EV owners, the system offers a convenient and user-friendly experience. A smartphone app or online dashboard allows users to monitor and control charging remotely. They can set preferences such as when they want their car to be fully charged, and the system will automatically handle the process in the most cost-effective and eco-friendly way. This takes the guesswork out of charging and ensures that the vehicle is always ready when needed.

Overall, this smart EV charging system represents a step forward in making electric vehicle charging more intelligent and efficient. By optimizing energy use, integrating with renewable sources, and giving users greater control, it not only benefits individual EV owners but also supports a more stable and sustainable electricity grid. As electric vehicles become more common, innovations like this will play a key role in ensuring that charging is both convenient and environmentally responsible.

 

[Zoidz]

Here is another way to access it.

https://patents.justia.com/patent/20250042285

I asked Chat GPT for a summary of the page I linked.

This patent describes an advanced electric vehicle (EV) charging system designed to make charging more efficient, cost-effective, and environmentally friendly. Unlike traditional chargers that simply supply power, this system can intelligently adjust charging based on various factors such as electricity demand, pricing, and renewable energy availability. By optimizing when and how an EV is charged, the system helps drivers save money, reduces strain on the power grid, and ensures that charging happens in the most sustainable way possible.

A major feature of this system is its ability to communicate with both the EV and the power grid. It can receive real-time data about electricity prices and demand, allowing it to charge the vehicle when rates are lowest. For example, if electricity is cheaper at night when overall demand is low, the system can delay charging until then. Additionally, it can adjust the charging speed based on grid conditions, slowing down or pausing charging during peak hours to help prevent overloading the electrical system.

Another important aspect is the system’s integration with renewable energy sources like solar and wind power. If clean energy is available, the charger can prioritize using it instead of drawing power from fossil fuel-based electricity sources. This reduces the carbon footprint of EV charging and makes better use of sustainable energy. The system can also store energy information and predict when renewable sources will be most available, further improving efficiency.

For EV owners, the system offers a convenient and user-friendly experience. A smartphone app or online dashboard allows users to monitor and control charging remotely. They can set preferences such as when they want their car to be fully charged, and the system will automatically handle the process in the most cost-effective and eco-friendly way. This takes the guesswork out of charging and ensures that the vehicle is always ready when needed.

Overall, this smart EV charging system represents a step forward in making electric vehicle charging more intelligent and efficient. By optimizing energy use, integrating with renewable sources, and giving users greater control, it not only benefits individual EV owners but also supports a more stable and sustainable electricity grid. As electric vehicles become more common, innovations like this will play a key role in ensuring that charging is both convenient and environmentally responsible.

This is a perfect example of AI Hallucination. The patent literally says none of this, lol.

 

[portdirect]

Here is another way to access it.

https://patents.justia.com/patent/20250042285

I asked Chat GPT for a summary of the page I linked.

This patent describes an advanced electric vehicle (EV) charging system designed to make charging more efficient, cost-effective, and environmentally friendly. Unlike traditional chargers that simply supply power, this system can intelligently adjust charging based on various factors such as electricity demand, pricing, and renewable energy availability. By optimizing when and how an EV is charged, the system helps drivers save money, reduces strain on the power grid, and ensures that charging happens in the most sustainable way possible.

A major feature of this system is its ability to communicate with both the EV and the power grid. It can receive real-time data about electricity prices and demand, allowing it to charge the vehicle when rates are lowest. For example, if electricity is cheaper at night when overall demand is low, the system can delay charging until then. Additionally, it can adjust the charging speed based on grid conditions, slowing down or pausing charging during peak hours to help prevent overloading the electrical system.

Another important aspect is the system’s integration with renewable energy sources like solar and wind power. If clean energy is available, the charger can prioritize using it instead of drawing power from fossil fuel-based electricity sources. This reduces the carbon footprint of EV charging and makes better use of sustainable energy. The system can also store energy information and predict when renewable sources will be most available, further improving efficiency.

For EV owners, the system offers a convenient and user-friendly experience. A smartphone app or online dashboard allows users to monitor and control charging remotely. They can set preferences such as when they want their car to be fully charged, and the system will automatically handle the process in the most cost-effective and eco-friendly way. This takes the guesswork out of charging and ensures that the vehicle is always ready when needed.

Overall, this smart EV charging system represents a step forward in making electric vehicle charging more intelligent and efficient. By optimizing energy use, integrating with renewable sources, and giving users greater control, it not only benefits individual EV owners but also supports a more stable and sustainable electricity grid. As electric vehicles become more common, innovations like this will play a key role in ensuring that charging is both convenient and environmentally responsible.

I did the same, and got some slightly more useful output from o1-pro:


1. Overview
This patent application describes a temperature-based charging control system for electric vehicles (EVs). Unlike traditional battery management systems (BMS) that often rely on absolute temperature thresholds (for example, stopping or slowing charging once a certain connector temperature is reached), this system continuously monitors how fast the connector temperature is changing (the “derivative” or “gradient” of temperature over time). By looking at the temperature’s rate of increase—rather than just its magnitude—this method aims to anticipate and prevent thermal events before they occur.

2. How It Works
1. Temperature Measurement
• The system places temperature sensors near the charging pins on the vehicle’s connector. These pins carry charging current from an external charger to the EV’s battery.
• Multiple sensors may be used, each capturing temperature readings from different pins (e.g., positive and negative pins)
.
2. Derivative (Rate of Change) Calculation
• The EV’s data processing system (or a charger’s processor) collects these temperature readings over time.
• It calculates how quickly the connector’s temperature is rising—or the “thermal gradient.”
• In addition, the system may compare rate-of-change measurements across different pins (i.e., detect if one pin is heating up significantly faster than the other).

3. Adaptive Charging Control
• If the derivative of temperature exceeds a predefined threshold—or if one pin’s heating rate is drastically higher than another—the system predicts an impending thermal issue.
• In response, charging current can be reduced (“derated”) or halted entirely before the temperature reaches dangerous levels.
• This approach helps avoid thermal runaway or damage to the connector and battery pack.

4. Learning and Tuning Thresholds
• The thresholds that trigger derating or shutdown can be determined through real-world testing or data analysis across multiple vehicles and chargers.
• These thresholds may be continuously tuned or updated based on observed conditions, improving predictive accuracy over time.

3. Advantages Over Traditional BMS Approaches
Most existing battery management systems focus on:
• Absolute Temperature Limits: They stop or slow down charging only after the connector or battery surpasses a specific high-temperature threshold (e.g., 80°C or 90°C). By that time, significant damage can already be underway.
• Battery-Centric Monitoring: They may prioritize monitoring cell or module temperatures inside the battery, paying less attention to connector temperatures or how quickly the connector heats up.

Key Differentiator:
This new system uses temperature rate-of-change at the charging interface itself. This proactive method predicts a potential fault sooner—rather than reacting after the temperature has already spiked—thereby preventing or minimizing damage.


4. Plain English Explanation
Think of it like watching water boil. Instead of waiting until the water is already at 100°C to turn down the heat, you monitor how quickly the bubbles are forming. If the bubbles start forming too fast (indicating a rapid temperature rise), you turn down the stove before it boils over. Here, the “boiling over” is equivalent to a thermal event at the charging connector.

By responding to the speed of temperature change rather than just a single temperature reading, the EV can gracefully and safely slow or stop charging. This keeps the connector from reaching temperatures that could cause damage or pose safety risks. It’s a more predictive and proactive safety measure than many current charging or BMS systems.

 

[mikehmb]

They implemented down-rating a few cycles back based on connector temperature, at least on L2, so this definitely jives.

I haven't read beyond what you posted here, so unclear to me if this applies to L3. Seems logical - you'd want to take into account a number of factors including the potential (ha!) of the input overheating.

Good find. I'll try to read it in full later tonight. The ChatGPT interpretation made me think I was reading the collateral from a junior marketing manager.

 

[ThirteenElectrics]

Anyone that didn't know this is what all good BMS does probably isn't really going to understand this. They'll still stop by to complain.

Slow charging is still a bad user experience, even if it’s the BMS causing it.

 

[Glembi2]

To see what the patent drafted thought would be the broadest protection available on filing, just go to the claims section:

“1. A system, comprising:
a data processing system comprising one or more memory devices coupled with one or more processors to:
receive a plurality of measurements of a temperature of a connector of a vehicle, the connector to charge the vehicle;
determine a derivative using the plurality of measurements; and
control charging of the vehicle based on the derivative.”

Google Patents is good but delays posting pre-grant publications (like the one above) and patents by a month or so.
https://patents.google.com/patent/US20250042285A1/en?oq=20250042285

Another good site I use for searching is Espacenet (the European Patent Office’s search portal):
https://worldwide.espacenet.com/pat...ublication/US2025042285A1?q=pn=US2025042285A1

Sponsored