Electronics > Projects, Designs, and Technical Stuff
What kind of variable load does an EV car present to the EVSE?
<< < (3/6) > >>
richard.cs:

--- Quote from: fcb on January 07, 2020, 01:44:36 pm ---If you are metering energy use inorder to charge £$ for the service, you are about to enter a whole world of s***.

--- End quote ---

I haven't looked up the rules for this, but it's conceivable that you could sidestep some regulatory complexity by designing it to always under-read by more than your worst-case error budget. It then becomes trivial to show that errors are always in the customers favour (electricity is cheap anyway, and you can just claw it back in the price markup).
max_torque:
There will be very little dynamic load tapering on single phase AC charging because the charging power/rate is so low.  Even on a 7kW system,which tends to be the maximum power rating of a single phase OBC, that is, relative to the battery of a modern EV, trivial to handle (where the battery is rated to deal with hundreds of kW during useage).  Hence, connect the EVSE, the car signals its presence, the EVSE closes it's contactors, and the OBC will pretty much straight away ramp to the requested charge current and hold that untill 100% SoC is reached in most cases (or the OBC charge scheduler otherwise terminates the charging).  Even with a hot/cold battery, just 7kW is unlikely to cause an issue with cell balancing or cell thermal effects.

Once you get to dc fast charging, at powers between 50 and 250 kW then yes, the real time dynamic capability of the battery cells starts to matter, and charge rates will vary with SoC for example
richard.cs:
The 22 kWh Zoe tapers off the charging in the last 1-2% of SOC, and then continues to draw low power around, 2 kW, for quite a long period of time (half an hour or more) as cell balancing or similar takes place. On connection it does a few tests taking 10-20 seconds, at least one of which is earth impedance, and then ramps up the power smoothly over around 5 seconds.
DBecker:
The EVSE typically has an incidental bit of line filtering as a side effect of the GFI/RCI sensing coil.

Power monitoring is either not done, done with current sense coils when the point is informational, or done with revenue-grade shunt resistors (e.g. Vishay WSL3637) when the result might be used to bill a customer.  A typical approach is using the CS5463 Cirrus Logic power monitor.

Every type I've examined has the power monitoring separated from the GFI safety circuit.
richard.cs:

--- Quote from: barycentric on January 07, 2020, 10:34:35 pm ---These are usually not rated for even continuous level 1 (120V/15A/1500W).

--- End quote ---

He did suggest changing the shunt, that would allow freely rescaling the current range.
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod