Really stupid EVSE question

GeorgeB

ElectroHydraulics engineer (retired)
Location
Greenville SC
Occupation
Retired
I don't have an EV. I don't have any EVSE at all. I do have curiosity. Let's not kill the cat.

I think I understand how circuit ampacity and OLP determine charging maxima. EVSE products often offer multiple settings, for example one may be rated up to 32 amps for use on a 40 amp circuit and have a 14-50 plug.

That same unit may be able to be set only allowing (example) 16, 20, 24, 28, or 32 amps allowing use on 20, 25, 30, 35, or 40 amp rated circuits.

HOW does it accomplish limiting? Maybe the larger question is DOES it limit, or only tell the EV over data connection to limit it? In theory, I can see how a CT, simple electronics, and thyristor(s) could do it, but ... ?
 

retirede

Senior Member
Location
Illinois
The EVSE tells the car what the max allowable charge rate is. The communication is one-way. The charger in the car actually controls the charge rate. Look up J1772 protocol for details.

Most EVSEs have internal thermal monitoring and will tell the car to derate when necessary. Likewise, the car has temp sensors in critical areas (such as the charge port, charger, and battery) and can throttle the charge rate based on those temps.

Your questions are not stupid at all.
 

augie47

Moderator
Staff member
Location
Tennessee
Occupation
State Electrical Inspector (Retired)
Definitely not a stupid question...
This thread sheds some light on Tesla charging

 

brycenesbitt

Senior Member
Location
United States
HOW does it accomplish limiting? .... In theory, I can see how a CT, simple electronics, and thyristor(s) could do it, but ... ?
The J1772 pilot signal is pulse width modulated and in a remarkably analog manner signals, one way, the maximum amps available.
The EVSE is free to alter the pulse train for any reason, such as heat or a grid emergency, to slow the car down.
Where the best place to control charge is quite unclear --- some cars, EVSE equipment and load management systems all can do it and there's no clear single standard or even place where charge rate is controlled.
 

retirede

Senior Member
Location
Illinois
The J1772 pilot signal is pulse width modulated and in a remarkably analog manner signals, one way, the maximum amps available.
The EVSE is free to alter the pulse train for any reason, such as heat or a grid emergency, to slow the car down.
Where the best place to control charge is quite unclear --- some cars, EVSE equipment and load management systems all can do it and there's no clear single standard or even place where charge rate is controlled.

I’m not sure what you mean. The charger in the car is always in control. It just uses information from external devices such as the EVSE to help determine the charge rate. J1772 is the standard. I know Tesla technically doesn’t use J1772, but the pilot signal control is essentially the same.
 

brycenesbitt

Senior Member
Location
United States
I’m not sure what you mean. The charger in the car is always in control. It just uses information from external devices such as the EVSE to help determine the charge rate. J1772 is the standard. I know Tesla technically doesn’t use J1772, but the pilot signal control is essentially the same.
The "Charger" or EVSE can be used to dynamically throttle the car's maximum charging rate, or to shut off power completely.
The dynamic time of use or grid support programs are often set up at the EVSE, not the car. Meaning the EVSE has the Internet connection and determines the charging window. So both the EVSE and the car have to independently be set to compatible schedules, as there's no communication beyond the pilot signal.
This can be messy. A car might be set to "charge as late as possible to be at 80% by 11am departure" but the EVSE is given information that the electric rates are forecast to be high from 9am onward. The EVSE has no way to accelerate charging overnight --- all it can do is offer the car maximum amps all night then shut off at 9am.
 

retirede

Senior Member
Location
Illinois
The "Charger" or EVSE can be used to dynamically throttle the car's maximum charging rate, or to shut off power completely.
The dynamic time of use or grid support programs are often set up at the EVSE, not the car. Meaning the EVSE has the Internet connection and determines the charging window. So both the EVSE and the car have to independently be set to compatible schedules, as there's no communication beyond the pilot signal.
This can be messy. A car might be set to "charge as late as possible to be at 80% by 11am departure" but the EVSE is given information that the electric rates are forecast to be high from 9am onward. The EVSE has no way to accelerate charging overnight --- all it can do is offer the car maximum amps all night then shut off at 9am.

That’s only an issue in the (somewhat rare) case that TOU rates are dynamic. I know at least one utility on the east coast will give you a discounted rate if you use their EVSE and let them control it. If you agree, you need to make sure you’re always plugged in so you can charge when available.
Everyone I know with a fixed TOU scheme sets the schedule in the vehicle to match the rate schedule and forgets it.
I also know of another utility that will give a discount if you give them your vehicle log-in credentials so they can control charging directly in the car. This is a disaster because it’s a violation of the terms of service of the vehicle software and has resulted in users losing access to the vehicle software.
I don’t know of any EV that enables you to schedule dynamically by finish time. I know some drivers have pined for this feature, but I’m not sure it’s worth the complexity.
 

brycenesbitt

Senior Member
Location
United States
That’s only an issue in the (somewhat rare) case that TOU rates are dynamic.
I'm a member of a private load shedding incentive program that amounts to the same thing. Look up "JuicePoints" from Enel X.
The program is run 100% in the EVSE.
The Chevy bolt allows setting a finish time and policy "as soon as" or "as late as possible". To work with JuicePoints it's best to set the car to charge immediately, then let the robots figure out when to charge.

For Duck Curve reasons I try to charge during daylight hours myself, even though it would be safer to start charging at midnight.
But a future version of the load shedding initiative will probably take care of that, once excess daytime solar becomes a real problem.
 
Last edited:

brycenesbitt

Senior Member
Location
United States
For those not familiar with the Duck Curve, California's grid faces very uneven generation with excess during the daytime,
and peaker generation plants running from 4-8pm or so, and perhaps again at midnight:

california_iso_duck_curve.jpg
See, looks like a Duck. Source: https://www.caiso.com/
Thus all the local interest in EVLMS systems (EV load management systems) in my area of California.
 
Last edited:
Top