Measured charging overhead in Kona?

  • Thread starter Thread starter TheLight75
  • Start date Start date
  • Replies Replies 7
  • Views Views 4K
TheLight75

TheLight75

Active Member
A few years ago, when I owned a Kona EV, it was casually thrown out that charging via an EVSE consumed 15-20% of overhead. Have there been conducted any actual scientific tests to confirm the charging overhead of the Kona EV?

When I say charging overhead, I mean the difference between the kWh delivered by the EVSE vs the kWh charge increase to the battery.
 
I did a test a few months back using a TP-Link energy-measuring plug adapter thing that I verified against another AC wattmeter. Accuracy is important of course. From that I get a total kWh value for the session. At the upper-right of the image I have the session total energy and time. The Torque Pro logging stopped unexpectedly after a shorter period, 4.28 h, so I had to assume a constant power input and just factor it.

The DC power output of the OBC can be determined from OBD data for pack voltage and amps, or alternatively another PIDs contain the product of those, labeled as "000_Battery Power". Integrate that power over time for cumulative energy. The dash readout appears to reflect this power value.

The DC input energy that the battery pack takes on can be determined by the change in OBD data CEC (cumulative energy charged) but the resolution is only 0.1 kWh so you need to charge up quite a bit to minimise that error. Note the steps in the red line.

I also pre-charged the 12V battery in Utility mode before the test to remove that source of error, shown in the small unreadable 12V system graph.

So, in this graph I have cumulative energy at each major point in the system, AC in on top, then OBC out in blue, and the red jiggle is CEC going into the battery after the LDC takes it share, about 0.2 kW.

To determine an LDC efficiency I have to assume that the only loads on the OBC are the LDC and the battery pack so I can deduce what the LDC draws. I assumed an efficiency value to produce the black line then adjusted that to match the black with the red line, ending up with 92% in the end.

Overall charging efficiency is about 79.5%, not terrible but not to be ignored either.

The last thing is that the overall efficiency of AC charging varies a lot by the amps that you're charging at and I have only one option at home, 230VAC 8A which turns out to be only 7.2 amps. The OBC could be up to 91% efficient at higher current and that 0.2 kW a smaller portion, both reasons to charge at high rates. But I doubt you could reach 90% overall.

Low power charging efficiency.webp
 
Last edited:
Wow! This is quite the analysis and I appreciate it. The reason I was asking is because I recently purchased a 2022 Bolt EUV and have noticed a sizable difference between the electricity cost-per-mile. When I had my Kona, it cost me $0.11/mile in electricity. But with my Bolt EUV, the cost-per-mile so far is about $0.07/mile. The same EVSE has been used on both cars and my utility cost-per-kWh is the same so the only explanation I could come up with is that the Kona had a larger OBC overhead.

Last night before charging the Bolt, I noted the totals hours on the EVSE and then again this morning after it was done. I also noted the battery level pre-charging so I could determine how many kWh's were added overnight. The EVSE delivered 37.3 kWh which added 34.45 kWh to the battery. This is an OBC efficiency of 92.36% which is definitely higher than what I got with my Kona.

KiwiME said:
I did a test a few months back using a TP-Link energy-measuring plug adapter thing that I verified against another AC wattmeter. Accuracy is important of course. From that I get a total kWh value for the session. At the upper-right of the image I have the session total energy and time. The Torque Pro logging stopped unexpectedly after a shorter period, 4.28 h, so I had to assume a constant power input and just factor it.

The DC power output of the OBC can be determined from OBD data for pack voltage and amps, or alternatively another PIDs contain the product of those, labeled as "000_Battery Power". Integrate that power over time for cumulative energy. The dash readout appears to reflect this power value.

The DC input energy that the battery pack takes on can be determined by the change in OBD data CEC (cumulative energy charged) but the resolution is only 0.1 kWh so you need to charge up quite a bit to minimise that error. Note the steps in the red line.

I also pre-charged the 12V battery in Utility mode before the test to remove that source of error, shown in the small unreadable 12V system graph.

So, in this graph I have cumulative energy at each major point in the system, AC in on top, then OBC out in blue, and the red jiggle is CEC going into the battery after the LDC takes it share, about 0.2 kW.

To determine an LDC efficiency I have to assume that the only loads on the OBC are the LDC and the battery pack so I can deduce what the LDC draws. I assumed an efficiency value to produce the black line then adjusted that to match the black with the red line, ending up with 92% in the end.

Overall charging efficiency is about 79.5%, not terrible but not to be ignored either.

The last thing is that the overall efficiency of AC charging varies a lot by the amps that you're charging at and I have only one option at home, 230VAC 8A which turns out to be only 7.2 amps. The OBC could be up to 91% efficient at higher current and that 0.2 kW a smaller portion, both reasons to charge at high rates. But I doubt you could reach 90% overall.

View attachment 17071
Click to expand...
 
Ultimately what matters is KwH consumed at EVSE vs Miles output.

I am getting about 3.9 Miles/ KwH with mostly highway driving with speeds upto 75 and average of 55 with level 1 regen. ( Mostly my wife)
I get about 4.1 Miles / KwH at similar speeds using level 0 regen and using left paddle to slow down. On local roads with speed upto 45 MPH I get about 5+ Miles / KwH again at 0 regen. Someday I will try my wife's route and compare difference between Level 1 regen and level 0 regen for a better comparison.
This is of course at summer temparatures of about 80F.

I have been charging at 16 Amps (3.8 Kw/H) at about half the capacity , since I imagine it would be better for battery in the long run. This though may lead to slightly increased loss at OBC due to longer charging times.

All KwH is as reported by EVSE App. But I have seen it closely matches with electric meter, after offsetting mostly refrigerator use overnight.
 
Last edited:
From what I've seen and heard, AC charging efficiency usually falls somewhere between 79-92%, depending on things like how fast you're charging and what’s pulling extra power.
 
Tylopwort said:
From what I've seen and heard, AC charging efficiency usually falls somewhere between 79-92%, depending on things like how fast you're charging and what’s pulling extra power.
Click to expand...
The idea that the Kona might burn more energy overhead than something like the Bolt EUV is pretty interesting, but to really compare, you'd need to look at stuff like the hardware differences and how the software handles charging.
When I was tracking my own charging data, I found it helpful to use an on-screen tool to measure things like changes in battery levels or energy graphs. Something like https://ruler.onl/ could make it easier to eyeball and compare those numbers.
 
You must log in or register to reply here.
Back
Top