AC Charging Speed / Efficiency

Negue · Feb 11, 2025

i-Miev charge at 230V 16A EVSE.

Start at 20% SoC (~2.2kwh) at 2°C , charging at 2.5kwh
+~3h
~80% SoC (~9kwh) at 10°C charging at 1kwh

OBDzero says that I charged ~7kwh , and official counter electric supply equipment says ~9kwh.
And with each hour in 1kwh charge time efficiency will drop lower.

Can we improve somehow AC charging speed ?

When I will upgrade to ~40kwh , 80% will be ~32kwh and for rest of 6-8kwh I will need to 8h :/

Negue · Feb 11, 2025

MickeyS70 · Feb 11, 2025

Negue said:
Can we improve somehow AC charging speed ?

The stock OBC is limited to around 3.7kW (AC). Best efficiency is achieved when your EVSE is supplying the full 16A.

Once a cell gets close to 4.1V, the charge rate drops to avoid ‘overshooting’, therefore having a well balanced pack is important.

If you want to achieve even higher AC rates, you would need to replace the OBC altogether.

Negue · Feb 12, 2025

OBC is ranked at 3.7kw (16A) but in reality use ~14.5A , bouth I-Miev and Leaf.
Miev OBC loses ~15% , cells resistance add some loses , electronics in charge mode use some 100-200w ...and means 20% loses :/

Over 80% SoC charging can have 30-40-50% loses and its outrageous

With what OBC do you recommend to replace the Miev OBC ?

WReed82 · Feb 12, 2025

Is this 20% loss typical of EVs?
Are there ways to lower the loss to 5 or 10%?

kiev · Feb 12, 2025

Negue said:
to replace the Miev OBC ?

The operating system of the miev is so tightly integrated with the hardware-- this is not feasible.

BTW, the energy or capacity of the traction pack can be counted in units of kWh, but the power units of the charger is kW.

Much handwaving about efficiency--measure the Output voltage and current, and divide this by the Input voltage and current, to get a good estimate of OBC efficiency.

MickeyS70 · Feb 12, 2025

Negue said:
Over 80% SoC charging can have 30-40-50% loses and its outrageous

If you look closer at your OBDZero data, it shows something odd: a HV battery voltage of 360.1V but a SoC of 81%?

As mentioned before, the BMU will reduce power once a cell comes near the 4.1V limit. Seems a battery capacity calibration is in order?

Negue said:
With what OBC do you recommend to replace the Miev OBC ?

No idea..

Negue · Feb 12, 2025

MickeyS70 said:
If you look closer at your OBDZero data, it shows something odd: a HV battery voltage of 360.1V but a SoC of 81%?

As mentioned before, the BMU will reduce power once a cell comes near the 4.1V limit. Seems a battery capacity calibration is in order?

No idea..

30mV delta it's OK.
4.10V*88cells = 360.8V

Seems ( but I'm not sure) that at 4.10v OBC change in mode :
Fixed Voltage, Variable Amps

In last 8 months didn't charge it.
In last 4 days I charged 2 times.

I will do with OBDZero a Capacity Test and perhaps will reset charging behavior.

@kiev

Charging:
Mains AC 1405w
Battery DC 1264w.
10% loses at half of charger capacity.

Battery:
Watts Out - 1024w
Another -19% loses ( others )

I'm wrong?

MickeyS70 · Feb 12, 2025

Negue said:
30mV delta it's OK.
4.10V*88cells = 360.8V

Correct, but obviously that can’t be @ 81% SoC

Negue said:
Seems ( but I'm not sure) that at 4.10v OBC change in mode :
Fixed Voltage, Variable Amps

Yes, initial the car will change at max current (whatever value the EVSE is set to) then slowly taper off (reduce Amps) closer to and stops altogether when the first cell reaches 4.1V.

Negue said:
In last 8 months didn't charge it.
In last 4 days I charged 2 times.

That explains the discrepancy

Negue said:
I will do with OBDZero a Capacity Test and perhaps will reset charging behavior.

OBDZero’s CAP 2 function is a great way to check if the BMU data is in sync with the actual battery capacity.

Note however, it cannot correct a mismatch.

kiev · Feb 12, 2025

To do the calibration, you drive the car down to under 2 fuel bars, then charge it fully without interruption.

For efficiency:
Your data was taken during the CV end of charging mode as the top cell was already at 4.1V. So some balancing at a max of 0.1A per active cell (burning off excess cell voltage thru resistance heating) was occurring to prevent exceeding the cell limit.

So the AC input was at only 6.3 Amps, which is about the lowest that the OBC operates. The DC output was 3.5 A, of which 2.84 went to the pack and 0.66 to the DCDC converter.

The OBC efficiency seems about 89% at these conditions; it might be +/- a few % under higher load conditions, but that is a fairly good efficiency for a variable-output PWM-switching boost inverter (AC in, DC out). i doubt you could find anything better for sale, on the internet or off the shelf, but if you do let us know.

ps 30 mV delta at the top end is huge (excessive), hopefully it will balance out during the calibration charge, but could be a sign of a weak cell.

MickeyS70 · Feb 12, 2025

kiev said:
The OBC efficiency seems about 89% at these conditions; it might be +/- a few % under higher load conditions, but that is a fairly good efficiency for a variable-output PWM-switching boost inverter (AC in, DC out). i doubt you could find anything better for sale, on the internet or off the shelf, but if you do let us know.

You’re correct, however, IIRC, charge efficiency is the ratio of the energy that ‘goes into’ at the battery vs ‘leaves’ the EVSE?

For the example above it would be less than 73% (1023W/1405W).

This can be improved a bit by setting the EVSE to a higher charge current (max 16A AC), but will increase the OBC’s internal temperature and therefore the risk of failure.

Negue · Feb 12, 2025

Internal temp is not a problem. Outside is 0°C

No more then 10°C from 20%-80%SoC

I connect to a 4.1kw EVSE but charging same as original Miev EVSE (16A)

MickeyS70 · Feb 12, 2025

Negue said:
Internal temp is not a problem. Outside is 0°C
No more then 10°C from 20%-80%SoC

What’s your average ambient temperature in summer?

Negue said:
I connect to a 4.1kw EVSE but charging same as original Miev EVSE (16A)

The only way to increase the charge rate above the 3.7kWh is to move away from AC and switch to DC, maybe when you’re done with the cell upgrade, you can start another project

:

It’s relatively simple to operate the CHAdeMON contactors externally; ‘all you need’ is a 360VDC source (e.g. a couple of DC power supplies in series). Set a current limit of 20A DC and you’re about doubling the charge rate while still being able to use a single phase domestic AC source to power the whole thing…

Negue · Feb 12, 2025

Agree

I supposed that talk about this days test where 0°C can easy suporte high AC charging.

I'm sure that BMS will limit charging when read high temp and will be no issues , only more time to finish charging.

I have a 20A EVSE and I was curious to see a improve of charging but sadly no. I hoped a 16A charging and not limited ~14.5A.

With Junsun EVSE used at Leaf 32A charger , if I set at EVSE 16A , charger use only 14.5A. If I set at 20A, use only 18.5A.

Seems chargers not use max Amps of EVSE can deliver.

I don't have experience to DIY a DC charge by myself.

kiev · Feb 12, 2025

Similar thread,
https://myimiev.com/threads/how-to-calculating-on-board-chargers-efficiency.5203/

MickeyS70 · Feb 12, 2025

kiev said:
Similar thread,
https://myimiev.com/threads/how-to-calculating-on-board-chargers-efficiency.5203/

Applying the same 'logic' the I-MIEV would have a charge efficiency of just under 70%?

This is close enough to the worst case scenario (OBC charging with 6A AC), but otherwise a good bit off the mark...

Negue · Feb 13, 2025

I say that at top charging session ( over 80%Soc) reading sensors, I suspect that efficiency drop near 100%SoC
Indeed I will confirm after I syncing discrepancy.

@kiev I will read it

njh · Feb 13, 2025

Not sure if this is useful or not, but I charge my Peugeot Ion using OpenEVSE, which sends data about the charge session to EmonCMS:

(This is the power that is leaving the house, so doesn't account for losses)

I also have SoC information being captured via WiCAN but I don't have that being graphed at the moment - it is something I should get setup. I am interested in what SoC it starts ramping down the charge rate at.

I believe there are two different SoC values - the actual one and the one that is displayed on the dial.

MickeyS70 · Feb 13, 2025

njh said:
I am interested in what SoC it starts ramping down the charge rate at.

Check the reported SoC just after 1 am but I gather the car is looking at cell voltage to start ramping down.
The graph looks like the SoC was high to start with, barely 30min before power starts reducing?

njh said:
I believe there are two different SoC values - the actual one and the one that is displayed on the dial.

Yes, the car uses a few different PIDs for SoC, therefore it would be more interesting to track the highest cell voltage…

What exactly is EV-temp 1?

kiev · Feb 13, 2025

njh said:
I charge my Peugeot Ion using OpenEVSE, which sends data about the charge session to EmonCMS:

Nice chart.
What temperature sensor is that--where is it located?

There is an interesting low frequency ripple in the power during the last 2 hours--little pulses about every 2 minutes. Maybe DCDC load adjust?

AC Charging Speed / Efficiency

Help Support Mitsubishi i-MiEV Forum:

Well-known member

Well-known member

Moderator

Well-known member

Well-known member

Moderator

Moderator

Well-known member

Moderator

Moderator

Moderator

Well-known member

Moderator

Well-known member

Moderator

Moderator

Well-known member

Member

Moderator

Moderator

Similar threads