Quantifying Charging Power and Onboard Charger Efficiency

Mitsubishi i-MiEV Forum

Help Support Mitsubishi i-MiEV Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Carsten

Active member
Joined
Sep 16, 2015
Messages
39
Location
USA, Vermont
I have my EVSE at 12A and charge from 240V. When I look at the CanIon, it never goes above 2.3kW & 6.7A. Don't we have a 3.3kW-charger on board?
I have a photo but don't know, how to insert it.
Regards,
 
Yes, we have a 3.3 Kw charger, but you can't get all of that with the Upgraded EVSE

Are you sure you have it set to 12 amps? Your 2.3 Kw reading sounds like maybe it's actually set at 10 amps??

Don
 
There are losses in the charger and in ancillary loads related to charging the car. Among other things, there are contactors, a coolant pump, and the DC-DC converter all pulling power before it reaches the battery. At full speed, you will probably only see 2.6 kW going into the battery on CaniOn. With EVSEUpgrade, I see 2.3 kW going into the pack.
 
Does this mean, that with a higher output EVSE I could get closer to the 3.3kW?
I have a dedicated line, that is good for up to 40A.
Would it be worth the investment?
Regards,
 
I've never seen (using CaniOn) more than -2.5 kW or -7 A while charging. I have a 230V 16A cable. At full blast it draws 14-15 A from the 230 V wall socket, which corresponds to 3.3 kW input power. The 2.5 kW is the most I've got into the pack from that 3.3 kW input, so the whole process is around 75% efficient. You can't get more than that using the built-in charger with any EVSE cable or station, except for CHAdeMO quick charge of course.

EVTEC advertises a portable CHAdeMO, which you could use to charge faster, but it's 16.000-18.000€ plus taxes:

http://www.evtec.ch/en/products/move_and_charge/
 
I've seen 2.5Kwh and 7.1 amps on canion using a juicebox EVSE which has huge wires and contactors in it which yeild virtually no voltage drop across the unit at the amp rates that the "i" will draw.

Also TED (which is very accurate +- 2%) has never shown more than 3.1 Kwh going to the car from the electrical panel so I am confused by the 3.3 KWh spec as well.

Also the canion input to the battery numbers don't add up. Think about it, if you have 10% SOC and you charge at L2 20+ amps available. Canion will never show more than 2.5Kwh charge rate. 90% of (lets be conservative 14.5 Kwh battery pack) is 13 KWh needed to get to full. 13KWh / 2.5KW/h = 5.2 hours. Plus you need to add at least another hour to compensate for the period when the charge rate tapers and balancing is done (which wastes some power). Plus the cells warm while charging which is due to less than 100% charge efficiency at the cells. So you really need to put more than 13 kwh back in to get full. So conservatively, it should take at least (with nonexistant 100% cell efficiency and no power dumped to balancing right) a minimum of 6 hours and 15 minutes to recharge. But it doesn't! It's about an hour off actual experience. Anyone had their "i" take more than 6 hours to recharge on full capacity L2??

So the canion input readings don't add up and seem to be up to 20% low.

I looked at the charger manufactures site and they claim 90+ % efficiency for the charger used in the "i". So the ancillary loads come off that spec and we don't know how big they are. But the efficiency can be calculated from the other end and it is quite a bit better than the ~75% that Canion is showing us.

Aerowhatt
 
I agree, the 2.5 kW shown by CaniOn just doesn't add up and 75% efficiency would be really bad. There must be an issue there. Should put a clamp meter on the battery wires at some point and see how many amps that reads.
 
By the way, when I did a full discharge-recharge test on my pack a couple of weeks ago, I got 13,117 Wh out, but fully charging the battery only yielded 11,250 Wh into the pack - further pointing that CaniOn doesn't show correct input kW when charging:

https://www.dropbox.com/s/no7fcpcfhuqvory/2015-08-31%2022.56.26.jpg?dl=0

If we take 11250 and divide by 2500 (as shown by Canion), we get 4,5. 13117 divided by 4,5 results 2915, which is probably close to the actual recharge watts using the integraetd charger. So instead of -2.5 kW the reality is closer to -2.9 kW and Canion shows about 14% too little.

Of course it may not be Canion's fault that the number is incorrect. It might be the vehicle. I can never get the input and output Ah to completely match on my DIY EV either. Making even a shunt based meter count big and small currents accurately at the same time is not easy.
 
Carsten said:
I have my EVSE at 12A and charge from 240V. When I look at the CanIon, it never goes above 2.3kW & 6.7A. Don't we have a 3.3kW-charger on board?
Aerowhatt said:
Also the canion input to the battery numbers don't add up.
The DC amp reading we get from Canion is about 1.0 amps low.

The on board charger takes alternating current (AC) from the EVSE and converts it to direct current (DC) going to the traction battery.
viEIEUZl.png


Heres's the AC from the EVSE.
(244.85 volts)x(13.31 amps)=3,259 volt-amps or about 3,250 watts taking in the power factor.
cwEJGVpl.png


Here's the DC going to the traction battery.
(352 volts)x(8.12 amps)=2,858 watts.
l0PQN9pm.png


Here's the DC when the cooling pump is running.
About 0.2 amps less.
ahYe8ySm.png


The charger efficiency is (DC power to traction battery)/(AC power from EVSE)x100 = (2,858 watts)/(3,250 watts)x100 = 88%
 
jsantala said:
Thanks, RObertC, that's just about as comprehensive as it gets. :)

I'll second that sentiment!

But, I couldn't get my TED data to jive with RObertC's comprehensive testing. So in the interest of peer review and my inability to stop thinking about something until it makes sense, or at least some sense.

I recreated his measurements and found almost identical numbers on the DC side. I also measured and added in the low voltage output on the charger. Even though it's relatively insignificant compared to the rest of the system.

On the DC to the pack I measured 8.14amps at 352 volts for 2865 watts out Then the low voltage DC is putting out 2.86 amps at 14.4 volts for an additional 41 watts. A total of 2906 watts output. Very close to to the previous analysis indeed. That's good!!

Now on the AC side of things is where the discrepancies are that had my TED data calling things into question. At 241 volts AC the charger was pulling 12.69 amps or a total of 3060 watts. This fit my TED data at the time which was 3058 watts going to the EVSE. So that's 3.06KW not the 3.3 KWH claimed for the charger or the almost 3.3 measured by RObertC. My observations yield a charger efficiency of 2906/3085 = ~94%

I think we are both right and the charger efficiency has been improved between the 2012 and 2014 model years. It's the same manufacturer (Nichicon) so nice job by them!

I think his data is accurate and I know mine is. I guess you can't complain if the improved charger doesn't draw 3.3KW when it's output is identical to it's ancestor which did ;) . I also learned from looking at the low voltage output of the charger that the coolant pump draws 67 watts when it is running.

Aerowhatt
 
Hi. I moved the relevant posts over to this new thread. Unsure as to which subforum to put this into...
Hope y'all are ok with this.
Aerowhatt said:
..I think we are both right and the charger efficiency has been improved between the 2012 and 2014 model years. It's the same manufacturer (Nichicon) so nice job by them!...
Wow, I would be surprised (pleasantly) by that. Anyone have access to an illustrated parts breakdown with part numbers to see if this has changed?
 
JoeS said:
Wow, I would be surprised (pleasantly) by that. Anyone have access to an illustrated parts breakdown with part numbers to see if this has changed?

If one adds in the low voltage DC output (making the assumption that it is similar to what I measured) for him (RObertC) we get 2858w +41w = 2899w DC output with 3250w AC input. So the 2012 charger efficiency becomes 2899w/3250w x 100 = 89.2% efficiency in RobertC' 2012 ,compared to 94% for the 2014.

I agree 94% is surprisingly high but well within what is achievable. Also since my DC measurements closely match RObertC's they are not very suspect. My AC measurements come from two independent sources TED reporting 3058 watts at the distribution panel. And a good clamp meter placed in the EVSE measuring AC amps combined with an AC voltage reading from the same point, showing 3060 watts. Should be reliable considering the methods and proximity of the independent values to each other.

The part numbers are likely the same. With the evolution in efficiency documented in the serial numbers instead.

I would love to hear from someone else with TED on their instantaneous watts out of the distribution panel to their 240VAC EVSE during the middle of a charge cycle.

Aerowhatt
 
Aerowhatt said:
If one adds in the low voltage DC output (making the assumption that it is similar to what I measured) for him (RObertC) we get 2858w +41w = 2899w DC output with 3250w AC input. So the 2012 charger efficiency becomes 2899w/3250w x 100 = 89.2% efficiency in RobertC' 2012 ,compared to 94% for the 2014.
If my math is right, for the 57.1 miles I drove last weekend and recharged, my wall-to-wheels efficiency would increase from 3.9 mi./kWh to 4.14 mi./kWh based on the efficiency difference between these two chargers. For the entire trip of 150.9 miles, efficiency would increase from 4.33 mi./kWh to 4.6 mi./kWh, a consumption decrease from 34.85 kWh to 32.81 kWh. That extra 2.04 kWh would provide an additional 9.4 miles of travel in the 2014 on the same energy as the 2012.
 
Y'know, I've been musing about this possible difference in efficiency between the 2012 and 2014 models - not from the mathematical perspective but from the regulatory and marketing and fiscal viewpoint. A couple of random thoughts -

If efficiency had been improved, wouldn't marketing have jumped on this and touted it?

Any efficiency change would have to be reported to the EPA, as their Monroney labels showing vehicle efficiency (wall to wheels) would now have to be changed. I wonder if there's a clause in the regulations which says that if you make an improvement you don't necessarily need to jump through hoops - or perhaps there's some percentage minimum which makes this unnecessary? Conversely, if a manufacturer does something that makes things worse, then they must be obliged to report it … and re-test? In any case, the paperwork alone (not to mention possible re-testing) could possibly shoot down the authorizing of such an improvement in the US by Mitsu management for a limited-production vehicle such as the i-MiEV. Sad, really, when you think about it - but I'm merely speculating.

Next time I'm at a Mitsu dealer I'll stop by the parts department and see what they show for a part number for the charger… maybe there's a hidden option for a 6.6kW unit lurking in their catalog? :evil:
 
JoeS said:
If efficiency had been improved, wouldn't marketing have jumped on this and touted it.

Next time I'm at a Mitsu dealer I'll stop by the parts department and see what they show for a part number for the charger… maybe there's a hidden option for a 6.6kW unit lurking in their catalog? :evil:

Mitsu marketing, that's genuinely funny :D

Aerowhatt
 
Maybe just a natural progression in power electronic performance? There was over two years span between the 2012 and 2014 models being built. A lot of stuff changed in that time. It could've been overlooked or decided to not be worth the effort to tout and re-test, sort of like the reported better ride of the 2014 model over the 2012 that Mitsubishi didn't advertise as "improved suspension" or similar.

Then again, we've only compared two chargers.
 
Back
Top