Inconsistencies promised WLTP range

[Peter WA]

This is a snippet from the Australian ZS EV brochure, download from the website just now.

Can anyone spot the inconsistency?

Ok here we go: 18.6 × 2.63 = 48.918. That means, for the WLTP range to be realistic with the given WLTP efficiency, our battery would need to have a usable capacity of 49kWh. Not a stated capacity of 44.5kWh and an actual usable capacity of 40kWh.

Irrespective of how the actually achievable range is affected by speed, temperature, wind, driving style, this shows that we are fighting a losing battle.

The question I have: any plausible explanation? Or simply fraud with the stated WLTP range?

Perhaps that number should be 215 km, which would match the actual real world experience of so many of us. Maybe it's not that the WLTP test is an unrealistic lab test, never seen in real life. Maybe MG just fudged some numbers? If they did, I believe in Australia at least, given our strong consumer laws, they'd be in a lot of trouble.

Oh, btw: I am a little over halfway in a 4500km road trip that tests the vehicle to the max. In perfect conditions, moderate temperatures and no wind or tailwinds, no elevation changes, I only manage to get close to the WLTP (and GOM) range if I drop my speed to a low 85km/h. Battery appears healthy, minimal difference between lowest and highest cell voltage.

 

[JodyS21]

Ah, I was about to say that is indeed impossible. However do have the actual answer for you!

It's because of the inefficiencies in the charger, that figure will be the amount of power needed to supply the onboard AC charger in order to fully fill the 44.5kWh battery.

I looked into this previously, as I wasn't and still aren't convinced about the figures/the way they're realistically calculated.
The only source I could find published was this, I wanted to find the figures actually published by MG officially but couldn't:

MG ZS EV

All details and specs of the MG ZS EV (2019-2021). Compare price, lease, real-world range and consumption of every electric vehicle.

ev-database.uk

WLTP Ratings​

Range	163 mi
Rated Consumption	295 Wh/mi
Vehicle Consumption	260 Wh/mi

Rated = official figures as published by manufacturer. Rated consumption and fuel equivalency figures include charging losses.
Vehicle = calculated battery energy consumption used by the vehicle for propulsion and on-board systems.

163 X 260 = 42.380kWh
163 X 295 = 48.085kWh

 

[Peter WA]

It's because of the inefficiencies in the charger, that figure will be the amount of power needed to supply the onboard AC charger in order to fully fill the 44.5kWh battery.

I agree there are inefficiencies. The problem is: when I wrote of an actual usable capacity of 40kWh, that is the measured value of energy supplied by a charger.

I.e.i recently ran the battery down to a (as per car display when it started charging) 3% level. When the charging went through 78% charge level the charger had supplied quite exactly 30kWh. That puts the total at 40kWh supplied for 0 to 100%.

I had multiple of those charging sessions, all at different 50kW chargers, all with almost exactly the same calculated battery size.

If we were to take inefficiencies into account, that would equate to only 36kWh actual usable stored capacity. 20% short of the promised 44.5kWh.

 

[JodyS21]

I agree there are inefficiencies. The problem is: when I wrote of an actual usable capacity of 40kWh, that is the measured value of energy supplied by a charger.

I.e.i recently ran the battery down to a (as per car display when it started charging) 3% level. When the charging went through 78% charge level the charger had supplied quite exactly 30kWh. That puts the total at 40kWh supplied for 0 to 100%.

I had multiple of those charging sessions, all at different 50kW chargers, all with almost exactly the same calculated battery size.

If we were to take inefficiencies into account, that would equate to only 36kWh actual usable stored capacity. 20% short of the promised 44.5kWh.

Ah, you're talking about DC charging though not AC.
They will be reporting the amount of kWh DC provided to the car.

I presume the "official" figures are using the provided AC onboard charger - that's where the power losses are doing the transformation from AC to DC.

BTW, my opinion is very much they should not be allowed to advertise 44.5kWh battery if we're not allowed to us 44.5kWh of battery capacity. They should have to state - only XX kWh usable.

 

[JodyS21]

MG seem very intent on not wanting to tell us what the usable battery capacity is.

 

[Kithmo]

I believe VW quote the usable battery size on their ID3.

 

[Cocijo]

Ah, you're talking about DC charging though not AC.
They will be reporting the amount of kWh DC provided to the car.

I presume the "official" figures are using the provided AC onboard charger - that's where the power losses are doing the transformation from AC to DC.

BTW, my opinion is very much they should not be allowed to advertise 44.5kWh battery if we're not allowed to us 44.5kWh of battery capacity. They should have to state - only XX kWh usable.

I believe VW quote the usable battery size on their ID3.

I believe VW quote the usable battery size on their ID3.

I think that there is also an argument that even the usable battery capacity is not actual real-world usable.
What I mean by that is for example;
If the top and bottom bms controlled buffers in a battery pack leave remaining say 40kwh ‘usable’ - there won’t be many people who will be comfortable using this full capacity - this means charging to bms full and running to bms empty.
What you really have is probably nearer 36kwh if you add in that you wouldn’t really want to go much below having about say 12miles of range left before charging.
(In the old world ICE car it might read as ‘mileage to reserve tank’. The big difference of course being you could also carry a can of fuel for emergencies if you ran out)!
Most trip planners have the functionality to build in a % of battery capacity left before predicted charge stop - but bear in mind that this may not be the bms allowed capacity.
Based on your approach to the risk of running flat - taking both bms controlled capacity and risk buffer into account this is really the real-world usable capacity.
So in reality there are 2 main constraints that need to be considered;
1/The ‘usable’ capacity based on how the manufacture configures their bms
2/The amount of that usable capacity that it is reasonable to realistically use without running a major risk of being stranded.
We start with A/actual battery capacity B/bms controlled battery capacity C/risk modified real-world usable capacity without running to completely flat
Using the ZS as an example if you achieve 3mpk (which is not untypical in cold weather or faster speeds etc) based on my very rough numbers - you would really have a risk rated real world range of 111miles.
I’m unsure if any official mileage tests even come close to taking this into consideration or ever will. By my estimates you can multiply most EV‘s total battery capacity by between 75-80% to get the useable low-risk real world capacity and then use that as the basis to calculate realistic ranges based on mpk.

 

[Jake]

Agree Cocijo. In a nutshell what you are indicating is on a multi-stop extended journey apart from the first and last legs of that journey your realistic range is 60% (80-20%) of usable capacity between charges. So for the sake of argument nigh on 100 miles at 4miles/kWh. That's what I would aim for stopping very 1.5 to 2 hours for a charge and comfort break.

 

[Mike]

There are a few unknowns. Even though there is a percentage when charging that the car reports, is it really? Is is linear?

I’ve tried in the past to work this out and never been successful. It also doesn’t seem to take into account regenerative braking on the in car display.

I do know that I have put in the car over 40kWh in the past, which I know will be less in the battery.

 

[JodyS21]

I read up a while ago about the WLTP testing routine and, it said something about the way it is calculated (for EVs) is using the power used to charge the car and not the power that is actually stored in the car.

So charge the car to 100%
Perform the test of X miles
Measure how much power is needed to charge it back up to 100% - not how much power is actually stored in the battery.
That must give you the 295Wh/mil figure.

They then must do a 0% to 100% charge to find out the amount of power it takes.
Multiply the 2 together to get the 163 miles range....


I believe this is why the "official" WLTP rating is:

MG ZS EV

All details and specs of the MG ZS EV (2019-2021). Compare price, lease, real-world range and consumption of every electric vehicle.

ev-database.uk

Rated Consumption 295 Wh/mi
Range 163mi
163 X 295 = 48.085kWh is the amount of power in AC they're saying it needs to charge the battery from 0% to 100%.

I wish MG would officially publish the WLTP testing calculations used, I cannot find them anywhere.

BTW
This would a charging efficiency of (44.5/48.085) 92.5%, or to put it the other way (48.085/44.5) 108%

 

[JodyS21]

There are a few unknowns. Even though there is a percentage when charging that the car reports, is it really? Is is linear?

I’ve tried in the past to work this out and never been successful. It also doesn’t seem to take into account regenerative braking on the in car display.

I do know that I have put in the car over 40kWh in the past, which I know will be less in the battery.

Regen braking def makes it all more complicated to work out!

 

[Peter WA]

Ah, you're talking about DC charging though not AC.
They will be reporting the amount of kWh DC provided to the car.

The biggest part of the losses is from putting energy into the battery to getting it out. That's why the battery needs cooling, not only the charger. The AC/DC conversion efficiency should be about 97%. If the overall efficiency is 90%, then 7% would be lost to the battery, in both AC and DC charging modes.

 

[Cocijo]

There are a few unknowns. Even though there is a percentage when charging that the car reports, is it really? Is is linear?

I’ve tried in the past to work this out and never been successful. It also doesn’t seem to take into account regenerative braking on the in car display.

I do know that I have put in the car over 40kWh in the past, which I know will be less in the battery.

How’s your leg? On the mend?

 

[JodyS21]

I've done lots more reading!
It's very hard to find out decent info on the WLTP test procedure.
I've found some very complicated documentation, but it's really scientifically complicated!!!!
Shame there's not an official more detailed document with in it layman's terms.

E.G

https://unece.org/fileadmin/DAM/trans/main/wp29/wp29r-1998agr-rules/ECE-TRANS-180a15e.pdf

I've found confirmation though that the power measurements/recharge is using Mains AC, so that means it must use the onboard provided AC/DC charger. Therefore using a DC charge the power readings wouldn't necessarily tally with the WLTP results.

From googling, it sounds like AC/DC converters are surprisingly usually very inefficient! 90% or worse isn't that unusual, new tech is probably helping.
- They keep this a bit quiet from consumers down they lol.

Electric Vehicle Charger Hits New Efficiency Level

Kettering University researchers have partnered with HELLA, an automotive electronics company, to develop a compact vehicle charger that makes a big step forward in efficiency...

www.electronicdesign.com

Calculating on-board chargers efficiency

In this article I’ll show you how easy it’s to estimate the on-board charger efficiency of an electric car by using WLTP ratings. I’ll use WLTP ratings in this article, but you can also use EPA or NEDC ratings, since they also measure plug-to-wheels consumption, this means that they include...

pushevs.com

 

[Peter WA]

From googling, it sounds like AC/DC converters are surprisingly usually very inefficient! 90% or worse isn't that unusual, new tech is probably helping.
- They keep this a bit quiet from consumers down they lol.

Yes, it's a waste of energy, but in a way comparable to how ICE engines only convert say 30% of the energy from fuel into motion. Another way EVs are a lot better, lower losses. Of course if electricity is made from fossil fuels, a similar waste occurs on the way from fossil to electricity.

I'll reply in more detail when I get home. An 2016 article I found was about a (then) new way of having only 3% conversion losses AC to DC, as opposed to a more typical higher value, they quoted 6% I believe. A lot lower than 10%, which would include the charging losses too. Those 3% are in line with what modern solar inverters achieve in the other direction, from DC to AC.

As to actual kWh charged: my OpenEVSE does document those for each charging session. The problem I have here on the road is that only the current is measured, the voltage is taken to be standard grid voltage. (Which is weird, since voltage is so much easier to measure than current)

Crossing the Nullarbor ymuch of that electricity comes from diesel generators. Putting strain on only one phase of the 3-phase supply will see voltage drops. If I assume that I charge on 230V after those drops rather than the 240V the grid is supposed to have, then the actual usable capacity using an AC charger, before losses, comes to about 44kWh.

Even if those generators were able to supply a steady 240V on one heavily loaded phase, that would only be 46kWh. Both figures well short of the 49 we should see on AC to charge a 44.5kWh DC rated battery.

I will have accurate voltage readings and therefore accurate kWh once I can charge back at my own home (more accurate at least, some losses still occur in the wires between my solar inverter and the car charger). If I was a betting man I'd put money on finding the same result.

 

[Gomev]

Did anyone ever get the advertised mpg from an ICE car ? Especially VWs, so hardly surprising we don't get advertised range.

 

[JodyS21]

Did anyone ever get the advertised mpg from an ICE car ? Especially VWs, so hardly surprising we don't get advertised range.

I wonder if ICE cars mpgs are more realistic now with the new WLTP test?

 

[Jake]

Did anyone ever get the advertised mpg from an ICE car ? Especially VWs, so hardly surprising we don't get advertised range.

Back in 2001 the original Honda Insight was rated at averaging 80mpg (I think). After I had got used to its characteristics I once managed to achieve 99mpg on a trip from Brighton to Bognor Regis and back along the relatively flat coast road. It was really hard work and at times there was a build of cursing traffic behind me (sorry if any of you were affected).

After achieving that figure I went back to more realistic driving and over a 10 year period and 60k miles I think the lifetime average was somewhere between 70 and 80mpg. I'll see if I can fish out my logging details (sad I know).

 

[Peter WA]

Did anyone ever get the advertised mpg from an ICE car ? Especially VWs, so hardly surprising we don't get advertised range.

Yes, I did. I had a hired VW Golf Diesel for a major tour of Europe in 2011. On days when I moderated the way I drove I got very close to the 4.1l or so per 100km that were promised.

There is a difference though between something highly variable like how many miles you personally get and something objectively verifiable like a WLTP figure.

If they calculate WLTP range on 44.5KWh battery capacity, but only give us 40kWH battery capacity, then that range is overstated by 10%.

Your individual driving style may further reduce the range you get, but that's not the point of the post I started

 

[JodyS21]

Yes, I did. I had a hired VW Golf Diesel for a major tour of Europe in 2011. On days when I moderated the way I drove I got very close to the 4.1l or so per 100km that were promised.

There is a difference though between something highly variable like how many miles you personally get and something objectively verifiable like a WLTP figure.

If they calculate WLTP range on 44.5KWh battery capacity, but only give us 40kWH battery capacity, then that range is overstated by 10%.

Your individual driving style may further reduce the range you get, but that's not the point of the post I started

You have said EXACTLY what my "concern" is.
I want to see official WLTP testing figures based on what is our definite USABLE battery capacity is.

If they could actually achieve the full 163miles on a WLTP test (not pro-rata) then so be it and that's that. I just honestly have my doubts, I wish I didn't but I do!
E.g. we loose power at low SoC - I'm not sure at that point the car would be deemed to be usable/passable on the WLTP test...