Very interesting data! Thanks! When comparing cars with other cars for a ranking it is important though to keep in mind that temperature and weather have a significant influence on consumption. A difference of 27 degrees Celsius results in a difference of 10% in wind draft, which is the main factor for consumption at highway speeds.
Thank you Chris. Please continue these tests. Could you please do some tests for losses at a home charger? Also if you get yourself a test-partner to drive 2 electric vehicles simultaneously to compare them in the exact conditions and see real world consumption, that would be great!
I really like the commentators who argue, that these numbers are useless because they’re not from a standardized test procedure. The same crowd argues that the WLTP numbers are useless because they not representing the real World. All the “if this” or “why not that” are meaningless. If you want to do your own real world test, do it and share the data. Every dataset helps. I think these numbers are one more data point that helps to understand the consumption of these particular models. No more no less. But if you have nothing to contribute just go with the data you get provided by this channel an all the other sources out there.
You are right. But the main issue with this data is that the charge limit is not strict enough. It is 50% or roughly 50%. Even if we think about a standard deviation, which 50 means between 49.5% and 50.5%, with a battery of 56kWH capacity, you may add or retract roughly 0.5kWh (1%), so you have a possible delta between 2 cars of 1kWh per test. At the end, there is no proven difference between the 5 first ranked (more or less). Which, on my point of view, is cool, because you can chose the car you like, they are (almost) all efficient. ANd that makes the numbers interesting to look at. Actual vs measured and power loss during charging are very interesting numbers, for sure. They shows how inefficiently some batteries are cooled.
@@francoisjeanmougin4405 You making my point exactly. You probably never did a standardized Test of EV consumption, according to the numbers you are stating. It really doesn't matter because the only way to get more reliable data is to have more numbers you than can use for a median score. As with all other test (also ICE) you always end up with a discrepancy of a few percent. the only way to even that out is more data (with a different discrepancy). Even in a Lab-Setting your result will spread over 1%. So what ever people like you want from one test, you only get through aggregated data of many test. So thank you for making this point so obvious.
Chris, considering some cars, if not all cars, heat the battery if temperature is too low, would doing the 130kph range test first then the 110kph be an idea? This should possibly heat the battery at higher speed, enough to eliminate some of the charging losses due to heating the battery.
This makes sense. To get the real fuel consumption, in litres or electrons, you have to take the measured total energy input versus kms travelled to get real energy consumption. On my fossil car, the fuel consumption readout is very inaccurate, consistently claiming better consumption than actual calculated consumption. It's worse if you zero the readout at each fill. Not surprising that the manufacturers are still being less than honest about their consumption figures and configuring the consumption readout to be optimistically inaccurate. Thanks, Chris, for pointing out the truth.
Even allowing for the higher real world consumption of the Tesla Model Y, it is notable that all the cars with lower consumption are significantly smaller than it. So, it remains very efficient for its size.
True, but even the iD4 has really good numbers for its size. Very close to the MY. Also the Mini cooper SE the eGolf and Hyundai ioniq show that a ICE platform can be an efficient BEV.
Interesting results Chris... I just checked my Tesla model 3 SR+ and I get 15.3% difference between used kWh / kWh added using a 7kwh charger. I believe Tesla doesn't count the cars computer consumption in its Wh/km numbers although it dose seem to count the HVAC consumption...
I doubt the computer uses that much power. However, knowing where that energy goes would be very interesting...and also knowing whether this actually points at higher consumption than reported or actual charging loss. I think it is more likely charging loss, or rather the BMS eating up KWh to heat the battery and stuff like that...
Our Tesla model 3 LR shows differences of 15-30!! % between what the car is showing and what Wallbox is measuring. The lower the consumption the higher is the difference - independent to outside temperature. So, I also think the computer or whatever is very hungry.... (... hardly never wakes up the Car via App. )
I have also noticed quite high charge losses on my Ioniq. Then I slow charge with my own wall box and use the onboard charger. Will raise a question in the Hyndai group on Facebook if more people have the same experience.
Ioniq will pre heat the battery if you come to the charger with the battery temperature lower than 21 C. This will reise the, so called charge loss, if you come to charger with lower battery temperature.
It seems right and reasonable. A conclusion would be that we need to add the charging data with battery working temperature. Now we don't know if the battery needs to heat up or cool down.
Cold battery don t perform well, hot battery degrade more, that s how lixx batteries works. According to the Hyundai thermal management, golden area is from 20-40 deg.C Under that temperature thermal management wil be heating up the battery, over that temperature, battery will be cooled down. But battery will be heated only while charging with more than 3,5kW.
Interesting numbers you capture here Chris. I also monitor this aspect, I keep a lot of data with my EV (a Model 3 LR) and I see “losses” between 5% and 15% when I compare the numbers reported as SuperCharging energy (on my tesla account) vs the energy that goes into the battery measured with Scan My Tesla tool during the sessions. Having said that, I noticed few aspects you might want to consider in your tests/findings: - Most of my measurements were performed during summer while I was sitting inside the car with AC ON/gaming so on. That adds easily 1-2kWh to the session depending on the length and climate outside which could mean a lot considering you only charge small amount of energy in your tests; - On short sessions (5% to 50%) but V3 (max 250kW), the losses are close to 12%-15% while similar sessions on V2 (max 150kW), the losses are around 7-8% “only”; - On long sessions (10% to 90%) on V3, losses are up to 10% while for the same session on V2 the losses are around 5-6%; - Two comparable sessions on V3 (14,4% to 100%) at 18,5 Celsius outside (I also did not stay as much inside the car during this session) -> 6,7% losses while at the same charger (8,1% to 78,1%) at 25 Celsius outside -> 10% losses. My point is that there are many aspects impacting this like: 1. Were you consistent with AC ON/OFF during the recorded sessions? 2. Outside temperature play a major role in this; 3. How did you use the car before charging (was the battery warm/cold before charging); 4. Some of the tested cars are optimizing charging sessions for fast charging by pre-heating the battery (during charging) or excessive cooling (instead of limiting the charge speed); 5. Do not take this personally as I think the point you are making is good but, only 60Km of driving/charging is way too short for the data to be trustworthy. If you would charge the cars on AC (which is what most of us do on day to day use), the table will look completely different but yes I know, you also have a life ;) Cheers, Florin.
@@Psi-Storm everything matters when the test period is so short. AC on/off was not just about the period while charging but also about the drive itself.
Our Tesla model 3 LR shows differences of 15-30!! % between what the car is showing and what Wallbox is measuring. The lower the consumption the higher is the difference - independent to outside temperature. So, I also think the computer or whatever is very hungry.... (... hardly never wakes up the Car via App. )
Very interesting results. I would have thought Ioniq would have won outright. Chris if Robert de Niro is like my Kona then you cant get the maximum charge rate until the battery heats up to 25C maybe losing kw's there, on flip side if battery gets to about 38C then it starts cooling. At a 50kw charger battery has to be 15C or higher to get max charge.
Wow, that was a lot of work done, we don't see such an effort frequently nowadays! :) Charging losses bias may come from BMS reporting. It is quite a task to determine the state of charge which is based on "out" voltage from the battery. It depends on several parameters such as current and temperature. Ideally repeatable SoC measurement needs batteries to idle and be at the uniform and equal T. What BMS has at highspeed charging is only instant voltage, current and Tmean/Tmax (with T distribution within +-5C over the battery). So when you see 50% SoCc, it is in fact the interpretation of several parameters with a bunch of mathematics simplified with assumptions. Once charge is stopped and battery temp equalizes, actual SoC might be 48% or 52%, the uncertainty range of +-1.2kwh for Kia eNiro for example, for which whole charging session was around 11 kwh (so the bias would be 10%). Still the answer is good- statistically charging losses are within 5-15%, what should be counted for better understanding of real consumption (plus it seems like tesla doesn't report non-driving consumption).
Good job. A lot of time I check difference between consumption and charge and I'm surprised how very big difference is between both score. For example yesterday i had quick charge from 33% (SOC) to 43% in my eNiro. On the display (ABB 40kW, 100A on Lidl - free) showed me 10,2 kWh. This is like my battery has more than 100 kWh capacity. It is good that you check real consumption. I see that many charge machines inflate the result (10-20% and more).
Interesting.. I am following my consumption as well. Hyundai Kona. For those 3000 km I drove in 2 months (let's call it summer :) ) Kona show's me an average consumption - 14.7 KWH / 100 Km. Data what I get from electrical company - gives me 16.35 KWH/ 100 km. 11 % more. But still - I am happy with the result.
Critical thinking will not be appreciated, massive effort Chris and thank you kindly for the access to your capacity to gain an accurate and deep understanding for us.
its appreciated, but the problem is that 1 time results are anecdotal, you need last least 3 runs with each car, plus one 100-0% to get max net discharge energy available. The 0,42 on the i3s is clearly some kind of measurement error. This result cant be true... because physics ;)
Thank you, very interesting :) It would be great if you could update these spreedsheets after each new car you test. I am curious about the ID3 Pure. :)
Very interesting data points. There's a lot of variability potentially at play, but ignoring those... I wonder have the Engineers in some of the cars put in assumptions into charging, Like Huyundai/Kia Engineers assuming someone at a fast charger is going to charge to >50% If Huyundai/Kia have coded in the assumption that you are going to charge to 90%+, and the charger is supplying more power than the batteries can peak charge - why not focus on maximum battery heating now to get an optimal charge time. It might explain the stepped charge profile of those cars - big drop-offs in current pulled as the aggressive battery heating finishes. Doesn't change the case for driver of those cars though. Maybe riding the maximum charge curve as the charger displays is actually a huge waste for those.
It would be interesting to relate adjusted consumption data to calculated net battery capacity. If we get closer to net capacity stated by manufacturer, it could be just car display showing incorrect consumption. However, other causes are possible, like charger showing more energy then actually consumed, or an actual energy loss on charging for example to run air condition in car while charging (though 20% on eNiro seems way too much).
Nice stats. In Germany that would make the price for a trip in an iD3 about the same as for a diesel golf. If you use 30ct kW. Those with solarpanels..... much less.
Nice info Chris. Really good of you to go to all of the effort to collate these data. I guess a few of the comparisons are a bit tricky with the huge variations in weather conditions. The eGolf for example looks amazingly good considering it is a legacy car converted to electric, but do you think this is just because it was so warm?
I know this wasn't the point of this Video...but seeing how close the consumption is between a massive Model Y and and the eUP is really fascinating...
That's the problem with Tesla. When you navigate to a HPC which is no SuC, than the battery heater is running during charging, which is not shown in the display. When you navigate to a SuC than the preheating is calculated within the consumption shown in the display. But nevertheless for 450 HP car it is still amazing low.
At colder temperatures, the thermal energy in the battery is also used very efficiently to warm the interior. Not helpful in summer, of course, but in the other seasons you also have a heat store, which is definitely an advantage.
In some respects I feel this fact must be a bit disappointing for VW. They have gone to all the effort to develop the MEB platform but compared with the eGolf, the ID3 does not feel like you are taking a leap into the future. I currently drive an eGolf (love it) and had a 5 day test drive of the ID3 as I assumed it would be my next car when the eGolf lease is up. However the ID3 left me feeling underwhelmed. As I was driving it I just kept feeling that it was a shame that it wasn’t better.
Chris, If I understood correctly you were using 50% state of charge as a reference point for each charging session. Is that percentage coming from EV's BMS? In that case what EV uses full charge of battery (100%) and what is the zero percentage? Car software can be build in that way it estimate those full and zero values using battery current conditions like battery temperature etc. Does it make sense if you stop charging in those tests using charging voltage and then charging to the same voltage after driving? I checked your eNiro test but I couldn't see (you just edited in that point) if the battery voltage is always the same when the SoC is 50%.
I drive a BMW 5 series Touring D. I average 7,2 l/100km combined, at a steady 130km/h that would be under 7 l/100km. Let's take a price of Eur 1.50/l. 100km costs me under 11Eur to drive 100 km. Ionity charges 0,79Eur /kWh. So if I look at a car with reasonable space such as the Hyundai Ioniq, at 130km/h that would use 22.8 kWh/100km which would cost 18Eur. Please correct me if I'm wrong, in case I misunderstood the figures. If the average EV costs more to drive than a large BMW Touring, where is the incentive to change? 11Eur vs 18Eur. I won't even mention range and time lost charging on longer trips. Edit: I just noticed in some countries Ionity charges per minute so then it depends how fast your EV can charge and not battery consumption that will determine the cost and you cannot control that charging rate by sensible driving. I am not anti EV, I am just thinking of the total cost to the driver which is often what matters most to many.
@@BatteryLife thanks, 30 cent sounds a lot more reasonable. I know there are options such as charging at home but many don't have that possibility, especially in city/urban areas. The cost of charging is often the factor that is missing in many EV reviews or it is assumed everyone can charge at home. Perhaps a review of average costs with /without plans or between various providers would be beneficial to many?
you have to take how complex it is.. using only 1 or 2 charges will not give a really accurate number. the car should be accurate but it does not show figure of what come to the battery and what is displayed by the charger unit since this figure is were the error and accuracy comes from if a car displays battery took in xx kWh and the machine displayed xx kWh provided then you could see what the difference was. but since it does not then accuracy will always remain off by a variable amount of power.
My diesel van (sorry, waiting for the ID.Buzz) reports 45mpg on the trip computer, but that never matches up to the miles and amount of Dinojuice that it actually drinks.
Interesting! If you charge slow at home will de differences between cars be smaller then? About low consumption I’m not sure but I think a velomobile averaging 50km/h can do it with about 5Wh/km. Maybe I calculated wrong but I think it should be correct.
Depends on how slow you charge. The battery management system also takes a bunch of power, with low charging speed it has to run much longer. If you charge slowly with home solar, the car uses 20% additional energy than charging with a 11kw home charger. But with solar the energy is mostly free, so it's not that deciding.
Very interesting data - thank you for posting this! About the Tesla Model Y - I think that this can be the only car in this list that heats up the battery to high temperatures while starting DC charging and during this time it can consume around 7-8kW of power. Usually this heating takes around 10 minutes, to this could consume around 1,3 kWh and this would explain the additional losses in this case. If the car would be charged with AC, this would not happen and the consumption would be lower. Probably this is why this energy is not calculated into the consumption measurement as in 90% of cases it does not occur and it would require the gauge to change when charging. You were charging on Ionity, so Tesla does not preheat the battery during the drive, but only after connecting the charger. When navigating to the Supercharger the car preheats the battery while driving and then the additional consumption is taken into account and calculated into the shown consumption. It's tricky, you know - what should they do with the energy used to heat up the battery on the charger - add it to the last trip and show higher consumption? What if the last trip was only 1km so the consumption would skyrocket to 10000kWh/km? This is an edge case that has no good and easy solution.
Also interested in this. ID4/enyaq is too big for my parents garage, I told them to get an eniro. Should I tell them to get something else? The ID3 is too small for them and they do not want an VW due to the dieselgate that they had to pay for (worth loss but no money from VW in austria).
There is no "real" consumption. It depends on what you want to determine. When you are on long trips you only care about battery to wheel consumption as this determines when to get to the next charger. When it comes to running cost then off course you have to include charging "losses". But in the "real" life of most EV drivers the range on single charge is more than enough to do "slow" charging at a wall-box. Under most conditions most EVs neither actively heat nor actively cool the battery when charged on AC. That's why it is the most cost-effective way of charging. So when talking about the "real" consumption, in a way that the BEVs are actually used, a wall-box charge test would be of much more importance for most. Charging at HPC speeds is not the norm for most, according to statistics. I'm the exception of the norm and have to charge at public DC fast chargers as I have no AC charger in reasonable range.
But not when it's cold. Far from it. I can squeeze out around 330 km in Summer, but right now it's closer to 220 km and in Winter it will be around 180 km. Heating requires a lot of energy, especially with a relatively small battery.
Interesting information but not surprising that cars consumption numbers are different than reality. Of course they want it to look better than it is. Some manufacturers are just better at maximizing the numbers than others. It is the same with fossil cars, numbers displayed are always better than reality and vary between manufacturers.
Thanks for doing this but I think your data may be flawed in several ways. 1. You need to charge to 100% to get a top balance of the pack so the BMS is more accurate. 2. By only charging to 50% your calculations become error prone due to rounding errors in distance and energy- I see this with my e-Golf capacity calculations. I only get consistent calculations when I charge pack to at least 90% and consume at least 80% of pack capacity. You are missing these two important constraints in your testing. 3. Do not uses DCFC as this introduces a lot of charging losses that do not exist if you use L2. L2 charging is most efficient and consistent because most cars will not need to run heating or cooling for charging, power (heat) loss due to charging varies with the square of the current. The on board AC charger is optimized for efficiency whereas DCFC is not. I believe these reasons are why some of your data makes no sense. Please try my methods and let’s see your results. Thanks!
When you look at your data and say”something looks strange”, that means something is wrong with the testing procedure. While DCFC may be relevant for some people, it introduces a lot of variables that you do not have with L2. Top balancing should enable BMS to be more accurate so you can trust BMS consumption data.
@@abraxastulammo9940 Probably due to running cabin heating and electronics that are not included in displayed consumption (trip meter) while charging (and also otherwise while in "P" mode).
Mmm, at 3:53 it says the ioniq consumed 163 Wh/km at 130 km/h but at 7:14 it says 207 Wh/km? 🤔 Am I missing something or is that an error? The consumption at 90 and 110 km/h are the same
Hi Chris, thank you for the very interesting measurements. However, I believe these kind of measurements rely too much on the accuracy of the BMS around the 50% SOC. I wished you had run some additional tests to ensure reproducibility by repeating some of the tests. My experience tells me that 50% BMS value is not always the same under different environmental conditions or under the influence of the experienced battery stress due to prior charge and discharge cycles. This could be an explanation for the weird outlier of the Kia measurement.
Besides the temperatures in the tests, in winter, the e-niro doesn't throw away energy for heating the battery up to 13°C or even more(?) every bloody day only for driving, not for charging.
@timemachine_194 But according to Battery Life's recent video VW has not fixed anything from user input, just trying to bugfix their release-SW... Oh well, I guess we will see if VW is clever and provides the ID.3 Pure test car with a new SW or if it will get horrible consumption again for everyone to see. 🤷
