Generally power converters like the AC/DC converters in the charger will be most efficient at or close to its maximum rated capacity. This is a guideline, but not a hard rule though. This explains why the three-phase charging was less efficient as it will use an additional module but at lower capacity compared to the same power at a single phase. Methodologically your biggest issue is in using the cars reported stored battery capacity as a metric for charging efficiency. There is no direct way of measuring the potential energy stored in a battery short of dismantling it and doing an analysis on a molecular level. The indirect ways by which we measure capacity is to measure the charge going in and the charge going out then compare the voltage to get to a power loss. To complicate matters there is also some charge loss though it is very small. In practice this means that the reported SOC is a guess by the cars software based on what you have done to them. A truer result would be obtained from measuring the discharge power compared to the charge power, The challenge with that is of course to know when you are back to a common state after a charge-discharge cycle as if the battery SOC is different at the start and end it would skew the results. Integrating over many repeated tests might help. Note that 100% SOC does not help, 100% is also just a guess by the BOM, there is no well defined 0% and 100% SOC that can be directly measured. My observation of the data leads me to think that the efficiency of AC charging might be a bit high, i.e. the BMS might overstate the SOC numbers a bit, as I would expect some percentage points loss in the battery and about 8% in the AC/DC converter. DC charging seems reasonable though. (Note that AC and DC charging efficiencies are not comparable here as the DC charging does not measure the efficiency loss in the charger itself, this is just the estimated charge loss of the battery. Note that battery charging loss is highly SOC dependent. It goes from very low at low SOC to ~5-15% at high SOC as the internal resistance increases. You can observe this by the hysteresis of the charge-discharge voltage curve for the battery (But is also dependent on C rate) TL/DR: As a method for discovering comparable efficiencies for a specific car this is OK, but do not use this methodology to compare cars. The use of the BMS' estimated SOC in the calculation makes numbers compared across different cars (potentially even of the same make and model due to production variation) highly suspect.
Very interesting!! I have always charged as slow as possible on AC because I figured there would be less loss, but I hadn’t thought about it keeping the car awake longer. Now I can save some energy so thanks! :-)
09:40 there is a huge misunderstanding going on with three phase AC It’s only 400V BETWEEN two phases, wich is only used for rotary current (eg to power an electric engine). But the charger uses the three different phases against neutral wich gives it three times 230V and NOT 400V. Calculation; 3 Phases * 230V * 16A = 11kW 400V would be: 400V * 16A = 6,4 kW
Yes, but in case of 3 phases, 1/3 of the current [amps] is required, as opposed to 1 phase, for the same power [watts]. Less current is less heat (loss).
dezz First of all that has nothing to do with the „400V“ claim Secondly, yes with 3 phases 1/3 of current would be needed to achieve the current of one single phase. But in reality you’re not demanding the same, but tripple the power (11kW) wich makes the load on the cables 3 * 230V * 16A
@@MrMoccachinoo That was a mistake or perhaps an oversimplification, but I don't think Bjørn don't know what 3 phase really means, at least what I wrote above. There were 7 kW 1-phase compared to 8 kW 3-phase.
dezz I was just talking about the 400V thing, wich is a complete misunderstanding. You get 400V if you measure between to phases. This is only used in devices with rotary movement to get constant 400V by overlapping 3 phases. (Eg saws, chipper, mason tools etc.) but NOT when converting into DC. It just has nothing to do with each other and so it should not be mentioned together with charging something. However you’re totally right that there phases can deliver the same power as one phase with 1/3 of current per phase
@@dezz00002 I'm under the impression that there are 3 charging modules in the Tesla, but I am not positive how the single phase current is split amongst them.
I don’t like this feature as it will encourage people to occupy a stall for longer than necessary. With public chargers, people need to leave as soon as soon and free up the space for someone else.
Thanks for this video Bjorn. I was actually doing these very same tests with my Kona just last week. Earlier this year I charged at a 350 kW charger. The charger delivered 22 kWh, but the car only received 18.4 kWh. This was on a 35 degree day after hammering on the motorway for 175 kms, so the battery cooling system was in full swing. So I've observed almost identical charging losses as you experienced here e.g. 50kW being the most efficient, 10 amp (2.3kw) the least efficient. Your theory about the length of charging time (that the car has to be awake) rather than the higher charge rate impacting on the efficency makes sense.
The 3 phase will be lower efficiency since it is running 3 converting circuits at a lower load. DC-DC conversion hasn’t a sweet spot of efficiency, and outside that it tapers off, as you can see in your results. With 3 phase, the load has to be specific for every phase itself, so every phase needs his own conversion circuit, and thus they run at a very low load. I would suggest that you find a car that can charge at 22kW, so that you can test 7kW vs 22kW, which should give a 7kW load on every phase and thus be comparable to 7kW efficiency
11KW AC 3 phase is 3 x 15 Amp x 240 volt with the car useing all three of its AC to DC converters. So the loss should be the same as 15Amp x 240 volt 1 phase 3.5ish charge useing one of the three AC to DC converters. But, the charge takes 3 times as long Bjorn describes that as takeing 3 tines as much car awake energy. I think of it as that plus 3 times as much keep battery and power management warm energy. My Chevy Volt has 15% loss based on what I measure as being suplied vs what car reports as being used. A great deal of that is the AC to DC loss, that thing gets very warm while chargring. Thanks Bjorn you have saved me heaps of time doing these tests for me. And dambn you Bjorn you have taken away from me hours of fun testing. :-)
Chris Avram those are also factors. But what I describe is certainly also a factor. Circuitry has 1 optimum spot that it is most efficient, outside of that it loses efficiency. You can learn about that by looking at the efficiency rating if PSU for desktop PC, most PSU hit their highest efficiency at 50-80% load, for a platinum rating it has a minimum of 92% at 50% load. When looking at 20% it requires a minimum of 90, and at 100% load a minimum of 89%. If he takes a car that has 22kW AC charging, he can compare the influence of faster AC charging aside of the 8nfluence of different phases. Here he thought 8kW 3 phase would be more efficient than 7kW 1phase but it wasn’t, and normally 8kW charges faster than 7kW so it should already have a head start by having less loss due to screen time/other electronics. The efficiency of the 8kW is still higher than the 2,3kW, which has around the same ampere load, and so that is where you see the influence of faster charging reducing the part lost due to other electronics/screen time. This is why 8 suspect that the true reason that 8kW and 11kW is less efficient than the 7kW is due to the charging electronics not running on their most efficient load. This is why I would like to see the test repeat with a car where he can compare 22kW vs 7kW, that would be a true comparison for 3 phase charging. It could be even more efficient than the 7kW charging, because of reduced screen time/other electronics. Or it could be close. To me it is a vital data point we need to know to have a real idea of all the possible AC charging
A 3-phase rectifier circuit is only slightly more complex than a 1-phase one (6 diodes/thyristors, instead of 4). Even if there were 3 full rectifier circuits, instead of only one, that wouldn't multiply the losses, either, as these are in parallel, not in series. Rather, in case of 3 phases, 1/3 of the current [amps] is required, as opposed to 1 phase, for the same power [watts]. Less current is less heat (loss).
dezz Yes I know why you should chose 3 phase. But how do you explain that 8kW 3 phase is less efficient than the 7kW single phase? Those extra losses will not be due to lower amperage as by the logic you are applying. The only way that could be true if the test isn’t flawed, is if there are higher losses in the circuitry itself that are more severe than the saved losses by lower amperage
I have also seen similar, For me the AC charging sweet spot is 24A 240v. 94-95%. (SR+3 32A Onbord max) Normal AC here in the US we have 120v, so 12a 120v, 83-85%. Amazing the difference.
I love the in depth testing, results, and great commentary. Thanks Bjørn! Keep the Nerd Level 1000 videos coming. Also, consider buying some calibrated test gear for more accurate AC and DC current and voltage measurements.
Very interesting video, I really enjoy these videos. 9:43 charging with 3 phase is not getting the charger higher volts (the car is not getting 400 V, it's only getting 230 V, but three times ) every phase is seperated in the Charger, so that there are 3 chargers with 230V times 16 A (u're right, the amps on every phase is lower). So maybe thats the reason why the 7 kW session was more efficient than the 8 kW Session, because at 3 phase the charger has the losses of all three inverters, while at 7 kW session theres only the loss of one inverter.
Good to have this. German ADAC reported 20% loss for Model 3 recently. But I assume they left the heater on or something. At least they lost their credibility years ago when faking tyre tests. Now it shows that they have no clue about anything.
Solid job and interesting findings! Did couple of similar calculations/observations with 24k eGolf - after 500+ km trip on 7.2kW single phase got 92% efficiency out of the battery (incl heat loss). Quite realistic, service manual claims 94% power transformation efficiency @32A. And checked once briefly measurements with VCDS - at 6A OBC claims "power efficiency" parameter close to 100%. So couple of hundred watts for keeping awake seem realistic as well.
Very useful analysis, uniquely so. Thank you. My interest: my 16a circuit in the condo parking garage is showing a near 20% charging loss, why? All your examples are a very short distance from delivery to the obc. In my situation the cable run is close on 100m. At 16a, there is a 4-5v drop, but at 13a there is no drop. Still doesn’t account for all the losses seen, and I wonder if the counter used by the condo may be incorrect. New installations here are all 3 phase 11kw, but depending on where in the garage cable runs can still be 100m, though there are less cable losses with 3P for long runs. For my needs the Tesla portable charger at 13a or 16a (blue connector) are more than adequate, so what ever the loss, a 3p11kw wall connector installation will never pay for itself. But I still want to find out where my unaccounted for 10% is going!
Excellent video. Would be interesting to compare charger AC input power to battery DC input power to obtain the efficiency of the charger. And then maybe compare cell voltage while charging to cell voltage while discharging to get a rough estimate of the charge/discharge efficiency. Not sure whether I'd factor in the cars standby draw into charging efficiency. Especially for Teslas who seem to constantly burn quite a lot of power
the higher losses comes from the voltage difference from the source. the battery is 400v. so a 230v supply needs a extra step to boost the voltage. there comes the addes losses from.
Very interesting theme. Please Bjørn, If it is possible and you have the chance to get an MG ZS EV to do the same tests, the MG community and I would be very grateful.
Great Video and to add to the Nerd Level: A German motorists club has tested charging loss by comparing the consumed energy in the car display and the energy on the meter on the connector of the wall box outside the car, and they found 25% charging loss for the Model 3 LR AWD, only about 9% for the SR+, and for most other cars. But what they did wrong IMHO: They always charged all cars from 0 to 100%, which on the Tesla Long Range models takes forever, but is much quicker with cars who have a large top buffer (which I think also the Model 3 SR+ has, because it doesn't "need" such a high capacity on paper). So, the Model 3 LR was left dangling on the charging cable probably for hours without significant further increase of SoC, while all the stand-by consumers were running, and the battery was probably only balancing, not charging anymore. This way, charging loss appears much higher on the LR than on the SR+, but in reality it's just a time effect. But why do I think the SR+ has a larger top buffer than the LR? Remember the Model S 75D, this was also a Tesla which was marketed as an explicit shorter-range Model S. They put in essentially the same battery as for the larger versions, but they software-restricted it, which gives you a cool top buffer and fast charging speed until 100%. The SR+ may not have the same battery as the LR, but still the physical battery might be a bit larger in relation to the usable capacity compared to the LR. The SR+ hasn't been marketed with the greatest range, so it doesn't need to have, and the other benefits of having a top buffer might be worthwhile.
This confirms what I have been thinking with my PHEV. Charging on the granny charger (10 amps at 240v AC) against changing on a type 1 16 amp at 240 Volts has always seemed quite a bit quicker on 16 Amps than the difference should be.
Power factor doesn't mean only a fraction of the power delivered. If the PF lower than 1, it means that the current and voltage were out of phase during that session, which is the normal for inductive loads, such as your car's rectifier. Resistive loads, such as an electric oven or heater have a PF equal to 1. That is usually not a concern for residential installations, but commercial and industrial customers have limits on how low their PF can get, or face an extra fee from the utility. You can trust the meter delivered the 13 kwh it says, after all it saw roughly 2 kw for over 7 hours.
My experiences from my trip to Spain and back with Ioniq28kWh in 2019, with 80+ charging sessions, is that Ioniq sees losses from 3%-12%. But it seems to vary more by the charging station than the actual car for the 28kWh.
I think the main takeaway is that, if you take away the extremes where you deliberately picked an efficient part of the cycle, there is really only about a 10% spread in charging efficiency regardless of how you charge. That's pretty insignificant in the grand scheme of things, and so you should just do what suits you best.
The German ADAC also made some test to measure the difference between car consumption and real current consumption. presse.adac.de/meldungen/adac-ev/technik/ladeverlust.html After driving on the test bench, all electric cars were always charged at the same 22-kW wallbox using the vehicle's type-2 charging cable and the same ambient conditions (23 degrees Celsius). Thus, each electric car was charged with the maximum charging power supported by the on-board charger'.
So they tested the Tesla with its suplied type 2 cable did they? It doesn't have one. Bjorn showed in this video that it is more efficent to charge with a one phase 32 Amp EVSE than with a 3 phase 32 Amp EVSE operating at 15 Amp per each of the 3 phases, this is the 11 KWatt limit of the Tesla.
Unfortunately the ADAC test is low on details. When you see that M3 LR and M3 SR+ have very different losses (25% resp 18%), although they use the same on-board charger and their battery, at least at AC charging as can be supposed from above description, probably behaves very similarly in terms of losses, you can question the test methodology. I rather believe in Uncle Bjørn :-)
@@michlanners1550Me too, I really like his videos! The ADAC calculates the difference between the displayed value of the on-board computer after the test and the amount of kWh that needs to be charged to reach the same SOC value as before the test run. But it would be nice if all manufacturers had to list the charging losses. It would be better for a detailed comparison (level 1000) of vehicle efficiency. In fact, the losses that have to be taken into account for charging are around 10-20% for every car. Why do manufacturers actually use kWh for battery capacity to represent capacity. kWh is power, not capacity, a battery has mAh or Ah but not kWh :) just by the way.
Great Video Björn, thank you. BTW: you can‘t charge a Tesla „at 400 Volt“. On a 3-phase circuit aTesla (and all other EVs that I’m aware of) runs 3 on-board chargers in parallel, each running at 230 Volt. You would only get 400 Volt if the charger would draw current between phases, but that is not the case. You run 3 chargers, one on each of the phases, each operating at 230 Volt between the respective phase an neutral.
Do you really think so? What is the difference between 3×230×16 and SQR3×400×16... not very much. Basically the same energy. Just a slightly rounding error. Or depending on how you description the problem. (230×SQR3) is about 400V so you can switch into SQR3×230×SQR3×16=3×230×16 Now you see that it does not matter if the on board charger is operating in 400V delta mode or in 230V star mode it is exactly the same amount of energy if the current drawn is the same.
Tesla could install a few of these Tesla City SuperChargers (70kW) on every SuperCharger site, to support people who do not need to charge at high speeds. (and / or want to take it easy on their battery)
The fan stops as the car assumes you will be driving and the air stream would pass through the front radiator (no longer needing the fan to cater for the air flow).
brill vid, really informative, Many thanks. Could you do an uncle Bjorn's top ten tips on saving money and efficient driving for newbs or the long distance daily commuter who could avoid a daily 5 min stop off top up charge session every day using a combined efficiency strategy. ? Best charging speed, best tyres and sizes, best speed to drive at dry, what speed to drive wet, best home charger, what things in the car suck the most energy. what would you do if you worked about 15 min beyond your cars comfort zone.
Hey, you have just answered a question. The charger at my local supermarket provides 6.4kw, as l have a 64kwh battery in my Kia l assumed an hour on the charging station would give me a 10% increase in battery charge. It never does, now l know why & have something more to geek out about.
Like you said: when you convert from AC to DC it's like a powersupply for a computer. It will only have high effecienty when you charge at maximum speed. (Max amps) The Zoe is really ineffecient if it get somerhimg lower than 22kw (32amp) 70-80%
i enjoyed this video - i am confused, but i kinda understand various charging sessions and their efficiency - I also think tesla is ahead in least charge wasted as compared to other EV cars. Awesome video - thank you :)
I also charge at 1kw (5A*220V) on my Model 3 LR and I was wondering about efficiency. Is the charger less efficient at low power? How about energy lost due to fan cooler? I also wonder if efficiency is not better when the battery is quite empty compare to quite full? I’m really interested in total efficiency comparison between 1kw, 2kw, 3.5kw and 7kw. Did somrlne make the test?
Franck Swordmaster Björn did do comparison for 2, 3.5 and 7kW in this video. I don’t think the charger is less efficient at lower power ( although that may also be a factor), but I think the main reason is because the slower you charge, the longer it takes, and the car is awake all that time, drawing background power of probably a few hundred watts. And with the lower charging power like 1kW, a bigger proportion of the power goes to keep the car awake.
There is also more loss in slower charging because of the conversion ac to dc. The lower Amps you charge at, the more often the conversion process of making 50 changes in +/- per seconds to dc has to be performed to charge the same amount of kWh with losses in every single ac-dc process.
My big takeaway from this is that the onboard charger has a horrendous power factor. Most domestic customers pay for true power (the 7kw). If you were paying by apparent power (like some commercial premises, or like some speculate smart meters will enable) you would be paying double. Interesting though, the power meter you are using says 13.19kWh which should be true power, as apparent power should be in kVA. It also means to charge from a battery backup unit or power wall for example you would need a unit rated at twice the power you want to charge at.
Many thanks Bjorn, was waiting for this one ! Now you have the Taycan to compare with :-) In the EPA, they MPGe is based on the the power needed for charging via AC charger - not the battery. Taycan is being panelized due to a very low AC charging efficiency. From your video, it looks that the DC charging efficiency is also pretty bad.
@@bjornnyland I remember from one of your video a huge discrepancy between IONITY kWh and the gained kWh. I would expect that a 400VDC charging efficiency to be even lower. Waiting for your video :-) EPA cares about AC charging efficiency as it is what most people will use. Hence the poor 69 combined MPGe.
I love these semi-technical videos. Well done ! A couple of questions: 1. How do you know if the DCFC is reporting output or input power ? 2. Where is the supercharger data ?
I charge my Ioniq PHEV at home on 240V (Canada). My Flo Home consistently reports a 14.6A draw for 3.5 kW, and whenever I've checked with Torque Pro / OBD it shows 3.08 kW going into the battery for a 12% loss. I can't speak for the Ioniq BEV which has a 7 kW onboard charger, it may be more (or less) efficient. Occasionally I can hear an electric coolant pump running, but I've never heard a fan.
Well, the loss is calculated by P_loss = R * I * I. For the cable R is proportional to the length and inversely proportional to the diameter of the cable. For home sockets the cable is pretty thin. The cable for a 11kW charger is a lot thicker. (For the AC DC comparison that doesn't really help because in addition to a diffenet cable gemometry the path in the car is different. )
Oh man, I like that sheet.. Initiation: well done! Follow up: with other cars! Which car has the best on-board-charger? Who's the efficiency king when grid to bat is to be or not to be!? Yeah, more of that sh..! (y)
except for 2 aberrations, most seem to be around 90% with car use varying it by a bit on either side. i'm sure different weather conditions would skew the tests from each charger again too if done on a hot day and a freezing day.
Sentry Mode does matter if charging runs for many hours! When the calculation precision comes to tenths of a percent it's a factor, driving up the loss figures for the 3,5 kW charge rate. Otherwise: I love the numbers, good stuff! DC charging has still some way to go in terms of efficiency...
Does it really matter? Sentry mode requires the car to be awake. Since it's already awake while charging, keeping some cameras active should only pull some extra 10-20 W.
@@bjornnyland True if the car is already awake, the difference is far less. If the difference is in the order of 20 W, it comes to 200 Wh (or 0,2 kWh) in ten hours charging time at 3,5 kWh - that's 0,5 %. All other factors being equal, which they rarely are... it's all geek factor 1000 after all ;-) ;-)
The number is an important one because it appears in other calculations. When someone asks me how much it costs to charge my car at home, I have to figure charging loss because all the consumption data I have is from the car. I just wonder if there is any difference between the fast DC chargers tested here and a Tesla Supercharger. You wouldn't expect any difference, but for us in the US, we rarely use 3rd party chargers.
On a DC there are also losses in charger that you can't see and in some cases even batteries that remove need for hi power conection to grid. So I'm sure 50KW is not the best...
You should test the standing consumption of it at standby/idle over a few hours. Then account for that in each of your tests. This will give you a more accurate indication of actual charger efficiency. The slower/lower you charge at, the lower the relative efficiency will be. Imagine charging at 0.5kW and the car is consuming 0.25kW just to do basic standby functions. The relative loss looks huge.
Thank you Bjorn excellent test on charging efficiency but I wonder what is best for battery health and longevity. At home I usually charge slowly at 16A x 240V to hopefully be nicer to the battery but I'm not so sure that is the case since the battery seems to take a charge more efficiently when it is 40C or warmer.
I wonder if its because at lower charging rate in your example, its at 230v - which is less than the pack voltage... so maybe it also needs to do some more work to step up the voltage vs 400v at higher power stated. Maybe check 7 vs 3.3 on 230v?
Excellent video. Would be good to calculate what the loss is like on solar charging, often mine charges at 1.4KW (6 amps) at home. I don't have scan my tesla and the in car display rounds everything up so it'd be hard to be certain. Maybe you could try charging on schuko at 6 amps by limiting it in the car?
Hey Bjorn, very interesting, nowhere could i find what is the best choice for home charging, considering efficiency and/or load of the grid. Now i mostly choose, when having the time, as low as possible to let my PV installation deliver most of the power ( 4.5 KW ) and the minimum load on the net, but maybe losses are much higher than setting it on max 3 x 16 amps ( 11 KW ). can you maybe do in the future also the lowest setting test 3 x 5 amps ???
Really nice tests. I assume charging on less than 11kW AC generates more loss because car is active and a higher % of current from the AC charger is used to power the car. Think I could see som 300W there, sometimes maybe less. On a 3.7kW AC charger that is up to about 8% extra that will add to the charging loss. In my experience with my MG I think less than 5%, but I have not noted any numbers, I think I have to. But of course, MG hides % of charge... And yes, I am more nerd'y than most
Hi Bjorn , would please let me know what you software / hardware you used to monitor the Ioniq S,O.C. . I have the budget 'efficiency King' 28kw version and its amazing. Thanks for all your hard work
As far as I know the efficiency on AC 400V charging is a bell graph (the efficiency at low kW and at 22kW is lower than at 7-10 kW) so you ha e the highest pick of efficiency arroung 7-11 kW but this is related to the physics of 400v AC charging system
I know the E-Niro drops the charge to 0.5kw when a door is opened for safety (so if you pull the plug after unlocking or something) I wonder if the Tesla does the same and that may explain the fan speed stopping if it’s cooling the charger. Might not but it was just a thought.
My last Schuko charging session at home 13A: Energy meter: 33 kWh (power factor was at 100% and power fluctuates between 3150W and 3200W) Teslafi energy used: 31,80 kWh Teslafi energy added: 29,63 kWh I suspect that my energy meter is over reporting kWh I will parallel check on Scanmytesla next time
Hello from one EXPERT. Nice video. Please continue with the energy lose after driving. How much we take out from the battery. This is the true efficiency.
Best efficiency of my Zoe is when charging with 43 kW AC. Everything lower is less efficiency. Especially when charging with one phase only, the efficiency is very bad. So as long it is at least 3-phase charging, efficiency is still tolerable.
I saw the same behaviour of the fan in our TM3P, yesterday. When I just parked and wait couple of minutes the fan kicked in and when I just open the door it stops. And the car was just sitting in the parking lot without charging in 20 degrees... Weird sh*t :)
Would be amazing if Tesla could just charge your car and then let it drive away to a close parking space once the charging is done if someone else comes in who needs the charger
@@roccosperanza so with Bjorn in the back seat in 10 minutes it comes on. If he opens the door, it goes off - car in use so don't dehumifify ...hops back in 10 minutes later it comes on. Thanks.
Yes, the same observation I did have. Lower stress to the phases (also in house), increases efficiency.. But, all in all, it depends on other conditions as well. What I miss, is that I can adjust the amps with the app, by smartphone. Hope, that it is coming soon
@@bjornnyland yes i saw, but at 1phase 30 amps, the car uses 2 charger units out of 3 and load them with 15 Amps each. When you charge AC 3 phase with 13 amps, the car uses the 3 charger units witch 13 Amps ant the efficenty is better. Greetings from Germany
When charging on AC there is an AC-to-DC rectifier in the loop, which has an inverted bell-shape efficiency curve. Look at PC PSUs, at low power they have poor efficiency, peak out towards 50% of max power and drop a bit at max power. An example here: www.anandtech.com/show/15950/the-nzxt-c650-650w-psu-review/2. The Power factor, PF correction or lack of, does not play a role in the efficiency game. Battery resistive losses are easy to compute: say the 3 is 46 parallel and 96 series (googled it) , the pack shoult have resistance of about 37mOhm at ambient for 18mOhm cell resistance. At 7kW DC you get 17A which leads to about 10W of heat losses. At high power DC charging undoubtedly heat losses in the battery pack matter more, also since you don't have the rectifier in the loop any more. Does this make sense?
Hej Bjørn, there is another source of lost, you didn't have attention to: how hot/cold were the charging stations, you used? I have heard that hot charging stations have up to 5% lost, which you will be get billed for. Can you please check if that is right og just a rumer? Thank you 😀
AC single phase: Power is delivered between phase and neutral. That means two wires with losses. You only get one phase. AC three phase: Power is only delivered between the phases. Nearly no current is on neutral. That means three wires with losses, but three phases too. -> Three phase charging should be more efficient in general. DC: The charger is in the cabinet, so most of the losses are not counted. The internal resistance of cable and battery is not that much. Problem on M3: The afterburner with up to 7kW heat generation. You can't switch them off. Especially on low DC power (20kW) you only get about 13kW into the battery. You'll be charged with 20kW because this is what you get. It would be even worse if the AC-DC losses are counted. -> It makes nearly no sense to charge on DC 20kW. HPC: Don't think about. If you charge with 100kw and more you'll just give a fuck about the 7kW afterburner. Furthermore the current is that high, the battery will be warm enough very soon to charge without battery heating.
I always assumed slower charging speeds would have less loss due to heat so would therefore be more efficient, but as a counter argument I did always wonder if the losses being spread over a long period of time perhaps hurts efficiency , sounds like there is not a simple answer as the results varied based on a number of factors. I wonder if the lower speeds work better on non Tesla cars which might use less energy when charging? I always try to charge quite slowly at home to maximise proportion of solar energy from by roof, now i'm worried i'm maybe losing some efficiency in doing so
4:26 I disagree. 3.5kW charger was less efficient because it had to bump the voltage to 400V, if the input were 400V like 11kW, then chargeing at 3.5kW would be a bit more efficient due to the lower current. But I suspect it will be not appreciable(unsure)
"This is Nerd Level 1000" - that's what I came for
Generally power converters like the AC/DC converters in the charger will be most efficient at or close to its maximum rated capacity. This is a guideline, but not a hard rule though. This explains why the three-phase charging was less efficient as it will use an additional module but at lower capacity compared to the same power at a single phase.
Methodologically your biggest issue is in using the cars reported stored battery capacity as a metric for charging efficiency. There is no direct way of measuring the potential energy stored in a battery short of dismantling it and doing an analysis on a molecular level. The indirect ways by which we measure capacity is to measure the charge going in and the charge going out then compare the voltage to get to a power loss. To complicate matters there is also some charge loss though it is very small. In practice this means that the reported SOC is a guess by the cars software based on what you have done to them.
A truer result would be obtained from measuring the discharge power compared to the charge power, The challenge with that is of course to know when you are back to a common state after a charge-discharge cycle as if the battery SOC is different at the start and end it would skew the results. Integrating over many repeated tests might help. Note that 100% SOC does not help, 100% is also just a guess by the BOM, there is no well defined 0% and 100% SOC that can be directly measured.
My observation of the data leads me to think that the efficiency of AC charging might be a bit high, i.e. the BMS might overstate the SOC numbers a bit, as I would expect some percentage points loss in the battery and about 8% in the AC/DC converter. DC charging seems reasonable though. (Note that AC and DC charging efficiencies are not comparable here as the DC charging does not measure the efficiency loss in the charger itself, this is just the estimated charge loss of the battery.
Note that battery charging loss is highly SOC dependent. It goes from very low at low SOC to ~5-15% at high SOC as the internal resistance increases. You can observe this by the hysteresis of the charge-discharge voltage curve for the battery (But is also dependent on C rate)
TL/DR: As a method for discovering comparable efficiencies for a specific car this is OK, but do not use this methodology to compare cars. The use of the BMS' estimated SOC in the calculation makes numbers compared across different cars (potentially even of the same make and model due to production variation) highly suspect.
I guess I'm a Big nerd, because I did like this video a lot! 😁
Very interesting!! I have always charged as slow as possible on AC because I figured there would be less loss, but I hadn’t thought about it keeping the car awake longer. Now I can save some energy so thanks! :-)
09:40 there is a huge misunderstanding going on with three phase AC
It’s only 400V BETWEEN two phases, wich is only used for rotary current (eg to power an electric engine).
But the charger uses the three different phases against neutral wich gives it three times 230V and NOT 400V.
Calculation;
3 Phases * 230V * 16A = 11kW
400V would be:
400V * 16A = 6,4 kW
Yes, but in case of 3 phases, 1/3 of the current [amps] is required, as opposed to 1 phase, for the same power [watts]. Less current is less heat (loss).
dezz
First of all that has nothing to do with the „400V“ claim
Secondly, yes with 3 phases 1/3 of current would be needed to achieve the current of one single phase. But in reality you’re not demanding the same, but tripple the power (11kW) wich makes the load on the cables 3 * 230V * 16A
@@MrMoccachinoo That was a mistake or perhaps an oversimplification, but I don't think
Bjørn don't know what 3 phase really means, at least what I wrote above.
There were 7 kW 1-phase compared to 8 kW 3-phase.
dezz
I was just talking about the 400V thing, wich is a complete misunderstanding.
You get 400V if you measure between to phases. This is only used in devices with rotary movement to get constant 400V by overlapping 3 phases. (Eg saws, chipper, mason tools etc.) but NOT when converting into DC.
It just has nothing to do with each other and so it should not be mentioned together with charging something.
However you’re totally right that there phases can deliver the same power as one phase with 1/3 of current per phase
@@dezz00002 I'm under the impression that there are 3 charging modules in the Tesla, but I am not positive how the single phase current is split amongst them.
This kinds of tests are priceless!
Thanks Bjorn.
The Ev expert that doesn't gives us bull.
Also, Taycan will have a new feature to allow limiting of charging speeds on DC, read about it yesterday
and the losses will be lower on the taycan thanks to the higher voltage.
My motorcycle, an Energica Ego DC fast charges, and allows me to choose the amps so I can slow it down if I want.
@@Pixelplanet5 In theory, yes. But needs to be verified :)
I don’t like this feature as it will encourage people to occupy a stall for longer than necessary. With public chargers, people need to leave as soon as soon and free up the space for someone else.
Please keep making videos forever Bjørn 🔥
I'm here for the "Nerd level 1000" stuff,
Actual numbers, measurements, standardised tests instead of feelings and opinions of exterior design.
Thanks for this video Bjorn. I was actually doing these very same tests with my Kona just last week. Earlier this year I charged at a 350 kW charger. The charger delivered 22 kWh, but the car only received 18.4 kWh. This was on a 35 degree day after hammering on the motorway for 175 kms, so the battery cooling system was in full swing.
So I've observed almost identical charging losses as you experienced here e.g. 50kW being the most efficient, 10 amp (2.3kw) the least efficient.
Your theory about the length of charging time (that the car has to be awake) rather than the higher charge rate impacting on the efficency makes sense.
By far one of the more interesting videos! Not nerd level 1000. This is why we love your videos! :) Tusen takk!
You should be hired by Tesla to be chief of consumption !! :D
Well Tesla doesn’t even need to pay him now :D
The 3 phase will be lower efficiency since it is running 3 converting circuits at a lower load. DC-DC conversion hasn’t a sweet spot of efficiency, and outside that it tapers off, as you can see in your results. With 3 phase, the load has to be specific for every phase itself, so every phase needs his own conversion circuit, and thus they run at a very low load. I would suggest that you find a car that can charge at 22kW, so that you can test 7kW vs 22kW, which should give a 7kW load on every phase and thus be comparable to 7kW efficiency
11KW AC 3 phase is 3 x 15 Amp x 240 volt with the car useing all three of its AC to DC converters. So the loss should be the same as 15Amp x 240 volt 1 phase 3.5ish charge useing one of the three AC to DC converters. But, the charge takes 3 times as long Bjorn describes that as takeing 3 tines as much car awake energy. I think of it as that plus 3 times as much keep battery and power management warm energy.
My Chevy Volt has 15% loss based on what I measure as being suplied vs what car reports as being used. A great deal of that is the AC to DC loss, that thing gets very warm while chargring.
Thanks Bjorn you have saved me heaps of time doing these tests for me. And dambn you Bjorn you have taken away from me hours of fun testing. :-)
Chris Avram those are also factors. But what I describe is certainly also a factor. Circuitry has 1 optimum spot that it is most efficient, outside of that it loses efficiency. You can learn about that by looking at the efficiency rating if PSU for desktop PC, most PSU hit their highest efficiency at 50-80% load, for a platinum rating it has a minimum of 92% at 50% load. When looking at 20% it requires a minimum of 90, and at 100% load a minimum of 89%. If he takes a car that has 22kW AC charging, he can compare the influence of faster AC charging aside of the 8nfluence of different phases. Here he thought 8kW 3 phase would be more efficient than 7kW 1phase but it wasn’t, and normally 8kW charges faster than 7kW so it should already have a head start by having less loss due to screen time/other electronics. The efficiency of the 8kW is still higher than the 2,3kW, which has around the same ampere load, and so that is where you see the influence of faster charging reducing the part lost due to other electronics/screen time. This is why 8 suspect that the true reason that 8kW and 11kW is less efficient than the 7kW is due to the charging electronics not running on their most efficient load. This is why I would like to see the test repeat with a car where he can compare 22kW vs 7kW, that would be a true comparison for 3 phase charging. It could be even more efficient than the 7kW charging, because of reduced screen time/other electronics. Or it could be close. To me it is a vital data point we need to know to have a real idea of all the possible AC charging
A 3-phase rectifier circuit is only slightly more complex than a 1-phase one (6 diodes/thyristors, instead of 4). Even if there were 3 full rectifier circuits, instead of only one, that wouldn't multiply the losses, either, as these are in parallel, not in series. Rather, in case of 3 phases, 1/3 of the current [amps] is required, as opposed to 1 phase, for the same power [watts]. Less current is less heat (loss).
dezz Yes I know why you should chose 3 phase. But how do you explain that 8kW 3 phase is less efficient than the 7kW single phase? Those extra losses will not be due to lower amperage as by the logic you are applying. The only way that could be true if the test isn’t flawed, is if there are higher losses in the circuitry itself that are more severe than the saved losses by lower amperage
@@hoedjeexplains3661 That is weird, indeed. I wonder the actual cause of this. It could be what you explained, I'm just not entirely sure.
You are the King Bjorn! Thanks
I have also seen similar, For me the AC charging sweet spot is 24A 240v. 94-95%. (SR+3 32A Onbord max) Normal AC here in the US we have 120v, so 12a 120v, 83-85%. Amazing the difference.
I love the in depth testing, results, and great commentary. Thanks Bjørn! Keep the Nerd Level 1000 videos coming. Also, consider buying some calibrated test gear for more accurate AC and DC current and voltage measurements.
Cheers from Oregon, I enjoyed your nerd level 1000.
Very interesting video, I really enjoy these videos.
9:43 charging with 3 phase is not getting the charger higher volts (the car is not getting 400 V, it's only getting 230 V, but three times )
every phase is seperated in the Charger, so that there are 3 chargers with 230V times 16 A (u're right, the amps on every phase is lower).
So maybe thats the reason why the 7 kW session was more efficient than the 8 kW Session, because at 3 phase the charger has the losses of all three inverters, while at 7 kW session theres only the loss of one inverter.
Good to have this. German ADAC reported 20% loss for Model 3 recently. But I assume they left the heater on or something. At least they lost their credibility years ago when faking tyre tests. Now it shows that they have no clue about anything.
Solid job and interesting findings!
Did couple of similar calculations/observations with 24k eGolf - after 500+ km trip on 7.2kW single phase got 92% efficiency out of the battery (incl heat loss). Quite realistic, service manual claims 94% power transformation efficiency @32A.
And checked once briefly measurements with VCDS - at 6A OBC claims "power efficiency" parameter close to 100%. So couple of hundred watts for keeping awake seem realistic as well.
Nerd Level 1000 - LOVE THAT!! Exactely that kind of tests are SO great!!
Thanx for that. I screenshot the last table (sheeeet) for further use.
No need to. You find the spreadsheet in the description.
Very useful analysis, uniquely so. Thank you. My interest: my 16a circuit in the condo parking garage is showing a near 20% charging loss, why? All your examples are a very short distance from delivery to the obc. In my situation the cable run is close on 100m. At 16a, there is a 4-5v drop, but at 13a there is no drop. Still doesn’t account for all the losses seen, and I wonder if the counter used by the condo may be incorrect. New installations here are all 3 phase 11kw, but depending on where in the garage cable runs can still be 100m, though there are less cable losses with 3P for long runs. For my needs the Tesla portable charger at 13a or 16a (blue connector) are more than adequate, so what ever the loss, a 3p11kw wall connector installation will never pay for itself. But I still want to find out where my unaccounted for 10% is going!
Thanks for making this!
Excellent video. Would be interesting to compare charger AC input power to battery DC input power to obtain the efficiency of the charger. And then maybe compare cell voltage while charging to cell voltage while discharging to get a rough estimate of the charge/discharge efficiency.
Not sure whether I'd factor in the cars standby draw into charging efficiency. Especially for Teslas who seem to constantly burn quite a lot of power
the higher losses comes from the voltage difference from the source. the battery is 400v. so a 230v supply needs a extra step to boost the voltage. there comes the addes losses from.
Watched them all from kenya Africa baby ...straight love bro
Good to see that my home charging is efficient (7 kW AC)
Very interesting theme. Please Bjørn, If it is possible and you have the chance to get an MG ZS EV to do the same tests, the MG community and I would be very grateful.
Does the car report charging speed on screen?
Or is there a good way to get it like "scan my tesla" and leafspy?
Great Video and to add to the Nerd Level: A German motorists club has tested charging loss by comparing the consumed energy in the car display and the energy on the meter on the connector of the wall box outside the car, and they found 25% charging loss for the Model 3 LR AWD, only about 9% for the SR+, and for most other cars. But what they did wrong IMHO: They always charged all cars from 0 to 100%, which on the Tesla Long Range models takes forever, but is much quicker with cars who have a large top buffer (which I think also the Model 3 SR+ has, because it doesn't "need" such a high capacity on paper). So, the Model 3 LR was left dangling on the charging cable probably for hours without significant further increase of SoC, while all the stand-by consumers were running, and the battery was probably only balancing, not charging anymore. This way, charging loss appears much higher on the LR than on the SR+, but in reality it's just a time effect. But why do I think the SR+ has a larger top buffer than the LR? Remember the Model S 75D, this was also a Tesla which was marketed as an explicit shorter-range Model S. They put in essentially the same battery as for the larger versions, but they software-restricted it, which gives you a cool top buffer and fast charging speed until 100%. The SR+ may not have the same battery as the LR, but still the physical battery might be a bit larger in relation to the usable capacity compared to the LR. The SR+ hasn't been marketed with the greatest range, so it doesn't need to have, and the other benefits of having a top buffer might be worthwhile.
The mistake they also did was that they measured both charging loss and discharging loss together. I often describe discharging loss as "heat loss".
This confirms what I have been thinking with my PHEV. Charging on the granny charger (10 amps at 240v AC) against changing on a type 1 16 amp at 240 Volts has always seemed quite a bit quicker on 16 Amps than the difference should be.
Solid work Bjørn.
Power factor doesn't mean only a fraction of the power delivered. If the PF lower than 1, it means that the current and voltage were out of phase during that session, which is the normal for inductive loads, such as your car's rectifier. Resistive loads, such as an electric oven or heater have a PF equal to 1. That is usually not a concern for residential installations, but commercial and industrial customers have limits on how low their PF can get, or face an extra fee from the utility. You can trust the meter delivered the 13 kwh it says, after all it saw roughly 2 kw for over 7 hours.
My experiences from my trip to Spain and back with Ioniq28kWh in 2019, with 80+ charging sessions, is that Ioniq sees losses from 3%-12%. But it seems to vary more by the charging station than the actual car for the 28kWh.
That's exactly what i always wanted to know. :) Thanks a lot.
This was right up my alley, so i guess im a Nerd, Level 1000.
Thanks for the all the testing and info.
I think the main takeaway is that, if you take away the extremes where you deliberately picked an efficient part of the cycle, there is really only about a 10% spread in charging efficiency regardless of how you charge. That's pretty insignificant in the grand scheme of things, and so you should just do what suits you best.
The German ADAC also made some test to measure the difference between car consumption and real current consumption. presse.adac.de/meldungen/adac-ev/technik/ladeverlust.html After driving on the test bench, all electric cars were always charged at the same 22-kW wallbox using the vehicle's type-2 charging cable and the same ambient conditions (23 degrees Celsius). Thus, each electric car was charged with the maximum charging power supported by the on-board charger'.
So they tested the Tesla with its suplied type 2 cable did they? It doesn't have one. Bjorn showed in this video that it is more efficent to charge with a one phase 32 Amp EVSE than with a 3 phase 32 Amp EVSE operating at 15 Amp per each of the 3 phases, this is the 11 KWatt limit of the Tesla.
Unfortunately the ADAC test is low on details. When you see that M3 LR and M3 SR+ have very different losses (25% resp 18%), although they use the same on-board charger and their battery, at least at AC charging as can be supposed from above description, probably behaves very similarly in terms of losses, you can question the test methodology. I rather believe in Uncle Bjørn :-)
@@michlanners1550Me too, I really like his videos! The ADAC calculates the difference between the displayed value of the on-board computer after the test and the amount of kWh that needs to be charged to reach the same SOC value as before the test run. But it would be nice if all manufacturers had to list the charging losses. It would be better for a detailed comparison (level 1000) of vehicle efficiency. In fact, the losses that have to be taken into account for charging are around 10-20% for every car. Why do manufacturers actually use kWh for battery capacity to represent capacity. kWh is power, not capacity, a battery has mAh or Ah but not kWh :) just by the way.
You could also look at losses from power lines, power plant losses...etc. Engineering Explained had a great video about this.
Great Video Björn, thank you.
BTW: you can‘t charge a Tesla „at 400 Volt“. On a 3-phase circuit aTesla (and all other EVs that I’m aware of) runs 3 on-board chargers in parallel, each running at 230 Volt. You would only get 400 Volt if the charger would draw current between phases, but that is not the case. You run 3 chargers, one on each of the phases, each operating at 230 Volt between the respective phase an neutral.
Do you really think so?
What is the difference between 3×230×16 and SQR3×400×16... not very much. Basically the same energy. Just a slightly rounding error.
Or depending on how you description the problem. (230×SQR3) is about 400V so you can switch into SQR3×230×SQR3×16=3×230×16
Now you see that it does not matter if the on board charger is operating in 400V delta mode or in 230V star mode it is exactly the same amount of energy if the current drawn is the same.
Tesla could install a few of these Tesla City SuperChargers (70kW) on every SuperCharger site, to support people who do not need to charge at high speeds. (and / or want to take it easy on their battery)
They are called Urban Chargers
Bjørn Nyland Right! Thanks. :-)
The fan stops as the car assumes you will be driving and the air stream would pass through the front radiator (no longer needing the fan to cater for the air flow).
brill vid, really informative, Many thanks. Could you do an uncle Bjorn's top ten tips on saving money and efficient driving for newbs or the long distance daily commuter who could avoid a daily 5 min stop off top up charge session every day using a
combined efficiency strategy. ? Best charging speed, best tyres and sizes, best speed to drive at dry, what speed to drive wet, best home charger, what things in the car suck the most energy. what would you do if you worked about 15 min beyond your cars comfort zone.
Hey, you have just answered a question. The charger at my local supermarket provides 6.4kw, as l have a 64kwh battery in my Kia l assumed an hour on the charging station would give me a 10% increase in battery charge. It never does, now l know why & have something more to geek out about.
Like you said: when you convert from AC to DC it's like a powersupply for a computer. It will only have high effecienty when you charge at maximum speed. (Max amps) The Zoe is really ineffecient if it get somerhimg lower than 22kw (32amp) 70-80%
i enjoyed this video - i am confused, but i kinda understand various charging sessions and their efficiency - I also think tesla is ahead in least charge wasted as compared to other EV cars.
Awesome video - thank you :)
I’m charging at about 1kW (12A 120V) so it looks like that is pretty inefficient. I expected that was the case but thank you for confirming.
I also charge at 1kw (5A*220V) on my Model 3 LR and I was wondering about efficiency.
Is the charger less efficient at low power?
How about energy lost due to fan cooler?
I also wonder if efficiency is not better when the battery is quite empty compare to quite full?
I’m really interested in total efficiency comparison between 1kw, 2kw, 3.5kw and 7kw.
Did somrlne make the test?
Franck Swordmaster Björn did do comparison for 2, 3.5 and 7kW in this video. I don’t think the charger is less efficient at lower power ( although that may also be a factor), but I think the main reason is because the slower you charge, the longer it takes, and the car is awake all that time, drawing background power of probably a few hundred watts. And with the lower charging power like 1kW, a bigger proportion of the power goes to keep the car awake.
There is also more loss in slower charging because of the conversion ac to dc. The lower Amps you charge at, the more often the conversion process of making 50 changes in +/- per seconds to dc has to be performed to charge the same amount of kWh with losses in every single ac-dc process.
My big takeaway from this is that the onboard charger has a horrendous power factor. Most domestic customers pay for true power (the 7kw). If you were paying by apparent power (like some commercial premises, or like some speculate smart meters will enable) you would be paying double.
Interesting though, the power meter you are using says 13.19kWh which should be true power, as apparent power should be in kVA. It also means to charge from a battery backup unit or power wall for example you would need a unit rated at twice the power you want to charge at.
MAINGAU will increase prices very much from 1.september. Charging at IONITY costs 73c/kWh from then on.
I love this kind of video! Nerd vids for the win!!
Many thanks Bjorn, was waiting for this one !
Now you have the Taycan to compare with :-)
In the EPA, they MPGe is based on the the power needed for charging via AC charger - not the battery.
Taycan is being panelized due to a very low AC charging efficiency.
From your video, it looks that the DC charging efficiency is also pretty bad.
Taycan's DC charging efficiency is actually outstanding. Even better than Model 3.
@@bjornnyland I remember from one of your video a huge discrepancy between IONITY kWh and the gained kWh.
I would expect that a 400VDC charging efficiency to be even lower. Waiting for your video :-)
EPA cares about AC charging efficiency as it is what most people will use. Hence the poor 69 combined MPGe.
I love these semi-technical videos. Well done !
A couple of questions:
1. How do you know if the DCFC is reporting output or input power ?
2. Where is the supercharger data ?
2. Supercharger reports in integers only. Too inaccurate.
I charge my Ioniq PHEV at home on 240V (Canada). My Flo Home consistently reports a 14.6A draw for 3.5 kW, and whenever I've checked with Torque Pro / OBD it shows 3.08 kW going into the battery for a 12% loss. I can't speak for the Ioniq BEV which has a 7 kW onboard charger, it may be more (or less) efficient. Occasionally I can hear an electric coolant pump running, but I've never heard a fan.
Well, the loss is calculated by P_loss = R * I * I. For the cable R is proportional to the length and inversely proportional to the diameter of the cable. For home sockets the cable is pretty thin. The cable for a 11kW charger is a lot thicker. (For the AC DC comparison that doesn't really help because in addition to a diffenet cable gemometry the path in the car is different. )
Oh man, I like that sheet.. Initiation: well done! Follow up: with other cars! Which car has the best on-board-charger? Who's the efficiency king when grid to bat is to be or not to be!? Yeah, more of that sh..! (y)
except for 2 aberrations, most seem to be around 90% with car use varying it by a bit on either side.
i'm sure different weather conditions would skew the tests from each charger again too if done on a hot day and a freezing day.
That ansers my question. Thank you for your work
I listen your channel for that kind of stuff. Really interesting video. Am I a nerd? 🤔 😅
Sentry Mode does matter if charging runs for many hours! When the calculation precision comes to tenths of a percent it's a factor, driving up the loss figures for the 3,5 kW charge rate. Otherwise: I love the numbers, good stuff! DC charging has still some way to go in terms of efficiency...
Does it really matter? Sentry mode requires the car to be awake. Since it's already awake while charging, keeping some cameras active should only pull some extra 10-20 W.
@@bjornnyland True if the car is already awake, the difference is far less. If the difference is in the order of 20 W, it comes to 200 Wh (or 0,2 kWh) in ten hours charging time at 3,5 kWh - that's 0,5 %. All other factors being equal, which they rarely are... it's all geek factor 1000 after all ;-) ;-)
The number is an important one because it appears in other calculations. When someone asks me how much it costs to charge my car at home, I have to figure charging loss because all the consumption data I have is from the car. I just wonder if there is any difference between the fast DC chargers tested here and a Tesla Supercharger. You wouldn't expect any difference, but for us in the US, we rarely use 3rd party chargers.
Nice observations and comment👍
On a DC there are also losses in charger that you can't see and in some cases even batteries that remove need for hi power conection to grid. So I'm sure 50KW is not the best...
You should test the standing consumption of it at standby/idle over a few hours. Then account for that in each of your tests. This will give you a more accurate indication of actual charger efficiency.
The slower/lower you charge at, the lower the relative efficiency will be. Imagine charging at 0.5kW and the car is consuming 0.25kW just to do basic standby functions. The relative loss looks huge.
Charging efficiency is of course counting everything. We're not only looking at charger efficiency but everything together.
Thank you Bjorn excellent test on charging efficiency but I wonder what is best for battery health and longevity. At home I usually charge slowly at 16A x 240V to hopefully be nicer to the battery but I'm not so sure that is the case since the battery seems to take a charge more efficiently when it is 40C or warmer.
I wonder if its because at lower charging rate in your example, its at 230v - which is less than the pack voltage... so maybe it also needs to do some more work to step up the voltage vs 400v at higher power stated. Maybe check 7 vs 3.3 on 230v?
I am such a noob - I should have watched all the way first ... 32 amp was covered :-)
But the efficiency at 230 V 30 A was higher than 400 V 11 A.
Bjørn Nyland yeah I realised this after actually watching it all the way through... as I said I am a nooob 😂
I Became an expert because of your videos
Lower voltage is also lower efficiency
Excellent video. Would be good to calculate what the loss is like on solar charging, often mine charges at 1.4KW (6 amps) at home. I don't have scan my tesla and the in car display rounds everything up so it'd be hard to be certain. Maybe you could try charging on schuko at 6 amps by limiting it in the car?
Hey Bjorn, very interesting, nowhere could i find what is the best choice for home charging, considering efficiency and/or load of the grid. Now i mostly choose, when having the time, as low as possible to let my PV installation deliver most of the power ( 4.5 KW ) and the minimum load on the net, but maybe losses are much higher than setting it on max 3 x 16 amps ( 11 KW ). can you maybe do in the future also the lowest setting test 3 x 5 amps ???
Really nice tests. I assume charging on less than 11kW AC generates more loss because car is active and a higher % of current from the AC charger is used to power the car. Think I could see som 300W there, sometimes maybe less. On a 3.7kW AC charger that is up to about 8% extra that will add to the charging loss. In my experience with my MG I think less than 5%, but I have not noted any numbers, I think I have to. But of course, MG hides % of charge... And yes, I am more nerd'y than most
Hi Bjorn , would please let me know what you software / hardware you used to monitor the Ioniq S,O.C. . I have the budget 'efficiency King' 28kw version and its amazing. Thanks for all your hard work
Good stuff 👍
As far as I know the efficiency on AC 400V charging is a bell graph (the efficiency at low kW and at 22kW is lower than at 7-10 kW) so you ha e the highest pick of efficiency arroung 7-11 kW but this is related to the physics of 400v AC charging system
Thanks!
I know the E-Niro drops the charge to 0.5kw when a door is opened for safety (so if you pull the plug after unlocking or something) I wonder if the Tesla does the same and that may explain the fan speed stopping if it’s cooling the charger. Might not but it was just a thought.
Aim for a v2 Supercharger where you’re the 2nd on the pair and hope the other car just arrived, so you’ll be restricted.
Supercharger app (like CovidApp) matches singular arrivals to extends charging time for a less stressed Restaurant digg in :o)
Battery cooling loop always runs when charging. Fan will run to cool as needed
Good stuff
I like your nerd level, keep it up. Your garage is so empty, where do you keep your car stuff? Car detailing, spare tires etc?
My garage is not empty at all.
My last Schuko charging session at home 13A:
Energy meter: 33 kWh (power factor was at 100% and power fluctuates between 3150W and 3200W)
Teslafi energy used: 31,80 kWh
Teslafi energy added: 29,63 kWh
I suspect that my energy meter is over reporting kWh
I will parallel check on Scanmytesla next time
*Schuko
Bjørn Nyland updated:) shame! I’m German...
Hello from one EXPERT. Nice video. Please continue with the energy lose after driving.
How much we take out from the battery. This is the true efficiency.
Watch this video:
th-cam.com/video/tUJAgWpgJzs/w-d-xo.html
Best efficiency of my Zoe is when charging with 43 kW AC. Everything lower is less efficiency. Especially when charging with one phase only, the efficiency is very bad. So as long it is at least 3-phase charging, efficiency is still tolerable.
But the chameleon charger in Zoe has poor 80 % efficiency, right?
How do you think the 800 volt and 900 v cars will work with charging efficiency? It should be better correct?
I saw the same behaviour of the fan in our TM3P, yesterday. When I just parked and wait couple of minutes the fan kicked in and when I just open the door it stops. And the car was just sitting in the parking lot without charging in 20 degrees... Weird sh*t :)
Would be amazing if Tesla could just charge your car and then let it drive away to a close parking space once the charging is done if someone else comes in who needs the charger
I wonder what would the efficiency be for 120V @ 12 amps 🙁.
Fantastic video. Thanks
That fan is because of the mildew smell from AC, it's turns on to dry off the condenser.
When does it come on or off?
@@cavramau I noticed about 10-15 minutes after parking.
@@roccosperanza so with Bjorn in the back seat in 10 minutes it comes on. If he opens the door, it goes off - car in use so don't dehumifify ...hops back in 10 minutes later it comes on. Thanks.
@@cavramau Yeah, I've tested too, as soon as you open the door it'll shut the fan off. I was opening from the outside though.
But you can see in the video that the condenser was off.
perfect video..thanks bjørn
Hy Bjorn, great Video,
I heard that the Model 3 best charging efficenty is at 3 phase 13 amps - 8.9kW. Perhaps you do another test at this speed.
The 8 kW test was quite close.
Yes, the same observation I did have. Lower stress to the phases (also in house), increases efficiency.. But, all in all, it depends on other conditions as well. What I miss, is that I can adjust the amps with the app, by smartphone. Hope, that it is coming soon
@@bjornnyland yes i saw, but at 1phase 30 amps, the car uses 2 charger units out of 3 and load them with 15 Amps each. When you charge AC 3 phase with 13 amps, the car uses the 3 charger units witch 13 Amps ant the efficenty is better. Greetings from Germany
When charging on AC there is an AC-to-DC rectifier in the loop, which has an inverted bell-shape efficiency curve. Look at PC PSUs, at low power they have poor efficiency, peak out towards 50% of max power and drop a bit at max power. An example here: www.anandtech.com/show/15950/the-nzxt-c650-650w-psu-review/2. The Power factor, PF correction or lack of, does not play a role in the efficiency game.
Battery resistive losses are easy to compute: say the 3 is 46 parallel and 96 series (googled it) , the pack shoult have resistance of about 37mOhm at ambient for 18mOhm cell resistance. At 7kW DC you get 17A which leads to about 10W of heat losses. At high power DC charging undoubtedly heat losses in the battery pack matter more, also since you don't have the rectifier in the loop any more. Does this make sense?
In the new Leaf (from 2018), the screen shows kW from the charger or kW that goes directly to the battery?
Hej Bjørn, there is another source of lost, you didn't have attention to: how hot/cold were the charging stations, you used? I have heard that hot charging stations have up to 5% lost, which you will be get billed for. Can you please check if that is right og just a rumer? Thank you 😀
It wasn't extremely cold or hot during my tests.
AC single phase: Power is delivered between phase and neutral. That means two wires with losses. You only get one phase.
AC three phase: Power is only delivered between the phases. Nearly no current is on neutral. That means three wires with losses, but three phases too.
-> Three phase charging should be more efficient in general.
DC: The charger is in the cabinet, so most of the losses are not counted. The internal resistance of cable and battery is not that much.
Problem on M3: The afterburner with up to 7kW heat generation. You can't switch them off. Especially on low DC power (20kW) you only get about 13kW into the battery. You'll be charged with 20kW because this is what you get. It would be even worse if the AC-DC losses are counted.
-> It makes nearly no sense to charge on DC 20kW.
HPC: Don't think about. If you charge with 100kw and more you'll just give a fuck about the 7kW afterburner. Furthermore the current is that high, the battery will be warm enough very soon to charge without battery heating.
Humidity control perhaps? Most cars have a sensor
Yes, very interesting 👍
I always assumed slower charging speeds would have less loss due to heat so would therefore be more efficient, but as a counter argument I did always wonder if the losses being spread over a long period of time perhaps hurts efficiency , sounds like there is not a simple answer as the results varied based on a number of factors. I wonder if the lower speeds work better on non Tesla cars which might use less energy when charging? I always try to charge quite slowly at home to maximise proportion of solar energy from by roof, now i'm worried i'm maybe losing some efficiency in doing so
4:26 I disagree. 3.5kW charger was less efficient because it had to bump the voltage to 400V, if the input were 400V like 11kW, then chargeing at 3.5kW would be a bit more efficient due to the lower current. But I suspect it will be not appreciable(unsure)
How do you explain high efficiency at 230 V 30 A then?
No, the car only uses three 230V chargers when charging on 3 phase, not 400V. The 400V are only between phases.