evse recommendations?

This is exactly what I do. I think the Tesla connectors are a great value (high power to cost ratio), and having one plus a J-1772 adapter means I can charge anything (except CHAdeMO, which is a standard I hope dies, anyway). We also own a Tesla, and actually have two Tesla HPWCs linked together, sharing a 60A breaker.

See this post and the one before it.
https://www.insideevsforum.com/community/index.php?threads/evse-recommendations.8985/#post-105001

 

A friend brought his Tesla over for a test charge on my Juicebox.

Worked like a charm. Pulled about 31a.

Soooooo ready!

 

Actually, a question.

My friend with the Tesla only charges to 70% SOC. Something about damaging the batteries if charging to 100%.

Is that a thing? Is it a thing with the SE? Or do you all just charge all the way to 100%?

 

BMW says you can charge to 100%, but this runs counter to most EVs. They do have some buffer built in so that the battery is never truly at 100% however all batteries lose capacity over time regardless of how much you charge the battery...so at some point the buffer will be gone, or the capacity that’s available will go down. Not sure what kind of analysis anyone’s done on the 2021 models to monitor battery degradation, but unfortunately BMW has not made this clear, probably intentionally.

 

The amount of thought and bickering in the Tesla community over the proper charging daily max is astonishing. After hearing all the arguments, I charge mine to 80% daily, then every few months to 95% and then let it sleep for 3 hours to balance the cells and calibrate the GOM, and 100% with no sleep time if I'm about to depart on a road trip.

But, there's something to be said for MINI's approach: "We'll set the top and bottom end for battery longevity; you just plug it in and let it fill to full."

 

Up front, word is to charge at will - no limits.

For in depth discussion, our very own Inside EVs gurus have a few words:

Out of Spec Motoring: MINI Electric Road Trip

InsideEVs Weekly Podcast: MINI Electric Road Tested

 

Actually, in the MINIs case it is fine. We have quite a large buffer at the top. Charging to 100% is really like charging to 85% (ish). The benefits of charging to 85% actual (100% indicated) and cell leveling (where all cells end up the same voltage) far outweigh charging to 80% actual and not leveling. The buffer is also designed to minimise the battery deterioration over time. We can drop 10-12% on the battery before even noticing a change. The MINI is based on the I3, but with more modern batteries. It doesn't have any reports of poor battery longevity.

For those wondering what cell leveling is, if all cells (batteries) are at the same voltage, they all provide the same power to the system. If a battery is at a lower voltage, it has to work harder to "keep up" with the other batteries and wears quicker. A Tesla gives you complete access to the top of the battery, and they suggest an 80% charge. It is swings and roundabouts. We pay for 32.6 and get 28.9, but the battery will have a gentler life, whereas others get what they paid for, but have to be careful not to harm the battery.

 

The buffer does not appear to be entirely at the top. There are two points of evidence for this: First, the battery cells are charged to 4.15V, which is pretty close to the commonly accepted Li-Ion maximum charge voltage of 4.2V. There is rarely more than 5% of capacity between 4.15 and 4.2V in my experience, but of course it depends on the exact chemistry. Second, the car reports two battery SoCs over OBD. One value is the SoC % reported on the user-facing displays, and the other value appears to be the true SoC % of what MINI considers to be the full battery capacity. They both start at 100%, and the user-visible SoC drops faster than the true SoC. That would indicate that the buffer is entirely at the bottom. I'll try to remember to do the math next time I discharge the battery to figure out how much extra % is at the bottom.

Also it seems unlikely that they would use the buffer to hide battery degradation. That would mean to keep the usable capacity the same as the battery degrades, the BMS would have to increase the maximum charge voltage and/or decrease the minimum discharge voltage, which would cause the battery to degrade even more rapidly.

 

Interesting info fizzit. I tried finding information on the MINIs cells. There is precious little. All I could find is 12 CATL modules for a total of 350.4V nominal, 93Ah and a weight (including harness/wiring etc) of 200kg. This would indicate each module has 8 cells (or several in parallel) of 3.65V nominal in series (eg. NCM 811). This gives 8*12*3.65 = 350.4V. The nominal charging voltage is 4.2V with a 4.25V maximum. Voltage to SOC is actually notoriously difficult in lithium batteries as it is non linear and depends on many factors. If, however, the batteries behave like other (smaller) batteries, 4.1V represents around 87% SOC, based on 100% at 4.2V. So if 4.15 is correct, you would say it is somewhere between 6.5 and 10% down (due to the non-linearity of the voltage drop). This is less than the 12% the battery doesn't have access to, so it would show a bottom buffer of between 2 and 5.5%.

When I said hiding the degradation, I mean from the user, not as in VW hiding results. You are correct, in that you can keep charging to the same voltage as the battery degrades - well until it can't physically reach the 4.15V per cell. One benefit of a large(ish) top buffer is that you still have regen at 100% charge. A Tesla owning friend of mine complains bitterly about this when he road trips and charges to 100%. Until the battery drops a bit, you can't regen. Tesla doesn't have a top buffer, but suggests only charging to 80%. Luckily the MINI is quite happy to use regen at "100%" as it is not 100%

 

Yeah, mine shows 398-400V hot off the charger for a cell voltage of slightly over 4.15V.

As for degradation, my thought was that MINI probably doesn't hide it from the user by reducing the buffer size to keep the usable capacity the same, because that would require changing the max/min voltages and damage the battery further. They can "hide" it by recalibrating the SoC % meter, so the car doesn't die before you hit 0% even if you have a degraded battery, but the degradation will still be betrayed by the reduction in range and capacity. Hopefully it will be a while before any of us notice that. The MINI reports an SOH (battery health) % over OBD, which is still at 100.0% in my car after 8k miles. But it also reports a battery capacity of 99AH so I'm not sure if the app I'm using is reading all the codes correctly LOL

 

That is very interesting. The only info on CATL cells I could find were for 50Ah, 3.65 nominal cells. If a module was 8s2p, that would fit nicely. Nominal voltage of 350.4V, 398.40V at cell voltage of 4.15, 100Ah pack and 163kg (just the modules) so 200kg with all the gubbins that goes into making the pack (wiring, cooling, harness etc). That would mean the MINI's battery is actually a little bigger than the advertised 32.6 and it probably has a somewhat bigger buffer. I wonder if there were different packs and that is why some people get significantly more range than others. I am hard pushed to go below 125miles (200km) no matter what I do, and regularly get 155miles (250km).

Anyway, I'm not trying to hijack the thread - as per the original question I think we can now safely say that charging to 100% is perfectly OK in the MINI.

 

@fizzit I think you should start an OBDII thread. You've inspired me so I'm going to get an OBDII scanner to see what can be seen. Do you know if the data you are getting is the same data that MINI uses to determine when a battery qualifies for a warranty replacement?

Autel was persuaded by an Inside EVs Honda Clarity forum member to create some beta software for their cellphone-based AP-200 OBDII scanner that reads special battery data for that car. The unique thing this beta software does that no other low-cost scanner could do (but other units can now finally do) is read the Battery Capacity number Honda uses to determine warranty battery replacement. However, because this beta software has caused Autel many support issues, I doubt they'd do it for the SE, too.

 

No, no. Keep it coming! Mongo like science.

P.S. I confirm the Tesla's regenerative braking is increasingly degraded above 85% or so; it's also temperature dependent, too.

 

Love all the discussions here about battery life - learned a lot! And sorry to digress from it, but does any of you know the length of the EVSE cable that comes with MINI? Thanks!

 

Charge and regen limits aren't just a Tesla thing. VW, Audi, Porsche and others use them as well. It's the best way imo to utilize the current battery tech to its fullest, while also being protective of the battery.

If the MINI is anything like other EVs, the majority of that ~12% buffer is at the bottom. Reason being, you run the pack down to 0 volts it's really shot. The Audi e-tron (known for a generous buffer) has about 1% locked away at the top (my older model has about 4%).

I'm all in favor of a charge limit, especially with how efficient this thing is. A typical trip might take as little as 3-4% battery. It's kind of an enthusiast option in a way.

 

The yellow cable is ~14', the plug & base are ~2'.

 

Thank you!!

 

Okay, I checked it out, and at 91.3% displayed SoC it shows 93.31% actual SoC in the app. That means that if the displayed consumption is (1 - 0.913) = 8.7%, and true consumption is (1 - 0.9331) = 6.69%, the actual battery capacity is (8.7 / 6.69) = 30% more than the displayed capacity, which seems crazy high. Perhaps the two percentages don't track each other linearly like I'm assuming. Otherwise we're carrying around 30% of dead weight. Even if the battery capacity is actually 100AH at 350V for 35kWh nominal, that's only a 21% buffer over the stated usable capacity of 28.9kWh.

 

Either I am very confused, or those numbers are way off. Check out this article comparing battery buffers.

https://pushevs.com/2020/04/04/comparison-of-different-ev-batteries-in-2020/

efficiency losses in charging?

 

It see,s unlikely that it is truly linear. A simple way to check is to take several measurements across the range.