You might consider charging slower

[markwj]

The Leaf also has a finish by X setting. An algorithm shouldn't be particularly hard.

1. The driver sets minimum start time (for TOU), stop time, ideal SOC, and ideal amps.

2. The car starts charging at the minimum start time and calculates when the finish will happen based on ideal amps and ideal SOC.

3. If the calculation in #2 is prior to the stop time, ideal amps are reduced (or increased if it appears that it won't finish in time).

4. At regular intervals the car checks the SOC and remaining time and adjusts the amps as required.

5. Car stops charging at the stop time.

FWIW, I've normally charged at less than the maximum--more to prevent overheating of the UMC than anything else. After the recent discussion about cell equalization, I reduced that to 16 amps. I'll see how it goes.

We've been working on something like this for the roadster, in the OVMS project. Looking at thousands of charging logs, from hundreds of cars, in a wide variety of environments.

The ramp-up and main charging parts of the curve are relatively easy to predict, with the variable being ambient / pack temperature (primarily accounting for cooling losses). The issue is predicting the ramp-down tail at the end of the charge - we are seeing a lot of variance there. The car controls that based on a bunch of factors hard to model (and even harder to predict). Changing charge limit at that point will have zero impact (as the limit is being controlled by the car during ramp-down). But, getting to within 30 minutes seems to be feasible, so that is what we are trying to do.

 

[boilerbots]

Just noticed this comment. Can you explain how my data looks fishy?

As for Tom - he knows more about Roadster than most anyone here. I think fishy is a strong word to use.

It is real data, with logs to back it up. Not just idle speculation.

I remarked that Tom's data looks fishy. I would like to understand the details about how you and Tom took these measurements.

When I quickly scanned through your thread, I could not see any details of your electrical system, the type of meter you used, the layout of wiring in your home or how you data was collected. All I remember seeing is that you where measuring watts using an external meter and the car's and you logged the data. You also where rightfully concerned about cooling and the ambient air temperature. Your data seemed to stretch many days of collection.

Can you give me the make and model of your power meter.

You said that you have a dedicated feed for your charging. If this is after your service meter, how long do you estimate the wire run and what is the gauge and type of wire?

According to Wikipedia, the roadster has 53kWh of capacity and can charge at up to 70A @ 240V equivalent to 16.8kW.

Also, in Hong Kong you have 50Hz power, we have 60Hz.

Can I see your daily logs vs. the summarized data? Perhaps during the short 32A chargers there was something strange happening, for example what if cell balancing kicked in and wasted a ton of power on one of those few days and never happend during the 70A charge. If the daily logs are available I would like to compare day by day efficiencies vs a lump sum.

To put some numbers up I would at least need a logging true power meter (that logs voltage, current and watts) and a logging volt meter. I would place the logging volt meter at my service entrance and the logging power meter as close to the car as I can get it. Then I would need a dual charger Model S for a complete study, I only have a single charger. It would also help to record the ambient air temperature and what the vehicle's cooling/heating system is doing.

Since the Model S will pull power from the wall when connected to prevent battery drain, you would need to analyze the data logs and extract the power consumed only when charging is active. During those times the climate control should not be activated.

Once everything is in place, drive a lot, log a lot and analyze the results. Repeat with different current limits.

This would be a lot of effort and some expense to put this together. I have better things to do with my time, I can't even believe I started this thread in the first place. Like I said in my first post, "The information above is free. Take it, or leave it I don't care." I can also stick to the general principals of physics and say that power=current^2 x resistance. The resistance in the "system" is basically a constant. So, your power losses due to resistance increase exponentially with current.

 

[markwj]

I remarked that Tom's data looks fishy. I would like to understand the details about how you and Tom took these measurements.

The actual quote was 'too' - ie; both our data.

To put some numbers up I would at least need a logging true power meter (that logs voltage, current and watts) and a logging volt meter. I would place the logging volt meter at my service entrance and the logging power meter as close to the car as I can get it.

Once everything is in place, drive a lot, log a lot and analyze the results. Repeat with different current limits.

This would be a lot of effort and some expense to put this together.

That is essentially what I did, and yes it was a lot of work and discipline over several months. I am glad I did it, as it gave us a lot of insight into real-world behavior. It would be great to see this done with the Model S.

Note that while I see more voltage drop on 70A charges than on 13A charges (dialing the current down on a 70A circuit), I see even more voltage drop at 13A on a standard household 13A circuit. The reason is that the 70A circuit has less other equipment on it and better wiring. My 70A wiring run is with wires rated to 100A and only about 12' from the utility feed.

My conclusions were that it is not as simple as charge slower. If it was, then a 1A charge would be more efficient than a 80A charge - which it is not. While I agree that the charge system electrical losses are in general greater the higher the current, what you are missing is that there are also fixed losses during charging that depend on the duration of the charge (primarily pumps, computer systems and HVACs) - the longer the charge the greater those losses. In my experience, such resistive losses are less than battery conditioning system losses, but beyond 32A the benefit tails off.

As you say, take it or leave it - but that is the result of the biggest study into this.

 

[stopcrazypp]

I can also stick to the general principals of physics and say that power=current^2 x resistance. The resistance in the "system" is basically a constant. So, your power losses due to resistance increase exponentially with current.

You have to take into account the powered delivered is linear with the current. That means percentage loss = (I^2 * R)/(I*V) = I * R / V, which means as a percentage it only goes up linearly. Plus there are also fixed losses that make transformers have higher losses at low loads, although the I^2*R losses you mention contribute to more losses at higher loads.

The fixed loss from air conditioning is going to be extremely important to determine overall charging losses. Air conditioning load is going matter more at lower charging rates than at higher ones.

For both cases there will be a "sweet spot" for efficiency because of those fixed losses, so I don't think you can say for certain lower power definitely is more efficient.

 

[Dien Kim Nguyen]

There is one very important fact that everyone seems to forget: higher current means more stress on the battery. Over time this stress leads to fracture of the graphite particles which hold the lithium -- hence, battery degradation. This is the same reason that when one uses SC -- the current slows down dramatically to prevent this kind of stress when about to reach capacity. When there isn't a good reason to charge fast, one shouldn't do it.

 

[markwj]

There is one very important fact that everyone seems to forget: higher current means more stress on the battery.

Looking at charge profiles, this impacts most at high levels of SOC - which is why the car tapers off the current towards the end of the charge. Tom's work on the roadster showed the ramping down of charge currents towards the end of the charge means that no matter the current limit selected, all are the same towards the end (for example, a 70A charge would ramp down and pass the 40A level at around the same point that a 40A charge starts to taper off).

Tesla Roadster Charging Rates and Efficiency - Tom Saxton's Blog

A dual charger, maximum 80A @240V is around 20kW. About 1/5th the rate of a supercharger (and they are proposing to increase the supercharging rate even higher - figures I hear are 90kW to 100kW today, with 120kW coming). So, how conservative should we be? 80A is five times as conservative a rate as the superchargers. 40A is ten times as conservative, and 20A is twenty times as conservative.

 

[boilerbots]

The actual quote was 'too' - ie; both our data.


My conclusions were that it is not as simple as charge slower. If it was, then a 1A charge would be more efficient than a 80A charge - which it is not. While I agree that the charge system electrical losses are in general greater the higher the current, what you are missing is that there are also fixed losses during charging that depend on the duration of the charge (primarily pumps, computer systems and HVACs) - the longer the charge the greater those losses. In my experience, such resistive losses are less than battery conditioning system losses, but beyond 32A the benefit tails off.

I am not missing anything. My post clearly talks about this under vehicle losses, cooling pumps, fans all add up. Can you charge without active cooling, seems possible, do I sleep in my garage every night to verify what is happening, nope.

When I charge at 40A I noticed the fans where running in the radiator and the charging cable was very warm.

Since I dialed back to 20A I haven't heard the fan running but I hear a noise that sounds like a very small circulation pump. It might run at some point.

Randomly when I check the vehicle the computer screen is on and I hear the pump noise but this is outside of charging, seems the car sometimes just wakes up and then goes back to sleep around 5 minutes later. This has nothing to do with charging.

My feed is more than 280 feet. First I have about 220 feet of 2/0 aluminum wire from my 200A service entrance meter to my main panel. Then I have another 60 feet of 2/0 aluminum to a sub panel, then a few feet drop to the 14-50 outlet. Unfortunately my house was built when 2/0 was allowed for 200A service, now they require 4/0 aluminum. Anyway this is acceptable for 150A service. What matters is the resistance per foot. If I take Southwire's value of 0.176 ohms per 1000 feet, and take my approximate length doubled, since their are two conductors I get 0.09856 ohms total. If I could charge at 80A, that would be 631W lost to wire resistance. If I charge at 20A, that is 39W. Multiply by charging times, lets say I need 20kWh every night to recharge (power delivered to the car). That equates to 20000/(240*80)=1.042 * 630.784 = 657.1Wh vs 20000/(240*20)=4.167*3.424= 164.3Wh every charge cycle. For me this is a delta of 492.8Wh every day, times 30 days is 14.8kWh lost every month to resistance in my electrical wiring if I where to charge at 80A vs. 20A.

In your data summary I saw 66.7% vs. 64.3% efficiency difference, so 2.4%. In my above example 2.4% x 20kWh is 480Wh saved however my wiring would have wasted 493Wh because of the increased power dissipation. Theoretically this is a wash if these efficiency numbers hold up for the Model S, also I am mixing 70A, 32A vs 80A and 20A in my example.

The big unknown is what is the charging efficiency of the Model S under every possible scenario?

 

[jaanton]

I look at voltage drops together with current. With my utility smart meter, it shows voltage and current. My Roadster shows voltage and current. I have The Energy Detective (TED) installed so I get detailed graphs of voltage and power.

Typically, my house voltage idle is around 248 v. With the car plugged in but not charging it reports the same. If I charge at 40 A my voltage at the car drops to about 244 v and at my meter is about 246 v. So I figure those 40 A cost me 2 v each before the meter and after. 2 * 2v * 40A is 160W of power lost to wiring, breakers, outlet connections and not going into the car. (Although the before meter loss isn't something I'm paying for! Shortly after I got the car, my utility replaced my power drop and that improved my voltage. The utility wasn't dumb. They knew my old drop was going to cost them when I start using serious power. A few months later they even added another transformer on my street. Definitely tell your utility you have a Tesla!)

When these detailed measurements are taken, ambient temperature should also be logged. When I do daytime charging in summer, the car cooling system comes on several times but that cooling system runs off the power into the car which is power not going into the battery. At night with the much cooler environment less active cooling is needed. The need for active cooling will probably sap a few percent and will be environmental so where you are in the world will make a difference.

Copper resistance is also temperature related. If you run your current at your wires capacity, it may get warm and probably have more resistance then when cold. You can see this with a meter on a toaster oven, initially maximum current flows and as the oven filament heats the current will drop a bit. I like to back off of full current just on the principle of not pushing the wiring to its limit.

 

[Merrill]

This seems like a silly question but I would like to know if my thinking is correct. The question is that if you charge at a slower amperage and it takes longer are you using the same amount of electricity as if you charged at a higher amperage and finished In less time. I assume that either scenario uses the same amount of kWh.

 

[pilotSteve]

I thought the 'range increased by slow charging' was kind of bogus, but I gave it a try. The last two 12 hour charges (240V@5A) resulted in my rated range increasing from 209 (where my slider ended up and its not been touched since then) to 212 and 214 today.

Maybe not such a crazy idea after all, thanks! Does make sense to me that slow charging will allow various banks of cell to stop charging before others hence leveling out all cells to a full charge due to long charge time.

Or maybe its just PFM (pretty fine magic) at work.

 

[boilerbots]

I bought a power meter to make some efficiency measurements and after only a few days I ran into a problem. I want to try and record the power used during charging and not just drawn from the wall when the car is sitting idle. To do this I unplug the car in the morning after charging is complete regardless of driving the car. I unplug the car on Saturday morning after charging from my Friday commute and the car sits the whole weekend. Monday morning the computer reports zero power used but the car is down 13 miles. I drove to work and home and then charged, Tuesday morning the car took an additional 4kW-hr of power to replenish those lost 13 additional miles but the computer trip meter does not report this in the totals.

So now it is nearly impossible to know how much power is drawn out from the battery to compare with the power that I measure going into the car unless Tesla makes a software change to fix this problem.

 

[markwj]

Boilerbots: nasty. On the roadster, we found the car display buggy and useless as well.

If you measure change in indicated miles (ideal miles?), you could at least get a comparison. For example, to add X miles uses Y kWh from the wall, repeat in different conditions.

 

[Cal1]

I just took possession of my car. We're retired and live in a pretty rural area. I anticipate many of our trips to our local town will be very short. On a given day I might put 30-40 miles and on others nothing. I wondered if it would be better to wait until the range has gone down to say 80 or 90 miles or still recharge every night regardless of how much range is left.

I realize we could be SOL if we suddenly wanted to go far but we typically plan our longer trips to bigger cities well in advance. (Still got ICE vehicles as a backup).

God I love this "wake up every morning with a full tank" thing! I really don't think anyone can really get it until they live with a EV. Doesn't take long but no amount of test drives or research will really bring it home.

 

[djp]

I anticipate many of our trips to our local town will be very short. On a given day I might put 30-40 miles and on others nothing. I wondered if it would be better to wait until the range has gone down to say 80 or 90 miles or still recharge every night regardless of how much range is left.

The charging screen has a slider that you can use to set a maximum charge level. If you're only driving 30-40 miles per day you could get away with dropping it to 50% and still charging every night, and bring it back up before you do a longer trip.

The battery will last longer by keeping it at a low state of charge and cool temperature, and doing shallow charging cycles.

 

[yobigd20]

Read this How to Prolong Lithium-based Batteries - Battery University

Your going to want to charge every possible chance to keep the charge cycle as shallow as possible to extend the life of the battery. The two worst things you could do are 1) discharge completely before recharging (thus making this a full charge cycle) instead of many small shallow discharge/charge cycles) and 2) Range charge to max and leave it sit there, especially in hot climates. Avoid those 2 things like the plague, and your battery will last a very very long time. Also, only really charge to what you need on a daily basis to keep your car around that 50% state of charge.

For example, if you only drive 50 miles daily, you could do something like set the charge limit to 60%. That should have you at about 159 miles range every day. Driving 50 miles would put you at 109 before charging again. That's about 41% battery (assuming 95kWh battery). So in this case you're always staying between 41-60%. AFAIK that's best case scenario. Try to always stay between 30-80% if you can. It's not bad if you need more or you drain more on long trips, but the shallowest you can keep your charge cycles on a daily basis would be the best you could possibly do on a daily basis to really make this battery last decades.

This is of course my opinion based on my own research and from talking to Tesla folks and other owners. There is no "official" guideline from Tesla afaik except for 1) plug it in every night and during long term storage and 2) don't range charge unless you really need it.

 

[wycolo]

> 2) don't range charge unless you really need it. [yobigd20]

Aren't you forgetting that the battery needs to be balanced on a regular, let's say monthly, basis? The phobia against Range Charging, like all phobias, can get overblown into a PROHIBITION such that it never is attempted. Might be better to adopt MicroSoft's old motto: "Where have you gone THIS MONTH (with your Tesla)?" and do a breakout of the tether, long range trip. [This is me talking to self here].
--

 

[AmpedRealtor]

> 2) don't range charge unless you really need it. [yobigd20]

Aren't you forgetting that the battery needs to be balanced on a regular, let's say monthly, basis? The phobia against Range Charging, like all phobias, can get overblown into a PROHIBITION such that it never is attempted. Might be better to adopt MicroSoft's old motto: "Where have you gone THIS MONTH (with your Tesla)?" and do a breakout of the tether, long range trip. [This is me talking to self here].
--

I disagree. I've spoken with multiple people at Tesla Ownership who have indicated that the "ideal" charge level for maximum battery life is 50%-62%. Ownership qualified the recommendation by saying gains would be "negligible", but that technically that is where the battery needs to be for maximum long term life. Of course those who think they know better than Tesla will quickly debate this and question the source, that unless it comes from an engineer the information is useless, blah blah. Yet nobody here is a battery engineer either, and many think they know better than what Tesla says or simply rely on the written marketing recommendation.

To each his own. We can come back and compare notes in a few years. This is a major capital investment, I'm going to do everything I can to take care of this battery. It's better to err on the more conservative side. One day when we go to resell our Model S vehicles, the battery capacity will end up being a critical factor in someone's purchase decision. If my car gets 25 miles extra range over another vehicle after 5 years, I consider that significant and more resale money in my pocket.

This is just what I have chosen to do for myself based on recommendations from those at Tesla who know much more about this car and its technology than I.

 

[liuping]

> 2) don't range charge unless you really need it. [yobigd20]

Aren't you forgetting that the battery needs to be balanced on a regular, let's say monthly, basis? The phobia against Range Charging, like all phobias, can get overblown into a PROHIBITION such that it never is attempted. Might be better to adopt MicroSoft's old motto: "Where have you gone THIS MONTH (with your Tesla)?" and do a breakout of the tether, long range trip. [This is me talking to self here].
--

Is balancing only done when you range charge?

 

[RodF]

I disagree. I've spoken with multiple people at Tesla Ownership who have indicated that the "ideal" charge level for maximum battery life is 50%-62%. Ownership qualified the recommendation by saying gains would be "negligible", but that technically that is where the battery needs to be for maximum long term life. Of course those who think they know better than Tesla will quickly debate this and question the source, that unless it comes from an engineer the information is useless, blah blah. Yet nobody here is a battery engineer either, and many think they know better than what Tesla says or simply rely on the written marketing recommendation.

To each his own. We can come back and compare notes in a few years. This is a major capital investment, I'm going to do everything I can to take care of this battery. It's better to err on the more conservative side. One day when we go to resell our Model S vehicles, the battery capacity will end up being a critical factor in someone's purchase decision. If my car gets 25 miles extra range over another vehicle after 5 years, I consider that significant and more resale money in my pocket.

This is just what I have chosen to do for myself based on recommendations from those at Tesla who know much more about this car and its technology than I.

Well said. Exactly what I am doing based on the low miles I need on a daily basis. Charging at 60%, using to 40% and plugging in at the end of the day. No hassle, and if happens to prolong the life of my battery, what the heck.

 

[wycolo]

> Is balancing only done when you range charge? [liuping]

Yes, I believe that has always been the assumption. And that means a 'full' range charge which means going ON for a few minutes then OFF, over and over until finally all segments of the pack are equal. Quite a ruckus just before a long trip if the contactor is clunking away under your bedroom while you are still trying to sleep. :frown:

I don't think TM Ownership is going to sway R & MS owners *not* to range charge on some kind of regular basis; this would be for Engineering to do. Keeping a balanced battery is rather basic. My R gets charged only every few weeks and I finally did a full range charge. Might have been the only one it ever got (over 7k miles on clock). Result was 235 miles, whew, which is very good. I plan to do full range charges every 60 days on the MS if long trips don't require this beforehand. Staying around 50% SOC is fine, but doing balancing is a separate requirement.
--