First CCS experience

[bwilson4web]

Hi,

Huntsville just got a high-speed, DC charger up and I visited it today with 91% SOC:

• Single charger, dual head: CHAdeMO and CCS.
• Two parking spots, next time, the front one to keep the plug-to-charger short.
• Parking meter, 30 minutes for $0.25, I put in two quarters.

The CCS plug is pretty large and the power cable at least 1" (25 mm) diameter but there is a top-mounted, steel cable that keeps it from drooping to the ground. I pretty well had to stretch it as far as possible. Here is the plug answering some questions about the contacts:

So everything ready, I hit "START" and after the initial negotiation, 17 A @392 VDC, 6.7 kW???? It was ~75 F (~23 C) and I'd only been doing local, city driving. This bothers me as I don't know if the car or the charger was the limit:

In contrast, my L2 EVSE easily starts at 7.22 kW, 30 A @240 VAC. If I use a nominal 92% efficiency for the onboard chargers, I get 6.7 kW. HUH???

Now it appears the charger is configured to run for only 30 minutes at a time, $0.25. Not a big problem as there are restaurants on the other side of the block. So as I was monitoring the SOC with my iPhone, it stopped charging a 99%:

I'm left with two hypothesis: (1) charger configuration limitation, or (2) BMW i3-REX limitation. But I have to admit disappointment. I thought the car would charge at something much higher than 6.7 kW. The claims of reaching 80% SOC within 30 minutes can only be reached if the car can take a higher rate.

As a test, I plan to run the SOC down, say 50%, and see if that changes the rate of charge. It may be that the peak charge rate requires an SOC below 80% and then taper to 6.7 kW for the remaining time. I am OK with this but I need to know.

Thanks,
Bob Wilson

 

[epirali]

Part of the issue is that with high speed charging the fast rate is only used until 80% of charge, then the charge slows down to not damage the battery. I believe you started at 90% charge and never really used the full rate.

 

[I33t]

+1

Go back with a near empty battery and notice the difference!

 

[TomMoloughney]

Admin Note:

Thread updated to correct the name of the charging system used. It's CCS, not CSS. CCS stands for Combined Charging System. I didn't want new comers here to get confused. We already have too many "standards".

Tom

 

[bwilson4web]

Thanks Tom. I'm trying to resist calling it the Franken-Plug (i.e., "New Dual Combo(CSS) And CHAdeMO Charger".) I don't have the resources to test the CHAdeMO part but I know a lady at work with a Leaf.

Second visit:

3:34 PM @37.8 kW

• 63.5% = (86.5% - 23.0% ) # Change in battery SOC
• 11.9 kWh = 18.8 kWh * 63.5% # Charge put into battery
• 12.9 kWh = 11.9 kWh / 92% # Estimated input power to DC charger
• $1.29 = 12.9 kWh * $0.10 / kWh # Estimated cost to charge
• ~2.1% SOC / minute # Measured charge rate, need full record!

No parking meter enforcement after hours and weekend.

Bob Wilson

 

[jadnashuanh]

Depending on battery temperature and current SOC (and the CCS unit's capacity), the i3 can accept up to about 50Kw/hr. Not all of the power going in actually charges the batteries. You might see more momentarily, but not sustained.

 

[bwilson4web]

Depending on battery temperature and current SOC (and the CCS unit's capacity), the i3 can accept up to about 50Kw/hr. Not all of the power going in actually charges the batteries. You might see more momentarily, but not sustained.

Agreed but in the meanwhile, I want to run a few more tests.

Since the charger display shows the progress, I'm thinking about using my iPhone 5s in time-lapse mode. Based on my last test, I'll probably need two, sessions, as the 63.5% gain is not enough to fully charge the car.

I'll probably run the SOC down to 6%, automatic REX enable mode, and then drive to the CCS. I'll park the car and let everything cool for an hour before starting the test(s). The goal is to charge from 6% to 99% with time stamps.

Bob Wilson

 

[jadnashuanh]

The last 20% may not be much faster than an EVSE.

There are some newer tech batteries that have shown faster charging rates in the labs, but they are not going to become available for awhile yet. BMW has said their next battery will have about 2x the original capacity. Whether they'll change the on-board charging significantly or just require longer charging times will be to be seen. Many people, at least at home, will not be able to support significantly faster EVSE units, and the current public network certainly doesn't. CCS units could as the standard evolves.

 

[bwilson4web]

The last 20% may not be much faster than an EVSE.
. . .

That is what I expect to find. My first test at 91% SOC showed a starting charge of 6.6 kW which is consistent with an L2 EVSE. The second test suggests this is what might have happened.

The second test started at 23% and ended at 86.5% and started at 37 kW. If it had sustained that rate, 37 kW * 0.5 hr = 18.5 kWh. Had that been the rate through the end, we would expect to find 18.5 / 18.8 ~= 98%. Solving a set of linear equations should give us a reasonable estimate of when the rate dropped:

86.5% * 18.8 kWh = 16.3 kWh
23% * 18.8 kWh = 4.3 kWh
16.3 - 4.3 = 12.0 kWh # charge

n = hours charging
37 kW * n = A kWh
6.6 kW * (1 - n) = B kWh
A + B = 12.0 kWh

(37 * n) + (6.6 * (1-n)) = 12
37n + 6.6 - 6.6n = 12
30.4n = 5.4
n = 5.4 / 30.4
n = 0.17 hr ~= 10.2 minutes

Check:
37 * .17 = 6.29 kWh
6.6 * (1 - .17) = 5.4 kWh
6.29 + 5.4 = 11.69 kWh != 12.0 kWh

This suggests there may be more than just two rates, 37 kW and 6.6 kW. This can be resolved by a time-lapse recording of the charging. However, I may need to include a factor for the 0.5 hr charge interval but that defeats me for now.

A spreadsheet model seeking a solution is another approach that can give a hard answer and perhaps reveal my mistake.

Bob Wilson

 

[epirali]

There may be many factors, including battery temp as you charge. Also some high speed chargers will not provide full rate if a second car is plugged into the second position

 

[jadnashuanh]

The ambient temperature affects not only the car's batteries, but the ability of the CCS unit to perform. Consider that they are a pretty significant high-voltage DC power supply that is not 100% efficient. They can get hot. Throw in unknown condition of their inlet air filters for their cooling circuit, and they may need to turn themselves down in current to survive, regardless of what they ultimately can produce or the car can accept. BMW was touting a 24KwHr CCS unit because it was less expensive to purchase, required less infrastructure to site it, and didn't really impact the overall charging rate all that much given the heating of the battery pack. Putting in at a slower rate means it can probably keep at that rate longer whereas a higher powered unit would have to throttle down to keep everything within temperature limits.

Unfortunately, for me, this is somewhat of a mute point as there's only one CCS unit that's convenient, and if I were there, I'd just go home and charge (barely a mile away)! Plus, I don't like their cost structure. There's a free one at a WHole Foods about at the end of my range, but if for some reason, it was inoperative, I'd be stuck for hours trying to find an EVSE I could use to then get back home. As a result, I've never tested my CCS input. I use my i3 as designed, for running around the city.

 

[bwilson4web]

Sad to say, technical difficulties with the iPhone led the recorded video to be unusable. It focused on the wrong point and the metrics were unusable. However, I did find the output is limited to 100 A. So the 375 V @100 A ~=37 kW.

The gross numbers were 7% start and 83% end on the first 30 minutes. The second 30 minutes reached 95%. By the time it was up 91%, we were in 6-7 kW range.

Tuesday is another day and I'll try it again. This time I'll do some field, quality checks to make sure we're good to go. I'll also rig up an aux charger for the cell phone to make sure it doesn't run the battery too low. Always something.

Bob Wilson

 

[michaelbmw]

Running up to high levels of charge on a DCFC is inconsiderate to other users. I can understand trying it once to evaluate performance, but in the time it takes you to charge from 80 to 99%, another user could probably charge from 0 to 80

 

[bwilson4web]

Running up to high levels of charge on a DCFC is inconsiderate to other users. I can understand trying it once to evaluate performance, but in the time it takes you to charge from 80 to 99%, another user could probably charge from 0 to 80

Agreed. My test was between 2:00-3:30 PM and I didn't see any other EVs. I work 6:00 AM to 2:00 PM.

Most of what I want for a charge is in the first 30 minutes but insights are limited by the 100 A charger capacity. The second 30 minutes is to further understand the charge taper. If someone else shows up before I start my second 30 minutes, I can move and let them get their charge. Then return and complete the experiment. But the last of the first 30 minutes and the whole of the second 30 minutes gives insights.

In particular, I'm interested in the transition zones between this 100 A DC charger (37 kW); L2 AC EVSE (6.6/7.2 kW), and; L1 (1/1.3 kW). This is only a rough order of magnitude:

• 0-60% - high power DC charger, limited by 100 A charger @355-400 VDC
• 60%-90% - L2 charger, 30 A @240 VAC
• 90%-100% - L1 charger, 12 A @120 VAC

I also want an accurate charging profile to measure how much charge was put in the battery. I will also run the car down to the original SOC and use the tripmeter to measure the discharge Ahr. These will bracket the battery capacity.

Hummm, on reflection, I can stop the charge at say 95 A. The duration tells me the best/worst case use of this charger. Then start a second, 30 minute charge to see how close it comes to 100%.

Bob Wilson

ps. I saw 'traffic enforcement' come by twice and give my setup a close look. No problem, I'd already put in enough quarters to be legal.

 

[bwilson4web]

This time I was able to record both, 30 minute charge sessions. I combined them and generated these plots:

More later,
Bob Wilson

 

[bwilson4web]

FYI, our local DC charger only brings the car to 97% +/- 0.5%. In contrast, the L1 and L2 chargers using the built-in AC-to-DC converters achieve 100%.

UPDATE:

With 'iRemote' broken, I can't tell the SOC until I get back to the car. Fortunately, I can use the clock:

• L1 @12A - ~4% SOC per hour, taper delay is insignificant.
• L2 @30-31A - ~20% SOC per hour, taper is evident in the last 15-20 minutes.
• Fast-DC @100A - below 65% SOC, it adds +50% in 30 minutes. Above 65% SOC, the taper seriously limits so 71% only went to 91% in 30 minutes at lunch.

Bob Wilson

 

[psquare]

I've used many CCS chargers over here in the UK. On long distance journeys, I try and charge with around 15-25% SOC left, regulated using the REx. I usually stop for 25 minutes, which brings the charge back to 85-90%. Charging speed tapers off after about 75% SOC. After that it's like a Type 2 charger. To charge using CCS from say 15% to 100%, it would take me usually around 45 minutes.