Charging the Roadster

[Palpatine]

For anyone considering an extension, I would recommend something more like 25 to 30 ft. It's a trade off between weight and space in the trunk versus coming up just a little too short and not being able to charge. For me anyway, charging takes priority!

The Roadster Foundry MC is 18 ft in length.
The pigtail adapter adds another 1 ft.
Adding an extension cord of 15 ft gives you 34+ ft of length with everything combined.

The cable coils nicely around the base wall of the trunch. So you shouldn't loose much packing space if you plan it.

 

[Tdave]

I chose the do it yourself route. It is a simple task for anyone capable of firmly tightening a few screws.

Best price for the connectors I found are here.

Female Connector for $37.95

Male Connector for $29.95

These prices are about 1/3 of my local electrical suppliers prices.

The cable required is SO 6/4 which was $ 2.18 a ft.

Total parts cost of $134 for a 30 ft. extension.

The connectors are very simple to wire. Just match color coded contacts to the color of the wire in the cable.

Don, where do you get 6/4 for $2.18 per ft? And how water resistant is the resulting cable with those ends you mentioned?

 

[vfx]

Eventually someone may introduce commercial recharge points with retractable cables built in. Although a double recharge point between each pair of parking bays would be simpler.

In the meantime, will Model S come with a recharge cable + reel as well as a spare tyre?

As I said here, having power running through a coiled cable is an incredibly bad idea. Coiling on the ground, in a reel, car or RV causes a heat buildup that will burn and melt the rubber coating to a crisp.

Perhaps the reel could be designed so it would not put out power unless full expended.

 

[Palpatine]

Don, where do you get 6/4 for $2.18 per ft? And how water resistant is the resulting cable with those ends you mentioned?

If you buy it 1,000 feet at a time you can get the price that low per ft.
Most of the stores only sell it 100 feet at a time and it is closer to $3 to $4 per ft for the cable.

 

[donauker]

Don, where do you get 6/4 for $2.18 per ft? And how water resistant is the resulting cable with those ends you mentioned?

Yale Electrical Supply Co. here in Lebanon, PA sells the CCI Royal SOOW 6/4 cable for 2.18 a ft. cut to any length desired. This rubber coated cable is oil and water submersible. The listed connectors are the same rubber gasketed Marinco Connectors used on your RFMC.

 

[SByer]

For those not interested in building-your-own, I found pre-assembled 50A extension cables here (StayOnline.com) at a reasonable price. The cable is noticeably thicker than the cable for the RFMC, even though it's still just 6/4 inside (assuming just thicker insulation) - I got the 10 ft. length, I would imagine the longer lengths would be quite a handful. For me, the 10 ft. extension is just enough to eek it's way inside to the drier plug, and when I (finally) get some form of 240v outlet in the garage, will make that process easier than without it.

The connectors aren't quite as nice as the RFMC ones, but they still mate well.

 

[ChargeIt!]

Roadster Bugs

I thought we might have discussed "bugs" in regards to the Roadster before ... but a quick search did not yield a suitable topic. I was thinking of putting it under "It's the Batteries, Stupid!", but that did not seem right either. Moderators, feel free to move ...

And before you ask ... NO it is NOT a photoshopped image. It was taken thru the driver's window of a model 2008 Roadster charging this morning ...

Don't you wish you had a battery THIS big ???

 

[shark2k]

That is not necessarily a bug. From what I recall of more informed members here, the battery needs to be kept cool so more energy could be used. I would guess the VDS would also include that as the energy drawn because it probably wouldn't be able to tell.

I'm just recalling this from memory, so hopefully a more informed member chimes in and let's me know if I'm just pulling stuff of my a** or if I'm somewhat right.

-Shark2k

 

[SByer]

And at 110v/12A, charging is particularly inefficient. I didn't think it was that bad, though.

 

[ChargeIt!]

Sorry ... no offense, but wrong on both counts.

At 120V & 12A charging 121kWh would take 84 hours. This was an overnight session. The charging was from an HPC at 249V (actual) with current limited to 12A. There was no heating involved, and actually only a little cooling at the start of charging after a long outing. The previous highest estimate of a completely empty ESS charging to full in Range Mode (which this was not ... started at 67miles IR in standard mode) I remember hearing about is between 63-66kWh. And that's taking into account all of the cooling during hot weather and other charging system overhead. (ESS at 54kWh rating plus 20-25%).

In any case, what's wrong is the 121kWh ... even at 249V/12A that would take more than 40 hours.

 

[TEG]

Maybe that's not a bug, ... maybe it is an "easter egg".

1.21 Gigawatts and all.

 

[ChargeIt!]

Would it be legal to put the "Danger" sign in the back window ??? :biggrin:

 

[Palpatine]

I agree that charging at 120v/15a is very inefficient. I have been using a MC 120 for the past few weeks because I have to ship out our RFMCs as fast as we produce them.

The MC 120 is fine and I start every day with a full charge because I rarely go below 50%. But here is my observation.

My car started charging yesterday late afternoon at 93 ideal miles. I think it just finished. So I went from about 50% to full standard. The VDS is reporting 34 kwh. So clearly that is more than 50% of the battery pack. The VDS must be measuring the amount of energy it pulled from the wall.

So about 1/3 of the energy used during charging was wasted on:
1) losses during transmission
2) running fans early in the recharge process
3) running the coolant pumps during the entire process

Also I went and checked my car at 8 am this morning it showed 162 ideal miles. So I was thinking I had another 30 miles to go for a full charge.

60-90 minutes later I was in the garage and looked again. It was showing 192 ideal miles and finished charging. I was only expecting perhaps 170 ideal miles after such a short amount of time.

So I think what happened is the VDS lost track of the battery pack. When the BMS detected full charge and stopped the process, then the VDS updated with 192 ideal miles.

This could get very messy if you were doing a bunch of partial charges without ever allowing it to finish and balance the cells.

 

[TEG]

...
So I think what happened is the VDS lost track of the battery pack. When the BMS detected full charge and stopped the process, then the VDS updated with 192 ideal miles.

This could get very messy if you were doing a bunch of partial charges without ever allowing it to finish and balance the cells.

The same sort of thing happens with my NiMH Ford RangerEV. I think it is rather tricky business to track exact SOC when the vehicle is put on and off charge repeatedly without ever going between full and empty. Just reading pack voltage is not accurate. With the RangerEV, there is an occasional, factory suggested "reset procedure" to drain to empty then charge to full a few times in a row so that the computer has a better idea of pack health and SOC. Yes, messy.

I suspect Tesla takes this area very seriously and has been doing everything possible to make their VDS computer as smart as possible to make the best estimation it can. Still, there are a lot of variables, some of which may not be available as input to their calculations. One thing it has to do is figure out how pack capacity changes as the cells age over time.

 

[ChargeIt!]

Thanks for the comments and sharing experiences; some good points.

But a little more info:
This Roadster is VERY rarely connected to 120V. And the erroneous tracking of Charging Sessions seems to happen mostly at 12A and 16A. Almost never at 24A (or higher). In all cases Voltage is between 242V - 249V on a 70A capable HPC. Also in error (when unusual kWh's are recorded) is the amount of hh:mm tracked as having charged (illogical and impossible in some cases).

Another "error" that occured recently was a Storage Mode session which started on plug-in with the HPC limited to 12A at 27miles IR and self-stopped at 32 miles IR (when it was checked a few hours later). Tesla Motors has been contacted.

Another "bug": Opening and closing the charge port door -- without actually charging -- wipes out the previous entries on the Charging Session History screen. Any one else experience this ?

 

[Palpatine]

Why are you only charging at 12 amps with a HPC capable of 70 amps?

I know some people with an HPC keep it set at 40 amps instead of 70 amps to minimize heat during the recharge process. But 12 amps would seem to be too slow and perhaps inefficient.

I have always thought that 240 volts and 30-40 amps is the sweet spot for efficient charging. Not too fast (heat) and not too slow (inefficient).

 

[Palpatine]

But a little more info:
This Roadster is VERY rarely connected to 120V. And the erroneous tracking of Charging Sessions seems to happen mostly at 12A and 16A. Almost never at 24A (or higher).

That appears to be the common factor. 12A to 16A leads to inaccurate info on the VDS. This is likely because a larger percentage of the energy consumed is not actually going into the battery pack.

I recall hearing that there is a 6A overhead just to manage the charging process (fans and coolant pumps).

If that is roughly accurate, then at 12A approx half of the energy in the process does not go into your battery pack.
At 16A it is a little better.
At 70A the overhead to run the fans and coolant pump is minor.

Because of the length of time when charging at 12A the fans and coolant pumps could be running 24/7 with driving and recharging all the time, contributing to inefficiencies.

When recharging at 40A to 70A there is a large amount of time that your Roadster just can sit quietly without running any of the cooling systems.

 

[chronopublish]

I've done a lot of charging at 120v/16a and I am familiar with the range estimate "drift" during charging. The longer you charge, the less accurate the SOC estimate becomes, always underestimating the real SOC. If you let the vehicle sit for about 10 minutes without charging it or driving it, the VDS will update with a more accurate number.

In the most extreme example, I charged the car from "emtpy" (0 miles in Standard Mode, 25 in Range) for 24 hours once. When I came back to the car, the VDS showed only 64 ideal miles. I drove the car about 10 miles, parked it, and came back to the car an hour later. The VDS now showed around 120 ideal miles!

 

[ChargeIt!]

I've done a lot of charging at 120v/16a and I am familiar with the range estimate "drift" during charging. ...

In the most extreme example, I charged the car from "emtpy" (0 miles in Standard Mode, 25 in Range) for 24 hours once. When I came back to the car, the VDS showed only 64 ideal miles. I drove the car about 10 miles, parked it, and came back to the car an hour later. The VDS now showed around 120 ideal miles!

WOW ! (Thanks for that info.)

That appears to be the common factor. 12A to 16A leads to inaccurate info on the VDS. This is likely because a larger percentage of the energy consumed is not actually going into the battery pack.

I recall hearing that there is a 6A overhead just to manage the charging process (fans and coolant pumps).

Thanks ... interesting info. But James, you still miss the point. The car was charging overnight (about 14 hours) to full (192miles), reporting 121kWh used. That's not possible from several perspectives (time, ESS capacity, ...). And it reported it had charged for 20+ hours. The 14 hours actually makes sense (with little overhead) charging at 10 MPHC from 67 to 192, but it should have used 12*249*14 = 41.8 kWh.

 

[ChargeIt!]

Why are you only charging at 12 amps with a HPC capable of 70 amps?

Must have happened once accidentally and caused some funky data. Then it was done for debugging/testing.

I have always thought that 240 volts and 30-40 amps is the sweet spot for efficient charging. Not too fast (heat) and not too slow (inefficient).

Agreed. Personally I think 24A at 240V is a great place to be, especially in ambient near 80F.