Live data from Hacker News

That Tesla Data: What It Says and What It Doesn’t

wheels.blogs.nytimes.com

161–170 of 281 posts

Re: That Tesla Data: What It Says and What It Doesn’t

#161
post #102

Earlier quoted context omitted.

The battery's heating system likely drained the power overnight. The battery pack must be kept above a certain temperature or it will become irreparably damaged.

Do you have a reference for that? I ask because while I can imagine there are limits the kinds of things I've not seen one for Tesla's battery packs. I see that typical LiOn type batteries have a 'one month at -20C (-4F)' sort of limit, car batteries are typically better than that, good to -40F or so. Edit: This article http://electronicdesign.com/power/operating-conditions-get-t... says "permanent damage at -50 degr…

Mostly correct, but I remember standing outside my 5th grade classroom window waiting for the teachers to unlock the door. At that time the school doors were only unlocked prior to class starting if there were blizzard conditions, not for cold. The classroom had a thermometer on the window that read -50C at the time. The only reason I remember it was because it was the first time I realized -40C = -40F.

http://en.wikipedia.org/wiki/List_of_Minnesota_weather_recor...

Re: That Tesla Data: What It Says and What It Doesn’t

#162
I put a graph together showing all the points/counterpoints and here's what I'm left with.

- not charging past 90% at the 1st station is defensible. Especially if Tesla warns it will shorten the battery's life.

- It'd be better if Tesla's software took temperature into account when estimating mileage range.

- Broder is being ridiculous about the cruise control and his speed estimates, and the temperature. He has partial accountability for the NYT graph. Turning the temperature down to 70 does not mean "sharply downward". He turned it down once more at around 250 miles, but that's far after the 182 that the article/graph indicates, and markedly different than "a little over 200 miles". He writes about how his feet were freezing and knuckles were turning white, but he didn't turn the heat down below 70 until (if my math is right) right about the time he got to Manhattan, which was mentioned afterward in his article timeline. On the other hand, it looks like it was turned down for most of the leg between Manhattan and Milford.

- The leg from Manhattan to Milford sounds legitimately stressful, but Broder claims low-speed cruise control that never happened. He tries to blame this on wheel size later, while simultaneously using Tesla's logs as showing that he wasn't driving very fast. That makes no sense.

- Tesla claims they told Broder to fully charge at Milford. He did not, and lists a lot of justifications for not following their advice.

- Broder complains that Tesla didn't tell him more about how to get the most out of charging stations, but Broder's a veteran at electric vehicles even if it's true that he hates them. And his defense for not charging overnight is laughable - plugging in overnight is not the same as plugging it at every Walmart stop. It's overnight in cold temperatures and the car was at less than 50% of its range. He's just being obstinate here.

- It's laughable to say that the Tesla fell short of its projected range when the projected range was from the day before, before not plugging it in on a cold night.

- Tesla and Broder directly contradict each other on whether they gave him the go-ahead to stop charging after an hour in Norwich. Broder's being vague in his wordings - he phrases it as if they approved him to unplug when they might have just said to plug it in for about an hour when they were trying to find him the plugin station ahead of time. Tesla, in turn, says that he unplugged over their objections.

- Broder just can't "account for the discrepancy" about the logs showing him driving close to 55 when he said he was limping along at 45, back from Norwich.

- Broder seems defensible on the parking brake - how is he to know not to turn off the car? - and the driving in circles, although it would be interesting to know just how clearly marked that supercharge station is. Musk is probably guilty of the "fundamental attribution error" here.

Anyway, even after Broder's latest response, it still reads like Broder was trying to stack the deck.

Re: That Tesla Data: What It Says and What It Doesn’t

#163
post #7

Overall I think Broder acquits himself well, and his explanation about circling the Milford parking lot makes sense. Also, the Tesla engineer he was on the phone with at the time should have been able to tell Musk that happened (or Tesla should have just asked Broder to comment before posting the rebuttal). His action here is still hard to understand: > The Tesla personnel whom I consulted over the phone – Ms. Ra and…

That stuck out for me and I think he is being dishonest with his delicate wording. They didn't clear him to go, as he suggests. They told him that it would be ready in an hour. He also explains that he drove past a station "because no one made him aware of it". That smells like wordplay too. He was probably aware of it, and any reasonable owner of the car would probably be aware of it, or be well equipped to look it…

Maybe a driver should be responsible enough to be aware of his vehicle's range, and able to locate refueling stations?

This is like saying "I ran out of gas because I couldn't find a gas station" when you're a block away from one. It's ridiculous and, in this 21st century, not an excuse.

Re: That Tesla Data: What It Says and What It Doesn’t

#166
Tesla employees give the guy a ballpark estimate that an hour of charging should get him enough juice. He charges the car an hour. The car says it doesn't have enough juice yet.

Instead of waiting until his tank is full enough, John instead starts driving. Why? Is he just stupid, and decides to ignore the gauge and risk his own safety based upon the ballpark estimate of a Tesla customer support rep? Or did he decide that now his ass was covered if the car did somehow end up dying, giving him his story? Either way, the story stinks.

Re: That Tesla Data: What It Says and What It Doesn’t

#167
Others have already noted parts of this, but Broder is very clearly only naming 3 people at Tesla: the CTO, a press spokeswoman (who's since moved on), and a product planner.

These are not technical support personnel.

I doubt that the specific operational advice that Broder received when he called Tesla's 24-hour customer support line was wrong or contradictory.

I can very much believe that if he placed a reporter's call to a press spokeswoman's cell phone late at night (as he claimed in his response of today), or to a product planner, or even to a CTO who's not in the weeds of the tech support database, that he might receive contradictory information.

I think this confusion between Tesla people who aren't technical giving Broder advice, and Tesla's tech support people (who aren't named in Broder's story) giving him advice, is a large part of the source of the contradictions.

Re: That Tesla Data: What It Says and What It Doesn’t

#168

Earlier quoted context omitted.

I drove a honda civic hybrid in Northern Ontario for two winter seasons. The batteries where NiMh and the cold was absolutely brutal on them. Indeed they'd recover a bit of charge when warmed up but it was usually very hard to tell if that was because of the engine running or because the batteries warmed up. (and sometimes even harder to tell because the displays wouldn't work at all below -25 celsius). Maybe the bat…

Are Tesla's customers all Tesla engineers? If not reproducing the run with a Tesla engineer is pointless.

How is having an engineer reproduce a failure condition under controlled circumstances pointless? That's how bugs are fixed.

Re: That Tesla Data: What It Says and What It Doesn’t

#169

I put a graph together showing all the points/counterpoints and here's what I'm left with. - not charging past 90% at the 1st station is defensible. Especially if Tesla warns it will shorten the battery's life. - It'd be better if Tesla's software took temperature into account when estimating mileage range. - Broder is being ridiculous about the cruise control and his speed estimates, and the temperature. He has part…

Most of this analysis is spot on in my opinion - the one area I disagree with you in is the portion where he allegedly drove in circles. The graph Tesla shows is misleading and completely blown out of proportion.

0.6 miles is not very far to travel, especially when 0.2 miles are accounted for by the offramp. A drive once around an unfamiliar parking lot in the dark, looking for a small charging station can easily account for 4 minutes and 0.4 miles. Every EV charging station I've seen hasn't had any signage or otherwise aside from what's written right on it, You can't assume he'd automatically know where it is.

Regardless of whether Tesla is right or the NYT is right (and I'm inclined to believe both sides are partially at fault), the fact that Tesla took such offense to a measly 0.6 miles I find laughable.

Re: That Tesla Data: What It Says and What It Doesn’t

#170

I put a graph together showing all the points/counterpoints and here's what I'm left with. - not charging past 90% at the 1st station is defensible. Especially if Tesla warns it will shorten the battery's life. - It'd be better if Tesla's software took temperature into account when estimating mileage range. - Broder is being ridiculous about the cruise control and his speed estimates, and the temperature. He has part…

I agree. He's willing to lie about how long he was at the Milford supercharger station by reporting 58 minutes while in reality it was 47 minutes. He's intentionally reducing range whilst reporting otherwise so it can't be a matter of time savings. He's trying to vandalize the reputation of the vehicle.
Post reply on HN