Thanks for all the info! Perfectly understood that you can’t answer everything.
It’s good to hear that you are keeping significant parts of the pack design and production internal. With the Bolt, GM outsourced big chunks of the drivetrain to LG. Outsourcing isn’t inherently bad, clearly, but I think that automakers that don’t treat their EV drivetrains with the same level of dedication as their ICE drivetrains will find themselves at a disadvantage.
As you note, UHSS is an industry trend and Tesla is no different. I thought these were some interesting pictures. The Model 3 uses aluminum, mild steel, HSS, and UHSS. It’s pretty easy to see that UHSS is used for the most critical areas, particularly around the roof and door framing.
Even between my Taurus and my 2004 Expedition, the things that get me are:
[ul]
[li]Needing to take the keyfob out to lock and unlock the door.[/li][li]Needing to insert a key to start the vehicle.[/li][li]The cruise control does not account for the vehicle in front of me.[/li][li]I keep looking at the center stack for the GPS that’s not there.[/li][/ul]
I have to imagine that we did. We used to get email invitations to look at torn down vehicles, and after a period of time, all of the parts were disposed of. Apparently now we have IT technology, though, that keeps a catalogue of all of these teardowns. I’ll try to have a look this morning.
Something from the article, though:
Actually, we do this for pretty much every make out there. And they tear our stuff down, too. And Tesla certainly does it, too, despite the implication in the article that they don’t. This isn’t all about drivetrains, but engineering and manufacturing techniques.
For example, we might know that a certain percentage of a component is subject to corrosion sooner than expected. What does Audi do to address that? Or maybe the inner door flanges on a Honda are really, really small, and manufacturing says they can’t weld them. How does Honda join these sections? Or, we spend $x on this adhesive… is it worth it? The Infiniti is using this cheap stuff, instead.
I’m amazed at how many people (who are supposed “experts” on this industry they cover) are totally ignorant of the concept of “competitive benchmarking.” Practically every company in the industry does this.
Mine seems to have a little bit of trouble when cars in the next lane are right next to mine about directly to the side - the vehicles on the display will occasionally pop in and out of existence at that location, or move around a lot, especially on the passenger’s side. I don’t know if it’s an issue with camera coverage/overlap, or calibration. I’m not too concerned about it, as blind spot coverage seems to be working pretty well. There have been a few times already where I was getting ready to change lanes and then in my peripheral vision realized that another vehicle was on the display right in the blind spot area.
One thing I don’t like at all about V9 is that they moved the homelink button to the far right top corner of the screen. This seems terrible from a UI/human factors standpoint, as I have long arms and I still have to reach quite a bit to hit it, plus its pretty far away from a driver’s typical line of sight when looking straight ahead at the road.
This would be fine for a setting/control that is only used when the car is parked, but I have it programmed to open my community gate, and another one for my garage door, both of which I’d prefer to use while the car is in motion. I could set it to auto open/close, but I don’t think that’ll work on my gate (since its not at my “home” location), and I don’t really trust the auto open/close on my garage door since the car has no idea what state the garage door actually is to start with.
I’m still getting marketing calls for Tesla solar products. I’m not really interested, mainly because I don’t think my house is a good candidate. But of course, the salesman insists it is.
If they call again – strike that – next week when they call again, I’ll ask them if I invest in a solar system with them whether I get moved up the priority list for the M3 SR. Then when they call two weeks later, I’ll just keep asking that question.
Understood–but note that the article is from 2014. The S was only two years old, the X was not out yet, and the 3 was a twinkle in Elon’s eye. Tesla sold less than 60k vehicles over their entire life at that point. I think there’s some validity to the idea that while Tesla may have been tiny (still are, really), they were still interesting enough to earn a teardown from Ford and others. And I think it speaks well of Ford, too, not being too proud to admit that Tesla did a few things right and are worth learning from.
Tesla’s EV competition in 2014 was pretty minimal and there probably wasn’t a lot to learn from the competition drivetrain-wise back then. But I’d hope and expect that they selected a handful of ICE cars just to learn about the rest of the engineering (the kinds of stuff that you mention).
Practically every company in every industry does it. Our company runs an automated testing lab–with a few units from the competition. We do a new test run every time they release a new driver. Sometimes we catch regressions in their drivers–performance loss, image corruption. A part of me wants to send them a note every time we find something broken :).
They could stand to make the touch target bigger, at least. You’re right; it’s far enough away that it’s a little tricky to touch accurately without looking at it. They could just expand the touch area by 4x or something and it would be easy to hit even without looking directly at it. Wouldn’t take any more room since it only pops up when you’re in the geofenced area.
As jz78817 said, the quote was from the article. I only brought it up because Balthisar thought that Ford might have done a teardown of Tesla’s packs, and the article quotes Mark Fields saying that they indeed had.
I think the biggest thing that makes the touch screen difficult to use isn’t that it’s flat without tactile feedback, but that it’s little targets that you have to hit as your hand and the screen bounce around. The heated seat button on my Suburban isn’t any bigger than the heated seat button on the screen, but once my finger is on the button on the Suburban I can press it multiple times without having to remove my finger from the button. On the screen I have to completely lift off the screen, retarget my finger, and then hit it again.
The new slider interface to set the climate control temperature is much easier to use than the old design, because I just have to target the right place on the screen, but then my finger is pressed to the screen the entire time I make the adjustment.
Anyway, much larger touch targets would make hitting them easier. The UI designers at Tesla need a simulator that can bounce them around as they try to use the screen.
I think the geo controlled homelink works very well. You might want to give it a try. I’ve only had it not work twice. Once I think it just didn’t trigger for some reason, and the other time I’m pretty sure I hit “skip” when I intended to hit navigate.
I think you can have a geofence for each entry. Just like how homelink opens and highlights the entry you are near. If you enable auto open for your gate, it should work for just your gate.
It works great on my garage door, because we only keep the one car in the garage, so it is almost always open when I leave (and needs to be closed) and closed when I get home (and needs to be opened). It’s rare enough to be out of phase, that I don’t think I’ve ever had to hit “skip”. If I had to hit “skip” all the time, because the door needed to be left open, or was already open, then I wouldn’t I’d use it.
As much as I like the auto open, the one habit I’m having a hard time breaking is trying to hit the button on the sun visor halfway down the driveway.
I don’t think so–if anything, I think the MR version might be the limited-time one. Tesla is working on a reduced-cost pack design for the SR version, but that’s not ready yet. The MR pack is just an LR pack with fewer cells, so it’s not as much of a savings (for Tesla) as it should be. I think the MR is a stopgap to get as many sales with the max Federal rebate as they can manage. And of course it captures a segment that couldn’t quite swing an LR, but can stretch a little beyond the SR.
No D.C. rebates that can put you under the wire? Looking around, it seems that there aren’t any ordinary vehicle rebates, but there are some incentives on charging and home chargers. Maybe some electric company rebates as well?
You don’t pay sales tax on EVs in DC, and that’s the only incentive. So it’s a significant savings - about three grand for this version - but that doesn’t make up for the thousands more in price relating to Tesla’s disappointing (but somewhat understandable) premium first sales strategy.
I already have a EVSE installed and got a tax credit two years ago for something like $500.
One of Tesla’s big shorts changed their mind on Tesla. The Model 3 sales numbers and the lack of short-term competition seem to be the key factors. A few zingers from the report:
*If you would have shown us the below chart five years ago there is no way we would have ever believed it. It looks like it is the competition that is taking the Ambien.
Like a magic trick, while everyone is focused on Elon smoking weed, he is quietly smoking the whole automotive industry. *
Tesla posts a profit! Woo! And not just inching over the line like I had expected; they actually posted very solid earnings of $312M GAAP. I need to read the whole thing but it doesn’t look like this was gamed at all; the Model 3 got great gross margins and they kept costs under control.
Tesla reports that it earned $1.15 a share, or $312 million, in the third quarter. This is substantially higher than analyst estimates this morning, which seemed to be all over the place but generally centered around a few pennies in earnings or losses per share.
It reports gross margins of 20% on the Model 3, substantially less than the 30% estimate that Munro estimated was possible with better manufacturing… but hey, I’d like to sell stuff at a 20% margin. Production was 4,300 per week across the quarter NOT including planned shutdowns.
There isn’t any mention of the expected ramp rate in Model 3 production other than affirming that Tesla intends to build 100,000 cars this year. Previously, there’s been estimates of growing to 6,000 and 10,000 Model 3s per week. I’m sure that will be discussed on the call.