Tag: Tesla

This is part of a series that covers my experience with a Tesla Model Y.

I went to Europe for two weeks. My wife joined me a week in and left our Tesla at the airport. Once she got on a plane, I double checked that Sentry mode was off and then went about our travels.

When we got back, about 8 days later, we got into the car and it had 64% battery charge. When my wife parked it, it had 65%. The usage shows 0.4kWh for the 8 days, with the car essentially shut down for the time.

Not a amount of usage for a week. I checked a couple times to see how things were changing in the week, but it didn’t seem like the charge was going down.

This was during a period of time when the weather was 70-80F, so that is a temperature that doesn’t affect the battery much. It will be interested to try this in the winter.

This is part of a series that covers my experience with a Tesla Model Y.

I saw this bet appear on Twitter between John Carmack and Jeff Atwood:

It’s quite a statement, and it leaves just over 7 years for this to come true. I do wonder if they’ve defined this well, as I don’t know if many people really think about what Level 5 means.

I ran across an article recently from someone who’s owned a model 3, titled: Tesla Full Self Driving Beta: How Close To Level 5 Autonomy? As I read it, I felt that a lot of the items it listed were things I see wrong with either the AutoPilot I have, or things I’ve seen from other users that are using the Full Self Driving (FSD). There is also an SAE infographic.

Disclosure: I haven’t tried the FSD, but I am interested in subscribing for a month at some point and seeing what I think.

What is Level 5?

Level 5 driving is full automation, with no need for a human. To me, this means I could put my infant in the car and it would make it to a hospital without me being in the car in any weather. I’m not sure how many people would want to go through that test, but that’s what it says.

It also says “They will be free from geofencing”, which would mean go anywhere. In terms of the bet above, I am guessing this means they can go anywhere in the city on any day of the year?

My Thoughts

I do think the self-driving is more possible in cities, where the mapping and understanding of the structures and roads is well known. Cities are generally well marked with regards to traffic indicators, lines, lights, etc., but not always.

I have always believed that self-driving needs to happen in a well-mapped, well drawn city. I don’t know that New York fits the bill. London has the congestion zone, which may or may not work, but really, I think a city like Chicago with relatively wide lanes, things well marked, etc., is a good choice. I wouldn’t pick Chicago because of the potholes and repairs in parts of the city, but maybe somewhere like Singapore or Melbourne would work well.

The key here is well marked and known. There will be weather, like snow or heavy rain, that will cause problems with vision based systems. It can also cause issues with radar as well. It already does with human drivers, but we are used to that. I don’t know we’d have the same tolerance for mistakes and accidents from self-driving cars.

I think it’s much more likely we get Level 4, which is mostly autonomous, but with some restrictions on when the self-driving works. For example, perhaps not in a blizzard.

I also think that outside of cities, it’s unlikely we’ll see self-driving anytime in the next decade. I think about my rural roads, some of which have no paint markers. Or mountain roads where the road disappears in the winter. I think about the West Ireland roads, driving from Cork to Baltimore, where the road literally isn’t wide enough for two cars and passing requires a little swerve into the bushes that are overgrown at the edge of the road.

I really don’t know that we can adequately model all these situations to make them self-driving level 5.

Personally, however, I’d be happy if we set some self-driving lanes, like half the highway where the car takes over for you and handles merging into the self-driving lane and out of it, after which a human takes over. Or perhaps some geo-fencing where most cars are self-driving (cities).

I’m not sure the tech is good enough to be unbounded, nor do I think most people would find this acceptable when there are mistakes.

A Good Bet?

Is this a good bet? Probably. There are already self-driving taxis in a few cities, including Beijing where there is no human at the wheel.

Plus, we humans have a history of letting technology loose before it’s really developed.

This is part of a series that covers my experience with a Tesla Model Y.

I wrote recently about the Tesla and the charge loss while sitting at the airport. While a little worrisome, I knew it was something I could live with, and even get around. A few of the parking places near the airport have long term charging, so it’s not the end of the world.

But it felt strange.

I don’t always plug in the car at home, and after a day or two, I might lose 1%, but nothing like the 8% we were losing each day. I decided to do a little investigating and see what other owners have experienced as I see plenty of Teslas parked at Wally Park, where I usually leave my car.

I saw more than a few people note that there are two reasons you usually use significant charge. First, people keep checking on their car in the app, and this prevents it from going to sleep.

Second: Sentry mode.

Sentry mode is essentially a combination car alarm and constant dashcam. The car is always recording things, which means the car and computer are always active. While the car isn’t moving, there is still power being burned when it is online.

Sentry Mode is especially active when it detects people near the car. In a busy parking lot at the airport, this means every 5-15 minutes, someone drives past. Certainly not longer than 30 minutes.

When I checked my TeslaMate stats, I could see the problem easily. In the image below, my latest trip is 4/28 to 5/2. On the right side of the image, the car is almost always asleep. If you look towards the left, from 4/14 to 4/18 was my second trip. This was the one where I lost 25% in three days. I can go back and see the trip before that where I lost quite a bit as well.

I turned off sentry mode between the last two trips and it made all the difference. We were gone 4 days this time and we lost around 1% of charge. I no longer worry about a two week trip out of the country with this car. I might be concerned for 30 days, but maybe not.

The vampire drain is shown below. You can see the difference in trips here.

A normal day at home has the car mostly asleep, which makes sense. Sentry mode is disabled at home (or work) if you set that. I had, but didn’t think about the airport. While it could help me detect who caused damage to the car if something happens, the lot tracks who is in and out and where they park, so they likely could help me pinpoint who damaged the car if I had an issue.

I’m glad to solve that, and more importantly, I hope more people realize the Sentry mode, while great, can eat up battery over time.