Tesla Talk

garo

If you are limited to 2.1kW it is not worth getting a Zappi and probably not even trying to charge the car. You will rarely have enough excess even if you draw only 5A - and yes you can charge the car at 1.2kW no problem as you propose. Note that 2.1kW would only produce that much during peak summer. Most of the time it would be much less. Your panels will be SSE right? So say May-August you get over 2kW output for anything from 4-6 hours on fully sunny days.



What you said about manually diverting the excess to the car is possible. i tried it with the immersion this summer as I was working from home and was modestly successful. It is PITA though and now that I have a Zappi and an EV I won't bother next summer. Also I think an EV doesn't like the power being switched off abruptly as much as a resistance heater like an immersion does. You lose about 15 seconds in the negotiation every time. As KCross said, anytime a cloud rolls by or the oven comes on you end up importing more than you are using. And clouds roll by rather often in this country!

AndyBoBandy

unkel wrote: »

Any link to those 540W panels?

There is not a hope they would be the same size as a standard panel. Simply not possible. The current standard of the latest panels is about 21% efficiency, or about 200wp per square meter of panel, or roughly 320W for a standard size 165*99cm panel. Any higher wattage panel is simply bigger than that.

No, I had simply & incorrectly assumed that there were higher density panels available, that would have been the same size, but just more expensive...

AndyBoBandy

garo wrote: »

Your panels will be SSE right? So say May-August you get over 2kW output for anything from 4-6 hours on fully sunny days.

SW will be the most likely orientation where I could get up to 6-7 panels...

SE would only fit maybe 2 panels, so not really an option...

May-August would be typically the months that we dont run the central heating, so a diverter in these months would be our best bet for any little excess we produce..

garo

Right and if you are getting a diverter you definitely won't have much left for the car. Go with the diverter only and forget about charging the car with PV. It will cost you more in trying not to export6 than just letting a small bit go to the grid. Only thing I'll say is make sure your immersion element is long and reaches to the bottom of your tank.

unkel

AndyBoBandy wrote: »

SW will be the most likely orientation where I could get up to 6-7 panels...

SE would only fit maybe 2 panels, so not really an option...

I saw your picture in the renewables forum. Not great but you just have to work with what you have. And yes 2 panels SE is certainly an option, these panels could have their own micro inverter. And perhaps put more panels on places you haven't thought of. Do you have a shed?

Also, don't get a diverter, it will be long dead before it can ever pay for itself with such a small system

unkel

padjocollins wrote: »

https://www.cnbc.com/2020/10/28/gms-super-cruise-tops-teslas-autopilot-in-consumer-reports-testing.html?__source=androidappshare

looks like gm's cruise will work on 200km of americas roads.

unkel wrote: »

Cruise is very good, but don't let a summary report like that fool you. Cruise is geo fenced with mapped roads, everything pre-defined. Autopilot works anywhere in principal as it is based on computer vision. This is a far superior approach - if it turns out that it can be safe enough (big if, bit of a gamble, the geo fencing approach is safer, but far inferior from an AI point of view)

Also I reckon points were given in that survey for driver monitoring. AP doesn't do any of that (apart from steering wheel holding), so the AP score is artificially low (and nothing to do with the actual self driving performance of the system

I had a bit more of a detailed look into this. It almost comical. Tesla has by far the best capabilities, but super cruise wins because it gets a lot of extra points because of the driver monitoring. Exactly as I reckoned

Didn't "Consumer Reports" come out with a very similar comparison test a few years ago with the same outcome? I get a sense of deja vu here...

I guess this part of their report says it all really



Here is the full article, read it for yourself

padjocollins

what was new to me was that there are a few others that have decent driver assist systems like gm , ford and audi at 8 . I know a good bit about tesla and waymo but hadn't kept up with others and their (mostly nvidia platform based ?) solutions
https://www.tesmanian.com/blogs/tesmanian-blog/tesla-fsd-super-training-computer-dojo-may-already-using-in-house-npu-chips

unkel

Oh aye, others are getting quite good too. But level 2 autonomous driving (basically lane keeping, active cruise control and autonomous emergency braking) is relatively simple. My €25k Hyundai had that in 2016 as standard and it worked quite well. All new cars in Europe that do not have it as standard in 2020 can't even get a 5 star NCAP safety score any more. There is progress here.

But the big challenge is the path from there to level 5 (truly full self driving). And that is where Tesla is very far ahead imho in terms of already having a million (or whatever it is) cars on the road with all the required hardware that are all collecting data for Tesla

ELM327

https://www.tesla.com/en_IE/careers/job/technician-dublin-67966

No salary quoted, I assume this is not going to be well paid?

unkel

That's interesting. I guess a "technician" does not need to be paid to the same levels as a qualified mechanic. Tesla presumably still have a few mechanics to work on or supervise stuff like brakes and suspension.

ELM327

Yes. I suggest a technician would do stuff like trim bits, tires etc. You dont need to be a mechanic for that.

unkel

I'd say they would do a whole lot more than that. Pretty much everything except maybe traditional mechanicals like brakes and suspension. Anyone in here apply?

ELM327

Problem is a base/entry level technician job would likely be a pay cut and a substantial one at that for anyone that owns a tesla. They dont even list the salary.

I wouldnt want to work for Tesla as I can imagine it would be hell.

unkel

Of course. I doubt more than a single employee from Tesla Sandyford own Teslas. Maybe the MD.

JPA

Looks like a entry level job, I bet you'd wash a lot of cars! Maybe work up to mechanic, although you wouldn't learn how to do an oil change and other basics.
Still, I think it would be an interesting environment.

fricatus

JPA wrote: »

Still, I think it would be an interesting environment.

I bet it would. Out of all the car makers, it would be the best place for an ambitious person to work. It's fast-growing and that bit different. There would be a defined career path for anyone working in Volkswagen or Ford I'd imagine, but in Tesla I would say you would get more responsibility (and salary) quickly if you proved yourself.

KCross

Interesting that the job spec specifically called out "door handle changes"..... those handles must really have been rubbish to mention it in the job description.

Are the replacement handles standing the test of time?

slave1

KCross wrote: »

Interesting that the job spec specifically called out "door handle changes"..... those handles must really have been rubbish to mention it in the job description.

Are the replacement handles standing the test of time?

AFAIK once the handles are replaced that's it, the weakness was a poor wire design where it was literally being worn away through constant use.

unkel

Aye and perhaps some people don't know how they work, but these door handles come out to hand themselves to you as you approach the car (when the key is somewhere in your pocket). This means with a car on your drive, this can happen dozens of times a day. Or ten thousand times a year. A bit like an electric window (in any car). Make it go up and down 10k times and it will fail. In a car that would take 10 years, with a Tesla door handle that could be in just a year or two. You can switch all this off, but isn't that just one of the charms from having a Tesla?

Approach the car, the door handle comes out. Grab it, sit in (the car adjusts its seat automatically), stick in D and press the play pedal. Simpler than any other car.

And what is usual with the Tesla vertical integration is that if they have found a weakness in a part, they don't just replace the broken part with another one of the same (like any other manufacturer would do), they try make the part better and replace the broken one with an improved one, as slave1 says.

slave1

Short free Kindle read, few bits I picked up that I did not know before and of course some mistakes, but grand for a spare hour to fly through

https://www.amazon.co.uk/gp/product/B08KTK5YCC/ref=ppx_yo_dt_b_d_asin_title_o03?ie=UTF8&psc=1

3d4life

https://www.eenewsembedded.com/news/researchers-find-serious-flaws-model-x-keyless-entry-system?fnid=133703

Where do these 'researchers' get funding for these antics ?

It must be great fun.

How can I get in on this ?

cnocbui

3d4life wrote: »

https://www.eenewsembedded.com/news/researchers-find-serious-flaws-model-x-keyless-entry-system?fnid=133703

Where do these 'researchers' get funding for these antics ?

It must be great fun.

How can I get in on this ?

You could start by getting a computer science degree.

3d4life

cnocbui wrote: »

You could start by getting a computer science degree.

Have one of them.

Next step ?

cnocbui

3d4life wrote: »

Have one of them.

Next step ?

Get another one (Phd) and then apply for a job at the University of Leuven.

Tom Dunne

3d4life wrote: »

Have one of them.

Next step ?

Get a PhD so you can do research and apply for funding to carry out that research.

fricatus

3d4life wrote: »

Where do these 'researchers' get funding for these antics ?

They're doing extremely valuable work if they're finding flaws in software before the bad guys do. They've got various government and industry partners, which is presumably where the money comes from (you can read more on Wikipedia). Part of their funding is even mentioned in the article:

Tesla confirmed the vulnerabilities, awarded the researchers a bug bounty and started working on security updates.

BTW, why the quotes around the word "researchers"? Do you not actually think that's what they are?

AndyBoBandy

It would be naive to think that there weren't teams of researchers out there trying to hack everything and anything from cars to smart phones, to TV's & Fridges etc..., all with the support of the manufacturers of the various items trying to be hacked.

How else do they pick up on and find vulnerabilities in their products??

3d4life

fricatus wrote: »

....BTW, why the quotes around the word "researchers"? Do you not actually think that's what they are?

Having been one and worked among many in the 70s & 80s I guess I'm just jealous that this generation of researchers seem to have unlimited budgets

Back on topic : Did Unkel post what revision of software his loaner Model X is on ?

cnocbui

3d4life wrote: »

Having been one and worked among many in the 70s & 80s I guess I'm just jealous that this generation of researchers seem to have unlimited budgets

I am intrigued as to how you got your computer science degree in the 60's. That would put you in your mid 70's these days?

3d4life

3d4life wrote: »

..... in the 70s & 80s......

cnocbui wrote: »

I am intrigued as to how you got your computer science degree in the 60's....

Do your sums again

( answer : in my late 60s )

FWIW I'm a refugee of core memory, punchcards, papertape, the ASR33 teletype. All stuff where you could see the data if things needed checking

You wouldnt believe the excitement there was when Intel came out with the 8008

Great times