Investor Engineering Discussions

[MC3OZ]

Sandy's first review of the fully disassembled 4680 pack:

Long story short: lots of part reductions, and "poka yoke," or assembly-mistake-avoidance design techniques.

It might be stating the obvious, but the 4680 structural pack seems superior in many ways to earlier generations.

It can act a structure and saves weight.

it seems easier to build, with likely fewer errors, and a higher yield,

It should be more robust and reliable.

 

[Ogre]

It's only speculation at this point and ignores the added engineering required for what is basically a party trick. I kind of hope they don't waste the time enabling this.

Big problem with this is that in most of the places it would be useful (very tight off road situations), it is also a very big chance something might go wrong mid-turn and you could get stuck.

 

[MC3OZ]

Big problem with this is that in most of the places it would be useful (very tight off road situations), it is also a very big chance something might go wrong mid-turn and you could get stuck.

I can see crab-walk occasionally being useful.

Situations where you might need a tank-turn can probably be negotiated by driving slowly in reverse, doing a 3-point turn, or being smart enough not to get into that situation.

Any driver today who gets into a situation where they need a tank turn is probably stuck.

A common problem I am aware of is, people go 4-wheel driving on a beach near where I live, think they can make it around the rocks before the tide comes in, and guess wrong. Their expensive 4-wheel drive goes floating out to sea. (And yes, a tank turn will not help at all.)

People also attempt to cross flooded roads, guess wrong, and their car goes floating down the river.

If the Cybertruck has even minimal ability to make some headway in shallow water, it may be able to get out of some of these situations.

 

[Ogre]

I can see crab-walk occasionally being useful.

Situations where you might need a tank-turn can probably be negotiated by driving slowly in reverse, doing a 3-point turn, or being smart enough not to get into that situation.

Any driver today who gets into a situation where they need a tank turn is probably stuck.

A common problem I am aware of is, people go 4-wheel driving on a beach near where I live, think they can make it around the rocks before the tide comes in, and guess wrong. Their expensive 4-wheel drive goes floating out to sea. (And yes, a tank turn will not help at all.)

People also attempt to cross flooded roads, guess wrong, and their car goes floating down the river.

If the Cybertruck has even minimal ability to make some headway in shallow water, it may be able to get out of some of these situations.

Crab walk, maybe. It's definitely useful for highway lane changes for example. Might be useful for getting around some hazards. More important, using it is unlikely to get you into a worse situation then you are already in.

I think "Tank Turns" or even "Diamond Turns" is just as likely to get you into more trouble then get you out. If you have enough room to 3 point turn then you can. Otherwise you can reverse pretty far... particularly since the Cybertruck has 4 wheel steering and a nice camera view. It is very hard to see things immediately beside you car and predict what your wheel path would be when you are spinning a vehicle.

Which brings up a much more interesting topic for 4 wheel steering, reversing! I can't believe they spent so much time on something most trucks will likely never do and didn't even mention the benefits this will have for people backing up a truck, particularly backing up to a trailer or reversing for more than a handful of feet. I had to back my Subaru about a quarter mile down a forest road the other day and it was a pain. 4WS would have been much nicer.

Considering the Model Y and even the Model 3 have been shown doing fairly decent in water, the Cybertruck should be just fine in shallow water.

 

[MC3OZ]

Seems to be a good source of information, worth watching.

 

[MC3OZ]

Tesla's SECRET NEXT GEN 4680 Battery & Machines​

 

[willow_hiller]

Still watching through this myself, but seems like another good video from the Limiting Factor:

 

[growler23]

Still watching through this myself, but seems like another good video from the Limiting Factor:

It was - still manages to impress me. And he doesn't waste your time with long pauses or unnecessary repetition.
Was a good overview of the evidence we have of Tesla's partial overcoming of the Dry Battery Electrode issues, some educated guesses about their progress on the remaining challenges, and good overall perspective relating to Tesla growth and the sustainability mission.
Highly recommend.

 

[Discoducky]

Main takeaway from this is that TLF was able to confirm that the anode is not only made with DBE, but it is more difficult and thus moving this process to the cathode does NOT require a breakthrough, but is just a matter of time.

Making a DBE cathode is the holy grail. Maybe Drew will talk about this at Q3 earnings...

EDIT: I should add this Jordan is so high quality in his delivery, that I had to pause, think, replay, pause, Google stuff, pause again, take notes...etc. So amazing.

 

[growler23]

Main takeaway from this is that TLF was able to confirm that the anode is not only made with DBE, but it is more difficult and thus moving this process to the cathode does NOT require a breakthrough, but is just a matter of time.

Making a DBE cathode is the holy grail. Maybe Drew will talk about this at Q3 earnings...

EDIT: I should add this Jordan is so high quality in his delivery, that I had to pause, think, replay, pause, Google stuff, pause again, take notes...etc. So amazing.

Agreed regarding his thought process, logic, perspective, evidence, graphics, and delivery. For those reasons he is now the first person I have supported on Patreon.

 

[MP3Mike]

New video from Cleanerwatt on the 4680 production status:

 

[MP3Mike]

New video from Munro on the 4680 cell after being taken apart:

 

[JRP3]

A bit overly dramatic about the "danger" of the cell disassembly.

 

[mongo]

A bit overly dramatic about the "danger" of the cell disassembly.

Yeah, the first step they mentioned would not make it safe. Sticking a load on it for a few days would...

 

[MC3OZ]

Some new info, not unexpected, or overly surprising.

Some of these engineering test runs could be attempts to make a cathode via a DBE process.

For the Cathode we know DBE doesn't currently work, perhaps the cathode plant at Austin can help:-

1. By possibly making cathodes via a wet process.
2. By possibly making cathode precursors in a form that is compatible with the DBE process.

Seems like 1. is far more likely than 2,, and is a safer option.

Should the DBE process eventually work for the cathode, then some of the new plant may no longer be required, but overall that is still a good result.
Or more likely even when DBE works for the cathode, being able to make cathodes via the wet process will still be handy.

 

[Discoducky]

In order to keep the FSD stuff out of the main investor thread, would folks like that stuff about FSD to be here instead? Just wondering...

Or maybe a new FSD for Investors thread or something?

In the meantime, I discovered an interesting bug with the non-real-time map data (I think).

I drove this road both ways twice to be sure and FSD slows to nearly a stop starting at the red lines and finishing right about where the railroad tracks used to be (see them at the bottom of the black line?). I plan to go back tomorrow and try again, this time without traffic so I don't have to do a disengagement and let it play out fully.

It is also doing something else strange on this road segment where it forces itself into the right lane in this road segment. I feel like it is doing this due to the road segment previously having a railroad crossing and the map data is possibly still has it, thus the non-real-time data is being fused and used primarily for path planning and controls.

Again, I'm not super confident about this yet, but plan to try again tomorrow.

I'm seeing more evidence of non-real-time map data possible being used, as my 3 new roundy's are also exhibiting strange lane changes that seem to be the path planner trying to force the car to stay on path by changing lanes 2 during the roundy. I also plan to do this more tomorrow as this is new behavior and unsettling as it is so abrubt.

Here's one of them. NOTE: Even Google maps are not updated with the new satellite data, but they are overlaying a faint transparent image of the new road data if you look at the image closely.

 

[MC3OZ]

thus the non-real-time data is being fused and used primarily for path planning and controls.

I always assumed that this was the case.

Also that:-

• Cars driving around update the maps.
• Maps are updated centrally then pushed to cars. (or pulled when needed).
• Map updates can occur more frequently than software updates (independent process).

I don't have anything solid to back my assumptions other than, it seems like a logical approach to take.

What I don't know is the granularity of the maps, and the frequency of updates.
For granularity, I assume a city like LA is broken down into multiple maps, but I don't know how many maps.

It will be interested to see if this problem eventually corrects itself, hopefully a map update will occur independent of a software update, and that will be strong evidence.

I'm happy to discuss technical aspects of FSD in this thread,

IMO the advantage of special threads is 2 fold:-
1. The main thread isn't cluttered up, and we avoid annoying some posters/moderators.
2. The conversation has a better flow, and occurs in more detail over a longer timeframe.

 

[mongo]

True the ribs might not allow the cell modules to drop out. Of course we've also seen the cells ejecting vertically through the top of the pack so it would seem the steel cover can also be burned through, as might the ribs.

Yeah, anything can melt. My recollection is the launching cells (or cell guts) have been in cases with frontal impact when the lead double stack modules get ruptured.
My thought with the electrical connection is that, even if the bottom melts, the modules/cells would be cantilevered off the midline rib if it's intact.
Also, the bottom current collecter is another layer of protection for the aluminum.

Recent fire was interesting in that it had quick response and daylight so provided preservation of structure and visibility. Interior was kept wet so no floor burnthrough/ heat release that way, may have increased thermal load on bottom?

 

[JRP3]

New LF video

 

[Discoducky]

In order to keep the FSD stuff out of the main investor thread, would folks like that stuff about FSD to be here instead? Just wondering...

Or maybe a new FSD for Investors thread or something?

In the meantime, I discovered an interesting bug with the non-real-time map data (I think).

I drove this road both ways twice to be sure and FSD slows to nearly a stop starting at the red lines and finishing right about where the railroad tracks used to be (see them at the bottom of the black line?). I plan to go back tomorrow and try again, this time without traffic so I don't have to do a disengagement and let it play out fully.

View attachment 854943

It is also doing something else strange on this road segment where it forces itself into the right lane in this road segment. I feel like it is doing this due to the road segment previously having a railroad crossing and the map data is possibly still has it, thus the non-real-time data is being fused and used primarily for path planning and controls.

Again, I'm not super confident about this yet, but plan to try again tomorrow.

I'm seeing more evidence of non-real-time map data possible being used, as my 3 new roundy's are also exhibiting strange lane changes that seem to be the path planner trying to force the car to stay on path by changing lanes 2 during the roundy. I also plan to do this more tomorrow as this is new behavior and unsettling as it is so abrubt.

Here's one of them. NOTE: Even Google maps are not updated with the new satellite data, but they are overlaying a faint transparent image of the new road data if you look at the image closely.

View attachment 854944

For my difficult/interesting/odd oddities, I have some new data. NOTE: Outside of weird lane changing, the current build of 10.69.2.2 is pretty great...the below are the 'rough edges' of the code that I'm trying to figure out...

The weird stopping is still occurring, but now the data seems to support that bad map data is still the root cause, but for another reason. I had to take video and play it back to spot the weirdness, but it does NOT occur in the same place, which points to dynamic traffic being the root and it wasn't until a car was pulling into the adjacent lane that it looks like the bad map data could be that it is feeding the real-time system that it is only a SINGLE lane (as opposed to two lanes of travel in the same direction) and the car will stop in its lane if a car is attempting to make a right turn in the lane to the right. Super weird to experience as my car came to a complete stop to let a Model S make a right into the adjacent lane.

The roundys also have new data as it seemingly will do completely different things given the same variables. Sometimes it wants to be in the left lane, sometimes the right, sometimes it tries to change lanes in the roundy, sometimes not, but it is always a bit different. At first, I dismissed this as the AI just doing what it thought was best, but then remembered that the map data could be off as the new lanes near the roundy's were MOVED about 30 feet or so and I think that the car is most likely using bad map data which has the lanes now off center AND when it snaps (Guaussian probability distribution) to a lane, it is sometimes picking one or the other, but gets too far off and it thinks it is departing the road so suddenly puts it back into the bad map data lane. I could describe this further, but this explanation is already too long...

My obtuse UPL does NOT get a creep wall and is thus still unsafe. It sucks as all my other UPLs are at 90 degrees and get a creep wall thus they have all been safe.

I have a new issue with an acute UPR and a 1 way bridge here, which it used to do well, but now will only complete if I set the speed to 20MPH or less. If it is set to more than that, it will not see the Stop sign, rolls it, completes the acute UPR, then speeds up to much to detect the red traffic light and tries to run the red light.