Investor Engineering Discussions

[JRP3]

I was curious so I just checked the power socket on my '22 Model 3 and it's showing 15.6V.

 

[BioSehnsucht]

It would be interesting to know if it was charging the LV battery at the time or not. It's not unheard of for ICE alternators to be up in the 15V range but it's not usually what they're supposed to be at.

 

[mongo]

It would be interesting to know if it was charging the LV battery at the time or not. It's not unheard of for ICE alternators to be up in the 15V range but it's not usually what they're supposed to be at.

Li isn't like a lead acid with bulk and float and large temperature compensation. The new one is really small capacity and low internal resistance, so the car DCDC is probably set up at fixed voltage with a charge current limit that would reduce voltage if the SOC was low.

 

[BioSehnsucht]

Li isn't like a lead acid with bulk and float and large temperature compensation. The new one is really small capacity and low internal resistance, so the car DCDC is probably set up at fixed voltage with a charge current limit that would reduce voltage if the SOC was low.

True, I am hoping/expecting that the operating range of the DC/DC for supporting a lead acid chemistry was such that it could also support a Lithium one easily, with just tweaks to software of the DC/DC unit. You surely wouldn't want to engineer a design so that it had to run to 100% maximum voltage just to reach charging voltage for the lead acid battery, you'd want to engineer it so you have a little wiggle room. But who knows, maybe that would require an entirely different chip/architecture to support more than 15V. Though since many automotive grade DCDC regulators for 12V devices are safe for up to 18V or 36V input, hopefully the reverse is true too, and they can support more than 15V for DCDC to charge the battery.

 

[mongo]

True, I am hoping/expecting that the operating range of the DC/DC for supporting a lead acid chemistry was such that it could also support a Lithium one easily, with just tweaks to software of the DC/DC unit. You surely wouldn't want to engineer a design so that it had to run to 100% maximum voltage just to reach charging voltage for the lead acid battery, you'd want to engineer it so you have a little wiggle room. But who knows, maybe that would require an entirely different chip/architecture to support more than 15V. Though since many automotive grade DCDC regulators for 12V devices are safe for up to 18V or 36V input, hopefully the reverse is true too, and they can support more than 15V for DCDC to charge the battery.

I doubt DC-DC needed to change. I think it's running a custom/ discrete bidirectional topology to also precharge the HV side. FETs and caps are probably 25V or higher on the LV side.

 

[JRP3]

With the fan on high it's 15.5V, heated wheel, front seats and rear defrost drops it to 15.3V, returns to 15.5 with everything off. This is a cheap 12V socket plugin reader so take all values as general. I might rig an extension so I can read it outside the vehicle with the armrest closed.

 

[willow_hiller]

Well now I know I can trust the coverage from CBC. One of the first fair and accurate representations of FSD Beta I've seen from a major news organization:

 

[mongo]

I suspect not... the titanium shield plate added to Model S is rather small (#2 in the pic below) and strategically placed to avoid damage to the front face of the critical battery pack, which if punctured is a fire hazard to occupants. The bottom of the rest of the pack has ballistic-grade aluminum armor, which has fared well.

The Rivian video linked showed only damage to the cosmetic plastic "panels" and non-recessed fasteners on the bottom of the truck surrounding the pack. The bottom surface of the pack itself appears to have held up well, and the material looks similar to the bottom surface of Tesla's pack.

Rather than expensive titanium trim pieces over the bottom of the non-fragile frame rails, I think the problem could be addressed with simply something more robust than plastic along with countersunk/recessed fasteners.


Model S Pack protection pic:
View attachment 799661
Key: (1) a hollow aluminum deflector bar, (2) a titanium plate and (3) a solid aluminum extrusion.

@Skryll
Not just front of pack protection
Tesla Adds Titanium Underbody Shield and Aluminum Deflector Plates to Model S

Tesla structural packs have more gap between pack bottom and critical bits. Cybertruck could use a beefier bottom plate, or secondary steel slid plate. (Possible off road package?)

 

[scaesare]

@Skryll
Not just front of pack protection
Tesla Adds Titanium Underbody Shield and Aluminum Deflector Plates to Model S

Tesla structural packs have more gap between pack bottom and critical bits. Cybertruck could use a beefier bottom plate, or secondary steel slid plate. (Possible off road package?)

This reply quoted me, but is directed at somebody else, so not sure if intended for me?

As that article states, "The titanium plate prevents sensitive front underbody components from being damaged and aids in neutralizing the road debris." As it mentions and the pic in my post shows, it's protecting the area in/on the front of the pack... which is where the pack that caught fire that spawned this change was pierced by the trailer hitch...

 

[mongo]

This reply quoted me, but is directed at somebody else, so not sure if intended for me?

As that article states, "The titanium plate prevents sensitive front underbody components from being damaged and aids in neutralizing the road debris." As it mentions and the pic in my post shows, it's protecting the area in/on the front of the pack... which is where the pack that caught fire that spawned this change was pierced by the trailer hitch...

Just joining the general discussion.
I was trying to point out the bars protect the rest of the pack (via deflection and jumping), not just the leading edge, but that is more the aluminum bits, not the titanium.

 

[JRP3]

Power screen tilt for S/X

https://www.reddit.com/r/teslamotors/comments/uhlmm9

 

[BioSehnsucht]

Will be interesting to see if they offer to retrofit existing S/X with the new interior, since apparently the structure was there for it, but just no actual motors etc to actually actuate it.

 

[mongo]

I'm thinking about this an an engineer and I'm wondering why an E/M clutch isn't just used to do this just like we do in aircraft. We have mechanical systems with a disengaging clutch for roll authority (ailerons) with a command by wire backup in case of a jam. Just reverse that sequence so you use command by wire primary for normal driving and when put in track mode you engage a mechanical clutch to regain the physical feel. Put the hard stops on the assembly at the steering rack and I think you've got the solution of best of both worlds. This also gives the safety backup if steer by wire fails due to a sensor failure or two.

Biggest issues I see are:
Realigning the wheel with vehicle direction
Slip ring integrity.
An intermediate planetary with motor (or motor in the steering wheel) can decouple steering from wheel and is also fail-safe.

 

[mongo]

Delta-V requirements are exponentially related to payload mass.

Incorrect use of the word "exponentially". This is a huge bugbear for me.
One form of Tsiolkovsky's rocket equation:

Note that it is actually linear in relation to total mass, the exponential relates to the exhaust velocity.

I think you two agree, but wording is ambiguous.
For a given deltaV, the inital mass is linear to the final mass.
For a changing deltaV, the inital mass changes exponentially to the deltaV scaling.
Double final mass, double inital mass.
Double final speed change, e squared inital mass.

IOW:
DeltaV requirement changes have an exponential impact on the mass (for a fixed final mass).
Payload mass changes have a linear impact on inital mass (for a fixed deltaV).

Going the other way, for a fixed amount of inital mass, final payload (fuel as mass fraction) is exponentially impacted by required delta V.

 

[Gigapress]

I think you two agree, but wording is ambiguous.
For a given deltaV, the inital mass is linear to the final mass.
For a changing deltaV, the inital mass changes exponentially to the deltaV scaling.
Double final mass, double inital mass.
Double final speed change, e squared inital mass.

IOW:
DeltaV requirement changes have an exponential impact on the mass (for a fixed final mass).
Payload mass changes have a linear impact on inital mass (for a fixed deltaV).

Thanks, I was going to say the same thing.

Holding all else equal, payload mass decreases exponentially with respect to linear increases in Delta-V requirements.

What I’m not sure of is how big the advantage is to being exactly on the Equator vs +/- 15 degrees. I suspect it’s probably worthwhile for making 1 million orbital flights in a year to send a big batch of ships to Mars.

Launching from 15 degrees costs 57 km/h for the Delta-V budget.

Earth’s axial tilt, atmospheric drag and staged design complicates the math beyond my understanding, but here is my feeble attempt at a first approximation:

​

Mo = Mf exp(dV / Ve)​

​

Mf = Mo/exp(dV/Ve)​

​

If dV requirement increases 57 km/hr, how much does Mf change?​

​

Mf,new/Mf,old =[ Mo/exp((dV+57)/Ve) ] / [ Mo/exp(dV/Ve) ]​

​

= exp(dV/Ve) / exp((dV+57)/Ve)​

​

= exp( (dV-dV-57)/Ve )​

​

= exp(-57/Ve)​

​

Ve for raptor engine is 3600 m/s = 13000 km/hr​

​

= exp( -57/13000)​

​

= 0.9956​

​

—> 0.44% decrease in final mass​

​

With final mass including the Starship itself constant at 80 tons and nominal payload of 100 tons, the loss in final mass is approximately 1 ton per flight.​

 

[Gigapress]

 

[MC3OZ]

My comments about single crystal cathodes relate to this patent.

Single crystal is one important part of the 100 year battery.

Limiting Factor researched a lot of great stuff like this before battery day.

When it wasn't included in battery day, it has been largely forgotten about.

 

[Doggydogworld]

TL;DW - due to sunk costs, minor 800V benefits aren't worth it for Tesla. Of course if Tesla were 800V and the others 400V it'd prove they were "5 years ahead", lol.

Why can't he let go of his 138S theory? The EPA docs say Y AWD, Y AWD LR and Y Perf packs are all 360V nominal. People drank way too much Kool-Aid at Battery Day and started imagining quantum leaps in energy density, charging speed, etc. Now with actual specs coming out they grasp at straws.

4680/DBE/structural pack is an attempt to cut costs. The other stuff they teased on Battery Day was a mix of long term R&D and marketing BS.

 

[Ogre]

My comments about single crystal cathodes relate to this patent.

Single crystal is one important part of the 100 year battery.

Limiting Factor researched a lot of great stuff like this before battery day.

When it wasn't included in battery day, it has been largely forgotten about.

TL;DW - due to sunk costs, minor 800V benefits aren't worth it for Tesla. Of course if Tesla were 800V and the others 400V it'd prove they were "5 years ahead", lol.

Why can't he let go of his 138S theory? The EPA docs say Y AWD, Y AWD LR and Y Perf packs are all 360V nominal. People drank way too much Kool-Aid at Battery Day and started imagining quantum leaps in energy density, charging speed, etc. Now with actual specs coming out they grasp at straws.

4680/DBE/structural pack is an attempt to cut costs. The other stuff they teased on Battery Day was a mix of long term R&D and marketing BS.

Tesla has said outright the 4680 currently shipping is not using any of the other advancements they talked about on battery day. They are focusing on scaling first then will introduce the rest of the changes over the next year.

 

[MC3OZ]

Tesla has said outright the 4680 currently shipping is not using any of the other advancements they talked about on battery day. They are focusing on scaling first then will introduce the rest of the changes over the next year.

Yes, If they do make single crystal cathodes it requires the cathode plant, construction has just started.

If you watch the video, making single crystal cathodes isn't overly difficult, but does involve some extra steps and extra expense.

Maybe the cells produced with this formula can last up to 10,000 cycles.

Applications

1. Grid energy storage.
2. RoboTaxi
3. Semi

My rationale for posting is that the 100 year battery was unsurprisingly single crystal.

Jeff Dahn does a lot of reseach into single crystal and Tesla has the patent.

I also remember single crystal from the chinese is expensive and low volume. The Tesla patent may be superior to the current production process.

I am not saying Tesla will definitely do this, just that on site production at Austin seems more likely than import.