Investor Engineering Discussions

[MC3OZ]

This is yet another new battery type:-

New Aqueous Zinc Battery Breakthrough - Real Deal or Hype?​

Worth watching some of the claimed specs seem to good to be true, by the way hydrogen ion are apparently used to boost the battery energy density is something I have been encountered before.

 

[Buckminster]

https://twitter.com/x/status/1748020327328682099

 

[mongo]

Typically each cabinet supplies up to 387 kW to 4 plugs, but when there are multiple cabinets they are linked in groups of up to 7 cabinets / 28 plugs

7? I've only seen installation details for 3 DC linked cabinets. Do 7 use a separate DC combiner switchboard?

 

[GhostSkater]

7? I've only seen installation details for 3 DC linked cabinets. Do 7 use a separate DC combiner switchboard?

Things might have recently changed, I remember seeing a video showing a string of 7 linked, but it was so long ago that I have no hope of finding it

 

[transpondster]

Things might have recently changed, I remember seeing a video showing a string of 7 linked, but it was so long ago that I have no hope of finding it

IIRC youtubers like Branden Flasch and Kyle Conner were claiming that only 4 cabinets are sharing,1 master with upto 3 slaves

 

[GhostSkater]

IIRC youtubers like Branden Flasch and Kyle Conner were claiming that only 4 cabinets are sharing,1 master with upto 3 slaves

Now I remember where I saw up to 7, it was in a exchange with Branden on X, not on video

 

[mongo]

Sure.... they don't have 1000's of thermistors... but if they are having to double the number of cells per brick, I had suggested, that I "suspect that means 2X temp sensors" in each brick... but just conjecture.

I think temperature monitoring would be based on maximum allowable thermal distance. Doubling cells per brick while halving the number of bricks gets you the same number of cells to monitor in roughly tye same physical arrangement leading to roughly the same number of temperature sample points.

If taking a fixed number of temperatures per interconnect, doubling voltage doubles the number of sensors.

 

[scaesare]

I think temperature monitoring would be based on maximum allowable thermal distance. Doubling cells per brick while halving the number of bricks gets you the same number of cells to monitor in roughly tye same physical arrangement leading to roughly the same number of temperature sample points.

If taking a fixed number of temperatures per interconnect, doubling voltage doubles the number of sensors.

Right... so # of thermistors would likely not care about pack voltage...

 

[mongo]

Right... so # of thermistors would likely not care about pack voltage...

Yeah, but weren't you saying the opposite?

The needed power dictates the number of cells, not the voltage. For a given KWh pack you need fewer bricks of cells in series, but you need double the cells in parallel (i.e. more current capacity) in each brick. I suspect that means 2X temp sensors to monitor...

Each brick might have more sensors, but there are less bricks and the number of BMS monitor/ balancing ICs is based on the number of bricks. Temperature sensing is isolated from cell voltage so isn't tied to cell arrangement.

 

[scaesare]

Yeah, but weren't you saying the opposite?

Not intentionally.... I had pointed out to @GhostSkater that if they implemented his conjectured 200V architecture, and thus halved the number of bricks, they'd have "to double the number of cells per brick, and had suggested, that I "suspect that means 2X temp sensors" in each brick...

If i managed to make that unclear, apologies.

 

[mongo]

Not intentionally.... I had pointed out to @GhostSkater that if they implemented his conjectured 200V architecture, and thus halved the number of bricks, they'd have "to double the number of cells per brick, and had suggested, that I "suspect that means 2X temp sensors" in each brick...

If i managed to make that unclear, apologies.

Ah, we are all in agreement then. I had read it as 2X overall.
In the interim time I also forgot the temperature sensors are likely isolated and so really don't care.

 

[GSP]

What 12 volt system in the Cybertruck? I haven't seen anything about a 12 volt system. (They do use 24v amplifiers because 48v ones weren't available yet.)

I expect that the Cybertruck will need to supply 12-14 V to its trailer connector, since this is used to charge trailer 12 V batteries. The X and Y have eliminated voltage supply to this pin when they switched to 15 V Li-ion batteries. However the larger trailers that the Cybertruck can haul will require 12 V batteries to apply the trailer brakes in the event of disconnection from the tow vehicle.

I have yet to see any info on Tesla’s trailer connector supporting this 12 V supply, or not, despite some Cybertruck owners testing with 10,000 lb. trailers.

 

[mongo]

I expect that the Cybertruck will need to supply 12-14 V to its trailer connector, since this is used to charge trailer 12 V batteries. The X and Y have eliminated voltage supply to this pin when they switched to 15 V Li-ion batteries. However the larger trailers that the Cybertruck can haul will require 12 V batteries to apply the trailer brakes in the event of disconnection from the tow vehicle.

I have yet to see any info on Tesla’s trailer connector supporting this 12 V supply, or not, despite some Cybertruck owners testing with 10,000 lb. trailers.

That's not a vehicle wide subsystem though. It's trailer specific outputs from the rear module just like the trailer brake and lights are.

 

[GSP]

That's not a vehicle wide subsystem though. It's trailer specific outputs from the rear module just like the trailer brake and lights are.

Yes, of course not vehicle wide. However there must be some sort of 12 V supply on the vehicle. Good point about the lights and trailer brake though. They could deal with 15 V from the X/Y, but not 48 V!

 

[mongo]

Yes, of course not vehicle wide. However there must be some sort of 12 V supply on the vehicle. Good point about the lights and trailer brake though. They could deal with 15 V from the X/Y, but not 48 V!

Ok, the idea of a vehicle wide for 12V legacy end nodes was what Mike was originally replying to.
Local buck converters for 12V loads makes the most sense to me. Trailer could use a common source and then add self protected switching devices for each load. Electric brake is likely it's own (safety and control) as would be trailer accessory power (current and voltage limited output).

 

[mongo]

OK, so on par with 2170's now by reducing deadspace. Given this was a "clean sheet" design, it would seem odd that they would plan on building a cell that was less volumetrically efficient to begin with. Hence my subsequent question:

Was the decrease of "deadspace" in order to get the same percentage of "active volume" per can as the 2170's, thus making each cell equivalent in terms of utilized volume and thus volumetric energy density?

-or-

Is the 4680 jellyroll less energy dense per cm^3, and thus they were forced to squeeze extra added "active volume" utilized in the 4680 can, making them equivalent, energy on a call volume level, but still actually less efficient on a "volume of active jellyroll" basis?

IOW: Did they just find a way to cram more jellyroll in?

First gen had the anode collector then bottom cap pushed into the can proper and crimped to seal.
Gen 2 laser welds the can bottom to the electrode instead and the gap is convex gaining a few mm of jelly roll height.

 

[scaesare]

First gen had the anode collector then bottom cap pushed into the can proper and crimped to seal.
Gen 2 laser welds the can bottom to the electrode instead and the gap is convex gaining a few mm of jelly roll height. View attachment 1012520
View attachment 1012518View attachment 1012519

Thanks @mongo... do as per my question above, did that give the 4680 the same percentage of total volume allocated to active material as the 2070's, or did that give the 4680's a greater percentage of volume allocated to active material to make up for less volumetrically efficient jellyroll?

 

[mongo]

Thanks @mongo... do as per my question above, did that give the 4680 the same percentage of total volume allocated to active material as the 2070's, or did that give the 4680's a greater percentage of volume allocated to active material to make up for less volumetrically efficient jellyroll?

I expect 4680 is more volumetrically efficient due to semi constant outer gap and void in the center. 4680 does lose some space due to the end shingle overlap and top insulator.

 

[mongo]

Nothing is 100% efficient.

Resistance heater...

 

[Tronguy]

Resistance heater...

Ha. I take your bid, and raise it: Unless said resistance heater is inside a sealed box, and maybe not even then, some heat will be lost to conduction/radiation/whatever and not delivered to what one desires.