Sponsored

jarross

Active Member
First Name
John
Joined
Jul 29, 2020
Threads
2
Messages
39
Reaction score
38
Location
Wyoming
Vehicles
Subaru Forester XT, Chevy Silverado
Occupation
Pharmacist
It says 440V in the subtitles. Also, they filed a patent showing the use of a 450V battery that could be charged at 900V. It is really simple. You just have two 450V packs in parallel to operate the vehicle, then switch them to series for charging. It wouldn't be 440 miles while towing. We know that for sure.
I'm pretty ignorant in this field: is that done to decrease dendrite formation during charging? Or to allow charging to happen faster? What's the advantage?
Sponsored

 

skyote

Well-Known Member
Joined
Mar 12, 2019
Threads
55
Messages
2,725
Reaction score
5,647
Location
Austin, TX
Vehicles
Jeeps, 2500HD Duramax, R1S Preorder (Dec 2018)
My understanding of it is that a gust of wind from a passing truck or whatever tries to push the trailer off the road to the passenger side. This puts huge counter clockwise torque on the tow vehicle which causes its front end to move in that direction. The driver sees this and put in a large steering input towards the right which puts a large clockwise torque on the tow hitch which tries to whip the trailer to the left. Picture just the right phasing here and its clear that you have a building oscillation with predictable results. Mechanical solutions try to get the pivot point up under the wheels (5th wheel, pivot point projection...) so the trailers input torque is reduced. In a vehicle with torque vectoring the motors can put the thrust necessary on each wheel in order to neutralize the trailer's torque input and the problem is this solved. No brakes would be used but acceleration could be automatically applied along with steering input. The issue I see is in the necessary sophistication in the anti sway control algorithms. It has to be pretty bullet proof. I think that there is tremendous potential to solve the sway problem with a vehicle that has computer controllable torque vectoring and steering but the algorithms must be super robust.
I remember seeing something from Ford about brake-based trailer sway control, and I agree that torque vectoring/independent wheel motor control could accomplish the same without braking.
 

thrill

Well-Known Member
First Name
billy
Joined
Jan 14, 2020
Threads
19
Messages
767
Reaction score
1,538
Location
South Carolina
Vehicles
i3s, (r1t)
I'm pretty ignorant in this field: is that [900V charging of 450V batteries] done to decrease dendrite formation during charging? Or to allow charging to happen faster? What's the advantage?
Dendrites are crystalline formations occurring when there are differing areas of energy (charge) or even temperature (unmixed liquids or metal to liquid interfaces) facilitating chemical reactions. A Lithium battery is vulnerable to this because of the difference in current carrying ability between the copper/aluminum anode/cathode of the battery and the lithium. Electrons only probabilistically travel to an open spot, and so there is inevitable leakage where the energetic electron may attach to something not contributing to the engineered purpose, i.e. a crystal begins forming. The dendrites cause problems in Lithium batteries because they're metal and can penetrate the polymer separator between the battery halves, causing a short and sudden large current flow, resulting it lots of heat (insert favorite mental image of EV burning by side of road). Controlling this is by careful selection of chemical composition, careful construction, and what we're getting better and better at, careful monitoring of the batteries characteristics (temp, voltage drop, current flow, rates of change, etc.) and removing that battery from contributing to the pack's total charge if it begins to behave differently than the other batteries around it - i.e. a sophisticated battery management system.

In a DC system, voltage and current have a linear relationship (leaving out the by design minimized effects of the system's inductance and capacitance once it reaches a steady flow), and noting that resistance (to current flow - so to charging) increases as the battery "fills", so doubling the voltage, while it may require additional insulation to prevent leakage, reduces the current by half. Power, of which leaked power is seen as heat (facilitating the dendrite formation) is calculated as voltage*current, or as current*current*resistance, or as voltage*voltage/resistance. One confusing point for most initially when seeing that is that the current is, by definition, how many electrons flow past a point per second, the voltage above is the voltage *drop* from beginning to end, which in a well constructed system is a small percentage of the total voltage. So, there's a lot to be gained in reducing heat loss (and hence probability of dendrite formation) by decreased current and increased voltage, as long as the system overall can be designed to manage the higher voltage and differing heat flow safely.

Splitting the battery packs into groups that can be charged in parallel means those groups can be charged all at once, instead of pushing through each one to the next, which is now quicker. However, there's no free lunch, so designing the system to manage the heat, which though reduced, is also spread over a larger area all at once, is part of the engineered effort.

We'll drive-by mention that the impedance in the battery actually matters too, as most charging systems use pulses of constant current but controlled variable widths, which means lots of rapid charge/no-charge cycles per second. Pulsing is liked because it reduces the tendency of polarization (a battery that charges better than discharge, etc.), and seems to keep the interfacial (between the cathode and the lithium, for example) resistance lower over the battery's life. I've read that some system designs also use pulse discharge for similar reasons.
 

jarross

Active Member
First Name
John
Joined
Jul 29, 2020
Threads
2
Messages
39
Reaction score
38
Location
Wyoming
Vehicles
Subaru Forester XT, Chevy Silverado
Occupation
Pharmacist
I think I may be following that. Sounds like it's a mixture of both faster charging and decreased dendrite formation?
 

ajdelange

Well-Known Member
First Name
A. J.
Joined
Aug 1, 2019
Threads
9
Messages
2,883
Reaction score
2,317
Location
Virginia/Quebec
Vehicles
Tesla XLR+2019, Lexus, Landcruiser, R1T
Occupation
EE Retired
I'm pretty ignorant in this field: is that done to decrease dendrite formation during charging? Or to allow charging to happen faster? What's the advantage?
Dendrites are formed when a lithium ion lands someplace other than where it is supposed to (in the center of a ring of carbon atoms) but gets an electron from somewhere anyway. It is now a lithium metal atom (not an ion any more) and it is conductive. Now along comes another one of these and easily gets an electron from the existing dendrite and gets turned to metal so the whisker grows (dendro means tree i.e. a little tree branch is growing). This is bad because 1) it takes lithium ions out of the pool available for holding charge and 2) the dendrite may eventually grow to the point where it penetrates the separator and reached the the other electrode shorting out the cell but I digress as dendrites have nothing to do with this.

Whether the car is 385 or 800 or 900V architecture the cells are still charged at around 4 V - they are arrange in series parallel configurations which put around that level of voltage on the individual cells regardless of the charger voltage.

Going to higher charging voltage conveys a few advantages the most important of which is that 350 kW delivered at 800 volts requires a current of 437.5 A whereas the same power delivered at 400 volts required twice that (975 A). This means that, for the same losses, the wiring for the low voltage system must have 4 times the cross sectional area and weigh, for a given length, 4 times as much. This is significant in the car and in the cable from the charger to the car.

Since the wire from the car gets so much hotter (4 times the heat if the same size wire is used) either the cable must be liquid cooled or bigger wire must be used to the point that it is so heavy and stiff it becomes unmanageable. Thus it is difficult to build a usable high power charger at 400 V. Tesla's limit is 250 kW and they are using liquid cooled cable. EA uses chargers that peak at 350 kW which means 437 A in the cable at 800 V and it is liquid cooled. The CCS connector seems to have a limit of 500 A meaning, at 385 V, a power maximum of 218 kW. Suppose you have, as does Rivian, a car with 385 V architecture and you want to charge it faster. What can you do? Answer: Divide the battery into two halves and put the two sub packs in series then apply 800V. The volatge across each of the packs is now 400 IOW to it it looks as if it is connected to a 400 V charger. Each cell sees the same voltage as when the vehicle is charging from a 400 V charger and, thus draws the same current.

Thus the reason for the Rivian battery swithching scheme is to give it access to higher power chargers all of which happen to be high voltage (800 V).
 

Sponsored

ajdelange

Well-Known Member
First Name
A. J.
Joined
Aug 1, 2019
Threads
9
Messages
2,883
Reaction score
2,317
Location
Virginia/Quebec
Vehicles
Tesla XLR+2019, Lexus, Landcruiser, R1T
Occupation
EE Retired
So, there's a lot to be gained in reducing heat loss (and hence probability of dendrite formation) by decreased current and increased voltage, as long as the system overall can be designed to manage the higher voltage and differing heat flow safely.
Let's look at this a bit more closely. Let's say we have 2 cells of 1 Ah capacity that we wish to charge in an hour. We can put them in parallel and push a total of 2 amperes into the combo, 1 amp for each, for a "pack" charging rate of 3.6*2 = 7.2 Watts or we can put them is series and push 1 amp through the string for a rate of 7.2*1 = 7.2W. The power absorbed by each cell is 3.6 W and as the current through each is the same 1 A the I^2R losses are the same in either case. Thus, plainly, there is no advantage to series (charger: high voltage low current) as opposed to parallel (Charger: low voltage, high current) here.

The point of the Rivian patent is that they do not want to charge at C. They want to charge at 2C (or, in any event some rate greater than C but lets stick with 2C to make the math simple) but nobody makes a 14.4 W charger at 3.6 V as that would require 4 A and the "cables" can't handle that. So they put the cells in series and push 2 amps. As the voltage is now 7.2 volts across the pair the power input to them is 2*7.2 = 14.4W. The current through each cell is now 2 A. The charging rate is 2C. This was the goal. Increase the charging rate from C to 2C and to do it without increasing the current in the wire connecting the charger. As the current is now double in each cell the I^2R losses are quadruple and there is actually more liklihood of dendrite formation but if you want to double the rate at which you charge a cell you have to accept that. Staying out of the high and low SoC bands reduces the liklihood of battery damage within acceptable limits and new chemistries improve the picture too but no matter how you slice it, 2C charging is harder on a battery than 1C charging.



Splitting the battery packs into groups that can be charged in parallel means those groups can be charged all at once, instead of pushing through each one to the next, which is now quicker. However, there's no free lunch, so designing the system to manage the heat, which though reduced, is also spread over a larger area all at once, is part of the engineered effort.
It doesn't matter whether you charge a cell at a particular rate in series or parallel. Charged at rC each cell will accept the same amount of charge (after equalization) and produce the same amount of heat.


We'll drive-by mention that the impedance in the battery actually matters too, as most charging systems use pulses of constant current but controlled variable widths, which means lots of rapid charge/no-charge cycles per second. Pulsing is liked because it reduces the tendency of polarization (a battery that charges better than discharge, etc.), and seems to keep the interfacial (between the cathode and the lithium, for example) resistance lower over the battery's life. I've read that some system designs also use pulse discharge for similar reasons.
I'm aware that pulse charging methods were used in older technologies in order to get at the open circuit voltage at the battery and monitor it's SoC but there are lots more sophisticated ways to do that these days (coulomb counting...). I am also aware that the charging current is controlled by PWM and that that implies the applied waveform is rich in harmonics but in a modern DC fast charger, supplied by 3ø mains, the rectifier output never goes to 0 and the switching rate is 40 kHz or above so that the harmonics are going to be 240 kHz and up and a few ufd should deal with them adequately. Do you mean this or do you mean the whole shebang is turned completely off for a period as they used to do with lead-acid, NiCd etc?[/QUOTE][/QUOTE]
 

Ssaygmo

Well-Known Member
First Name
Sage
Joined
Sep 11, 2020
Threads
3
Messages
123
Reaction score
171
Location
Ca
Vehicles
Volt
@Ssaygmo be careful if you actually do that. In CA over 10,000 lbs puts you in a different category and you will be held to the law under CVC 22406 and 22407. Basically this means that the speed limit for you is 35 MPH, not even 55 as jjwolf120 says. If you can actually get your truck up to 70 MPH while towing at max capacity you will be in for a 35 MPH OVER the speed limit ticket, which can also be reckless endangerment. Basically you could lose your license for doing that as well as endanger everyone else. I am by all means not telling you how to drive, but what you described if you do tow over 10k lbs is careless at best and criminal at worst.
I understand what. you are saying, and I personally only tow a few thousand pounds at most, but if you drive i5 a few times up and down cali you would understand the frustration at big rig drivers going way faster than they are allowed yet far slower than traffic, causing huge traffic pileups on the highway trying to pass at a few miles an hour faster, 70mph rig on the right being passed at 71mph by a rig on the left on a 2 lane highway. It is infuriating.
 

Ssaygmo

Well-Known Member
First Name
Sage
Joined
Sep 11, 2020
Threads
3
Messages
123
Reaction score
171
Location
Ca
Vehicles
Volt
That said (and I know you could calculate it) even without a trailer the grapevine kills batteries, a full 50 mile volt battery goes dead in 10 miles or so of uphill. I haven't done it in a full electric car, but having DC fast charging stations at the base of steep long grades should certainly get good business for convenience stores/transitioning gas stations along the routes.
 

ajdelange

Well-Known Member
First Name
A. J.
Joined
Aug 1, 2019
Threads
9
Messages
2,883
Reaction score
2,317
Location
Virginia/Quebec
Vehicles
Tesla XLR+2019, Lexus, Landcruiser, R1T
Occupation
EE Retired
That said (and I know you could calculate it) even without a trailer the grapevine kills batteries, a full 50 mile volt battery goes dead in 10 miles or so of uphill.
It really isn't that bad at all. Here's the picture driving up a 3% grade for an hour in traffic which requires frequent speed adjustment.

5500 lbs Speed 50 to 20 Avg 41.9 mph, 3.0 ± 0.0 %grade , Wh/mi Total: 744.6, Drive: 42.1, Slip: 79.9, Drag: 144.5, Roll: 81.4. Gravity: 271.3, Inertial: 32.4; Recov. On, Brake 16.4

The numbers represent the watt hours per mile that go for drive train losses, wheel slip loss, drag, rolling resistance, gravity and inertial. As you can see the consumption attributable to taking the truck up the hill. 271 Wh/mi is the greatest of the individual loads at 36% of the total which is 745.

By comparison on level ground but with the same start/stop constraints

5500 lbs Speed 50 to 20 Avg 41.9 mph, 0.0 ± 0.0 %grade , Wh/mi Total: 397.5, Drive: 27.2, Slip: 51.7, Drag: 123.3, Roll: 68.9. Gravity: 0.0, Inertial: 39.7; Recov. On, Brake 24.8

Note that the total load is now 398 Wh/mi which is a bit more than half so that says your range driving continuously uphil on a 3% average grade will be a little more than half (53%) of what it is on the flat. The good news is, of course, that that you will see what the bear saw that is the other side of the mountain which is downhill. Here is the picture with a grade of -3%:

5500 lbs Speed 50 to 20 Avg 41.9 mph, -3.0 ± 0.0 %grade , Wh/mi Total: 79.0, Drive: 16.1, Slip: 30.6, Drag: 78.1, Roll: 43.3. Gravity: -296.4, Inertial: 41.2; Recov. On, Brake 41.3

Here the total average consumption is only 79 Wh/mi (I have seen negative numbers in actual trips) with gravity supplying 296 Wh/mi. If you consider the trip up the mountain and back down your average consumption is (744.6 + 79)/2 = 414.5. This is only slightly greater than the flat terrain value of 397.5 and is attributable to the fact that not all the energy delivered by gravity on the way down can be recovered. The motors are less than 100% efficient and the power is coupled into the motors via wheel slip so there are losses from both of them.

This all depends on regenerative braking of course but when it is available hills and stop and go traffic become less of a burden than when it isn't. Note also that trailers don't have regenerative braking so that hauling a trailer up a 3% grade puts large demands on the battery and you don't get it back when you go down the other side.

I haven't done it in a full electric car, but having DC fast charging stations at the base of steep long grades should certainly get good business for convenience stores/transitioning gas stations along the routes.
Thus whether a charging station is necessary or highly desirable at the foot of a long up hill haul depends on what's at the top. Drivers will use extra juice going up the hill but if a negative grade commences at the top of the hill its almost as if the terrain were flat. If, OTOH, the crest of the hill represents the start of an extensive high elevation area then he wont get the uphill expenditure back before he potentially runs out. So a charger at the top of the hill might be a good alternative too.
 

trickflow

Well-Known Member
First Name
Andrew
Joined
Mar 26, 2019
Threads
1
Messages
120
Reaction score
135
Location
SoCal
Vehicles
Tesla Model 3 performance, Toyota 4runner
I understand what. you are saying, and I personally only tow a few thousand pounds at most, but if you drive i5 a few times up and down cali you would understand the frustration at big rig drivers going way faster than they are allowed yet far slower than traffic, causing huge traffic pileups on the highway trying to pass at a few miles an hour faster, 70mph rig on the right being passed at 71mph by a rig on the left on a 2 lane highway. It is infuriating.
I get it and I drive the grapevine a few times a year. It is super annoying. They are only supposed to stay in the 2 right lanes, but they do not. Then take all the left lane drivers that won't move over to let people pass, the road rage can be plentiful!
 

Sponsored

RivFly

Member
First Name
Paulo
Joined
Oct 5, 2020
Threads
0
Messages
22
Reaction score
31
Location
WA
Vehicles
Subie Crosstrek, Honda Africa Twin, Husq 310R
Occupation
Asphalt Durability Specialist
The said in a tweet that 11k pounds will reduce range by about 50%:
That doesn't mean much. The rolling resistance is the smallest impact on range. An 11,000 low profile trailer will have considerably less impact on range as a travel trainer with solar panels and air conditioners on the roof.
For me, it is less about the configurator and more about not being open and transparent about missing self stated deadlines.
Aerodynamics and rolling resistance, all factors in towing. Heck.....try driving around your current vehicle with a chair on the roof and see what kind of range you get. its amazing how fluid dynamics plays a part......and all that without even factoring rolling resistance!

As for the others complaining about configurators being delayed and rollout times blah, blah, blah.........LOL..........you guys should read the comments on the Ford Bronco forums!!:CWL: Seems auto forums are auto forums.....horses of of course.......:CWL:

Build it right first is better to me then being first to market. I'm looking at this for usefulness as a pickup first and range is my biggest concern! I need something that can go semi-serious off road, tow and have range!
 

kanundrum

Well-Known Member
Joined
May 2, 2020
Threads
217
Messages
3,976
Reaction score
12,107
Location
Washington, DC
Vehicles
Giulia QV, R1S (S00N)
Occupation
IT
Clubs
 
Not to bring back a old thread but I was watching the testing video again and I am hoping to see a "Cold Weather Testing | Progress Report by Rivian" video as this time would be the perfect time to get some real impacts on their battery cell architecture and cold weather testing!
 

ajdelange

Well-Known Member
First Name
A. J.
Joined
Aug 1, 2019
Threads
9
Messages
2,883
Reaction score
2,317
Location
Virginia/Quebec
Vehicles
Tesla XLR+2019, Lexus, Landcruiser, R1T
Occupation
EE Retired
Dramatic loss of performance in cold weather was a large part of early BEV FUD. And there was something to it. But with BMS that carefully regulate battery temperature and clever schemes for capture of waste heat it is, I believe, a thing of the past. At least around here. I can't speak for conditions up north as I flee those regions to avoid them. Note that there have been some recent reports of problems with Octovalves in Quebec.

Around DC you probably won't notice much if anything. I actually get slightly better consumption here in the winter than I do up north in the warmer summer weather.

There is one huge cold weather factor that definitely does have a profound effect on consumption and that is snow, slush or water on the roadbed. Of course one can have water on the road in summer too. Water will take my consumption from around 280 Wh/Mi (X LRP) up to 350. That business we had around here a week or so ago (some snow accumulation on the Beltway) took me up to 535, that is, it almost cut my range in half. So don't worry so much about the cold itself. Do think about precipitation when planning.
 

BigE

Well-Known Member
First Name
Eric
Joined
Nov 19, 2020
Threads
40
Messages
770
Reaction score
1,327
Location
North Carolina
Vehicles
Rivian R1T, Honda S2000
Dramatic loss of performance in cold weather was a large part of early BEV FUD. And there was something to it. But with BMS that carefully regulate battery temperature and clever schemes for capture of waste heat it is, I believe, a thing of the past. At least around here. I can't speak for conditions up north as I flee those regions to avoid them. Note that there have been some recent reports of problems with Octovalves in Quebec.

Around DC you probably won't notice much if anything. I actually get slightly better consumption here in the winter than I do up north in the warmer summer weather.

There is one huge cold weather factor that definitely does have a profound effect on consumption and that is snow, slush or water on the roadbed. Of course one can have water on the road in summer too. Water will take my consumption from around 280 Wh/Mi (X LRP) up to 350. That business we had around here a week or so ago (some snow accumulation on the Beltway) took me up to 535, that is, it almost cut my range in half. So don't worry so much about the cold itself. Do think about precipitation when planning.
Hi AJ...I was reading some of your past posts on this tread and appreciate your insight. I have a reservation for a Tesla CyberTruck and Pre-order for R1T. I think both have pros & cons for my situation so hopefully, I'll be able to actually see both before making a decision. What are your thoughts on Tesla Battery technology vs Rivian. Tesla has been doing this a long time now, is moving to their 4680 cells for CyberTruck...I'm just wondering, will the R1T be like old tech when it finally gets into production vs Tesla constantly moving the ball forward? Or, will Rivians' LG produced cells be just fine? Curious if you have any thoughts? Thanks
 

ajdelange

Well-Known Member
First Name
A. J.
Joined
Aug 1, 2019
Threads
9
Messages
2,883
Reaction score
2,317
Location
Virginia/Quebec
Vehicles
Tesla XLR+2019, Lexus, Landcruiser, R1T
Occupation
EE Retired
Lots of thoughts but they are pretty much the same as yours. I think we are in a typical high tech ramp up. Manufacturers will be leapfrogging one another until such time as all the cars have 800 miles range based on million mile batteries. Obviously Tesla has more resources to throw at battery research than Rivian and a little more time so I think it is reasonable to assume that the CT will be a bit ahead of the game in this department relative to Rivian. When Rivian announced a vehicle with 400 miles range I leapt. But that was a couple of years ago. Four hundred miles is sort of the minimum acceptable at this point. Is this enough for me to cancel my Rivian reservation and wait longer for a CT? No, no, no!
Sponsored

 
 




Top