Sponsored

Billyk24

Well-Known Member
First Name
William
Joined
Mar 21, 2020
Threads
4
Messages
108
Reaction score
39
Location
PA
Vehicles
Ford C-Max Energi, Premium Mach-E ordered
Occupation
health care
My feeling is that if we're on the cusp of going with 900v charging, and that's what future Rivians will work from, it would be a shame to miss that mark with just the earliest of vehicles. We'll always be close to the "next big thing", but this would seem to negatively delineate the LE R1's from the later versions of the same line. It's possible that it will hurt resale or trade-in values for those who plan to sell or trade them in. For those of us who plan to be buried in them in forty years, maybe it's not such a big deal. Let's call it "Under-roading".
KIA EV6 on sale the second half of 2021 will feature a 800V battery pack of 77.8kWh capacity. They claim 10-80% charging in 18 minutes! -- The Kia EV6 | Kia Global Brand Site | Movement that inspires ---and 60 miles in 4.5 minutes. This is with 350kW chargers. I do not know the charging curve. I do not know the charging time with 150kW chargers or even with those 62.5kW chargers that are popping up all over the state of Michigan.
Assuming this information is accurate, it makes me move away from vehicles with 400V battery packs as some of us will road trip in an EV.
Sponsored

 

DucRider

Well-Known Member
First Name
Gary
Joined
Oct 21, 2019
Threads
17
Messages
1,652
Reaction score
3,143
Location
ORegon
Vehicles
Polestar 2, Ioniq, R1S
I'm not overly concerned about 900V battery architecture per se, but the 50% bump allowed by 900V charging is a good thing on several levels. Not just the time savings, but also freeing up chargers faster (this will become an issue at some locations) allowing more vehicles to charge on the same amount of infrastructure.

450V vs 900V packs are different only in the number of cells in series. When it comes right down to it, each cell is 4.2V. Connecting 108 or 216 in series gets you the different pack voltages. Implementing actual 900V battery packs vs splitting the 900V into two 450V feeds when charging will be transparent to the owner. There are some relatively minor weight and efficiency gains in motors, wiring, and a few other components by doubling the voltage, but on a relatively inefficient EV like the Rivians, it is not as critical.

I'd take GMs claims of 350kW charging with a grain of salt. GM (and many other OEMs) spec the charger type required to achieve their maximum charge rate. As one example, the Kona EV states xx miles in xx minutes on a 100 kW charger. Many people assume this means it will charge at 100 kW when in actuality the max it will get is 77 kW. Since there is no spec for a 77 kW charger, they indicate the type/rating of the charger needed.

Rivian will never be very good at the xx miles per xx minute game. They are not very efficient when compared to most EVs. Charging at a higher voltage will allow for faster charging, but it is not required for the pack to be higher voltage for this.
 

DuckTruck

Well-Known Member
First Name
Duck
Joined
Dec 23, 2020
Threads
33
Messages
2,343
Reaction score
6,228
Location
PNW
Vehicles
Corvair, BMW325, Acura Legend, XC60, '16 Caddy ELR
Clubs
 
I'm not overly concerned about 900V battery architecture per se, but the 50% bump allowed by 900V charging is a good thing on several levels. Not just the time savings, but also freeing up chargers faster (this will become an issue at some locations) allowing more vehicles to charge on the same amount of infrastructure.

450V vs 900V packs are different only in the number of cells in series. When it comes right down to it, each cell is 4.2V. Connecting 108 or 216 in series gets you the different pack voltages. Implementing actual 900V battery packs vs splitting the 900V into two 450V feeds when charging will be transparent to the owner. There are some relatively minor weight and efficiency gains in motors, wiring, and a few other components by doubling the voltage, but on a relatively inefficient EV like the Rivians, it is not as critical.

I'd take GMs claims of 350kW charging with a grain of salt. GM (and many other OEMs) spec the charger type required to achieve their maximum charge rate. As one example, the Kona EV states xx miles in xx minutes on a 100 kW charger. Many people assume this means it will charge at 100 kW when in actuality the max it will get is 77 kW. Since there is no spec for a 77 kW charger, they indicate the type/rating of the charger needed.

Rivian will never be very good at the xx miles per xx minute game. They are not very efficient when compared to most EVs. Charging at a higher voltage will allow for faster charging, but it is not required for the pack to be higher voltage for this.
DucRider,

Thank you for the clarity you always bring to bear on all technical issues. I know I sure appreciate it. If the ability to configure the battery either way is fairly similar, why would one not configure it in the 900v architecture. You mentioned many benefits of going this route and that 900v will be used by many of the EVs we will be competing with in sales and for charging time at the "pump".

Is this only an issue when using L3 DCFC charging in a public setting, where EVs are competing for charging time, or am I missing something?

Thanks again!
 

Billyk24

Well-Known Member
First Name
William
Joined
Mar 21, 2020
Threads
4
Messages
108
Reaction score
39
Location
PA
Vehicles
Ford C-Max Energi, Premium Mach-E ordered
Occupation
health care
but on a relatively inefficient EV like the Rivians-------how are you defining this? The Rivian is "huge"-wider/taller/heavier as in an aero brick. The physical components of the vehicle are going to hinder it when referring to distance traveled per kWh of usage. Or are you using a different reference?
 

DucRider

Well-Known Member
First Name
Gary
Joined
Oct 21, 2019
Threads
17
Messages
1,652
Reaction score
3,143
Location
ORegon
Vehicles
Polestar 2, Ioniq, R1S
but on a relatively inefficient EV like the Rivians-------how are you defining this? The Rivian is "huge"-wider/taller/heavier as in an aero brick. The physical components of the vehicle are going to hinder it when referring to distance traveled per kWh of usage. Or are you using a different reference?
Yes,
Consumption is higher on the Rivian or any other similar vehicle. I expect the Hummer will be worse. The physical characteristics are the primary factor impacting the efficiency of a vehicle and the battery voltage in not part of that equation.

If the ability to configure the battery either way is fairly similar, why would one not configure it in the 900v architecture. You mentioned many benefits of going this route and that 900v will be used by many of the EVs we will be competing with in sales and for charging time at the "pump".
900V systems require different motors, inverters and other components. RJ expressed the desire to use them, but thought availability of those components was going to be an issue in an early interview. We are not privy to their decision making process, but they apparently committed to 400V (nominal) at some point during the design process.
Their charging patent essestianly allows for "on the fly" reconfiguration of the pack from 450V under normal operation to 900V when charging. Here is a simplified example:
Rivian R1T R1S New Official Rivian Adventure Network and Charging Station Info 1619364461420


The same amount of energy is available in either configuration (V x Ah). One potential downside to 900V architecture is when charging from "400v" DCFC equipment, the EV will need to have the circuitry to double the voltage to charge the battery. The "800V" Taycan can only charge at 50 kW @ 400V unless you order an option that allows for 150 kW. There is some debate as to the necessity of this since virtually all (with very few exceptions) DCFC equipment that provides 150 kW can do so at 920V.

As a note - pack voltages are often lumped into two categories "400V" and "800V". Both actually cover a range of actual voltages depending on pack architecture. Two voltage designations come into play - nominal and max. A fully charged Li Ion cell will be 4.2V, nominal voltage the average voltage it can provide from fully charged to fully discharged (AJ would provide a highly technical and slightly more accurate explanation but "average" will suffice here). Nominal voltage will be between 3.6 and 3.7V. Nominal voltage gives you the capacity/energy of a battery, but max voltage is what is relevant to the charger you are using.

The two most common "400V" configurations are 96s and 108s (number of cells in series).
96s yields a nominal voltage of about 350V and a fully charged voltage of about 400V (Tesla, Chevy, etc).
108s yields a nominal of about 400V and a max of about 450V (Rivian uses this layout)
Other configurations are possible - Honda uses 84s for 311V nominal 353V max in their Clarity series.

The Taycan "800V" is 198s for a nominal voltage of 723V and max of 832V.
I think it likely that they Hyundai Ioniq 5 is 192s for 710V nominal 800V max. Likely at least part of the reason they are getting 232 kW max on the specified 350kW charger from early reports - Hyundai specifies either 150 kW or 350 kW chargers when making references.

GM is utilizing the switching scheme we all thought was covered in the Rivian patent for it's 800V system in the "800V" Hummer:
HUMMER EV will have the ability to take advantage of the industry’s fastest 350-kilowatt DC fast chargers4. This vehicle will offer a GM-estimated 350+ miles7of range based on preliminary testing, with a unique ability to switch its battery pack from its native 400-volts to 800-volts for charging. A disconnect unit and mechanization within the pack enables the battery to switch from “parallel” to “series,” allowing it to add nearly 100 miles of range in 10 minutes of charging8.

Or:

On the battery side, however, GM says that the Ultium propulsion systems underpinning the Hummer EV are indeed fully developed. The so-called double-stack battery has 24 modules in all—two levels, each with 12 modules—and while it runs at 400 volts it charges at 800 volts, allowing 100 miles of recovered range in just 10 minutes, from low states of charge.
“It’s just not manageable to have the huge power cord at 240 kilowatts at 400 volts,” said Tim Grewe, GM’s director of battery cell engineering. “So it’s much more rational to go up to 800 volts to make that happen.”
Grewe said that essentially amounts to treating the two levels as separate packs—running them in parallel for vehicle operation, but allowing them to be placed in series during DC fast charging. As part of GM’s from-scratch Ultium modular approach, the smart cell-module assembly has the cooling integrated into it, where each module essentially has a plus and a minus that allows for coolant flow.

Given the Hummers modules reliance on 24 cell groupings, it is likely they are going to be 96s/192s (usage/charging).
 

Sponsored

SeaGeo

Well-Known Member
First Name
Brice
Joined
Jan 12, 2021
Threads
47
Messages
5,235
Reaction score
9,677
Location
Seattle
Vehicles
Xc60 T8
Occupation
Engineer
Not just you, I agree for the exact reason you pointed out.
The thing is there's an assumption here of *when they enable* more. There is nothing saying that will actually happen with the R1 series.

To come at this from another route, if they averaged say... 275 kw over 20 minutes instead of ~190 kw, that's 200 miles in 20 rather than 140 miles. Driving for almost another hour after charging for the same timespan isn't nothing. On a say... 600 mile trip, that's at a minimum 17 fewer minutes if charging. Keep in mind that it's unlikely that the 140 miles are highway miles.

Realistically the R1T "large" pack probably couldn't pull 275 for 20 minutes, but just pointing out that flipping the way we look at the problem can make the change how you view the impact of the decision.

Definitely not the end of the world either way, but 800 or 900v architecture can make a decent difference in trip duration. Especially if you have more than 100ish miles to the next charger.
Yes,
Consumption is higher on the Rivian or any other similar vehicle. I expect the Hummer will be worse. The physical characteristics are the primary factor impacting the efficiency of a vehicle and the battery voltage in not part of that equation.


900V systems require different motors, inverters and other components. RJ expressed the desire to use them, but thought availability of those components was going to be an issue in an early interview. We are not privy to their decision making process, but they apparently committed to 400V (nominal) at some point during the design process.
Their charging patent essestianly allows for "on the fly" reconfiguration of the pack from 450V under normal operation to 900V when charging. Here is a simplified example:
1619364461420.png


The same amount of energy is available in either configuration (V x Ah). One potential downside to 900V architecture is when charging from "400v" DCFC equipment, the EV will need to have the circuitry to double the voltage to charge the battery. The "800V" Taycan can only charge at 50 kW @ 400V unless you order an option that allows for 150 kW. There is some debate as to the necessity of this since virtually all (with very few exceptions) DCFC equipment that provides 150 kW can do so at 920V.

As a note - pack voltages are often lumped into two categories "400V" and "800V". Both actually cover a range of actual voltages depending on pack architecture. Two voltage designations come into play - nominal and max. A fully charged Li Ion cell will be 4.2V, nominal voltage the average voltage it can provide from fully charged to fully discharged (AJ would provide a highly technical and slightly more accurate explanation but "average" will suffice here). Nominal voltage will be between 3.6 and 3.7V. Nominal voltage gives you the capacity/energy of a battery, but max voltage is what is relevant to the charger you are using.

The two most common "400V" configurations are 96s and 108s (number of cells in series).
96s yields a nominal voltage of about 350V and a fully charged voltage of about 400V (Tesla, Chevy, etc).
108s yields a nominal of about 400V and a max of about 450V (Rivian uses this layout)
Other configurations are possible - Honda uses 84s for 311V nominal 353V max in their Clarity series.

The Taycan "800V" is 198s for a nominal voltage of 723V and max of 832V.
I think it likely that they Hyundai Ioniq 5 is 192s for 710V nominal 800V max. Likely at least part of the reason they are getting 232 kW max on the specified 350kW charger from early reports - Hyundai specifies either 150 kW or 350 kW chargers when making references.

GM is utilizing the switching scheme we all thought was covered in the Rivian patent for it's 800V system in the "800V" Hummer:
HUMMER EV will have the ability to take advantage of the industry’s fastest 350-kilowatt DC fast chargers4. This vehicle will offer a GM-estimated 350+ miles7of range based on preliminary testing, with a unique ability to switch its battery pack from its native 400-volts to 800-volts for charging. A disconnect unit and mechanization within the pack enables the battery to switch from “parallel” to “series,” allowing it to add nearly 100 miles of range in 10 minutes of charging8.

Or:

On the battery side, however, GM says that the Ultium propulsion systems underpinning the Hummer EV are indeed fully developed. The so-called double-stack battery has 24 modules in all—two levels, each with 12 modules—and while it runs at 400 volts it charges at 800 volts, allowing 100 miles of recovered range in just 10 minutes, from low states of charge.
“It’s just not manageable to have the huge power cord at 240 kilowatts at 400 volts,” said Tim Grewe, GM’s director of battery cell engineering. “So it’s much more rational to go up to 800 volts to make that happen.”
Grewe said that essentially amounts to treating the two levels as separate packs—running them in parallel for vehicle operation, but allowing them to be placed in series during DC fast charging. As part of GM’s from-scratch Ultium modular approach, the smart cell-module assembly has the cooling integrated into it, where each module essentially has a plus and a minus that allows for coolant flow.

Given the Hummers modules reliance on 24 cell groupings, it is likely they are going to be 96s/192s (usage/charging).
Well said Duc!

One other benefit (I think) of 800/900v systems is needing less amperage for a given kw. A lower amperage produces less heat (more energy efficient), with heat loss = I^2*R*t. So if you want to get to say... 250 kW from a 400v system, you need 625 amps. 312.5 amps at 800volts. The 400v system would produce 4x more heat. So if you want to manage heat loss, both in charging cables, and in the EV system, I *think* going to an 800v system is more efficient.

That's assuming the same resistance. A little rusty here, but if I remember right, wire sizes need to increase for additional amperage, which would increase the resistance, and increase heat loss.

None of this is a huge total loss compared to say an ICE, but if you're trying to maximize system efficiency, a higher voltage system is superior. Which, if I understand correctly, is part of the reason bulk power transmission lines are high voltage systems that are then stepped down locally.
 
Last edited:

cwoodcox

Well-Known Member
First Name
Corey
Joined
Mar 31, 2021
Threads
0
Messages
182
Reaction score
201
Location
Montreal, QC
Vehicles
2012 Ford Expedition
Occupation
Software Engineer
I love @DucRider’s whole post because it’s so well-informed and is likely exactly right, but I still think that Rivian will never release any of this because most people just don’t care. It over-complicates the product, and as long as they can make good on the charging performance claims they really don’t need to.

We’ll find out exactly how it works once someone gets a hold of one and does some measurements. ?
 

Gshenderson

Well-Known Member
First Name
Greg
Joined
Sep 28, 2019
Threads
13
Messages
1,229
Reaction score
2,767
Location
Park City, UT / Kemmerer, WY
Vehicles
2015 Tesla S 85D, 2019 4Runner TRD Offroad, R1T
I love @DucRider’s whole post because it’s so well-informed and is likely exactly right, but I still think that Rivian will never release any of this because most people just don’t care. It over-complicates the product, and as long as they can make good on the charging performance claims they really don’t need to.

We’ll find out exactly how it works once someone gets a hold of one and does some measurements. ?
Once they get closer to shipping, they will likely want to do another “launch” event (perhaps virtually) with some level of PR and media buzz. Trade rags are all going to be asking these questions, so it’s going to be hard for them not to answer them.
 

SeaGeo

Well-Known Member
First Name
Brice
Joined
Jan 12, 2021
Threads
47
Messages
5,235
Reaction score
9,677
Location
Seattle
Vehicles
Xc60 T8
Occupation
Engineer
I love @DucRider’s whole post because it’s so well-informed and is likely exactly right, but I still think that Rivian will never release any of this because most people just don’t care. It over-complicates the product, and as long as they can make good on the charging performance claims they really don’t need to.

We’ll find out exactly how it works once someone gets a hold of one and does some measurements. ?
That's an interesting take on Rivian, which assumes they're proposing to be equivalent to Apple in hiding specs. That's a totally different approach than other car manufacturers (other than tesla). From a demographics standpoint I suspect Rivian may have more people that care than a traditional manufacturer. In the end it doesn't matter much, but I wouldn't blindly assume Rivian has the same philosophy to technical specs that Tesla does. For example, we know the Rivian's torque and horsepower numbers (or at least have an idea).
 

SeaGeo

Well-Known Member
First Name
Brice
Joined
Jan 12, 2021
Threads
47
Messages
5,235
Reaction score
9,677
Location
Seattle
Vehicles
Xc60 T8
Occupation
Engineer
Once they get closer to shipping, they will likely want to do another “launch” event (perhaps virtually) with some level of PR and media buzz. Trade rags are all going to be asking these questions, so it’s going to be hard for them not to answer them.
This is where I've been leaning as well lately.
 

Sponsored

Billyk24

Well-Known Member
First Name
William
Joined
Mar 21, 2020
Threads
4
Messages
108
Reaction score
39
Location
PA
Vehicles
Ford C-Max Energi, Premium Mach-E ordered
Occupation
health care
The thing is there's an assumption here of *when they enable* more. There is nothing saying that will actually happen with the R1 series.

To come at this from another route, if they averaged say... 275 kw over 20 minutes instead of ~190 kw, that's 200 miles in 20 rather than 140 miles. Driving for almost another hour after charging for the same timespan isn't nothing. On a say... 600 mile trip, that's at a minimum 17 fewer minutes if charging. Keep in mind that it's unlikely that the 140 miles are highway miles.

Realistically the R1T "large" pack probably couldn't pull 275 for 20 minutes, but just pointing out that flipping the way we look at the problem can make the change how you view the impact of the decision.

Definitely not the end of the world either way, but 800 or 900v architecture can make a decent difference in trip duration. Especially if you have more than 100ish miles to the next charger.


Well said Duc!

One other benefit (I think) of 800/900v systems is needing less amperage for a given kw. A lower amperage produces less heat (more energy efficient), with heat loss = I^2*R*t. So if you want to get to say... 250 kW from a 400v system, you need 625 amps. 312.5 amps at 800volts. The 400v system would produce 4x more heat. So if you want to manage heat loss, both in charging cables, and in the EV system, I *think* going to an 800v system is more efficient.

That's assuming the same resistance. A little rusty here, but if I remember right, wire sizes need to increase for additional amperage, which would increase the resistance, and increase heat loss.

None of this is a huge total loss compared to say an ICE, but if you're trying to maximize system efficiency, a higher voltage system is superior. Which, if I understand correctly, is part of the reason bulk power transmission lines are high voltage systems that are then stepped down locally.
800V battery requires less amperage, resulting in less heat, meaning the charging curve can be flatter or sustained at a higher level than a 400V battery pack. Thus significantly shorten charging times.

What is the cost difference between 400V and 800V battery packs? The difference is due to?
 

DucRider

Well-Known Member
First Name
Gary
Joined
Oct 21, 2019
Threads
17
Messages
1,652
Reaction score
3,143
Location
ORegon
Vehicles
Polestar 2, Ioniq, R1S
800V battery requires less amperage, resulting in less heat, meaning the charging curve can be flatter or sustained at a higher level than a 400V battery pack. Thus significantly shorten charging times.

What is the cost difference between 400V and 800V battery packs? The difference is due to?
Same amount of energy is going into the same number of cells for a given kW rating. On the module/cell level there is no difference in heat related directly to voltage. On the charge cable and the wiring to the pack, the wire size can be reduced and there is some cost/weight saving there.

No real cost difference between a 400 and 800V pack. The wiring leading to/from a 400V pack would need to be beefier for 400V, but little to no cost difference in the pack itself.
Inverters and motors will be more expensive (and Rivian uses four of each).
 

SeaGeo

Well-Known Member
First Name
Brice
Joined
Jan 12, 2021
Threads
47
Messages
5,235
Reaction score
9,677
Location
Seattle
Vehicles
Xc60 T8
Occupation
Engineer
800V battery requires less amperage, resulting in less heat, meaning the charging curve can be flatter or sustained at a higher level than a 400V battery pack. Thus significantly shorten charging times.

What is the cost difference between 400V and 800V battery packs? The difference is due to?

The head is in the wires, not the pack for the voltage. That being said, you can have a flat curve with a 400V pack (see the e-tron). If the pack design sucks and doesn't manage heat well, it doesn't matter whether you use 400v or 12 kV, you'll get throttled either way. The Audi pulls of a flat 150 kw because they have really good cooling.

The R1 large pack and max pack are large enough that I don't think they'll have much problem managing the heat for 20ish minutes. I'm more curious what happens after the 20 minutes. Duc probably knows better of the top of his head, and had some great posts back in 2020, but I don't think the current EA chargers really allow you to pull much more than 200kW with a 400/450v system. So I'm assuming it's a fairly flat curve for that first 20 minutes. 140 miles implies about 45% of charge gained in 20 minutes. If they pull off a truly flat curve at say 190 kW to 80% like the etron that's about 30 minutes from 10% to 80%. This is a low enough c-rate (just 1.4) that I think that's totally reasonable. The e-tron's C-rate is about 1.6.

However, if Rivian pulled off the Ioniq 5 charging curve (based on c-rate) it'd take say... 20 minutes. vs. 30 minutes for a flat 190 kW to go from 10% to 80% for the large pack, and 40 minutes for the max pack.

Just for fun I've plotted up charging time vs %charge and miles gained assuming you plug in at 10% for an Ioniq5 type charge curve based on c-rate vs. a straight 190 kw charging curve. I also tossed in what a max pack curve could look like. Interestingly, the max pack at an Ioniq5 type C-rate would max out the 350 kW charging rate to start. I've assumed you plug in at 10% in these charts. Just interesting to see the *potential* impacts of architecture when accounting for the battery sizes and comparative C-rates. The ioniq5 is what I'd assume to be a best case scenario.

As I think it was DucRider that pointed out, while it may not save a ton of time for each person, the quicker charging time does definitely help with charger wait times. And for really short charging intervals (for example, plugging in for 10 minutes at a low state of charge).

Rivian R1T R1S New Official Rivian Adventure Network and Charging Station Info 1619386172265

Rivian R1T R1S New Official Rivian Adventure Network and Charging Station Info 1619386457914
 
Last edited:

Gshenderson

Well-Known Member
First Name
Greg
Joined
Sep 28, 2019
Threads
13
Messages
1,229
Reaction score
2,767
Location
Park City, UT / Kemmerer, WY
Vehicles
2015 Tesla S 85D, 2019 4Runner TRD Offroad, R1T
Same amount of energy is going into the same number of cells for a given kW rating. On the module/cell level there is no difference in heat related directly to voltage. On the charge cable and the wiring to the pack, the wire size can be reduced and there is some cost/weight saving there.

No real cost difference between a 400 and 800V pack. The wiring leading to/from a 400V pack would need to be beefier for 400V, but little to no cost difference in the pack itself.
Inverters and motors will be more expensive (and Rivian uses four of each).
Are there safety considerations at play with 400v vs. 800v? If things go wrong, would 800v have the potential to be a worse outcome? My limited experience with electrical engineering as all been on the low voltage / low power end of the spectrum.
 

DucRider

Well-Known Member
First Name
Gary
Joined
Oct 21, 2019
Threads
17
Messages
1,652
Reaction score
3,143
Location
ORegon
Vehicles
Polestar 2, Ioniq, R1S
The pack voltage makes no difference in all these comparisons - C-rate, charge curves, etc.
800V charging allows for higher kW (and C rates), but the Hummer and (eventually) the Rivian will utilize 800V charging on 400V packs (or 900V on 450V for the Rivian).
But, as an example, the Rivain charging at 450V and 450A would be a touch over 200 kW. Charging at 900V and 225V yields the same charge (200 kW) and C rate.

Are there safety considerations at play with 400v vs. 800v? If things go wrong, would 800v have the potential to be a worse outcome? My limited experience with electrical engineering as all been on the low voltage / low power end of the spectrum.
Higher voltages will arc easier and therefore require more spacing/insulation between conductors. RJ at one point stated the entire vehicle was designed with that in mind to allow for a future change to 800V components.
Sponsored

 
 




Top