Tesla Owners Online Forum banner
Status
Not open for further replies.

V3 Supercharger vs. New Tesla Plug vs. Double Tesla Plugs

11K views 42 replies 10 participants last post by  4701 
#1 ·
I was just thinking about the tweet from Elon about the possible new Supercharger going past 350kw. I was thinking about how this much higher amperage/voltage would affect the current Tesla connector and how it might have to change. The current connector is rated for ???. However looking at the sizes of these plugs (combo/CCS and Chademo) note the possible wire guage differences. It would be nice to keep the plug the same but if we are to double the current would Tesla move to double connectors(keeping the old one but two of them) or update to a new connector. I do not see how they could increase the guage of the connector and maintain the plug shape/socket. Thoughts?

 
See less See more
1
#2 ·
They tried liquid-cooled cables...

I don't think they're going to be changing the North American Tesla plug standard for 350Kw anytime soon. Im thinking at this point Model 3 was designed to handle any new higher power standards given the design timeframe of the car and the new battery pack systems. Model S and X will eventually have the new cells/pack in time. Europe imposes the type 2 connector on Tesla so anything can change over there.

Lastly, these large plugs (anything over the type 2) just don't fit in the Tesla taillight housing so where would they put the connector? Their design aesthetic pretty much dictates they will stay within current plug confines.
 
#4 ·
This is actually something that I have been wondering about quite a bit. What is the theoretical power limit of the Tesla connectors (or any of the other connectors for that matter)? We know that 120kW is the Supercharger limit for V2, but not necessarily the connector. Does the connector even have a power limit? That is, is it the physical connector that matters or other parts of the specification? Do the amount of pins matter?

I would love to have someone who is an expert on the subject comment on these things.
 
  • Like
Reactions: Badback
#5 ·
This is actually something that I have been wondering about quite a bit. What is the theoretical power limit of the Tesla connectors (or any of the other connectors for that matter)? We know that 120kW is the Supercharger limit for V2, but not necessarily the connector. Does the connector even have a power limit? That is, is it the physical connector that matters or other parts of the specification? Do the amount of pins matter?

I would love to have someone who is an expert on the subject comment on these things.
The connector has a CURRENT limit. As current passes through a connector, some power(heat) is lost at the contact point. The connector has some small resistance. The heat (power) loss is equal to current^2*resistance (I^2R). The contacts heat up as current increases. Some of this heat is conducted to the body of the connector, which will eventually melt, long before the metals in the contacts melt. Usually resistance will increase with temperature which will result in a run away temperature rise.

So, the current limit is that which will compromise the safe use of the connector.
 
G
#8 ·
It appears Tesla US plug is running near max current it can handle in ideal conditions, 335A.
Same is for non-US solution, aka enhanced Type2.
Actually often 335A is not available. And some of those instances it appears to be the coupling limitation.

When Elon mentioned 350kW, he mentioned charger, not the stall. That's a big difference.

I'm betting that Model 3 will have "enhanced Combo2" style port, that supports Type2 enhanced SC plug
and also CCS-s that will be soon available. It's bigger than S/X but M3 taillight design appear to have bigger flap too.

I can't imagine what happens in US. Maybe M3 will keep the regular port on drivers taillight and add "not enhanced Combo1" to other side.
In near future CCS will be more available than ChaDeMo today. Everywhere, except Asia I believe.
 
#9 ·
Seems like the best answer is a much higher charge voltage, to reduce the current that needs to flow through the connector for any given power level. After the connector (inside the car), you can step the voltage back down (and the current up) at the point it interfaces the battery. For instance, a 120 kW charger at 400 V nominally transfers a bit more than 300 A. A 1200 V system could transfer that same power using just 100A. Or more likely for the original 300 A it could provide over 350 kW.

I also think another key to fast charging is to have a flexible cell switching architecture that lets you charge in one configuration (more of the cells in parallel, charging all at once with independent current paths) and then the car runs in another configuration (more of the cells in parallel). This is in keeping with Tesla's design philosophy, which can be summed up as "electronics and software are cheap." (To wit: this is why they use AC induction motors rather than DC motors).
 
  • Like
Reactions: Tom Bodera
#10 ·
Problem with Higher voltage would be that the arc length between the contacts is higher. This may be a problem with the distance between the two connectors on the current plug (400v vs. 1200v).

Anybody have any technical idea if two separate existing tesla connectors could be used in parallel for increased power transfer?
 
G
#13 · (Edited)
Unfortunately electricity is not that simple. Arch is more complicated than that.
Especially DC arch. It is not self-extinguishing like AC arch.
Numbers you gave are for dielectric medium. But arch itself is excellent conductor.
Engineering is more complex and 1200V is definitely NOGO for Tesla SC connector.
Coupling will not be able to exctiquish 1200V DC arch if locking mechanism fails.
And this is the main presumption - locking mechanism can never be trusted 100%.
I've seen multiple ChaDeMo chargers that have connectors heavily burned.

Ok', I'll stop talking and just demonstrate, what is the difference between
220V AC and 220V DC arch under load.
 
#14 ·
Unfortunately electricity is not that simple. Arch is more complicated than that.
Especially DC arch. It is not self-extinguishing like AC arch.
Numbers you gave are for dielectric medium. But arch itself is excellent conductor.
Engineering is more complex and 1200V is definitely NOGO for Tesla SC connector.
Coupling will not be able to exctiquish 1200V DC arch if locking mechanism fails.
And this is the main presumption - locking mechanism can never be trusted 100%.
I've seen multiple ChaDeMo chargers that have connectors heavily burned.

Ok', I'll stop talking and just found a demonstration, what is the difference between
220V AC and 220V DC arch under load.
Wow. Great info and presentation.
 
G
#17 · (Edited)
Don't Tesla's include arc detection in their control circuitry?
The fact that arc exists means something is wrong with management electronics (for example damage occurred and control compromised). Any non-isolated detection mechanism is not reliable in this scenario.
Usually loss of pilot/control signal would stop the current. How many milliseconds SuperCharger needs to stop the flow after pilot signal is not in range is most likely only known by few Tesla engineers.
But having SC disconnected doesn't mean the end. If arcing is already happening it might also happen on car side as well.
I don't have information is the "onboard charger bypass contactor" capable to disconnect 100-1000A current. Most likely not. It would be gigantic in size. The fuse in the pack might not blow (designed for drivetrain protection, not supercharging, 1300A or 1500A depending on pack version).

That relay most likely separates contacts in normal atmosphere with only few millimeter distance with nothing for arc suppression.
Detecting DC arc is not enough. Disconnecting load will not work and opening regular relay will only make new arc in the relay.
If Tesla has another fuse on the DC input line, then this will solve everything. Would be awesome to know the value.

Thinking exercise:
Imagine 1200V Tesla. You have an accident at supercharger. Somebody drove into the stall and ripped the plug out of your car.
Supercharger de-energizes in fraction of a second. But arc is already stable at the vehicle port side. How would you extinguish the
arc if it draws below 1500Amps. And there are no additional fuses besides the one in the pack rated for 1500A made out of inconel.
Video above visually shows how long the arc will be in open air if was initiated at 220V 12A.
 
#18 ·
Relays? Don't they use high voltage IGBTs such as the ones JB has discussed from time to time on both sides of the connection? IGBTs can handle well over 4KV at 90A, in parallel if necessary to meet current requirements.

The essence of a control circuit is that it isolates the circuit when disabled or shorted. Just like you can't shock yourself by cramming snow or fingers inside the charging port of your Tesla, you can't sustain an arc after the plug is removed.
 
G
#19 ·
The essence of a control circuit is that it isolates the circuit when disabled or shorted
There are no semiconductors between li-ion cells and supercharger inverters (a dozen in each box) when DC-charging.


If you stop the current flow before arc jumps into direct contact between positive and negative sides there will be only a slight carbon residue at the place where two conductors were separated.

IGBTs are used in powertrain inverters. They can't be used as fuses. They need to be fused as one of their failure modes is short circuiting.
 
#20 ·
My main point is that if you've got arcing at the connector, you've got a system failure— whether you used a relay or a solid state component, or 400 V or 1200 V. As you've shown, arcing has to be prevented either way. And from an overall heat, cost, and efficiency perspective, high voltage would help. You just need efficient voltage conversion and fast protection circuitry.
 
G
#21 ·
All these connectors do prevent arcing on their own at 400V. I'm pretty sure. But non of those prevent at 1200V.
In addition to any protection circuit, connectors (plugs and sockets) are REQUIRED to be capable suppressing the arc at the voltage/current they are rated. CHAdeMo and Mennekes without optional pins are both rated to 500V max. Tesla connector most likely too (looking at the distances between +/-).
I'm interested in CCS DC pins. They are further apart and deeper inside than Tesla pins and Mennekes main part.
Chademo plug has even arc suppressing covers that are retracting with springs, CCS does not.
It appears CCS DC pins (US and EU both) might get rated for 1000V (they only way to get 350kW rate).

The problem is not the charger protection . The problem is vehicles protection. Battery can not be disconnected while under load with "electronics" nor relays (arc is a load too).

I'm still waiting for solutions:
Thinking exercise:
Imagine 1200V Tesla. You have an accident at supercharger. Somebody drove into the stall and ripped the plug out of your car.
Supercharger de-energizes in fraction of a second. But arc is already stable at the vehicle port side. How would you extinguish the
arc if it draws below 1500Amps. And there are no additional fuses besides the one in the pack rated for 1500A made out of inconel.
Video above visually shows how long the arc will be in open air if was initiated at 220V 12A.
So the arc is right here:
 
#22 ·
If the charger control system failed AND the car control systems failed AND you could start an arc across the terminals of the car, YES, it would be bad.

But even assuming those simultaneous failures happened, how would start that arc, since the terminals never come in contact with each other?

Are you assuming you could start an arc between the cable and the car and then "walk that over" by drawing it across and touching the opposite contact? Remember, unless you walk it over and make direct contact with the other terminal, the arc would have to come within a fraction of a millimeter to allow the arc to transfer to the second terminal, because even 1200 V isn't enough to jump even a 1 mm air gap. However, the connector is designed with recessed poles and it only fits one way, so there's no way to rotate it and touch the other pole, or even get the arc within 1 mm of it. So even with multiple control failures in the system, you can't use the charging connector to start an arc.
 
  • Like
Reactions: Red Sage
G
#23 · (Edited)
1) arc is extremely happy with jumping across different sources/conductors.
So when somebody drives into a SC stall and rips the plug out of the vehicle while charging, there will be 2 arcs between terminals that just got separated: one arc between positive socket terminal and positive plug terminal and another arc between negatives. As soon as plug is separated further than the distance Tesla socket +/- terminals are from each other, 2 arcs will join into one IF voltage is high enough. Arc will always choose the path of least resistance.


2) there is no control system on car side. I think I already mentioned that. It's not specific to Tesla. Everything is bypassed if charging with DC station.

Like I said, it is not legal to suppose arc detection mechanism will deal with all problems. Coupling must be able to extinguish the arc.

PS: that boomerang shaped triangular piece of plastic (under my spark drawing) between the terminals is for arc suppression. I think plastic tips on positive/negative terminals (both plug and socket) are also helping with arc suppression. Ground and pilot signals have neither of those.

because even 1200 V isn't enough to jump even a 1 mm air gap -
air at room temperature with no ionization and no metal vapors from conductors. Air near arc is not "air".
I would recommend dropping that 1mm arc "I'm feeling safe near 1200V DC" thing. It is true only in no arc situation. Just another cool demonstration of 300V DC arc with extremely high resistance path (lightbulb). You can clearly see that handheld stick doesn't create arc until conductors pretty much touch. We, on the other hand, have opposite situation. We have our plug and socket mated until accident happens and THEN start pulling terminals apart.
 
G
#24 ·
I just though to estimate timing. Vehicle travelling at 10m/s (36km/h 22mph) hitting SC stall will result plug moving away from the socket at the speed of 1cm per millisecond.
Not a single arc detection system can detect arc that fast. Fastest way is to use optics and this still takes 2-5ms.
SuperCharger control circuit will also need at least 10-100 milliseconds to execute the stop command.
For that time we already have two separate arcs that are 2-5cm long. Long enough for them to join. Game over.
https://static1.squarespace.com/sta...8a4/t/563cb0e7e4b029091adefb3c/1446818026345/
This is realistic distance of the plug 10ms after loss of contact.

If we assume pilot signal loss will initiate shutdown (SC control system fully functional),
it will still take those 10ms to stop the current from flowing.
Imagine that plug position (photo) while there are two arcs between appropriate terminals. They will DEFINITELY JOIN as distance between them is much smaller. Anybody, what is the distance between +/- terminals on US Tesla plug/socket?
 
G
#28 ·
http://www.abb.com/cawp/seitp202/f8e7a4a7e7496bddc125792900742d15.aspx
First link on Google about arc detection.

There is no information for how long is DC station allowed to continue providing current after pilot signal/ground loss.
I found something about adjustable value but nothing about requirements to SC, SAE Combo, CHAdeMO.
Adjustability had something to do with false positives. Pilot signal has square wave oscillating at 1kHz, one cycle per 1ms.
One interruption due to any electromagnetic interference would kill charging immediately, which is likely not wanted.
Also possible to detect ground loss. Again, found no information during 1h googling about power supply delay requirements.

Do I need to explain SC plug being torn off the vehicle at speed 1cm per millisecond?

Somebody please measure approximate distance between positive and negative terminals on Tesla socket.
EU version pins 16mm, plus additional depth minus pin diameters, so around 22mm minimal short circuit arc length.
 
#32 ·
If my car gets hit while Supercharging, I'm absolutely certain the occupants of my vehicle will be out before the car can burn to the ground. My next action is to check the condition of the person who hit me, followed by a 911 call. Either way, we're out of the car and if it burns to the ground, I will be thankful that my family wasn't harmed and that I have insurance. I will be hopeful that the other person has insurance as well.

Finally, I'll wonder how long the wait is to get the replacement Model 3....I did reserve on 4/5/16 and don't know how many unfilled reservations would be in front of me if it happened shortly after I received it.

As a side note to Trevor: Kudos on fixing the software font to change the way the three shows up when I post a message from the way it looks while I'm typing it in. You're the man! :cool:
 
G
#36 ·
Incorrect. Current per cell is the same as the voltage for most lithium cells in the same no matter the pack voltage. Somewhere between 3-4.5V.
Actually yes. Current is what causes heat in the charging cable and circuit. There is almost no problems with that heat as they have efficiency way above 99%.
 
#37 ·
Incorrect. Current per cell is the same as the voltage for most lithium cells in the same no matter the pack voltage. Somewhere between 3-4.5V.
"Current per cell is the same as the voltage for most lithium cells" is nonsense. For instance, a typical cell might have a mean voltage of 3.6V and might have a max discharge of 10A and normal (rated) discharge of 1A. There is no equivalence between current and voltage, nor does that even make sense. I think you meant to make an off-the-cuff comment about the frequent comparisons between capacity and max current outputs and charging currents, sometimes expressed in the form "1C" as in a 3.5 Ah capacity of battery can deliver a max of 3.5A of current.

Actually yes. Current is what causes heat in the charging cable and circuit. There is almost no problems with that heat as they have efficiency way above 99%.
C'mon. 1% of 120kW is 1.2kW. That will melt metal. Yes, conductors have very little resistance. But we're talking gobs of power, so unless you up the voltage a lot (and lower the current), you have to start talking about active liquid cooling and the like. Raising the voltage is a better way to do it. Liquid cooling is expensive and requires more maintenance.
 
#39 ·
"
There is no circuitry between DC charger and cells."

So then it's your professional opinion that you can open the charging port of a Tesla and measure the cell module voltage with a voltmeter?
 
  • Like
Reactions: Steve C
G
#40 ·
Why did you even mention efficiency? It doesn't matter if you drink two half liter bottles of coke
simultaneously or be a smart-arse and pour those two into one jug and hope for the best.
You still drink the same amount of nasty crap.
Problem with charging is not input voltage. It's the heat. And NOT CIRCUITRY HEAT.

The circuitry in the car can then easily step the voltage down
You can't cut 800V DC in half with some wire contactor and a multimeter :tmi::unamused:
 
#41 ·
If you think there is not circuitry between the charging port and the battery, then you believe that you can put a voltmeter on the connectors and measure the module voltage. Correct?
 
G
#43 ·
If there is no circuitry between the charging port and the battery, then you can not measure voltage with a voltmeter at the charging port.
If there is circuitry between the charging port and the battery you still can not measure voltage with a voltmeter at the charging port.

Tech CEO should know that.
 
Status
Not open for further replies.
You have insufficient privileges to reply here.
Top