They don't use relays to disconnect the power in the car. It's all solid state switching, in the inverter I expect. The EVSE has a relay which is only switched when there is no power flowing. Relays are faster than you can yank the connector, but they pit when operated under power and have a finite lifetime in those cases. Semiconductors can operate indefinitely if kept within spec.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
They don't use relays to disconnect the power in the car. It's all solid state switching, in the inverter I expect. The EVSE has a relay which is only switched when there is no power flowing. Relays are faster than you can yank the connector, but they pit when operated under power and have a finite lifetime in those cases. Semiconductors can operate indefinitely if kept within spec.
Eno Deb
Active Member
You should really catch up on what was written earlier. Sudden disconnect is a potential safety issue (e.g. it could cause arcing). It may be useful for the car or charger to prevent that by locking the plug. That's what is being discussed here.
AlanSubie4Life
Efficiency Obsessed Member
Good point. But in any case it clearly happens. Just gotta be quick on the draw.
In any case seems very prudent to have a locking mechanism for DC charging, for this reason.
Then the model 3 has yet another issue that needs to be addressed. The bad part is not really safety since it would be hard to be injured unless you had your other hand right by the connector (which is entirely possible I suppose) but that it wears the contacts which should last a lifetime if they aren't pitted by this sort of malfunction.
This is an EV 101 sort of problem. It literally should never happen.
AlanSubie4Life
Efficiency Obsessed Member
That's entirely possible. All the more reason for the DC connector design to be robust to car "lagginess".
Or just fix the damn car! The design in the EVSE is done to prevent a need for anything like software which could have delays or bugs. What I recall is just simple hardware. Even with that, in the J1772 interface there are two safety features so both would have to fail to have a safety problem which is standard procedure. The fast DC connectors have the same sort of pin length feature that forces the car to shut down current before the high current DC pins are disconnected. But this should not be software. There should be no "lagginess" involved. If Tesla designed an interface that doesn't ALWAYS immediately shut down the power draw on losing the pilot signal or receiving the unlatch command from the button, it is a faulty design and needs to be fixed.
Using a car controlled lock on the connector is something that can break and not do what it is supposed to do. It's fine if used, but it can't be used alone.
Eno Deb
Active Member
It's not just about detecting that the plug is being removed. If high power loads are suddenly stopped, this can cause very high voltage transients (as capacitive and/or inductive energy stored somewhere in the electric path is released). That's why a ramp is used when the power is turned off orderly. If the user manages to push the button and disconnect the last-break pins before the ramp-down is complete, arcing can occur. For J1772 the risk has apparently been deemed manageable, since there is no requirement to lock the plug. The question is whether that is still the case for DC charging with its different characteristics.
AlanSubie4Life
Efficiency Obsessed Member
I didn't say they should not address the issue in the car, if possible. However, that alone might not be enough...when it comes to DC charging & associated adapters, I think it would probably be good to have multiple levels of protection. Just as long as they aren't a huge pain...
This is missing the bigger picture. You are looking at only how long it takes for a signal to be detected. And you are probably right, that it is only a few milliseconds. But have you been watching the charging screen when you touch the "Stop Charging" button? It does take almost a second for the current measurement shown to lower itself down to 0 amps. That is what is being talked about here. When the car gets the signal either from the touch screen or from a press of the J1772 latch button, it does that medium speed ramp down of the current to be easier on the components, and that's where this delay is coming from, not in waiting to receive the signal.
israndy
Supercharger Hunter
I am guessing you saw they are BACK in stock again, again.
I was looking at all the places that I can go with my CHAdeMO, and noticed that I can go to Europe and China. When is the next EV AROUND THE WORLD trip planned? I can take my Model 3 now. That's a big advantage over any NA CCS adapter. The EU CCS adapter doesn't even plug it into my car.
This will be so much faster than trying to use my 220v adapter over there. I just did an empty to 90% full charge with the new chaDEmo adapter in less than an hour and a half. Facebook's, for FREE, SCHWEET!! Starts to drop off quickly right as you get to 90%, but it was ~48kW until then.
If only there were better route planning tools using a mix of charger formats. Specifically, need to come up with the CHEAPEST route.
NA CCS adapter won't help you in the EU - the top half of the CCS plug is different in the two regions, a J1772 without the power pins here, and a Mennekes Type 2 without the power pins over there. To charge a US Tesla using CCS in the EU, you'd need yet another adapter.
CHAdeMO is the same everywhere, though somewhat limited in power...
In J1772 AC use the power goes from zero to max and back to zero 120 or 100 times a second. Ramping the power is not the problem. With the DC load it does not take large fractions of a second to ramp the power up or down.
There is not likely any significant inductive load on the DC input. The switching supply drawing power switches at rates of 1000's of times per second to allow the use of small inductors and capacitors lowering the size, weight and cost of the supplies. Even a slow switching supply operates at 20 kHz.
The issue is interrupting the current while it is still in use, not an issue of not being able to ramp the current down quickly enough. For that the two safety circuits should be adequate.
Tesla does have a lock on releasing the charging connector. Sometimes it can get stuck. When that happens you have to open the rear of the car and start disassembling it to reach the mechanical bits that you can manipulate to force the connector to be released.
Then on top of that, there are situations where the car in general can just get hung up and nothing electrical works. Remember, these cars are less smart cars and more like a cell phone with wheels and a very big charging cable. So again, climbing through the trunk so you can remove the charging cable and have the vehicle towed.
That's enough of an inconvenience for me to live with two levels of protection and not require a third.
If you are talking about a CCS adapter, then the car would only lock the adapter. The CCS connector would still work like a CCS connector.
There is nothing on the display that is actually real time. It's all being passed through multiple levels of microprocessors and the biggest time loss is in the UI itself.
How long do you think they have to ramp down the current and why exactly? Which components need to be treated so delicately? Nothing in the power path is incapable of being switched off without harm faster than you can blink your eyes. Heck, most of the electronics is there in order to switch the power on and off at high rates, much higher than you can even hear typically. If they were switching at rates below 20,000 Hz you would be able to hear the supplies singing like banshees. Even with a DC input, the voltage to the battery has to be adjusted so they aren't charged too fast. They keep the power losses low by using switching supplies. They switch off and on at very high rates. In less than half an AC cycle the power goes to zero if that is desired.
Rocky_H disagreed with this post, but hasn't said anything about why. Rocky, care to explain what it is you disagree with?
vdiv
Chief Grump
Don't know what quantities Tesla makes available with the CHAdeMO adapter, but the constant in/out of stock indicates that people are buying it.
Out of stock again.
The model 3 has a pull release cable for the charge port accessible from the trunk, you don't have to disassemble anything. I have a YouTube video that goes over charging adapters and cables and also shows the release cable. Easy peasy.
There is CCS Combo in a lot of places the Tesla network doesn't reach. Can't get to Gaspé, Qc ... Can´t get to Rouyn-Noranda. If I click CCS-Combo in abrp. I can get to both places totally direct. This lack of CCS-combo is already a pain in Canada.
There is not a single supercharger on I26 in Columbia (and no, the ones that add more than 20 minutes of transit time off the interstate do not count). There is an Electrify America station right in the middle, though with 250KW support.
There is also no supercharger in Charleston, SC, but there is an EA station with 350KW ports.
KootsChewt
Member
Same here in BC. At the moment a CHAdeMO adapter would allow me to cover these routes and not have any difference, but the higher-powered Petro-Canada CCS stations are coming online now. Looking to the future, I don't want to sink $600 into an adapter that seems destined for the anals of history.
On another note, when I searched "Tesla CCS adapter" today to see if any new info had popped up in the last few months, I found a link to a CCS adapter on the Tesla Canada website! Sadly, it is for the Type 2 adapter for Europe
Tesla CCS Combo 2 Adapter
And for fun I made a meme and posted on Twitter to try and get Musk's attention... if you are are on Twitter, maybe repost and see if we get any traction?!
Kootenay EV Family on Twitter
