Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

J1772: Please explain the "signal" charging stations place in their current feeds...

This site may earn commission on affiliate links.

TOBASH

Member
Supporting Member
So, Tesla Roadster had an 80 amp bug where it would not charge after sensing a signal in the charger feed.

Please explain what that signal is exactly.

Searched and maybe my search terms were incorrect.

If there is/are threads on this I apologize. Please direct me to them.

Thanks,

T
 
Thanks for the link. I will research what is over my head, but this begs the question... Can't a plug be connected to a computer to falsify the amperage signal?

Specifically, say a Nissan Leaf connects to a 72 amp system, can't an interposed computer falsify the signal?

Now, would that leave the Nissan Leaf open to getting fried by higher amperage OR would the Leaf be able to self regulate an only use what it needs without overloading?

Thanks,

T

P.S. - Now that I own a roadster I want to dive into understanding the electronics and the basics of charging. Please bear with me.
 
Thanks for the link. I will research what is over my head, but this begs the question... Can't a plug be connected to a computer to falsify the amperage signal?

Specifically, say a Nissan Leaf connects to a 72 amp system, can't an interposed computer falsify the signal?

Now, would that leave the Nissan Leaf open to getting fried by higher amperage OR would the Leaf be able to self regulate an only use what it needs without overloading?

Thanks,

T

P.S. - Now that I own a roadster I want to dive into understanding the electronics and the basics of charging. Please bear with me.
For most everything except DC fast-charging (in other words, ignore Superchargers and CHAdeMO for a second), the "charger" is inside the car. The thing outside the car is the "EVSE", which is basically just an intelligent switch to connect/disconnect the car's charger to/from its 120VAC or 240VAC power source.

The point of the signaling is to tell the car's on-board charger how much continuous current it can SAFELY pull. That way the same plug/receptacle can be used for everything from the big 240V 70/80A EVSE stations, all the way down to the small 120V 12A units that plug into a typical household outlet.

The car gets to decide how much current to pull from the EVSE to charge its battery, and for how long, so it can't get maliciously "fried by higher amperage". But if the car doesn't pay attention to the EVSE and pulls more current than what the EVSE signals is available, it could pop a circuit breaker or potentially cause a fire. Conversely, if the car plays it safe and always uses less current to charge, then time would be wasted because the car could have been charged faster. That's why the signaling is there.
 
Understood and thanks. Yes, I was referring to the internal car charger and not the external Supercharger type system.

Just how is the signal encoded into the charge current?

How does the car "read" it?

I read a Wikipedia article but I don't understand yet where the signal is hidden.

Thanks.
 
Just how is the signal encoded into the charge current?

How does the car "read" it?

From that Wikipedia article:

The {EVSE} sends a 1000 Hz square wave on the contact pilot...

The {EVSE} station can use the wave signal to describe the maximum current that is available from the charging station with the help of pulse width modulation: a 16% PWM is a 10 A maximum, a 25% PWM is a 16 A maximum, a 50% PWM is a 32 A maximum...

Here's another link and image that covers J1772 Basics from the Open-EVSE site (which provides open-source designs for, and parts to build your own EVSE station)
Duty Cycle Max Current
< 3% Error
3% - 7% Digitial Com Required
10% - 96% 6A - 80A
>96 Error
9018558031_dc6f8d9c8c_b.jpg
 
And the "bug" that was preventing charging above 70a was that the "duty cycle" of the pilot signal got so extreme (96% on, 4% off) that the pilot signal reader had trouble figuring out just how much was on and how much was off. Duty cycles any higher trigger an error state, so basically it would false trigger the error state and shut off. Tesla revised how the signal was read and the duty cycle % calculation was done to make it more accurate even if the pilot signal was a little noisy or degraded.

Originally J1772 didn't have charge rates so high, and Tesla was involved in revising / extending the spec to support up to 80A charging (the 51A-80A "mode"). The original engineers who conceived of the initial J1772 scheme probably never expected it to be used this way. So basically there is a "hack" involved that mostly worked but had a "rough edge" at the extreme.
 
Thanks for the link. I will research what is over my head, but this begs the question... Can't a plug be connected to a computer to falsify the amperage signal?

It could be. The signal is generated in the first place by some kind of microcontroller device.

Specifically, say a Nissan Leaf connects to a 72 amp system, can't an interposed computer falsify the signal?

Yes. But this is hardly going to happen by accident.

Now, would that leave the Nissan Leaf open to getting fried by higher amperage OR would the Leaf be able to self regulate an only use what it needs without overloading?

The pilot signal tells the car how much current is available from the supply.

If the car isn't able to take that much current (or doesn't want to right now due to its state of charge etc.) it will draw less.

So in your particular example, a Leaf can't draw more than 30A (depending on model), so if the true signal was saying 30A and your hypothetical box in the middle was falsifying this to read 72, then nothing would happen. If the true signal was saying something less (maybe 10A), then the Leaf would draw 30A when it should be drawing only 10A. Similarly, a Roadster connected to a 30A point that had been falsified to 72A would now draw 72A.

In a properly installed system, the pilot signal indicates the current that can be drawn safely without overloading the wiring; there should then be a circuit breaker to protect the wiring from short circuits etc., so the effect of inserting your box would normally be that the breaker then trips.

Understood and thanks. Yes, I was referring to the internal car charger and not the external Supercharger type system.

Just how is the signal encoded into the charge current?

How does the car "read" it?

I read a Wikipedia article but I don't understand yet where the signal is hidden.

Maybe you are not understanding that there is an extra wire in the cable and an extra pin in the connector to carry the pilot signal.

Note also that the pilot signal has two functions - the pulses communicate the available AC current, but the car is also required to present a particular resistance between the pilot pin and earth - the EVSE passes a current through the pilot pin and back via the earth pin/wire, and uses this to verify that the earth wire has not become disconnected.
 
Also, many 70A Tesla / Clipper Creek units were changed to J1772 plug, and many LEAF have connected to these just fine and charged at the full rate the car's onboard charger supports (3.3kW or 6.0 kW depending on which one that LEAF has). So, having a pilot signal well above the max the car supports is no problem. The only real problem was the touchy detection of the 80amp level, so many of those EVSEs were set for 70A pilot signal to make sure that the car can read it reliably.
 
If I'm not mistaken, some EVSE also have current transformers so they can measure the current actually drawn by the car. If it exceeds the allowed level, it drops the pilot signal causing the car to stop charging. If that fails, it can just cut the power to the car. These measures allow a degree of safety greater than relying on the circuit breaker.