Bi-Directional Charging

[VolklKatana]

I looked for this article on the forum here, but i didn't see it mentioned in any threads I've been watching. I think this is going to be one aspect of what is going to make battery day a historic one for Tesla. This should make headlines in both the automotive industry as well as the energy and utility sectors. I am glad I reserved the Tri-motor with the larger battery now!

https://electrek.co/2020/05/19/tesla-bidirectional-charging-ready-game-changing-features/

Advertisement

 

[ajdelange]

I used to participate heavily in a brewing forum. For whatever reason things like this were referred to as "momilies' which, in that forum at least, referred to a rumor that despite being baseless came to be accepted as true because of its frequent repetition. Another example on this forum is that the CT will be equipped with a distillation apparatus derived from a slip of the tongue on Elon's part in an interview. In this case someone has found the autopsy report of a old model S charger which includes a schematic. Here's what has resulted;

"Electrek has learned that Tesla has already prepared its onboard vehicle charger for bidirectional charging.

Marco Gaxiola, an electrical engineer who participated in a Model 3 teardown for a Tesla competitor, reverse engineered the electric car’s charger and found it to be ready for bidirectional charging."

What Elextrek should have learned is that the basic topology of an electronic device capable of doing what the onboard charger is capable of doing is inherently bi directional. It would, IMO, be difficult to design it in a way that isn't inherently bi directional. I haven't read Gaxiola's report but if he actually said is that the charger is "ready for bidirectional charging" I would be very surprised because at the time that charger was designed the NEC had a requirement that vehicle charging equipment contain provisions to prevent backflow of energy to the grid.

The other side of the coin here is that given that the architecture of the Tesla charger is intrinsically bi direction Elon or Straubel or anyone else who works for Tesla is definitely entitled to say that while perhaps the company is not interested in supporting it at the present time their product certainly could support it at some time in the future if they wanted to. Perfectly true. And if you read the article you will see that this is more or less what they said.

Excellent example of the quality of modern journalism!

I am glad I reserved the Tri-motor with the larger battery now!

The charger shown in the schematic is not the charger in the recent S or X production but whatever is more or less has to be bidirectional as stated above and there is potential big benefit to CT owners with the big battery and that is that the same hardware used to charge the battery can serve as the inverter which supplies the bed AC outlets and can handle, while functioning as an inverter, about the same power levels as it does when serving as a charger. This means that we could have as much as 48 A of 240V available in the bed and that if it takes an hour to add 25 miles range while charging that we can draw 48 A for an hour at a cost of only 25 mi range. Clearly this would be a fantastic feature to have as nobody is going to run an 11 kW load continuously in the field. A welder could work all day in the field and have plenty of juice left over to travel a good distance home.

Yes, very cool but that does not mean that Tesla will do it this way.

 

[VolklKatana]

I think it is going to be very interesting to see in the end. If battery costs are low enough, would be incredibly cool to see them work other options into the vehicle, like the possibility of solar being standard. Put those two together, bi directional and solar, now youre powering peoples vehicles, and homes for free. That's a game changer, period

 

[ajdelange]

I'm afraid solar on the vehicle is another momily. I can't cover my house's use with 58 panels on my roof. Do we really think one or 2 of those on my truck would?

Now if I could power my house with 11 kW (and according to my history I can 97% of the time) and the charger on my X was configured to do inverter duty as well as charger duty I could drive over to the SC and load up then drive home and the charge would run the house for 2 days. That would be powering my house for free (X buyers get free super charging).

 

[VolklKatana]

While i understand your point, for someone like myself who has no solar right now, its at least a step in the right direction. some solar is better than none, right?

 

[ajdelange]

Absolutely! I charge my cellphone and noise cancelling earphones exclusively from solar. I just don't fancy that I'm saving much on my electric bill by so doing.

I guess what I really want to get across is that one sees all sorts of momilies on these web sites and that one ought, before accepting them, to do a couple of calculations. In case it isn't obvious I find doing those calculations quite entertaining in this plague year.

 

[TyPope]

I'm afraid solar on the vehicle is another momily. I can't cover my house's use with 58 panels on my roof. Do we really think one or 2 of those on my truck would?

Now if I could power my house with 11 kW (and according to my history I can 97% of the time) and the charger on my X was configured to do inverter duty as well as charger duty I could drive over to the SC and load up then drive home and the charge would run the house for 2 days. That would be powering my house for free (X buyers get free super charging).

But what if each cell was twice as efficient as new cells are now? (and supposing you lived where the sun actually shines. Holy crap, 58 panels is a lot!

 

[cybrtrk_maybe]

But what if each cell was twice as efficient as new cells are now? (and supposing you lived where the sun actually shines. Holy crap, 58 panels is a lot!

WOW, I agree; I thought I had a lot of panels with 38.

 

[Ehninger1212]

But what if each cell was twice as efficient as new cells are now? (and supposing you lived where the sun actually shines. Holy crap, 58 panels is a lot!

Yeah, SunPower's panels are at 410 watts per panel. The Calculation i made in another post for how many kWh you could gain if the whole CT roof was a panel was assuming a 265 watt panel, which is the industry average.

 

[ajdelange]

But what if each cell was twice as efficient as new cells are now? (and supposing you lived where the sun actually shines. Holy crap, 58 panels is a lot!

Well if Bill Shockley were wrong (panels twice as efficient would be better than the Shockley limit) and the inverters were more efficient than they are and if my panels could track the sun and if the sun didn't go down every night and head south for the winter and if was never cloudy or hazy here then the 58 I have installed would have covered more of my need than they did. As it is they've handled 47% of my consumption since I installed them last June.

In the earlier post I tried to convince people to work some numbers. Look up the insolation for your area. Don't forget to include a factor for the difference between the optimum elevation for your area and vertical (or at whatever elevation you think the panels will be oriented), Figure out what the probable area of panels you could get onto the roof of the CT and use whatever you think panel and converter efficiencies might be (but don't violate the Shockley limit).

If it helps, I live in a mesic climate zone at moderate latitude and each of my panels (representing last year's technology) produced power at an overage rate of 32.4 W. That's 260.3 kWh per panel per year.

 

[Ehninger1212]

Well if Bill Shockley was wrong (panels twice as efficient would be better than the Shockley limit) and the inverters were more efficient than they are and if my panels could track the sun and if the sun didn't go doen every night and head south for the winter and if was never cloudy or hazy here then the 58 I have installed would have covered more of my need than they did. As it is they've handled 47% of my consumption since I installed them last June.

In the earlier post I tried to convince people to work some numbers. Look up the insolation for your area. Don't forget to include a factor for the difference between the optimum elevation for your area and vertical (or at whatever elevation you think the panels will be oriented), Figure out what the probable area of panels you could get onto the roof of the CT and use whatever you think panel and converter efficiencies might be (but don't violate the Shockley limit).

If it helps, I live in a mesic climate zone at moderate latitude and each of my panels (representing last year's technology) produced power at an overage rate of 32.4 W. That's 260.3 kWh per panel per year.

Excellent description of how location will have an effect on solar panel production.

Sunpowers panels, which produce 410 watts, are currently to most efficient commercially available panel. They Sit at 22.8%. If Tesla was able to get there CT solar roof to reach that level of efficiency you would be looking at almost 1.5kWh of energy production. Given a rough estimate total roof size of 63 SqFt. Of course, as you have pointed out, that would be at optimum sun angle and weather conditions and not conducive of the average gain most people will see. This might be something someone living in Arizona will be able to achieve. This also does not take into account the fact there there is going to be a lot of efficiency loss between the panel and the battery pack.

This is why i think Tesla's claim of 15 miles per day is not just a shot in the dark. The real factor will be how much this option cost. Also if Elon views this as to complex he will probably axe it right away.

 

[TyPope]

Well if Bill Shockley was wrong (panels twice as efficient would be better than the Shockley limit) and the inverters were more efficient than they are and if my panels could track the sun and if the sun didn't go doen every night and head south for the winter and if was never cloudy or hazy here then the 58 I have installed would have covered more of my need than they did. As it is they've handled 47% of my consumption since I installed them last June.

In the earlier post I tried to convince people to work some numbers. Look up the insolation for your area. Don't forget to include a factor for the difference between the optimum elevation for your area and vertical (or at whatever elevation you think the panels will be oriented), Figure out what the probable area of panels you could get onto the roof of the CT and use whatever you think panel and converter efficiencies might be (but don't violate the Shockley limit).

If it helps, I live in a mesic climate zone at moderate latitude and each of my panels (representing last year's technology) produced power at an overage rate of 32.4 W. That's 260.3 kWh per panel per year.

I've done all that. 24 panels for 75% of my daily electrical use. But, that's down here and with currently available panels. With all your "IF"s there, maybe you'd do better if you moved closer to the Sun. LOL

 

[TyPope]

Well if Bill Shockley was wrong (panels twice as efficient would be better than the Shockley limit) and the inverters were more efficient than they are and if my panels could track the sun and if the sun didn't go doen every night and head south for the winter and if was never cloudy or hazy here then the 58 I have installed would have covered more of my need than they did. As it is they've handled 47% of my consumption since I installed them last June.

In the earlier post I tried to convince people to work some numbers. Look up the insolation for your area. Don't forget to include a factor for the difference between the optimum elevation for your area and vertical (or at whatever elevation you think the panels will be oriented), Figure out what the probable area of panels you could get onto the roof of the CT and use whatever you think panel and converter efficiencies might be (but don't violate the Shockley limit).

If it helps, I live in a mesic climate zone at moderate latitude and each of my panels (representing last year's technology) produced power at an overage rate of 32.4 W. That's 260.3 kWh per panel per year.

Check this out:
Cool article Ty found

 

[Aces-Truck]

I looked for this article on the forum here, but i didn't see it mentioned in any threads I've been watching. I think this is going to be one aspect of what is going to make battery day a historic one for Tesla. This should make headlines in both the automotive industry as well as the energy and utility sectors. I am glad I reserved the Tri-motor with the larger battery now!

https://electrek.co/2020/05/19/tesla-bidirectional-charging-ready-game-changing-features/

If you re-check out the link, they've updated it with a video that shows the charging hardware has Diodes in a key place, that would need to be Transistors, in order to do Bidirectional charging. So not something that could be turned on, in a software update in the model 3.

 

[ajdelange]

Well that was certainly interesting! I would like to sit down with that guy though as he said a couple of things that don't seem to make sense. '

He pointed out that there are three "modules" in this unit and that in the standard 3 only 2 are populated, Looking at the input end I see three single phase bridges. In Europe the mains power is 3 phase. In the US, though it is not perhaps widely understood, we have one or 2 depending on how we connect. In the cars the 3rd wire isn't used so it is single phase. Thus it appears the architecture for US production is two modules connected in parallel and in Europe it is one for each of the three phases. This is not to be confused with the schematic attached to the Electrek article which shows three modules each of which can connect to a biphase or 3 phase supply, You could buy the cars then (actually up to just a little over a year ago) with 2 modules for 48 A charging or 3 for 72A,

Next he refers to "power factor correction" and goes on to describe components which would be used to boost DC voltage and in that description says that this is what they are for. Power factor has nothing to do with this. The load presented by a battery charger is a battery and as such has unit power factor. May all be slips of the tongue but in ascribing the DC voltage boost to "power factor correction" he ignores the fact that voltage is trivially boosted by the transformers to which he assigns only the role of isolation (which, of course they also do). Then the frosting on the cake comes when he says the semiconductors, which he admits he cannot identify, are diodes because the PCB is silkscreened with a D rather than a Q. He obviously knows a hell of a lot more about this assembly than I do because he's got it lying on the bench in front of him but it seems those "diodes" have a lot of "input" pins for a diode. I can't say they aren't diodes because all those pins could be tied together but there are advantages to having switches in the final rectifier bridge such as the ability to regulate voltage and current flow to the battery.

[EDIT]I just remembered that sometimes a transistor is connected as a diode and, if I'm remembering things correctly, some times the PCB would be silkscreened with a D if that were done.

So while there is little doubt in my mind that this PCB is not bidirectional capable I'm not sure I buy this man's "proof". He could certainly remove one of those devices and determine what sort of a device it is, Had he done that and found those things to be diodes I'd say OK to his thesis.

Even though this design is substantially different from the one of the schematic in the original article the architecture is basically the same

Mains ==> Bridge -->Bridge==>Transformer==>Bridge-->Battery

==> means the power flow is AC, --> means DC. The power flow can clearly be reversed to

Mains <== Bridge <--Bridge<==Transformer<==Bridge<--Battery

if all the bridges use active devices. For unidirectional only the DC to AC bridges need them. Thus the Tesla architecture is intrinsically bidirectional though neither the Model S charger nor the Model 3 one seem to support it.