Knightshade
Well-Known Member
Not really... and they still have the headline at the top of their page, with the "correction" being just a few sentences halfway through the original article.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
MASTER THREAD: Powering house or other things with Model 3
- Thread starter Vines
- Start date
-
- Tags
- powerwall solar home
Not really... and they still have the headline at the top of their page, with the "correction" being just a few sentences halfway through the original article.
israndy
Supercharger Hunter
I know a lot of people are upset with Fred and his ilk for being click baiters. I am friends with Marco, the tear down engineer who did the presentation with the results of the Ford tear down at our local EV meeting 9 months or so ago. He probably showed the Electrek guys the same info. He had pictures of the full disassembly of the charger and the schematic he had drawn, and typical charger and typical inverter schematics. He really seems to know what he is talking about when it comes to electronics. And as Ingineer mentioned the package for the diodes is basically identical to the one for the transistor. Basically the only difference is the label on the circuit board.
As a reporter you have to rely on experts to form the basis of your stories. They went to journalism schools (hopefully) so they don't have to know everything about tanks and politics and in this case schematics. A good reporter will second source (or more) all info, but in this case who do you go to for that second source? Tesla will neither confirm nor deny. Do you NOT report it? It certainly was interesting and opened a lot of conversations about the electronics of the cars. I don't think Ingineer could have been talked into that deep dive if the article hadn't been published.
I call it a win, and now I know two famous people that are in my EV group.
As a reporter you have to rely on experts to form the basis of your stories. They went to journalism schools (hopefully) so they don't have to know everything about tanks and politics and in this case schematics. A good reporter will second source (or more) all info, but in this case who do you go to for that second source? Tesla will neither confirm nor deny. Do you NOT report it? It certainly was interesting and opened a lot of conversations about the electronics of the cars. I don't think Ingineer could have been talked into that deep dive if the article hadn't been published.
I call it a win, and now I know two famous people that are in my EV group.
The more I think about this issue, I think it is best solved with a dual battery system. This is common in the overlanding community (Jeepers). The basic idea is that your loads are connected to a dedicated battery. That "load" battery is charged by the vehicle's systems when the "engine" is running. There are devices available that will only connect the two when the voltage is high enough to indicate that the alternator is charging the vehicle battery.
For the Model 3, I would instead use a PWM charge controller to charge the load battery from the vehicle. The best and most convenient place to safely tap the Model 3 electrical system is the supply for the Subwoofer. It's a 30A 12V circuit. The charge controller will limit the draw to the rated amperage, even if the voltage difference would normally draw significantly more amperage. I have done this with a small 12V AGM battery for a my electric cooler. In that case, I used a 10A charge controller so that it would not overload a "cigarette lighter" 12V power point in my other car. If you're going to use a 120VAC inverter to power a refrigerator and other things in an emergency, 120W (10A 12V) is not going to be enough to keep the load battery charged, hence the 30A subwoofer tap.
For the Model 3, I would instead use a PWM charge controller to charge the load battery from the vehicle. The best and most convenient place to safely tap the Model 3 electrical system is the supply for the Subwoofer. It's a 30A 12V circuit. The charge controller will limit the draw to the rated amperage, even if the voltage difference would normally draw significantly more amperage. I have done this with a small 12V AGM battery for a my electric cooler. In that case, I used a 10A charge controller so that it would not overload a "cigarette lighter" 12V power point in my other car. If you're going to use a 120VAC inverter to power a refrigerator and other things in an emergency, 120W (10A 12V) is not going to be enough to keep the load battery charged, hence the 30A subwoofer tap.
n2mb_racing
Active Member
Sounds good, although inefficient. I was thinking approximately the same thing. You'll have conversion loss going from HVDC --> 12V, then 12V --> 12V, then 12V to 120V AC. But, you have 75 kWhr, so it doesn't really matter much.
What kind of 10A charge controller are you using?
I have one just like this: Amazon - Charge Controller
They are widely available in 10 / 20 / 30 / 50 / 60A versions from a variety of manufacturers.
However, this is for small loads on a 35Ah AGM battery like this: Amazon - 35Ah AGM Battery
A battery this size is easy to carry. If you're going to use an inverter over 500W, I would recommend a larger battery, like 60Ah or more.
n2mb_racing
Active Member
Neat. I was thinking of using the iSDT Q8 charger that I already have, but each time the input power cycles, I would have to manually restart the 12V battery charging.
Q Series - ISDT
A balance charger like that is clearly overkill for charging a 12V AGM battery. Now, if you wanted to build a 4S LiFePO4 battery for this, it would be much lighter for the same Ah capacity and the Q8 would be an excellent solution.
n2mb_racing
Active Member
Yeah, I have it to charge other things. It is nice that it can charge anything. But, I would have to manually restart the charging when the power goes off in the car.
I do have a tiny and cheap 4S lifepo which is great, but too small for this application, it can only deliver 20A peak:
https://www.amazon.com/gp/product/B07X5G2FFW/
Just saw that they added a 16ah version for only $46. That's really cheap for 206 whr Lifopo.
bwilson4web
hit the spot
My first modification to our 2003 Prius was to add a 1.1 kW, modified sine wave inverter:
In 2016 we upgraded from 100A to 200A service and added a 16 kW, automated, natural gas powered, emergency generator. This was sized to handle the house and EV load. But the fuel cost, $0.48/kWh, makes it impractical for anything but emergencies.
Today, it would be more affordable to go with a Tesla roof and power wall.
Bob Wilson
- 4d 6h operation at 2 gal/day during the 2011 tornado induced power outage
- typically used 1-3 times per year for 2-18 hr due to poor neighborhood power
- Prius quiet with engine cycling and very low, 1/100th, carbon monoxide risk
- operated power tools around property
In 2016 we upgraded from 100A to 200A service and added a 16 kW, automated, natural gas powered, emergency generator. This was sized to handle the house and EV load. But the fuel cost, $0.48/kWh, makes it impractical for anything but emergencies.
Today, it would be more affordable to go with a Tesla roof and power wall.
Bob Wilson
n2mb_racing
Active Member
See elsewhere in this thread. You can't connect an inverter to the Model 3 battery, since it is not connected directly to the DC/DC converter. The 12V battery has limited charging current, so you will quickly drain it.
On the Model 3, you need to connect the inverter directly to the DC/DC converter, under the rear seat.
On the Model 3, you need to connect the inverter directly to the DC/DC converter, under the rear seat.
He is comparing the cost of upgrading his service and installing an automatic nat-gas generator as he did in 2016 against solar + Powerwall.
I concur that installing solar + Powerwall is a better solution than a large whole-house generator with automatic transfer. However, that's really not the subject of this thread, which is powering some limited things from your car.
Tesla roof + PW maybe better in some ways but that does not mean it's more affordable. I was wondering about the specifics of how Tesla roof + PW that can provide 16KW power is more affordable.
n2mb_racing
Active Member
Maybe because it pays for itself eventually? I was thinking of getting solar after we replace our roof. With the recent price drop to $1.49 watt, that seems like it would pay for itself quickly.
However, I don't think the powerwall would pay for itself, at least where I am located in NC, since the difference between peak and off peak rates is not that high. So, I probably wouldn't get a powerwall, unless I was really concerned about having backup power.
israndy
Supercharger Hunter
Where I am there are no EV rates for electricity. So the Powerwall is silly. If I have extra solar power I'll put it in the car. I am working on building a big UPS to drive the TV as it's the thing using the most power after the sun goes down. Most everything else is run during the day, even the hot tub is set to not kick on until the sun is generating a lot of power.
Backpacker42
Member
I asked an engineer who worked at Tesla about that. He said they actually considered it, but decided at the time it was considered an edge case use. I wonder whether the thinking might change as a result of the rolling blackouts we had in California due to fire hazard mitigation. I imagine that there are also substantial safety issues here - you wouldn't want to energize your house if for instance, power had been cut so PG&E personnel could do repair work down the street (or if you ended up powering your entire neighborhood) if you were still connected to the grid.
Tesla would likely put in circuitry that will sense outside power and switch off power from the car to the house. Pretty much how a automatic transfer switch is used in a whole house generator setup, the home owner wouldn't even need to do anything. A notification that you got outside power back would be your indication that power is now normal.
FalconFour
Member
Wow, this thread has some... interesting tangents to it.
My journey to "powering the house with my Tesla" was mostly forged independently, as I couldn't really find the info I was looking for with the angle I took it with. I had access to a gratuitously oversized 4kW / 12v pure sine inverter from work that wasn't being used, and I hunted around for a way to make it work. I discovered the PCS (DC-DC converter) lugs under the rear seat, and proceeded to hack the hell out of it, crashing (as software, not cars) the car several times in my tinkering. Biggest takeaway so far is that you really can't break the car -- it's extremely sensitive, and if it doesn't like anything that's happening, it plorps back into its shell of safety and turns off the affected system.
That is to say, the terminals under the seat are harmless - you can lay a wrench across them with the car running, and you'll barely get a tiny spark before it shuts off. Instant reaction, thanks to MOSFET soft-fuses/eFuses on both sides of the converter that immediately disconnect both ends (the main bus & the DC-DC itself) of the PCS and drop the voltage to zero - while the rest of the car remains running (but not driveable), powered by the 12v battery. Disconnect the negative terminal of the 12v battery, clonk as the car powers off, wait 30 seconds, reconnect, good as new again, try a different angle this time. (unlike most devices, waiting 30 seconds seems essential here - if you don't wait long enough, some systems may not fully reset and you'll have a minor heart attack thinking you broke something!)
It can supply slightly north of 2.5kW - I've peaked it at 2.9kW I think, but 2.5kW average. Monitored with Scan My Tesla.
Here's a more-or-less complete brain-dump of everything I've learned about sucking power out of the Model 3: Controlling Tesla's Sleep Mode ~-or-~ Lucid Dreaming for Robotaxis : teslamotors (link to a specific comment reply there - not the thread itself)
If I can find a good enough chip to serve the purpose, I might just build a diode/precharge circuit that makes this more effortless... because right now, whenever an export system is installed and running in the car, stuff goes sideways if the car ever decides to go to sleep - and a 12v reset is needed to restore the car to normal. A diode and precharge circuit would be a simple board, but needs a thicccc MOSFET to allow around 200A to flow through without excess heat. Sad to say that my EE skills taper off around the ballpark of transistors... I'm more a digital guy.
But yeah, hopefully that helps add a new angle to these thoughts. No charge controller needs to be involved, but a buffer battery is almost necessary for some loads!
Now back to mining Zcash in my back seat...
My journey to "powering the house with my Tesla" was mostly forged independently, as I couldn't really find the info I was looking for with the angle I took it with. I had access to a gratuitously oversized 4kW / 12v pure sine inverter from work that wasn't being used, and I hunted around for a way to make it work. I discovered the PCS (DC-DC converter) lugs under the rear seat, and proceeded to hack the hell out of it, crashing (as software, not cars) the car several times in my tinkering. Biggest takeaway so far is that you really can't break the car -- it's extremely sensitive, and if it doesn't like anything that's happening, it plorps back into its shell of safety and turns off the affected system.
That is to say, the terminals under the seat are harmless - you can lay a wrench across them with the car running, and you'll barely get a tiny spark before it shuts off. Instant reaction, thanks to MOSFET soft-fuses/eFuses on both sides of the converter that immediately disconnect both ends (the main bus & the DC-DC itself) of the PCS and drop the voltage to zero - while the rest of the car remains running (but not driveable), powered by the 12v battery. Disconnect the negative terminal of the 12v battery, clonk as the car powers off, wait 30 seconds, reconnect, good as new again, try a different angle this time. (unlike most devices, waiting 30 seconds seems essential here - if you don't wait long enough, some systems may not fully reset and you'll have a minor heart attack thinking you broke something!)
It can supply slightly north of 2.5kW - I've peaked it at 2.9kW I think, but 2.5kW average. Monitored with Scan My Tesla.
Here's a more-or-less complete brain-dump of everything I've learned about sucking power out of the Model 3: Controlling Tesla's Sleep Mode ~-or-~ Lucid Dreaming for Robotaxis : teslamotors (link to a specific comment reply there - not the thread itself)
If I can find a good enough chip to serve the purpose, I might just build a diode/precharge circuit that makes this more effortless... because right now, whenever an export system is installed and running in the car, stuff goes sideways if the car ever decides to go to sleep - and a 12v reset is needed to restore the car to normal. A diode and precharge circuit would be a simple board, but needs a thicccc MOSFET to allow around 200A to flow through without excess heat. Sad to say that my EE skills taper off around the ballpark of transistors... I'm more a digital guy.
But yeah, hopefully that helps add a new angle to these thoughts. No charge controller needs to be involved, but a buffer battery is almost necessary for some loads!
Now back to mining Zcash in my back seat...
n2mb_racing
Active Member
Wow, this thread has some... interesting tangents to it.
Now back to mining Zcash in my back seat...
Nice!
I think the relay precharge circuit mentioned above is pretty good and simple. Just a big 12V relay from Amazon and a single resistor, and you're set. The resistor can be mounted on the relay, so the wiring is very neat. No need for external connections, just heavy +12V power in and +12V power out to the inverter. And a connection to ground.
FalconFour
Member
Unfortunately it only solves one of the two problems: precharging (which is honestly a minor issue; a non-issue with a buffer battery). The bigger issue I see is preventing back-feeding the PCS from the buffer battery when the car wants to sleep. Though I guess that's an architecture question - to use, or not to use, a buffer battery. In my case, the buffer battery is just an extra-large capacitor, not at all related to the idea of "constantly charge a battery and draw from it", but just as a capacitor. Maybe a huge capacitor could take its place.
Precharging gets the inverter to start just fine, but when you have a compressor- or AC motor-based load, those motors tend to draw a near-short-circuit current when they start, causing an inrush surge (long after the inverter was started) that also trips the PCS in the car. Those surges are absorbed just fine by the buffer battery, but I don't know if a capacitor would have similar characteristics to buffer such a large jolt... thus the battery.
Optimally I'd just have a board that acts as a combination diode and relay - diode to keep it only in one direction; relay to close after precharge. A MOSFET could do that, but I can't quite wrap my head around how it can handle both tasks. A relay (precharge) and diode (back-feeding protection) might just be the only complete solution...
Similar threads
