diy solar

diy solar

Solar to electric vehicle-direct

Keeping 25-50 gallons of gasoline around is problematic. Even if you choose non-ethanol (and can get it) it's not ideal. 100LL (airplane fuel) is probably better for longer storage, but cannot be used in anything EFI that isn't a simple engine.

I solved for this by buying 100lb propane tanks. That fuel keeps forever but has it's own "storage risks". Best bet is a large underground propane tank and a dual-fuel or propane generator.

Yes you would definitely need to keep cycling it through the vehicle. Maybe even keep the extra 25 gallon tank in the bed and use it as an intermediary tank.

It is a bigger problem in CA / NY and similar states that require more aggressive fuel processing vs TX. The processing that is required here in CA makes the fuel much less stable than 20 years ago and no amount of stabilizer seems to solve this problem.

Perhaps a diesel hybrid truck would be better as those might be more available in TX vs over regulated CA.
 
The newer Teslas have done away with the 12 volt batteries. Just had the high voltage batteries. Must use the converters for the 12 volt and 24 volt side.
I don't believe that is correct. My Model Y purchased in November 2022 still uses a 12 volt battery. Tesla has announced they are moving to a 48 volt system which will save a lot on copper wiring.
EDIT: A subsequent post revealed that since late 2021 Tesla is using Lithium batteries which may be 16 volts.
 
Last edited:
It is a bigger problem in CA / NY and similar states that require more aggressive fuel processing vs TX. The processing that is required here in CA makes the fuel much less stable than 20 years ago and no amount of stabilizer seems to solve this problem.
Even in CA, you still have general aviation airports. Many of them have self-serve fuel, so you can buy 100LL, which your generator will run on well and is designed for long term storage. Just don't put it in your car.
 
If you did this, the level 3 would need to communicate with the car. These batteries are in excess of 300 volts, and Tesla has at least two standards at around 310 and 340, depending on battery type.

It’s not that these Level 3 chargers aren’t available, it’s more that they are $13k + when I’ve seen them.

Hugh voltage DC with arcing or voltages in excess of 48 volts is no joke.
Educate me, I know Tesla owners on 100A (240V) circuits. I assume these are "level 3"? Or am I using the wrong language here?
 
Educate me, I know Tesla owners on 100A (240V) circuits. I assume these are "level 3"? Or am I using the wrong language here?
Level 3 is DC charging.

Also not discussed here but worth mentioning. DC charger requires the shore side equipment to handle all the responsibility of charger, including communications with EV, DC DC converter to provide CC and CV behavior at appropriate points in the cycle. DC EVSE also needs to adapt the output voltage to the car’s HVDC voltage. Except in edge cases like where the car has a higher voltage than the station, in that case the car has a unidirectional DC DC boost converter (may also need to be a charge controller)

Compare that to L1/L2 AC charging. The Car is in the driver seat there, uses the onboard charger to draw current up to what the EVSE offers. And then the onboard charger handles communications and DC DC / charging algorithm.

The technical complexity for an L2 EVSE vs a L3 DC charger could well be 100X lower. You can probably make an L2 EVSE with steampunk era technology.
 
The newer Teslas have done away with the 12 volt batteries. Just had the high voltage batteries. Must use the converters for the 12 volt and 24 volt side.
Teslas still have low voltage batteries, but they've switched from lead-acid to lithium in some models. I think the voltage is higher, over 16V when fully charged, but don't quote me on that. This DIY page has a few bits of info.
 
Again, I don't see how this is possible without a "battery buffer". PV is simply too unstable due to cloud coverage.
Capacitor can be used as a "buffer". Why is "stability" a problem? Seems like the car would charge when the voltage is high enough, and not charge when the cloud comes over. Say the panels put out 250v. The electrons hit a wall of 450v at the car and don't go anywhere.

Note: I'm not saying a battery would not be helpful to smooth out charging. I'm just saying it may not be "required".
 
Even in CA, you still have general aviation airports. Many of them have self-serve fuel, so you can buy 100LL, which your generator will run on well and is designed for long term storage. Just don't put it in your car.

What happens if you put it in your car ?
 
Capacitor can be used as a "buffer". Why is "stability" a problem? Seems like the car would charge when the voltage is high enough, and not charge when the cloud comes over. Say the panels put out 250v. The electrons hit a wall of 450v at the car and don't go anywhere.

Note: I'm not saying a battery would not be helpful to smooth out charging. I'm just saying it may not be "required".

Well there’s also the matter of the off the shelf charge controllers being designed to work with a stiff power supply rather than something as flaky and variable as solar panels. The DC DC converter might not even be able to start up if you connect solar panels to the input.

So then potentially you would need to design much of the DCFC from scratch. Maybe you can get the communications off the shelf, I think there is a $1500 comms dev kit for CCS. And then use a blocking diode and fuses on a 400V solar string straight into the traction battery after it is convinced to open its contactors; if that works then you don’t need to be a power electronics engineer to have a working charger. It probably won’t work robustly.

($1500 can be spent in other ways to counteract the DC to AC to DC inefficiency)
 
($1500 can be spent in other ways to counteract the DC to AC to DC inefficiency)
I don't even know what that inefficiency is. Even if it is 20% it would take a lot of charging to pay for the cost of the equipment mentioned to allow DC charging at home. Then you have to have a really good reason to want to charge with DC when AC EVSEs are readily available and can charge most EVs from empty to full, overnight.
 
I don't even know what that inefficiency is. Even if it is 20% it would take a lot of charging to pay for the cost of the equipment mentioned to allow DC charging at home. Then you have to have a really good reason to want to charge with DC when AC EVSEs are readily available and can charge most EVs from empty to full, overnight.
Someone out there almost certainly has posted the public DC fast charger efficiency curves for various cars, measured in the wild.

I think DC charging will become fairly common with V2X, there's like 4 DC chargers waiting to be released. The value is in the inverter / DC-DC converter to allow the house to draw power from the EV, not in the DC charging which comes for free once with that. Since the traction battery of different EVs vary in voltage, it follows that any DC coupled V2X will need DC-DC since the shore side HVDC system will also vary in voltage.

But I will bet that few of these will be optimized for "efficient" DC to DC conversion from solar panels.

I don't know what the share is between DC and AC coupled with the rest of the house systems. DCbel is DC coupled but in their walled garden. Emporia V2X (targeting 2024 but I bet they'll target whenever CCS2 shows up) is an AC coupled grid forming 1741SA inverter.

I doubt anyone here can say what a open standard HVDC coupled system will look like today, as compared to 48VDC DC coupled which has decades of pedigree.
 
Getting back to maybe the original question. Using the OBC + EVSE has similar issues with AC coupling of GTI with grid forming inverters. It's a dual version of the problem.

Namely your solar is enthralled to a high power component (a sink in this case) that operates somewhat independently of your inverters/solar output.

Potentially J1772 have a more standardized/sane contract compared to a 1741SA GTI in how it responds to a power change request, I think the EV is supposed to react within a fixed period of time if the charge current is cut by the EVSE. And the EVSE (if appropriately designed for this use case) can open the circuit in an emergency to prevent the house-side system from browning out.

With EV charging opening the relay probably has fewer consequences than opening a safety relay for AC coupled GTI since the EV OBC doesn't give a crap if power goes away.

Potentially a communicating (in the sense that it talks to the HOUSE and the EV) DC charger aware of house battery SoC and house PV output can know how much power is safe to send to the car, and control this with extremely low latency since it has direct control over the CC/CV parameters vs dealing with the blackbox behavior of the car OBC... but there's a lot of Ifs and Theory here, and as far as anyone knows here this kind of high complexity device is vaporware.
 
Any reason why you can't put 48V batteries in series, assuming a charger would do much higher voltages... It'd be easiest to DC to DC charge if you can get the voltages closer.

This is covered in a couple other threads. 48V BMS are generally FET based, which are limited to blocking maybe ~60V. If you series together a couple 48V packs, take out a notepad, draw out the circuit, and look at what the voltage is with 96V and some failure modes. One of the FETs is going to need to block more than its rating.
 
I printed out an email I just got from Enteligent to a PDF. It is attached. I believe it might be of interest here...
 

Attachments

  • Big Things Are Happening At Enteligent.pdf
    1.5 MB · Views: 13
I printed out an email I just got from Enteligent to a PDF. It is attached. I believe it might be of interest here...
Thanks.. $7M is going to to fast. I try not to beta test, but glad someone is working on building products like this.
 
I currently charge my Leaf with solar about 95% of the time with a Growatt SPF3000TL LVM-ES and a single EG4 LifePower4 battery with 6 URE 365 watt panels wired in series. I typically charge at 12 or 16 amps at 120v with a Zencar adjustable EVSE. I spent about 4 grand with cables, racking, disconnect, array ground, etc.

I plan to upgrade to a Growatt 5000ES in the future for level 2 charging.

Both of the "ES" version Growatts will work batteryless but having a battery or the grid to smooth out the power is much easier.
 
Back
Top