diy solar

diy solar

DC solar directly to EV (no AC!)

i want to build a 80-200W PV, 1000-2000 Watt hour storage Vehicle Integrated Photo Voltaic LiFePO4 charging system.

for now any system will require stationary condition to safely deliver charging.
 
This guy explains why it's not possible to connect solar directly to the HV system:
 
I seriously need to add solar panels to a van/truck sized electric vehicle.

Does anyone here have the expertise, or can recommend a professional or company that could actual pull this off?

I can handle the physical mounting and wiring of panels.
What I really need help with is engineering the electronics to integrate with the existing EV.

I would have ~ 6 kW of solar - possible configurations:
  • one string of 1016 V, 5.9 A
  • two strings of 508 V, 5.9 A

The electric vehicles I am looking at have 525V, 400V, or 320V batteries.
So a MPPT that can handle those high battery voltages would be required.
Or, a DC to AC to DC conversion :(.

And, of course, I want the solution to be as efficient and convenient as possible.
  • Direct DC charging would be more efficient that AC conversion.
  • Charging while driving is ideal. Other suboptimal possibilities include charging a buffer battery then charging the main battery while stopped.

How can this be done?
  • Direct to battery/BMS? A max of 6 kW should not be a big strain on the batteries compared to the normal draw of the motor, and influx from regenerative braking and high power charging.
  • Via the existing charger? Could the existing charger be made to accept the fluctuating solar input?

I really appreciate any tips to help me realize this!

Thanks!
 
I seriously need to add solar panels to a van/truck sized electric vehicle.

Does anyone here have the expertise, or can recommend a professional or company that could actual pull this off?

I can handle the physical mounting and wiring of panels.
What I really need help with is engineering the electronics to integrate with the existing EV.

I would have ~ 6 kW of solar - possible configurations:
  • one string of 1016 V, 5.9 A
  • two strings of 508 V, 5.9 A

The electric vehicles I am looking at have 525V, 400V, or 320V batteries.
So a MPPT that can handle those high battery voltages would be required.
Or, a DC to AC to DC conversion :(.

And, of course, I want the solution to be as efficient and convenient as possible.
  • Direct DC charging would be more efficient that AC conversion.
  • Charging while driving is ideal. Other suboptimal possibilities include charging a buffer battery then charging the main battery while stopped.

How can this be done?
  • Direct to battery/BMS? A max of 6 kW should not be a big strain on the batteries compared to the normal draw of the motor, and influx from regenerative braking and high power charging.
  • Via the existing charger? Could the existing charger be made to accept the fluctuating solar input?

I really appreciate any tips to help me realize this!

Thanks!
AC charging would not be possible while driving, since the EV's onboard systems don't allow that (try putting your EV into Drive while it's plugged in...).
DC charging might work, but you would have to bypass the charging port and go straight to the battery...

In any case, it's probably not feasible with an "off the shelf" EV. Maybe if you did your own EV build or conversion.

The HV wiring in an EV is well isolated. I don't see how you will connect hundreds of DC volts from outside the vehicle into the car, safely.

Sorry to burst your bubble...

The only way it could be done is by having a buffer battery and inverter onboard. That could charge up while driving or being parked and when stopped you could use that to do AC charging (with losses and inefficiencies included).

But take the time to do some math: How much added weight would you be hauling around all the time, just for a few added KWh a day? In the end, you might find out that your EV range will be decreased, almost as much as you can gain from solar.

A better option would be to install solar at home and use that to charge your EV while it's parked.
 
it's been done many times ;). It depends on your budget of course and design boundaries. How many panels are the 6kw ? a simplest and probably most dangerous is a simple block diode, no control, no smart, nada; just need the PV voltage to be higher than the EV pack. But don't do it

 
Thanks @meetyg!

For me, I am past the "if/why" to do this. I am doing this out of necessity, not to save money – this vehicle will be my off-grid home and I will not have access to grid power nor fossil fuels. So, I have to get this done one way or another.

In the worst case, I know I can achieve this with a buffer battery and charge the main battery via inverter while the vehicle is stopped. But this hacky solution makes me cringe.

One of the "off the shelf" trucks I am considering does have wide open access to battery packs, BMS and high voltage cables.
And if I do a Tesla swap or other conversion, I would have access to them too, although this likely will be a more costly solution that with an "off the shelf" EV.
I would get a professional to connect the proper gauge wires safely to the high voltage lines.
This part seems doable.

The part that I am struggling with is understanding how the existing EV electronics behave.
This doesn't seem to be common knowledge and this is why I am reaching out here for help.

Does anyone know exactly how the EV electronics behave?
Here is my speculation.

Consider the following state of the vehicle...

A) The EV is off and NOT connected to a charger.

There are no heavy currents (motor, regen, charger) in or out of the battery in this state.
Could an MPPT be connected directly to the battery or BMS in this state?

An MPPT normally handles the various charge cycles of the battery at the different levels of charge and will slow down when battery is near full and not overcharge.

Is the BMS still on and operating? If not, would it have to be switched on somehow (even though the rest of the vehicle is off) so that it could so it job of balancing and safety monitoring? Or, could the MPPT charge the battery safely without the BMS?

I noticed that the BMS has two set of cables. One set is the plus and minus high-voltage lines connected directly to the battery and a bundle which I assume is sensing and control signals between the BMS and all battery modules. The other set is two high voltage lines to the motor controller. I see another three phase lines between the motor control and the motor itself.

Would the MPPT be connected directly to the batteries or to the other side of the BMS?

B) The EV is off and charging from shore power.

Being able to charge from solar and from shore power at the same time is not a critical requirement for me, but just for completeness let's consider this state.

The charging connector would getting a AC or DC charging station.
Obviously the BMS will be on and doing its job of balancing and monitoring safety (current and temperature are within limits).
The built-in charge controller may have an expectation of the current level expected. For level 3 DC charging the power level probably is negotiated and could be changed dynamically.

Does the charge controller adjust its behaviour based on information from the BMS?

The MPPT would be adding a relatively smaller but potentially fluctuating amount of current.
I suspect that the built-in charge controller would monitor the shore power and shutdown if parameters were outside expected range, but the built-in charge controller probably won't be aware of any extra current going into the battery from the MPPT.

The BMS will see the sum of the currents from the built-in charge controller and the MPPT.
Would the extra unexpected current from the MPPT possibly trip some safety limit?

C) The EV is on and possibly being driven.

The BMS would be broadcasting safe operational limits (over CAN bus), specifically the maximum amount of current in and out of the battery.
Other control systems would be expected to respect these limits. For example, the acceleration rate of the vehicle may be limited, or the power recaptured by regenerative braking may be capped.
If the BMS would detect that the safe operational limits were being exceeded what would it do?

What would happen in this scenario?
  • BMS says max current into the battery is X.
  • Regen braking obeys and caps the current it produces to X during strong braking.
  • MPPT is adding an extra current of Y.
  • BMS detects a current of X + Y into the battery. Y is relatively small but still, it could trip a limit.

It would be nice to be able to charge while driving. Can it be done safely?

Any insight into this would be appreciated.

Thanks!
 
it's been done many times ;). It depends on your budget of course and design boundaries. How many panels are the 6kw ? a simplest and probably most dangerous is a simple block diode, no control, no smart, nada; just need the PV voltage to be higher than the EV pack. But don't do it

I only wish I could build my own vehicle from scratch, but sadly this is beyond what I could reasonably achieve legally.

I would be limited to integrating with a commercial vehicle. Proprietary electronics seem to be a challenge with this route.

A Tesla swap or other conversion would be possible. Electronics may be well know and could be easier to deal with.

There are many resources of how to configure solar panels. I'm not worried about that.
I would use (smart) bypass diodes to handle shading, and an MPPT charge controller for efficiency. The MPPT would be specced to handle the high battery voltage.
 
MPPT is Maximum Power Point Tracking on the PV side, nothing to do with charging per se; and MPPT is not a charger. A charger can have MPPT in the front end, connecting to the PVs. You keep mentioning Tesla, is this van a Tesla ? if it's a "tiny house" then wouldn't it be parked most of the time ? charging when EV off is heck easier than while on, but the EV BMS would be off too.
 
If you go on a forum for a popular EV, many random things have been figured out. I found out recently that Bolts are a popular donor car for EV parts to squish into another chassis. Thus many things about the Bolt’s HVDC system have been documented to support this.

I believe the contactors on the traction battery are disconnected whenever the car is in full off state. Tying into the battery in a way that bypasses the contactor probably escalates the overall challenge by quite a bit.

There are currently very few entry level CCS chargers, however a couple are coming onto the market this year. Maybe you can hack one of the DC platforms. Dcbel would probably be able to go straight from panels to CCS. Or hide the AC conversion from you, heh.

Development kits for CCS/DC charging are quite expensive as I understand, though I’ve mostly been talking with others merely peeking into that world rather than working in it. And those people aren’t the AliExpress experimenter types. There were some recent posts on this forum about chargers of appropriate power level.

Charging the car while driving seems like a terrible mode to use even if you get it to work, unless you have an intimate understanding of the interaction with the other things on the DC bus. Or custom power electronics in strategic places. The consequence of bad reverse engineering is pretty brutal given that the car is moving.

I could be wrong though, maybe someone has mapped out the behavior/state machine for an operating car.
 
As someone else pointed out, charging a car DC to DC is dangerous. But it is something that we all want to do because DC to AC to DC is just redundant and you lose as much as 40% efficiency.

I think the best way to accomplish this is to wait for the cars that has built in solar panels. I've read about some that are in the works, but nothing coming yet. I imagine it would be rather easy to tap into the solar panels and connect other DC supply sources to it.

What if Tesla, or Nissan, or Chevrolet provided a special port (maybe just MC4's) and an internal MPPT controller to attach solar panels on your own? Whichever company does that first, I will be buying their car.
 
Whichever company does that first, I will be buying their car.
There is no reason for them to include a MPPT to support a significantly larger solar panel than whatever engineering performance art meme inspired them to include a solar panel on the car. And the MPPT for the type of panel that fits on the car may not have an operating range (voltage or current) that you care for.
 
Thanks to Tesla and ipandee it’s possible to build one. https://tesla-cdn.thron.com/static/UMDJDV_North_American_DC_Charging_Connector_Datasheet_HTKQS6.pdf?xseo=&response-content-disposition=inline;filename="North-American-DC-Charging-Connector-Datasheet.pdf"
You need enough panels to supply 500v at 200-400 amps.
Then you need a charge controller from ipandee thanks Mike G for restarting the high voltage revolution. https://www.ipandee.com/large-mppt-solar-charge-controller/

And voila! DC fast charging for your vehicle when the sun shines. Add a high voltage battery bank and another charge controller and you can charge at night :) theoretically. Good luck!
 
Thanks to Tesla and ipandee it’s possible to build one. https://tesla-cdn.thron.com/static/UMDJDV_North_American_DC_Charging_Connector_Datasheet_HTKQS6.pdf?xseo=&response-content-disposition=inline;filename="North-American-DC-Charging-Connector-Datasheet.pdf"
You need enough panels to supply 500v at 200-400 amps.
Then you need a charge controller from ipandee thanks Mike G for restarting the high voltage revolution. https://www.ipandee.com/large-mppt-solar-charge-controller/

And voila! DC fast charging for your vehicle when the sun shines. Add a high voltage battery bank and another charge controller and you can charge at night :) theoretically. Good luck!
Well the ipandee might not cut it but you get the idea.
 
Well the ipandee might not cut it but you get the idea.
It might be easier to start with a hybrid solar inverter designed for HVDC batteries and figure out how spoof its communications to work with a car.

Because the solar inverter is designed to work with variable input at a more reasonable input power level than that ipandee
 
Thanks @meetyg!

For me, I am past the "if/why" to do this. I am doing this out of necessity, not to save money – this vehicle will be my off-grid home and I will not have access to grid power nor fossil fuels. So, I have to get this done one way or another.

In the worst case, I know I can achieve this with a buffer battery and charge the main battery via inverter while the vehicle is stopped. But this hacky solution makes me cringe.

One of the "off the shelf" trucks I am considering does have wide open access to battery packs, BMS and high voltage cables.
And if I do a Tesla swap or other conversion, I would have access to them too, although this likely will be a more costly solution that with an "off the shelf" EV.
I would get a professional to connect the proper gauge wires safely to the high voltage lines.
This part seems doable.

The part that I am struggling with is understanding how the existing EV electronics behave.
This doesn't seem to be common knowledge and this is why I am reaching out here for help.

Does anyone know exactly how the EV electronics behave?
Here is my speculation.

Consider the following state of the vehicle...

A) The EV is off and NOT connected to a charger.

There are no heavy currents (motor, regen, charger) in or out of the battery in this state.
Could an MPPT be connected directly to the battery or BMS in this state?

An MPPT normally handles the various charge cycles of the battery at the different levels of charge and will slow down when battery is near full and not overcharge.

Is the BMS still on and operating? If not, would it have to be switched on somehow (even though the rest of the vehicle is off) so that it could so it job of balancing and safety monitoring? Or, could the MPPT charge the battery safely without the BMS?

I noticed that the BMS has two set of cables. One set is the plus and minus high-voltage lines connected directly to the battery and a bundle which I assume is sensing and control signals between the BMS and all battery modules. The other set is two high voltage lines to the motor controller. I see another three phase lines between the motor control and the motor itself.

Would the MPPT be connected directly to the batteries or to the other side of the BMS?

B) The EV is off and charging from shore power.

Being able to charge from solar and from shore power at the same time is not a critical requirement for me, but just for completeness let's consider this state.

The charging connector would getting a AC or DC charging station.
Obviously the BMS will be on and doing its job of balancing and monitoring safety (current and temperature are within limits).
The built-in charge controller may have an expectation of the current level expected. For level 3 DC charging the power level probably is negotiated and could be changed dynamically.

Does the charge controller adjust its behaviour based on information from the BMS?

The MPPT would be adding a relatively smaller but potentially fluctuating amount of current.
I suspect that the built-in charge controller would monitor the shore power and shutdown if parameters were outside expected range, but the built-in charge controller probably won't be aware of any extra current going into the battery from the MPPT.

The BMS will see the sum of the currents from the built-in charge controller and the MPPT.
Would the extra unexpected current from the MPPT possibly trip some safety limit?

C) The EV is on and possibly being driven.

The BMS would be broadcasting safe operational limits (over CAN bus), specifically the maximum amount of current in and out of the battery.
Other control systems would be expected to respect these limits. For example, the acceleration rate of the vehicle may be limited, or the power recaptured by regenerative braking may be capped.
If the BMS would detect that the safe operational limits were being exceeded what would it do?

What would happen in this scenario?
  • BMS says max current into the battery is X.
  • Regen braking obeys and caps the current it produces to X during strong braking.
  • MPPT is adding an extra current of Y.
  • BMS detects a current of X + Y into the battery. Y is relatively small but still, it could trip a limit.

It would be nice to be able to charge while driving. Can it be done safely?

Any insight into this would be appreciated.

Thanks!
OK, let's try to answer your questions, but first, have a look at this video:

I've seen other EV owners do similar stuff (with a Nissan Leaf) on YT. This guy is special in that his build has folding panels. Probably not what you need but still interesting...

Now to your questions:
Case A:
Every EV battery pack has a BMS and contactors built-in. The contactors will close (making a connection) when the "ignition" switch is ON. They are powered by the 12v system (battery).
So when the EV is OFF, the HV battery is basically isolated from the car. This is an important safety measure.
The BMS is always on, or might go into a "sleep" mode, but it's still there and somewhat active.

Unless you pry open the HV battery pack (not easy and not recommended) , and get a direct connection to the BMS (before the contactors), you won't have access to the HV battery.

Case B:
When charging (either AC or DC) the contactors are engaged. However, AC and DC charging differ:
With AC, the charging is done and controlled by the vehicle's on-board charger. This OBC gets AC from the wallbox/EVSE and converts it to DC, pumping that into the HV battery. There is some communication going on between the OBC and the EVSE: Mainly negotiating the maximum charging current allowed by all factors: EVSE max charging capability, the CCS cable's max charging rating etc...
But again, the OBC controls the charging process and rate (up to the maximum negotiated).
Balancing usually starts at a high state of charge (SOC), say at the last percent or so of the full charging.

With DC, the charging is controlled by the DC charger. There is of course also some communication going on between the EV and the DC charger. This communication is crucial, because the EV must tell the charger what charging current is acceptable at any given moment. Factors like battery SOC / Temperature and more affect the charging rate.

Case C:
It depends on how the specific EV is programmed to work. It might or might not be a problem. For example, the BMS might think that the additional current coming from the MPPT SCC is simply additional regen.

Sounds like it might make things go wacky though...

In short, like I said, tapping into the HV battery is not easy and definitely NOT safe!

If the EV truck/van will be your offgrid home, you will need a "house" battery anyways for your offgrid loads. You could size that big enough (if you have the space and money) so that you could use some of that battery to do some AC charging daily. The house battery will recharge from solar power whenever that's available, either while driving or parking.
Although there are losses, it's a much simpler and safer solution than going directly to the HV battery. Another plus is that you won't be violating your EV's warranty. Most EVs have an 8 year HV battery warranty, which is very important.
 
Unless you pry open the HV battery pack (not easy and not recommended) , and get a direct connection to the BMS (before the contactors), you won't have access to the HV battery.
depends on the car, the BMW i3 has it's safety contactor box external of the battery.
There is some communication going on between the OBC and the EVSE: Mainly negotiating the maximum charging current allowed by all factors: EVSE max charging capability, the CCS cable's max charging rating etc...
But again, the OBC controls the charging process and rate (up to the maximum negotiated).
"Communication" is a interesting description for a simple 1khz PWM signal which originates from EVSE. During AC charging the car does not talk to the EVSE on the J1772 Standard. There is no negotiation, nothing it's one-way aside of the car closing a contact when you plug in.

The EVSE modifies a PWM signal to "tell" the car how fast it can charge. That cars OBC ramps up the charge until it matches that highest current (or below if the OBC doesn't have the capability)
In essence - a AC EVSE is nothing more the a fancy extension cord which tells the car it's maximum amps.

But as you state DC is different. That is two way communication. There is also a second contactor - which enables power to DC charge port.
While AC charging only turns on the battery contactor - DC turns on the second.
Maybe you can hack one of the DC platforms. Dcbel would probably be able to go straight from panels to CCS. Or hide the AC conversion from you, heh.

I recommend looking at the J1772 whitepapers and the documentation about the charging standard. It's a lot of reading but it tells you how DC charging and Vehicle to X over the port are working.

DC charging and discharging is decent simple in terms of the J1772 standard. But the cars Firmware must allow it. If they never had anticipated that the battery voltage goes down while DC is connected - some wizard programmer might have set a out of bounds condition and triggers an error.
 
Although there are losses, it's a much simpler and safer solution than going directly to the HV battery. Another plus is that you won't be violating your EV's warranty. Most EVs have an 8 year HV battery warranty, which is very important.
this is the single reason why we don't have Vehicle to Grid - V2X.

An outdated warranty law which works on years and miles. Two metrics which are pretty irrelevant for EVs, but easy to sell to customers and regulators.

A better Battery warranty would be Cycle Life - with a meter on the Dash. (like all home energy systems tell you somewhere the cycles they went through) with cycles there would be value attached to home energy use. Currently we just don't know how it would impact the calculation of EV lifespan. My $500 chins battery BMS tells me cycle life and my warranty is tied to cycle life + calendar age. What I do with the energy stored? Driving or using it my house - they don't care.
 
depends on the car, the BMW i3 has it's safety contactor box external of the battery.

"Communication" is a interesting description for a simple 1khz PWM signal which originates from EVSE. During AC charging the car does not talk to the EVSE on the J1772 Standard. There is no negotiation, nothing it's one-way aside of the car closing a contact when you plug in.

The EVSE modifies a PWM signal to "tell" the car how fast it can charge. That cars OBC ramps up the charge until it matches that highest current (or below if the OBC doesn't have the capability)
In essence - a AC EVSE is nothing more the a fancy extension cord which tells the car it's maximum amps.

But as you state DC is different. That is two way communication. There is also a second contactor - which enables power to DC charge port.
While AC charging only turns on the battery contactor - DC turns on the second.


I recommend looking at the J1772 whitepapers and the documentation about the charging standard. It's a lot of reading but it tells you how DC charging and Vehicle to X over the port are working.

DC charging and discharging is decent simple in terms of the J1772 standard. But the cars Firmware must allow it. If they never had anticipated that the battery voltage goes down while DC is connected - some wizard programmer might have set a out of bounds condition and triggers an error.
Yeah, I didn't want to go into too much detail.
I have actually read a summary of the CCS2 spec. Many DIYers make thier own AC EVSE with an Arduino. That is what OpenEVSE is based on. Not too complicated.

Just wanted to make my point that its not "plug an play".
 
I recommend looking at the J1772 whitepapers and the documentation about the charging standard. It's a lot of reading but it tells you how DC charging and Vehicle to X over the port are working.
Which specific whitepapers? If I randomly picked a J1772 documentation or whitepaper, I doubt it would say much about CCS and V2X because that is highly tangential.
 
Back
Top