Speak EV - Electric Car Forums banner

1 - 14 of 14 Posts

·
Registered
Joined
·
138 Posts
Discussion Starter #1
I have solar panels on the roof (of the house)
so I'm guessing it's going to be cheaper to charge during the day (while the sun shines)
or am I just dreaming?
 

·
Registered
Joined
·
130 Posts
No, you're not dreaming, but the paradigm is slightly different: your roof might peak at 4kW but your car will likely want to charge at 7kW and people often like to charge up at night if they have economy 7 or a similar new time-of-use tariff with a smart meter. However, there is a charger called a Zappi which I have for diverting the excess solar (which would be exported) into the car. Realistically this amounts to more of a trickle than a full charge and would at most charge my leaf 40 to 60% in one day.
 

·
Registered
Joined
·
923 Posts
I've been working through the challenges with getting the most from our surplus PV generation into the car. It's not at all as easy as it seems, as the lowest allowable charge current from an EVSE is 6 A, so about 1.4 kW, and not all cars will actually go that low, it seems. Add in that excess PV generation can jump up and down a lot, as house loads switch on and off, or as clouds drift over, and then chuck in that the car onboard charger tends not to track changes in available charge power very quickly (can take tens of seconds to react) and it starts to get more complex than it first looks to get the most out of charging from excess PV generation.

As I also charge using E7, I've opted to set the algorithm that controls daytime charging from excess PV generation so that the car will charge for as long as the cost of charging is lower then the cost of charging from E7. In essence that means that the car will use a mix of grid power and excess PV generation during the day, with the EVSE turning off (and staying off for a time, so as not to upset the car charger) if the proportion of PV charge is less than about 50% of the total.

It's been a damned sight more challenging to write the code for this new EVSE than it was to write the code for my hot water heating system power diverter!
 

·
Registered
Joined
·
1,165 Posts
We've had solar PV for 4 years. We switched to TIDE (basically E7) a couple of years ago, so moved to overnight charging at 5p/kWh. That tariff has now shot up in price, so we are moving to Octopus GO, which is also 5p/kWh overnight, but only for 4 hours. Still enough for most charging requirements. We've now switched, but are awaiting a smart meter installation to get the cheap overnight rate. This might be a couple of months away, due to incoming SMETS2 meters.

In the meantime I'm back to solar PV charging using the Zappi I picked up on eBay recently :).
 

·
Registered
Joined
·
923 Posts
I plug my Kona in using the three pin granny charger and set it to trickle charge at 1.2kw which takes up pretty much all of our solar excess.


Sent from my SM-G965F using Tapatalk
Interesting that you can get down to a lower charge rate (1.2 kW) than the protocol permits.

The minimum charge current is set at 6 A, which for a nominal 230 VAC mains supply would be 1.38 kW. In reality, the UK mains voltage tends to be higher than 230 VAC, as when we harmonised with the EU we just offset the voltage tolerance to +10%/-6%, so that our existing 240 VAC supply was compliant with the 230 VAC requirement. 240 VAC remains pretty much the average voltage here, so the minimum charge power allowable within the protocol would then be about 1.44 kW.

I will admit that I got to one stage when I was writing the code for the new EVSE, and testing it, where I seriously thought about just having two or three switched power levels and doing away with the wireless link that receives import/export power. I'm still not 100% convinced that just setting a low charge current, with a trigger to turn the EVSE on when export exceeds the lowest charge power, and a time out to turn the EVSE off after a sustained period with no PV generation, might not work just as well.

I doubt that all car onboard chargers behave the same way to varying the pulse width of the CP signal to signal more or less current being available, either. The i3 has a definite slow ramp up of charge current that lags tens of seconds behind any change in the CP pulse width, but I've no easy way of knowing if that's typical or just the way BMW have implemented current sensing.
 

·
Registered
Joined
·
1,006 Posts
Re varying charging current through EVSE PWM: The protocol allows 6A-60A but I have found that my EVSE (ABL Sursum EVCC controller) allows a PWM value of 48 - 4.8A - which my old Outlander was happy with until one annual service where they must have done an update and after that it drew 6A minimum regardless. I have not tried this with my Leaf.

The other thing I found odd was that when I sent the code to tell th car that charging was not available (to try avoid having to use a relay to drive the contactor to ... etc.) the Outland would sometimes never recover and required being unplugged, turned on and off again. Again, not tried with the Leaf, but may do later as the weather gets better.

In reply to the OP, I used to do this and I had the right access and tech to make sure I only sent the "spare" solar to the car, but unless you want to DIY you should look at the Zappi - but this is only useful if your car is likely to be plugged in and charging in the daytime. If you will, even most of the time, be charging overnight then you will either need to just use c cheap overnight tariff or invest in home battery storage.

I went with a couple of Tesla Powerwalls in the end, two both for capacity and also to have the inverter output to charge at full rate. There are losses everytime you charge one battery and discharge it - so I have set up a slightly different version of my home-brew solution for those days during this summer that I will cycle to work and leave the car, on the basis that using the solar directly will save those losses when it can.
 

·
Registered
Joined
·
322 Posts
I like to keep it simple (aka crude) If it is wall to wall sunshine I plug in my granny charger at 10Amps. If it is a bit hazy or early or late in the day but the sun is shining I set the car to charge at 6Amp.
 

·
Registered
Joined
·
923 Posts
The Prius Plug-in would sometimes throw a wobbly if the CP PWM was stopped and taken to the ready state of +12V. Not sure why, as it wasn't consistent. Sometimes it would just flag a "charge failed" message and only restart after having been unplugged and plugged back in again. Other times it would happily just pause charging and then restart when the CP PWM came back up.

I get the feeling that there's a fair bit of inconsistency in the way different OBCs handle both turning charge on and off and in the way they respond to varying the CP PWM duty cycle on the fly. I doubt that the idea of varying charge rate dynamically was uppermost in the minds of those that first put the protocol together, which may explain the differences.

I'm installing a house battery system, but don't intend to use it for charging the car, I just want to make the house pretty much grid-free for about 2/3rds of the year, and better utilise our generation capacity. Not really cost effective, but it will give us a separate backed up emergency circuit, which will be handy during power cuts (we get a fair few here for some reason).
 

·
Registered
Joined
·
510 Posts
Interesting that you can get down to a lower charge rate (1.2 kW) than the protocol permits.

The minimum charge current is set at 6 A, which for a nominal 230 VAC mains supply would be 1.38 kW. In reality, the UK mains voltage tends to be higher than 230 VAC, as when we harmonised with the EU we just offset the voltage tolerance to +10%/-6%, so that our existing 240 VAC supply was compliant with the 230 VAC requirement. 240 VAC remains pretty much the average voltage here, so the minimum charge power allowable within the protocol would then be about 1.44 kW.

I will admit that I got to one stage when I was writing the code for the new EVSE, and testing it, where I seriously thought about just having two or three switched power levels and doing away with the wireless link that receives import/export power. I'm still not 100% convinced that just setting a low charge current, with a trigger to turn the EVSE on when export exceeds the lowest charge power, and a time out to turn the EVSE off after a sustained period with no PV generation, might not work just as well.

I doubt that all car onboard chargers behave the same way to varying the pulse width of the CP signal to signal more or less current being available, either. The i3 has a definite slow ramp up of charge current that lags tens of seconds behind any change in the CP pulse width, but I've no easy way of knowing if that's typical or just the way BMW have implemented current sensing.
Just checked this, and it goes lower...1.1kw



Sent from my SM-G965F using Tapatalk
 

·
Registered
Joined
·
645 Posts
At the moment i have PV and a Immersun diverter for the hot water and 2x EV. My overall plan is that at the start of the day i allow the diverter to heat the water until sufficient excess PV reaches 2.4kw when i switch on the granny cable and leave it switched on until later in the day when it drops back to the 2.4kw. and it reverts back to the HW This makes no allowance for passing clouds or house use but i feel this is small beer in the scale of things. In summer any excess over the 2.4kw heats the HW.
I have doubts about Zappi/car charger following all the vagueries of the PV output and house use however if the electrical load is great enough then you need to use the PV and grid together and therefore you have used all the available PV and no advantage is to be gained. Only for a few days in summer will the PV be generating enough to do the EV,do the HW and do the house loads. To then worry about the little bit you exporting is not worth doing anything.
I note on the Immersun there is the facility to operate a relay which can be programmed and my gut instinct is that this would be a better way and relieve me of manual imput and allow the car to charge at full available excess PV. The relay can be programmed to only assess the situation every so often say 15mins but this seems more than adequate. What do you think?
I have 2 EVs so that one is operating as a battery and largely preventing export so until TOU tariffs and V2H/G happen this is near enough as good as it gets.
 

·
Registered
Joined
·
154 Posts
I've been working through the challenges with getting the most from our surplus PV generation into the car. It's not at all as easy as it seems, as the lowest allowable charge current from an EVSE is 6 A, so about 1.4 kW, and not all cars will actually go that low, it seems. Add in that excess PV generation can jump up and down a lot, as house loads switch on and off, or as clouds drift over, and then chuck in that the car onboard charger tends not to track changes in available charge power very quickly (can take tens of seconds to react) and it starts to get more complex than it first looks to get the most out of charging from excess PV generation.

As I also charge using E7, I've opted to set the algorithm that controls daytime charging from excess PV generation so that the car will charge for as long as the cost of charging is lower then the cost of charging from E7. In essence that means that the car will use a mix of grid power and excess PV generation during the day, with the EVSE turning off (and staying off for a time, so as not to upset the car charger) if the proportion of PV charge is less than about 50% of the total.

It's been a damned sight more challenging to write the code for this new EVSE than it was to write the code for my hot water heating system power diverter!
Throw a powerwall into the mix and then try to decide whether excess goes to powerwall, water heating or car charging......
 

·
Registered
Joined
·
923 Posts
The biggest single problem is the variability in the way vehicle OBCs behave when the EVSE switches charge off before the vehicle is fully charged. There doesn't seem to be any hard and fast way to determine how any vehicle might behave, other than by testing multiple times.

In theory, just switching the PWM off and taking the CP high should tell the OBC to stop charging, before the EVSE contactor opens, so gracefully ending the charge period without flagging any errors. In practice it seems more complex, as the time between the cessation of the PWM signal and the OBC turning off can be variable, and if the EVSE contactor opens too soon there is a risk that the vehicle will flag a charging error and then refuse to charge until unplugged and plugged back in again.

I had this happen a few times with the Prius Plug-in, although it seemed almost random as to whether the car would flag a charging error. One thing that was consistent was that just turning the power to the EVSE off would always flag a charging error, but oddly it wouldn't always need to be unplugged and plugged back in to reset that error.

The i3 behaves differently, in that it seems to always want to check that the EVSE is working before it will use the in-car charge timer (the car requests charge, shortly after being plugged in with the timer set, waits until power starts to be delivered, then turns the charge off until the set time). I've yet to fully test how the i3 behaves when the CP is turned off mid-charge, and then taken high. Hopefully this will allow a graceful charge shutdown, with no charging errors flagged. I need to determine how long to wait between the PWM turning off and the EVSE contactor opening, as it seems clear that the contactor needs to stay closed for a short period in order to allow the OBC to turn off.
 
1 - 14 of 14 Posts
Top