Off grid solar

[Manodamountain]

I live in the sticks off grid. I have a 6.5k solar system. I’m wondering if anyone has an idea of how long it would take to charge a CT with 110v? Without adding far more capacity to my system there is no way I’d be fast charging.

Sponsored

 

[JBee]

Assume about 2.5miles per kW per hour, but you're probably limited to what your 110V circuit is rated at, and whatever else is on that circuit. So maybe 5miles per hour max. It will take a while.

I'm offgrid with 30kW but that is 3phase 415V. Hopefully the CT will have some PV inputs that might help.

 

[ajdelange]

I live in the sticks off grid. I have a 6.5k solar system. I’m wondering if anyone has an idea of how long it would take to charge a CT with 110v? Without adding far more capacity to my system there is no way I’d be fast charging.

You wouldn't be L3 charging but you could charge L2.

There are a couple of things to consider with a solar system vis-a-vis a BEV fueled by solar. The first is how many miles you can collect in a day. With a 6.5 kW system in a place that has 6 hrs (summer time) FSE (Full Sun Equivalent) per day you can collect 39 kWh/da. Assuming the CT to be conservatively operated it's probably going to require 450 Wh to move it one mile so the 39 kWh corresponds to 39/.450 = 86.7 miles per day. That is if you don't use electricity for anything else which I assume you do.

The next consideration is as to how fast you can charge the truck. This depends on your system battery. If you have 2 Powerwalls, the minimum Tesla will accept in an order for an off-grid system, they will discharge at a 10 kW rate (that's the current ones - the newer ones are supposed to do faster than that). This means that you can charge the truck at up to 10 kW (if there are no other loads on the system). That's 45 miles per hour. But note that 2 Powerwalls have capacity of 2*13.5 = 27 kWh (60 miles). This is only 12% of a TriMotor battery. They will not even hold an entire day's worth of collection (39 kWh) from your array. That may or may not be a problem. If the batteries are full and the sun is shining you put the truck on to charge and the energy from the array goes directly to the truck rather than to charge the Powerwalls. If you have set the truck charger to 10 kW and the array produces 6.5 kW then all the panels' production will go to the truck and the additional 3.5 kWh will come from the battery. If you have full Powerwalls from yesterday and you want to charge today (assumed sunny) then you can take the 39 kWh from the array and the 27 kWh from the Powerwalls for a toatal of 66 kWh. Note that this is only 33% of the capacity of the Tri Motor battery. Also note that you are using 2 day's sun to get this. As the sun is the only source you are limited to the 39 kWh (87 miles) that the sun can provide each day.

A Powerwall is a 240V device. This means you can use Level 2 chargers with it. If you have 3 Powerwalls their 3*13.5 = 40.5 kWh combined capacity is enough to hold a days production and their 3 inverters will produce 15 kW which is more than the 11.5 kW maximum charging rate of the truck. You could load a full day's production in 39/11.5 = 3.49 hrs but note again that a full day's production is only 18% of the TriMotor's battery. The fastest you can do with a Level 1 charger is 1440 W. To load a day's production into the CT at that rate would take 27 hours. You'd always be behind the curve.

What really drives all this is the size of your system relative to the number of miles you need on a daily basis. If you need more than 87 miles per day there is no way that you can make it on a 6.5 kW system. If, OTOH, you only go to town once a week even though it's 250 miles one way you can do it. Theoretically anyway (we haven't talked about conversion losses, round trip losses or phantom drain). I'm doing an off grid system with 5 Powerwalls and 13 kW (AC) of panels. It is, with the other loads, going to be very marginal. But it isn't in the boonies. It is right next to the main house which is grid tied, has more panels and is backed up by a generator.

I would think long and hard about the practicality of a CT with a solar system as small as 6.5 kW or indeed any solar system with out any backup. I would definitely have a generator as Tesla recommends.

I've tried to throw numerical example in here in the hope that you will be able to do your own calculations for your own particular circumstances.

 

[JBee]

You can also just buy about 8kW or so of PV panels and a off-grid "batteryless" inverter for solar charging straight off the sun, and use the CT battery as the "buffer" for shorter trips. Which will cost you less than getting just one powerwall... or get three 8kW setups for the price of two powerwalls.

Powerwalls aren't cheap, and I think you need to buy solar from Tesla now to get them? It actually works out cheaper to buy another CT than to get the same battery capacity in Powerwalls.

 

[ajdelange]

Powerwalls aren't cheap, and I think you need to buy solar from Tesla now to get them? It actually works out cheaper to buy another CT than to get the same battery capacity in Powerwalls.

I used Powerwalls as an example rather than as a recommendation and I did that because I know their specs and, conceptually, at least how to cobble them into a system. Yes, they are expensive, yes, you must buy them as part of a system with panels, yes the lead time is months and months and no, your solar provider probably doesn't know how to hook them up.

You can also just buy about 8kW or so of PV panels and a off-grid "batteryless" inverter for solar charging straight off the sun, and use the CT battery as the "buffer" for shorter trips.

You can't use the CT battery as a buffer because you can't connect to it. You must go through a charger which will shut down when the power supplied to it drops or is too irregular. Here's what solar production looks like around here on an early July day.

Of course the biggest disadfantage of a batteryless system is that you must charge when the sun is shining. No coming in from a morning errand and topping off with the sun that fell while you were out galavanting. No coming home from work and plugging in for an overnight charge.

Yes, there are lots of things you can do. My personal goal is to achieve the same level of convenience I have with the charging arrangements in my grid tied house. I'm making choices, because of this, that are costing me more than I could get by with. Each must make his own decisions.

However, if you need more miles per day than your array can supply, you are stuck. BEV use a lot of electricity.

 

[JBee]

No worries your info was correct, extensive and valid just a bit out of budget for many. The data points are useful as a reference.

As you can see I don't believe in brand loyalty and I'm know to be a cheapskate, but only because I look to improve the effectiveness of my monetary levers.

There's actually another off grid thread were I mentioned how you would have to integrate a batteryless inverter, but I'll explain it in detail here instead. Note all of the below is for off-grid installs only.

The only thing stopping you with a batteryless setup is that you have to make sure the EV is sucking less than the solar can give. There's three ways to do this.

So in the first one you would need to feed the Tesla app a charge rate setting based on a solar light sensor from a wifi weather station. (Can all be done in software then) By leaving a bit of a overhead, you can then throttle the EV charge rate in realtime to follow PV input. Weather or system prediction can then determine the extent of PV headroom, and bring the headroom closer. If it runs into a fault then the consequence is just a inverter trip, upon which it will restart, and you'll have to restart the EV charge.

Or another way is to load up the PV input to its max input power rating, that can typically handle at least 50-60% more than its 240V output rating. Which means the PV side has some more headroom to keep the 240V output constant, so the EV charge rate doesn't have to be modulated much, when running without a battery. So the trick here would be to prioritize the PV input side of the system build, to such a degree that it is always more than the L2 charging rate of around 11.5kW on the 240V output side. Obviously adding 2sets of 8kW batteryless inverters ($1400 ea) with 10-12kW of PV each (at under $0..45c/W) will sustain the 11.5kW charger even going down to less than 40% of rated solar input. Essentially our solar PV headroom becomes our "buffer" here, and generally improves system performance by allowing more hours of coverage at over the L2 rate each day on sunny days, or even in winter with low angles of incidence. Only cloudy weather would require EV rate throttling. The inverters don't care, so long you don't exceed the PV input specs.

Sure you don't get to use all the energy collected, but PV panels are no longer the most expensive part of the system.

So thirdly, of course ideally, you'd run a small battery pack at least to do some buffering on the inverter side. This should be enough to allow for short weather disruptions. But as above the best way to sustain charging performance, is to ensure enough power can be collected by the PV modules, most of the time. That in turn allows for the batteries, if you have some, to be sized so their maximum current rating is sufficient to supply the charging load, but not that it has enough energy capacity by itself to charge the car completely. Doing it like that means that you can save on the most expensive part of the system, namely the batteries. Works out much better financially, but also means the batteries, how ever many you have, will not be cycled as much either. So a win win.

Lastly on the subject of using the CT as a buffer. As we know the CT will have L2 charging input, as well as a 7-10kW inverter output. So there's no issue for a typical household to operate within those two constraints. During the day, as outlined above, the solar would exceed the EV charge rate, with the small battery acting as a buffer to keep the headroom or supply the household startup loads too if required. During the night the CT can act as the supply for the house via the CT inverter output. Ideally the CT would run both L2 charging and inverter out at the same time, but it probably won't. But that's no biggy, you simply wire the CT inverter output into the inverter generator/network input (via a switchover for extra geny if you want) and hey presto you have a household supply from the CT battery instead.

Obviously using only one CT this is all range sensitive, and depends on how far you have to travel each day or each week, the weather, season etc, and for that matter how far you are from the nearest SC fast charger. If you could use a second CT (even just a SM) it's easier to use one CT as a backup battery, or alternate between CTs. But if you only have one CT, and you make that TM, it stands to reason that it's possible to use some of the CT range for household use at night too because you won't be travelling 500miles a day, during the day, and still expect to charge the CT during the same day with solar as well.

In summary, this simply means that it is unnecessary, and cost prohibitive for most to use Powerwalls for home, off grid charging, with the same convenience as a house on the grid. You're actually better off setting up a batteryless system and alternating EV's (don't all have to be a CT, but might as well get a SM for the price and pack size and inverter output rather than a M3). And if you're on the grid, it doesn't really make sense at all financially, as you are better off plugging it in there instead. Same goes for if you have a SC or similar nearby, or L2 charging ability at work, might as well get it there too.

Sorry long post...

 

[ajdelange]

No worries your info was correct, extensive and valid just a bit out of budget for many. The data points are useful as a reference.

Not worried at all and most interested in other peoples' perspectives and ideas.

The only thing stopping you with a batteryless setup is that you have to make sure the EV is sucking less than the solar can give.

Aye, there's the rub.

So in the first one you would need to feed the Tesla app a charge rate setting based on a solar light sensor from a wifi weather station. (Can all be done in software then) By leaving a bit of a overhead, you can then throttle the EV charge rate in realtime to follow PV input.

I have done this manually in order to be sure that all the power going to my car is from the sun but the grid is behind me if I don't react fast enough. Doing this automatically immediately comes to mind. I have a CT on the solar so there's my source of info on how much the panels are producing, the power meter has an API so the production is available to software and the Tesla has an API which can be used to set the charge rate. There are two problems: the first is that I wouldn't have a clue as to how to access those APIs and the second is that the Tesla API isn't officially blessed. Tesla could pull it at any time (they already disabled its ability to read battery current). But indeed a software system is possible. Best implementation of this would be modification, by Tesla, to the charging system hardware and firmware to gracefully accept wide swings in available power input. This conflicts with some of the systems safety features. As so many people seem to want to charge from solar I'd like to see them incorporate, perhaps as an option, a separate solar charger with the ability to plug in at least a couple of kW of panels.

Or another way is to load up the PV input to its max input power rating, that can typically handle at least 50-60% more than its 240V output rating.

I think you mean the inverter here. It should be able to handle panels with capacity several times its output capacity as it regulates by gating the solar input at what ever duty cycle is necessary to maintain the output voltage. The limitation would be the panel string voltage. If the panels can produce, in full sun, more voltage than the semiconductors in the DC/DC conversion circuit can withstand they will pop.

Which means the PV side has some more headroom to keep the 240V output constant, so the EV charge rate doesn't have to be modulated much, when running without a battery. So the trick here would be to prioritize the PV input side of the system build, to such a degree that it is always more than the L2 charging rate of around 11.5kW on the 240V output side. Obviously adding 2sets of 8kW batteryless inverters ($1400 ea) with 10-12kW of PV each (at under $0..45c/W) will sustain the 11.5kW charger even going down to less than 40% of rated solar input.

Only if there is 11.5 kW of sun being collected. For example I have 29 kW (DC) of panels with 26 kW of inverter behind them. As you can see from the plot of yesterday's collection there are times when output goes down to well below 11.5 kW. Modern inverters have MPPT. They will load the arrays to extract the maximum power they can for the given level of insolation. If you have a 5 Ω load (11.5 kW) and 20 kW of solar available the inverter(s) will gate the array at less than 100% to maintain 240V thereby pushing 48 A (single phase) through that load and that load will receive 11.52 kW. As the solar input drops to 15 kW the inverters duty cycle must go up to maintain 240V but the load still gets 11.52 kW. When the solar input is 11.52 kW the duty cycle has to go to 100% to maintain 240V (and 11.52 kW) but if it declines further the inverter cannot increase duty cycle further. It cannot maintain 240 V and the load no longer gets 11.52 kW. Now if the car senses the voltage dropping and reduces the load to try to hold voltage at 240, which it could do, it could follow the sun down below 11.52 kW but, AFAIK, it doesn't. It interprets the voltage fluctuation as a fault and shuts down. Thus unless you set the charge level really low you are going to have frequent charger shut downs and setting the threshold low means lots of solar wasted. Now for those of you who live out beyond the black stump where there is day after day of cloudless sky this may not be so much of an issue but for those of us in coastal mesic areas clouds are, as my picture shows, a big issue. In May I had 5 cloudless days, in June, 4. But no doubt about it, more panels means you can collect more sun no matter how you manage them.

Essentially our solar PV headroom becomes our "buffer" here, and generally improves system performance by allowing more hours of coverage at over the L2 rate each day on sunny days, or even in winter with low angles of incidence.

There is no point, from the BEV charging perspective, in exceeding the L2 level in a batteryless system as the truck cannot charge faster than 11.52 kW. I a battery buffered system the extra is stored for charging at a more convenient time or for serving other loads.

Batteryless systems bother me as an engineer mostly because it wastes collected solar energy. The panels are inexpensive and the batteries expensive and that definitely goes into the trade equations but I think it is true that what is holding back further adoption of solar world wide is the current lack of economical storage means.

Another aspect of this, which, again may not apply to you back of the black stump folks, is real estate. Many of us don't have hectares of land to put arrays on which means we are effectively limited to roof space. If I thought my wife would tolerate another 45 panels on the lawn I'd have another 45 panels on the lawn.

Sure you don't get to use all the energy collected, but PV panels are no longer the most expensive part of the system.

I'll note again that this really goes against decades of engineering training but I fully recognize that one doesn't have to make a crusty old engineer happy in order to have a workable solution in a particular application.

So thirdly, of course ideally, you'd run a small battery pack at least to do some buffering on the inverter side. This should be enough to allow for short weather disruptions.

I'd say that at my level of understanding of the current Tesla charging system this is a sine qua non. And of course a batteryless system is not possible if other electrical loads must be operated at night unless they are backed up by a generator, windmill, water wheel etc.

But as above the best way to sustain charging performance, is to ensure enough power can be collected by the PV modules, most of the time. That in turn allows for the batteries, if you have some, to be sized so their maximum current rating is sufficient to supply the charging load, but not that it has enough energy capacity by itself to charge the car completely.

The sizing of the battery bank is another subject in itself. In a charging only application the driving parameters are the number of miles you need to take from the system per day, the weather (how many sunless days in a row you will encounter) and, of course, the cost of the batteries which, as you have noted, is the biggest cost in the system. In a more realistic application, such as where grid connection is impossible, the charging load isn't the only load and the other load may completely dominate. I find my charging load is about 5% of my total load but I don't drive that much and I heat with heatpumps.

Doing it like that means that you can save on the most expensive part of the system, namely the batteries. Works out much better financially, but also means the batteries, how ever many you have, will not be cycled as much either. So a win win.

Were money no object or batteries cheaper the obvious solution is to add more and more battery. Clearly your array must provide, on average, your entire average load but the more battery you have the shallower the charge/discharge cycle.

Powerwall (and other battery backup) adverts always show the neat installation of a Powerwall or 2 against the side of a McMansion. This just isn't realistic. I don't live in a McMansion but it would be impossible for me to run it completely on solar. I heat with electricity and in the winter get perhaps 3 hrs MSE. My average load Nov - Feb is right around 137 kWh per day. I'd need 157 panels to supply that and to store a days worth 10 Powerwalls ($80,000 worth). That's for 1 day. Where I live the probability of 2 days in a row at nearly full solar output is 48%, the probability of 3 days in a row 28% and the probability of 4 days in a row 8%. Note that this is northern Virginia - not the foggy coast of Maine. So I'd want a few days worth of storage i.e. 30 powerwalls. Out of the question

Lastly on the subject of using the CT as a buffer. As we know the CT will have L2 charging input, as well as a 7-10kW inverter output. So there's no issue for a typical household to operate within those two constraints.

An off grid house isn't typical and a house with a BEV isn't typical. The average consumption in the states is 30 kWh/da. The average American drives 10,000 miles per year. At 250 Wh/mi (and note that a CT is going to pull almost twice this) his electric consumption would rise to 36.8 kWh/da (increase by 22.4%). BEV are owned not by the average American but by the better off than average American and as a consequence lives in a larger house which consumes more than the average amount of electricity. Assuming you have a Solar system which supplies 11.5 kW and 6 hrs FSE and enough battery to carry you through the short cloud outages you would be able to collect 59 kWh (85% charging efficiency) per day. That's not bad by any means but it wouldn't cover my daily utilization of 86 kWh/da in the transition seasons (no heating or A/C loads). I emphasize that my house is obviously larger than average but it is no McMansion by any stretch of the imagination.

During the day, as outlined above, the solar would exceed the EV charge rate, with the small battery acting as a buffer to keep the headroom or supply the household startup loads too if required. During the night the CT can act as the supply for the house via the CT inverter output. Ideally the CT would run both L2 charging and inverter out at the same time, but it probably won't.

I'd be surprised if you couldn't run loads off the bed outlets while charging but I'd be even more surprised were there no clause in the warranty forbidding using the truck to feed a house.

... this is all range sensitive, and depends on how far you have to travel each day or each week, the weather, season etc, and for that matter how far you are from the nearest SC fast charger.

In summary, this simply means that it is unnecessary, and cost prohibitive for most to use Powerwalls for home, off grid charging, with the same convenience as a house on the grid. You're actually better off setting up a batteryless

You can't say better without specifying the criterion of optimality. It is often the weighted sum of several things with cost obviously being one, frequency of being without power another, available daily range yet another and convenience yet another. And of course one's ability to maximize his optimality metric depends on his driving mileage needs, his desires (such as convenience and/or the desire to run for some number of sunless days), his financial means, the geometry of his house and the property it sits on, the climate in his locale and how easily his wife can be conned. If you put heavy weight on cost then one is led towards the sort of things discussed here. If heavier weight is put on convenience, reliability of supply, infrequent need for generator backup, staying within Tesla's warranty and such then he is driven towards more battery and more capital outlay for them.

And if you're on the grid, it doesn't really make sense at all financially, as you are better off plugging it in there instead.

My goal is to test by experiment the validity of my hypothesis that it is possible to build a 3 car garage and charge up to 3 BEV within in it in the same way i.e. with the same level of convenience using only the solar cells on its roof. "Convenience" here means you drive in at night, through normal electric garage doors, plug into normal EVSE go back to the main house and close the doors via WiFi and turn on the carriage lights for some family member expected home later. IOW just as you would in a grid connected McMansion. Is this sane? No, of course it isn't. But there ain't that much sand left in the hourglass and, I am told, I can't take it with me.

Sorry long post...

I am the last person to whom you should feel you have to apologize for a long post but, of course, there are all the other readers. As I said at the outset here I like to see what others are thinking.

 

[JBee]

Cool beans! ?

I completely agree that every system needs to be optimized within it's constraints.
I also agree that everyone has different constraints.

I'm also happy that you can see the merit of going batteryless, if the constraints allow. In particular I find "alternative" and more dynamic methods more appealing, especially in getting demand to meet supply, via automation. Like setting the CT charge rate to meet available PV input. As mentioned previously, it would be nice to have a "body builders" port to interface the CT with whatever is connected, or at a minimum a power/energy interface and API.

Despite making the longest post list for the forum, you missed a response on one subject, and that is on the viability of purchasing and using a CT as a PV household and BEV charging battery bank. Or to alternating between driving two CTs. Much cheaper than buying that many powerwalls in any case.

Good fun sharing ideas!

(BTW I have developed a variation of this for "wireless virtual microgrid" a while ago, for third world regions where cost is the dominate constraint, and where PV and storage are movable elements within the system, and can be shared to meet demand.)

 

[ajdelange]

I actually don't see much merit in batteryless systems at all as without a battery you don't have the ability to time shift or collect all the energy available. I do understand that there may be circumstances where economic considerations make a battery impossible and in those circumstances one has to get by with what is available.

I didn't specifically directly respond to the notion of using the truck as a battery but I would never do that unless there were no alternative. As I noted in an earlier post I think Tesla will probably declare such a use of one of their vehicles as a violation of warranty. They do so now but then their current cars don't have 120/240 inverters. Musk and Straubel have, in the past, said that this is not what trucks are for. Market conditions may change their minds eventually, though.

Charging of their vehicles from solar is a different matter. There are solar charging stations commercially available. These all use batteries - and are expensive. But I think the market is clearly for solutions that allow time shifting.

There are multiple You Tube videos by guys who have built their own solar charging stations. I have not seen one that does not use batteries.

You can, as noted, charge a Tesla from nothing but solar arrays and an inverter. Here's another picture of a couple of days collection showing what clouds do.

It's plain from this picture that there are opportunities on either of these days to charge a Tesla, even at a high rate. You actually wouldn't have to do anything. You could start charging at 0945 on the 20th at a 10 kW rate. You'd be fine until around 11:45 when the car would shut down. But as long as the 240 is there it will try to start again. When it starts a charge it ramps up. I don't really know what happens if voltage starts to drop before it reaches the desired rate. I think it is doing this ramping to test the supply and so I think it will shut down again. But it will keep trying and after that cloud passes it will be able to draw the desired 10 kW with voltage in range and the charge will resume. On the 21st (the solstice) you'd be fine at 10 kW from 0900 to 1430. For daily use replacement charges these windows are probably sufficient. They are for me. This is actually how I charge (except that the utility would step in rather than the charge terminating if solar can't do the full 10 kW).

Automate this through an API? Feasible, yes, but I don't think Tesla will make that possible or easy. You obviously can put a CT on the solar line (that's how I get these pictures) and you can send a command to the car through the unofficial API to change the SoC charge limit but you can't send a command to control the charge rate. I think I may have said you could in a previous post. If I did I was wrong. I wish you could but AFAIK you can't. In any case this is rather a clumsy way to manage this and there is clearly a better one which we have discussed and that is to have the panels connect to the battery directly through a DC/DC converter. We've mentioned before connectors in the bed for plugging in solar panels in the field. This would require Tesla to put the DC/DC converter in the truck. This would be a definite nice-to-have but I doubt Tesla will do it as there wouldn't be enough market for it. Can a third party connect directly to the battery? Yes, by building a box that looks like a DC fast charger. This requires reverse engineering the Tesla to SC interface which might present some challenges. One might have better luck reverse engineering the CHAdeMO interface as that's probably public information.

In any case a DC fast charger communicates with the vehicle telling it how much power is available and the vehicle responds by telling the fast charger to increase its voltage until the amount of current the vehicle wants is observed. If it can't get what it wants it accepts what it can get. I'm guessing that it will continue the charge down to pretty low levels (a kW or so) before it terminates a charge and I'm guessing that is the case because I have observed it. A Supercharger in Kingston NY started out at 97 kW and tapered down to 11 kW within 5 min and then down to 1 kW over the next 9 minutes (while I was fiddling around buying unhealthy eats in the shop). SoC was about 47% so clearly the car was not asking for this taper. It was imposed by the SC which was obviously faulting and limiting its output power. The point is the car seemed quite happy to keep charging at this low rate evidently indefinitely.

Thus what we need for batteryless charging fom solar is a third party (or could be Tesla but I doubt they do it because of limited market) box to which one connects solar panels and looks to the truck like a fat charger. This arrangement is the same as one has with a Jackery or Yeti power pack. One plugs solar panels into these devices which load the panels at the MPP and set DC/DC converter output voltage to whatever is necessary to deliver the available power to the battery. When the sun is behind clouds that isn't much, of course, but whatever it is goes to the battery. It goes without saying that this arrangement is going to be more efficient than inversion to AC outside the car followed by rectification, inversion and then rectification again inside the car.

What's the catch? Money, of course. There are DC chargers available to the home market but they are not inexpensive.

Just another thought.

 

[JBee]

Thats a good point. Until now I haven't had the opportunity to play with a Tesla or observe it's charging operation.

I agree that a PV to DC charging pathway would be better. (probably via CHAdeMO to access othe BEVs) The only benefit with a 240V batteryless version would be that it would allow other 240V consumers. But they could come from the CT inverter instead.

I think I saw a charging cable device like that already actually. It would allow for extra PV charging in the bush as well. Time for some R&D! Thanks.

As for how Tesla will lockdown on household use I don't know. I think that is not an option anymore with competitors already allowing house connections. I think they'll have to add it. It's our storage so why can't we use it as we like? Powerwall isn't competitive anymore, so I think their strategy will be forced to change as competitors come to market with options their customers are looking for. Powerwall/SolarCity will go niche otherwise. They've already tipped the coolade with including a inverter in the CT, now they just have to bring it full circle and include a PV charging option to one up the rest. I'd be happy to see a 10kW DC PV input connector as a part of their inverter setup. That would do for most I think.

Also regarding using a CT as a battery, is this because of cycling? The load will be much lower than driving, and the cycle depths too. And you're still getting the CT thrown in for free, for the same capacity price of a Powerwall. What am I missing?

 

[ajdelange]

I agree that a PV to DC charging pathway would be better. (probably via CHAdeMO to access othe BEVs)

CHAdeMO is probably the worst choice as it is on the way out. All fast chargers work in more or less the same way: a regulated DC power supply. The differences in the standards are in how the communication between the vehicle and the charger takes place and what the format is. There are units one can buy that will fast charge CHAdeMO and CCS. Thus the comm protocols are in the public domain. A Tesla can be charged from a CHAdeMO (and, recently, CCS) station which means that there are people out that that have cracked the code. These people could build the device I am proposing.

The only benefit with a 240V batteryless version would be that it would allow other 240V consumers. But they could come from the CT inverter instead.

I was intentionally a little vague about other inputs besides solar. One can envision the proposed box as having AC input. The trend is towards panels with microinverters. These are almost necessary for residential use as it is rare to have a building with roof surfaces oriented in exactly the right direction and to be free of partial shading from trees, other parts of the building etc. If you have microinverters there is no DC interface. They tie into the building's 240V bus. Now this brings us to a critical question and that is the question of anti islanding. If this box is connected to the house 240 V bus then it must, when the grid goes down, isolate the bus from the utility. It must also be able, when the utility is down, be able to clock the microinverters and curtail them when the vehicle battery becomes full. Essentially we are talking the functions of a Powerwall here with the vehicle battery providing the storage and this proposed box the gateway functions. $

As for how Tesla will lockdown on household use I don't know. I think that is not an option anymore with competitors already allowing house connections. I think they'll have to add it.

They may or may not. Tesla knows that V2H/V2G for the Ford product is basically marketing hype. They know some will find the requisite charging device appealing because it allows charging at up to 19 kW (not, BTW, via DC as I just found out) and, of course, allowing their truck to replace the functionality of the Powerwall competes with the Powerwall.

It's our storage so why can't we use it as we like?

Because to do so will impact battery life which means Tesla will either have to reduce the battery warranty period or swallow more warranty claims.

Powerwall isn't competitive anymore,

If you know of something that competes with Powerwall please let me know! I can't even get an estimate as to when I will get my Powerwalls. I am ready to consider any alternative that will meet my requirements. Tesla has the market by the short and curlies and they know it. Yes, others are trying to catch up but as with the cars and trucks the "Tesla killer" has yet to appear. A company that is no longer competitive does not raise the price of its non-competing product. Tesla has recently raised the Powerwall price (and the prices of their cars).

so I think their strategy will be forced to change as competitors come to market with options their customers are looking for. [Powerwall/SolarCity will go niche otherwise.

That may all happen some day but that day is not today. They can't keep up with demand for PW.

They've already tipped the coolade with including a inverter in the CT,

There is a real need for that for working people. Tesla leads again by making it 240 whereas Rivian is limited to 120. Ford follows suit.

now they just have to bring it full circle and include a PV charging option to one up the rest. I'd be happy to see a 10kW DC PV input connector as a part of their inverter setup. That would do for most I think.

I doubt it will be 10 kW. Maybe a couple of kW. 10 kW is a lot of power and requires substantial cooling. A fairly large number of people would like to carry a couple kW of panels in the bed, set them up and charge at a campsite. But that fairly large number is going to be a small fraction of buyers. Lots of people would like to be able to charge from their solar systems at home but the very large preponderance of those will be grid tied so there really is only a tiny market for those who want to charge from solar with no grid connection. And most of those will have a Powerwall system or something like which will enable use with HPWC or UMC. So I don't see a market for a 10 kW solar interface. This is where a third party will have to step in. And it will be expensive.

Also regarding using a CT as a battery, is this because of cycling? The load will be much lower than driving, and the cycle depths too.

If I tried to do this at my house the cycle depths would be much greater than for driving. Driving is only 6% of my electrical load. So yes, it is because of the constant cycling which means that if I only drive it to church on Sunday that the battery will degrade much faster, on a per mile logged basis, than if I don't use the battery for load leveling/time shifting.

What am I missing?

Just a little more rumination.

 

[Manodamountain]

Assume about 2.5miles per kW per hour, but you're probably limited to what your 110V circuit is rated at, and whatever else is on that circuit. So maybe 5miles per hour max. It will take a while.

I'm offgrid with 30kW but that is 3phase 415V. Hopefully the CT will have some PV inputs that might help.

Sounds to me like you’ll be fine. You’ve got better than city power!

 

[Manodamountain]

You wouldn't be L3 charging but you could charge L2.

There are a couple of things to consider with a solar system vis-a-vis a BEV fueled by solar. The first is how many miles you can collect in a day. With a 6.5 kW system in a place that has 6 hrs (summer time) FSE (Full Sun Equivalent) per day you can collect 39 kWh/da. Assuming the CT to be conservatively operated it's probably going to require 450 Wh to move it one mile so the 39 kWh corresponds to 39/.450 = 86.7 miles per day. That is if you don't use electricity for anything else which I assume you do.

The next consideration is as to how fast you can charge the truck. This depends on your system battery. If you have 2 Powerwalls, the minimum Tesla will accept in an order for an off-grid system, they will discharge at a 10 kW rate (that's the current ones - the newer ones are supposed to do faster than that). This means that you can charge the truck at up to 10 kW (if there are no other loads on the system). That's 45 miles per hour. But note that 2 Powerwalls have capacity of 2*13.5 = 27 kWh (60 miles). This is only 12% of a TriMotor battery. They will not even hold an entire day's worth of collection (39 kWh) from your array. That may or may not be a problem. If the batteries are full and the sun is shining you put the truck on to charge and the energy from the array goes directly to the truck rather than to charge the Powerwalls. If you have set the truck charger to 10 kW and the array produces 6.5 kW then all the panels' production will go to the truck and the additional 3.5 kWh will come from the battery. If you have full Powerwalls from yesterday and you want to charge today (assumed sunny) then you can take the 39 kWh from the array and the 27 kWh from the Powerwalls for a toatal of 66 kWh. Note that this is only 33% of the capacity of the Tri Motor battery. Also note that you are using 2 day's sun to get this. As the sun is the only source you are limited to the 39 kWh (87 miles) that the sun can provide each day.

A Powerwall is a 240V device. This means you can use Level 2 chargers with it. If you have 3 Powerwalls their 3*13.5 = 40.5 kWh combined capacity is enough to hold a days production and their 3 inverters will produce 15 kW which is more than the 11.5 kW maximum charging rate of the truck. You could load a full day's production in 39/11.5 = 3.49 hrs but note again that a full day's production is only 18% of the TriMotor's battery. The fastest you can do with a Level 1 charger is 1440 W. To load a day's production into the CT at that rate would take 27 hours. You'd always be behind the curve.

What really drives all this is the size of your system relative to the number of miles you need on a daily basis. If you need more than 87 miles per day there is no way that you can make it on a 6.5 kW system. If, OTOH, you only go to town once a week even though it's 250 miles one way you can do it. Theoretically anyway (we haven't talked about conversion losses, round trip losses or phantom drain). I'm doing an off grid system with 5 Powerwalls and 13 kW (AC) of panels. It is, with the other loads, going to be very marginal. But it isn't in the boonies. It is right next to the main house which is grid tied, has more panels and is backed up by a generator.

I would think long and hard about the practicality of a CT with a solar system as small as 6.5 kW or indeed any solar system with out any backup. I would definitely have a generator as Tesla recommends.

I've tried to throw numerical example in here in the hope that you will be able to do your own calculations for your own particular circumstances.

Thank you so much for input. I really don’t leave very often and towns with tesla chargers are within an hour. I don’t have power walls but iron edison 300 ah batteries. But from your analysis I’m getting a better idea. I do have a 7k Honda backup that rarely gets used. I actually have to remember to fire it up to keep the gas fresh and start battery charged.

 

[JBee]

Sounds to me like you’ll be fine. You’ve got better than city power!

Yeah I am I know. I was saying that the CT should really have the option to plug PV in directly for off-grid use like yourself. Thats the options we've been trying to figure out here, on how best to go about it.

 

[Manodamountain]

Assume about 2.5miles per kW per hour, but you're probably limited to what your 110V circuit is rated at, and whatever else is on that circuit. So maybe 5miles per hour max. It will take a while.

I'm offgrid with 30kW but that is 3phase 415V. Hopefully the CT will have some PV inputs that might help.

I do have 240v capability and when sun is shining seems like i have tons of extra juice. I also plan on adding hydro power supplement for winter (if it ever rains in California again). From what you all are saying i get the feeling I could manage an electric vehicle. Maybe some additions and tweaks and excuses to head out to super chargers! Can’t wait for my first tinder date picnic at a super charger!

Sponsored