Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Powerwall 2.0 Backup Runtime Extender

This site may earn commission on affiliate links.
Keep in mind that the reviews include all the models on the page. It looks like quite a few of the reviews are for the larger 1000W split-phase output model. I actually like that one because it has a screen and a variable limiter that makes it easier to use with a battery pack or stepped up source like this application. However, it's nearly $300. If I was more serious about this project I would buy that one, or one of its clones that is also on Amazon.

Great observation. It actually looks like there are no reviews for the 600w model.

Unfortunately I have no way of using the split phase with my current setup so single phase is my only option.
 
Many thanks to miimura for starting this thread and bringing to my attention the possibilities.

I have unsuccessfully struggled to get it working with M215s. I have branched off and ordered two 500w 12vdc -> 240vac grid tie inverters. It'll be a few weeks before I can test. If I come across some solution to the MPPT problem, I'll revisit M215s.

My blog on the subject:
Willie: Micro Grid and augmentation
The first of the 12vdc -> 240vac grid tie inverters arrived yesterday. I tested overnight. Input: 14vdc and 250 watts; power was significantly lower at 12vdc (that is, with the EV DC-DC off). Output was 211 watts into 240vac. Disappointing efficiency especially compared to M215s. To make the project worthwhile, IMHO, it will take about 3 of those units at a cost of about $300. Disappointing cost especially compared to M215s.

Still in the works is testing a 36v battery in parallet with boosted 12v fed into M215s to overcome the MPPT problem apparently created by a 12vdc booster. I'll also explore starting M215s on 36v and then operating on 12v.
 
Take a look at this document

https://teslamotorsclub.com/tmc/att...esla-powerwall-2-0-en-us-2018v1-0-pdf.432747/

I'm not sure if you knew this or not, but the IQ6 and up models support curtailing when AC coupled. Meaning that the PW can control their output by changing the micro-grid frequency. This would allow you to leave the extender plugged in all the time and it would only produce when the PW signals it to do so. What I don't know is how the IQs would behave being powered off a battery instead of a module. But then again Enphases new Enlighten system is set up exactly like this.
 
Take a look at this document

https://teslamotorsclub.com/tmc/att...esla-powerwall-2-0-en-us-2018v1-0-pdf.432747/

I'm not sure if you knew this or not, but the IQ6 and up models support curtailing when AC coupled. Meaning that the PW can control their output by changing the micro-grid frequency. This would allow you to leave the extender plugged in all the time and it would only produce when the PW signals it to do so. What I don't know is how the IQs would behave being powered off a battery instead of a module. But then again Enphases new Enlighten system is set up exactly like this.

AFAIK, all grid tie inverters can be turned off by the PW through frequency shifting; M215s are an example.

For the purposes being discussed here, the PW is highly unlikely to want to turn off inverters; that mostly happens after the PW battery is fully charged and there is no place to use surplus PV power. That is, during the day. Here, we are trying to provide extra power to slow the discharge of the PW battery. Understood to be at night. During the day, you would want to be recharging your supplemental source.

In unusual conditions, the PW will turn off M215s whether panel or battery powered.

But yes, if someone left the supplemental power running for days on end, the PW would turn it off and on.
 
If you look at the document you'll see there is an option to use an Envoy along with CTs which will signal the Micros the amount of power being consumed which in turn tells the micros how much power to generate based on demand (I don't use that mode though, I let the micros always generate 100% output and store the excess into my 60kWh battery pack). If you had enough micros connected to a DC supply (be it a car, generator, hydro, wind, or more batteries) you could zero out all consumption from the PW itself, even use excess to recharge the PW then let freq shift turn them off. Although if you secondary source is batteries you'll have to figure out a way to recharge those as they'll empty out before your PW does and the micros won't backfeed.

What you're trying to accomplish is nothing new. AC coupling has been around for a while, I have it on the PW I built, Solar will power my house loads and any excess power will flow into the batteries instead. For example in the screenshot below, Solar is generating 14kWh, my AC is using around 4.5kWh and the rest is being sent to the battery.

https://i.imgur.com/Z5TZBHu.jpg

Here's the whole build thread.
Building my own pseudo(Powerwall)
 
For the purposes being discussed here, the PW is highly unlikely to want to turn off inverters; that mostly happens after the PW battery is fully charged and there is no place to use surplus PV power. That is, during the day. Here, we are trying to provide extra power to slow the discharge of the PW battery. Understood to be at night. During the day, you would want to be recharging your supplemental source.
This is exactly my thoughts on the application. Most people have relatively large solar compared to Powerwalls, so the Powerwalls fill early in the solar production day. If you don't increase your consumption by doing something like running A/C or charging an EV during the day, that energy will be wasted when the grid is down and the Powerwalls are full. So this "Runtime Extender" would charge during the day and discharge only at night and only during a prolonged grid outage.
 
This is exactly my thoughts on the application. Most people have relatively large solar compared to Powerwalls, so the Powerwalls fill early in the solar production day. If you don't increase your consumption by doing something like running A/C or charging an EV during the day, that energy will be wasted when the grid is down and the Powerwalls are full. So this "Runtime Extender" would charge during the day and discharge only at night and only during a prolonged grid outage.
Why not just add an additional PowerWall, and increase the reserve percentage? That get you the desired result, without the complexity of manually managing an extender device.
 
This is exactly my thoughts on the application. Most people have relatively large solar compared to Powerwalls, so the Powerwalls fill early in the solar production day. If you don't increase your consumption by doing something like running A/C or charging an EV during the day, that energy will be wasted when the grid is down and the Powerwalls are full. So this "Runtime Extender" would charge during the day and discharge only at night and only during a prolonged grid outage.

What you need is a power bank that charges from an AC source (aka from the solar or the grid during the day) and then outputs DC for your GT inverter. That in its own is not hard, use one of these
Yeti 3000 Lithium Portable Power Station w/ WiFi Control | Goal Zero
Connect the charger during the day. Then on the DC side (12v output) connect your GT inverter to AC couple with the PW during the night. A couple caveats, the charging cable CANNOT be connected to the PW output source while the 12v output is powering the AC coupled GT inverter as I'm pretty sure it'll fry the transformer inside, to avoid this use a contactor with aux connections.

Secondly and the most important part that you guys keep glossing over. GT inverters output at 100% ALL the time and that power has to go somewhere. This means that if you're running your extender at lets say 500W in the middle of the night it will recharge your PW if there isn't sufficient loads to absorb power being sent out by your extender. If this happens your PW will not be happy and most likely shutdown to prevent overcharging the internal battery and causing a fire.

To get around this frequency shift is used when AC coupling, most GT inverters will stop producing if the sine wave gets too far outside 60hz, this is called curtailment. Most of the large inverter manufacturers have special grid profiles that are configurable, for example on my Enphase envoy I can configure what percentage of power to drop the GT inverter output by, per each hz the micro-grid produced by the PW inverter goes up. This allows the PW to control extra power without drowning and having to shut itself off.

All that being said, Are those cheap Chinese GT inverters compliant to UL1741? If they are they should stop producing around 62.5hz.
 
Why not just add an additional PowerWall, and increase the reserve percentage? That get you the desired result, without the complexity of manually managing an extender device.

This.. Soo much this. If your PW is too small for your needs just get another one. I can guarantee you. If your place catches on fire the fire marshal will notice your out of code wiring. He'll then go check with the city to find a permit was never pulled (because Tesla doesn't support this) and they'll pass that information to your insurance company who will deny paying out a single cent to you.
 
  • Like
Reactions: Janus
This.. Soo much this. If your PW is too small for your needs just get another one. I can guarantee you. If your place catches on fire the fire marshal will notice your out of code wiring. He'll then go check with the city to find a permit was never pulled (because Tesla doesn't support this) and they'll pass that information to your insurance company who will deny paying out a single cent to you.

The answer is because I have a $30k piece of hardware that performs the exact same function as a large home generator and another Powerwall with my Volt and the risk of a catastrophic fire is close to 0.

My Powerwall setup is pure backup so my reserve is already 100%. With one Powerwall I'm good for a day of outage, more if the sun is shining. For your garden variety hours-long outages, I'm good. If we get a particularly bad snow storm here in the pacific northwest or an earthquake one day won't cut it and I won't be able to rely on solar during the winter months to ride me through.

My goal here is to come up with a $650 solution that will give me more piece of mind in either of these extreme circumstances. I don't want to ever have to use it but if something happens I'm going to be damn well happy that I have the option. Having another $6500 on a Powerwall sit idle in my backyard doesn't have the ROI.
 
Why not just add an additional PowerWall, and increase the reserve percentage? That get you the desired result, without the complexity of manually managing an extender device.
For under $500 I can tap the EV battery energy that I normally drive around every day and can keep my house running during an extended power outage. I have two Powerwalls, but my solar is only 4.32kW and in the Winter it generates as little as 5kWh on a sunny day in December. Obviously, rainy days will drag that down even lower.

Spending more money on another Powerwall or two is not economically reasonable for disaster preparation. This kind of Runtime Extender is reasonable for that purpose. Obviously, the first order of business after a disaster is to change behavior like turning off non-essential loads. Also, if a big storm is coming, I will raise the Reserve. The Powerwalls are only 27kWh and my cars are about 40 kWh and 75 kWh, so it makes sense to me to use some of that energy to keep my food from spoiling and my furnace running as needed.
 
  • Like
Reactions: mckemie and gpez
For under $500 I can tap the EV battery energy that I normally drive around every day and can keep my house running during an extended power outage. I have two Powerwalls, but my solar is only 4.32kW and in the Winter it generates as little as 5kWh on a sunny day in December. Obviously, rainy days will drag that down even lower.

Spending more money on another Powerwall or two is not economically reasonable for disaster preparation. This kind of Runtime Extender is reasonable for that purpose. Obviously, the first order of business after a disaster is to change behavior like turning off non-essential loads. Also, if a big storm is coming, I will raise the Reserve. The Powerwalls are only 27kWh and my cars are about 40 kWh and 75 kWh, so it makes sense to me to use some of that energy to keep my food from spoiling and my furnace running as needed.

Exactly.

An additional PW will cost at least $5k and will be used a very small fraction of the time. Once a year. Once a lifetime. Never. Preparation is my goal. Cost effective preparation.

We all have EVs that have large batteries compared to a PW. Cost for this supplemental power solution will be a few hundered dollars for likely ~500 watts.

In my case, as single PW would easily get me through most nights while off grid. The exception is the very hotest part of the year, currently a couple of weeks. It is easy to imagine that that couple of weeks might expand to a couple of months with increasing temperatures. I typically do not run off grid execpt to test things such as this supplemental power project.

With the low efficiency and relatively high cost of the cheap chinese GT inverter, I am revisiting the EV 12v -> boost -> M215. I got 120w last night. Adjustment of boosts are obscure and generally not documented. If I find something that goes 200+ watts with a M215, I will inform.
Or maybe someone has suggestions on better quality and documented boosts?

I have a successful soltuion:
36v golf cart batteries -> M215s
that gives me cost effective 400-600 watts.

A couple of years ago, I bought a bunch of 4ah 36v batteries on eBay for $20. That's about $140/kwh. They have served me very well in golf carts. I just shopped for more and found nothing less than $50.














;
 
With the low efficiency and relatively high cost of the cheap chinese GT inverter, I am revisiting the EV 12v -> boost -> M215. I got 120w last night. Adjustment of boosts are obscure and generally not documented. If I find something that goes 200+ watts with a M215, I will inform.

You'll have to excuse my ignorance but just to clarify "boost" here means a step up inverter to take 12v to 24v or beyond, correct?
 
@mckemie thanks for the update.

Ultimately I decided that the tech just isn't there at a price point that makes sense for me to go the grid tie route. I ended up grabbing a 1200w Giandel inverter and a quick disconnect cable (links below). The quick disconnect is similar to what the evextend site is selling but at less than half the price. Even though its not grid tie it will still accomplish the net goal of moving some of the load from the Powerwall during an extended outage - with it I can easily power my fridge, freezer, gas hot water heater, and furnace blower plus some lights while the Powerwall soaks up whatever sun is available. Additionally I could use it in other situations such as camping or if a neighbor needs it for a more localized outage (or if we have plenty of sun and they need some juice).

Rough Country 7 FT Quick Disconnect Winch Power Cable Compatible w/Any Standard Size Winch RS107 -https://www.amazon.com/gp/product/B07G79K5SF

Power Inverter Pure Sine Wave 1200Watt 12V DC to 110V 120V with Remote Control Dual AC Outlets and USB Port for CPAP RV Car Solar System Emergency - https://www.amazon.com/gp/product/B07G36Z2TH
 
A little more testing. Some more insight. A few conclusions. No revelations. Down at the bottom.

Willie: Micro Grid and augmentation

I've now done more than 5 over night tests. Maybe 10. So far, no hardware failures or fire threats.
The M215 spec sheet says the MPPT voltage range is 27 - 39 Volts. That's why it doesn't work with the 24V boost converter. So, I would recommend the same type of DC-DC boost converter I used, but the 36V version, if you want to go straight from the EV instead of the golf cart.

This eBay listing has 360W, 720W, and 1080W versions. 1080W is ~$130 shipped from China.
The same thing on Amazon is ~$190 but still has 3-6 weeks lead time, probably shipped from China too.
 
The M215 spec sheet says the MPPT voltage range is 27 - 39 Volts. That's why it doesn't work with the 24V boost converter. So, I would recommend the same type of DC-DC boost converter I used, but the 36V version, if you want to go straight from the EV instead of the golf cart.

This eBay listing has 360W, 720W, and 1080W versions. 1080W is ~$130 shipped from China.
The same thing on Amazon is ~$190 but still has 3-6 weeks lead time, probably shipped from China too.


Thanks, I just ordered a 1080w. gpez pointed me toward other, persumably higher quality boosts but I got sidelined when they tried to hit me with ~$60 shipping.

As you observe much identical stuff is sold on Amazon, eBay, and AliExpress. On the 12v grid ties, I ordered one on Ali, the other on eBay. The seller contacted me to inform that I had ordered two of the same at different prices; did I want to continue with both orders.

I'm been attempting to boost to 36v. The cheap boosts I've tried are sadly lacking in documentation. When I finally got some documentation, it did not match the boards well. So, my principle trouble may be that I'm flying blind trying to guess the use of the trimpots. Sometimes they seemed to work, boosting to ~36 volts. But, not with any stablity.

Last night's test was interesting. The two 12v grid ties going to 240vac. For the first time, I used a Tesla. When I would turn an imiev on, the voltage would go from 12v to 14v giving me confidence that the DC-DC was on. Last night, the Tesla battery was 12v but I pressed on knowing the Teslas are notorious for running down traction batteries with vampire loads. I figured the DC-DC was only doing 12v or would switch on as the lead battery became depleted. I ended up with a low 12v battery and the car complaining. The 12v grid tie inverters give ~200 watts with 14v but only about 130 watts with 12v. So, since I think I pulled more energy than the lead battery will hold, I concluded that the Tesla DC-DC was likely throttled to a few hundred watts. Tonight, I'll try "dog mode" or something to see if the DC-DC gives me more.

It turns out that this is the most stressing time for a PW. We still need air conditioning all night but the nights are longer than our traditional night time air conditioning season of July and August. We at about 17 September days of 100 deg. 20-25 days looks likely. Though it is not expensive to use grid power at night, I am currently doing the augmentation to try to avoid grid power.
 
Thanks, I just ordered a 1080w. gpez pointed me toward other, persumably higher quality boosts but I got sidelined when they tried to hit me with ~$60 shipping.

As you observe much identical stuff is sold on Amazon, eBay, and AliExpress. On the 12v grid ties, I ordered one on Ali, the other on eBay. The seller contacted me to inform that I had ordered two of the same at different prices; did I want to continue with both orders.

I'm been attempting to boost to 36v. The cheap boosts I've tried are sadly lacking in documentation. When I finally got some documentation, it did not match the boards well. So, my principle trouble may be that I'm flying blind trying to guess the use of the trimpots. Sometimes they seemed to work, boosting to ~36 volts. But, not with any stablity.

Last night's test was interesting. The two 12v grid ties going to 240vac. For the first time, I used a Tesla. When I would turn an imiev on, the voltage would go from 12v to 14v giving me confidence that the DC-DC was on. Last night, the Tesla battery was 12v but I pressed on knowing the Teslas are notorious for running down traction batteries with vampire loads. I figured the DC-DC was only doing 12v or would switch on as the lead battery became depleted. I ended up with a low 12v battery and the car complaining. The 12v grid tie inverters give ~200 watts with 14v but only about 130 watts with 12v. So, since I think I pulled more energy than the lead battery will hold, I concluded that the Tesla DC-DC was likely throttled to a few hundred watts. Tonight, I'll try "dog mode" or something to see if the DC-DC gives me more.

It turns out that this is the most stressing time for a PW. We still need air conditioning all night but the nights are longer than our traditional night time air conditioning season of July and August. We at about 17 September days of 100 deg. 20-25 days looks likely. Though it is not expensive to use grid power at night, I am currently doing the augmentation to try to avoid grid power.
The adjustable board-type CC-CV boost converters are a completely different architecture and they almost never have enough heat dissipation to do the wattage that they claim. I use those as lithium battery chargers at limited power, usually about half of their spec. In any case, I don't know which unit you've tried or if we're even talking about the same things.

Anyway, when you're looking at this kind of specialized equipment there are not that many that are truly different, they are just relabeling or reselling the same stuff. Here's the "same" 1080W 12->36V step up converter on AliExpress for less than $127 including slightly faster DHL shipping.
https://www.aliexpress.com/item/33023035544.html
 
The second night test of the Tesla (Model 3) was not successful. I ran the car in "Dog Mode" hoping that would allow the car to give up more DC-DC power. But, with a demand of less than 500 watts, the car's 12v voltage continued to decline prompting the car to eventually complain about low battery voltage. Perhaps the car is too smart to allow the DC-DC to produce more power than the car is using? It is not a problem with an imiev and probably other EVs.

I will research and test "camping mode" but I'm not optimistic. Suggestions solicited.