60kW Whole House Powerwall build

[LPG]

I'm a bit behind on updates to this thread but so far I've gotten my solar AC coupled and everything is working fine. Today being the first day after Dorian the sky was just about cloudless and I was able to test the max output of the solar going into the pack.





The inverter and pack were able to take about 12kW of AC (less on the pic due to AC->DC losses). However, at one point the AC solar jumped to 13kW and the inverter took it fine, but it did light up the overload LED, no alarm went off though (I think the manual states that once the audible alarm is heard then the inverter is not happy).

Either way I solved that issue by just turning down the AC, I figured may as well use the power to cool the house rather than waste it by turning off a branch of micros. I plan on setting up an API connection to the ecobee and as a secondary in case internet is out the water heater using a contactor and MQTT. Talk about first world problems because I'm generating too much power.

I've also gotten a bit further on my custom control dashboard where I can control charging, take the house off the grid, etc. I have run into an issue that the Rpi I was using for this doesn't have enough resources so it looks like I'll be moving it to a VM.

 

[WuggyBuggy]

Beautiful work man! Very very nice!

 

[LPG]

I've been debating adding some cooling to the pack since it's getting above where I want it. Still within range but I'd rather cool it but I can't seem to find any info on how many gpm I need in order to move fluid through all the modules. I guess I should go look in some EV conversion forums.

 

[swoozle]

Very, very nice. Thank you for the excellent write-up. It is inspiring and will be put to use soon. My plans are similar; the use of Volt batteries instead of Teslaisthe major difference. I received my Sigineer 12kW inverter straight from China last week and found your write-up this week.
In particular your use of the transfer switch and hacking the charging dial switch on the inverter is exactly what I was looking for.
Thanks!

 

[LPG]

Very, very nice. Thank you for the excellent write-up. It is inspiring and will be put to use soon. My plans are similar; the use of Volt batteries instead of Teslaisthe major difference. I received my Sigineer 12kW inverter straight from China last week and found your write-up this week.
In particular your use of the transfer switch and hacking the charging dial switch on the inverter is exactly what I was looking for.
Thanks!

Here's a breakout of the rotary switch. To enable charging on the lithium profile, connect 1 and 8 together into one of the common pins. For me I use a relay in the C pin of the relay I connected a common pin from the switch. On the NC side of the relay I connected 1 and 8 together. This makes it so that when the relay energizes both 1 and 8 get connected to common.






On a side note, keep in mind you don't need the ATS panel I installed. The sigineer has an ATS built in. I simply wanted a way to bypass the inverter for maintenance or if something failed to not end up without power.

 

[LPG]

Yesterday I pointed my thermal camera at the battery cables while my solar system was at full output and I noticed some disturbing temps. The cable I'm using is EXRAD XLE which is rated to 390A the cable itself while hot is still within NEC allowed temps.





Then I moved over to the current sensor and I think I may have to find a way to make a better connection here.





You can see where the connection between the cable and the current sensor is causing the whole cable to heat up.






The funny thing is that the connection to the module is nowhere near that hot (there's a lot of reflection on the picture below but you can see the connection that goes to module is blue/purpleish). The positive cables are also not hot either.






The heat has turned the lug black






For a comparison, here's the positive side.

 

[PAF]

is there a manual with a guideline how you should connect / mount the shunt?
a Quick and Dirty solution would be to add a metal plate to one side of the Shunt and use it as a Heatsink.
but i dont understand why the Lug is now black. maybe the connection itsealf is not good.

 

[Korishan]

Usually a hot connection like that is due to a loose or poor connection. Perhaps take the lug back off, hit the contact points with a brass wire brush (either manual or rotary) to get it nice and clean. Then use some alcohol to make sure there is no residue on the surfaces. Tighten the bolts down nice and tight.

I would not put a heatsink on the shunt. This could throw off it's measurements. However, if the shunt is getting hot, perhaps you need to upgrade the shunt to a larger rated one. I haven't read everything here, but depending on the BMS you're using, you might be able to put two shunts in parallel and make the adjustments needed in software to compensate. Two shunts in parallel will report 1/2 the amps used. So the calibration would need to be double the values read. But I'd say replace with a larger one, less work/hassle/chance of goofup.

 

[LPG]

There is a datasheet but it just states that when connecting to a busbar you should make a connection to the whole plate.

https://www.isabellenhuette.de/fileadmin/Daten/Praezisionsmesstechnik/Datasheet_IVT-S.pdf

I may temporarily remove the shunt and connect directly to the bus bar to see if the temp gets that high again, if they don't then it's on the sensor, if they do then it could be the lug itself.

---

Usually a hot connection like that is due to a loose or poor connection. Perhaps take the lug back off, hit the contact points with a brass wire brush (either manual or rotary) to get it nice and clean. Then use some alcohol to make sure there is no residue on the surfaces. Tighten the bolts down nice and tight.

I would not put a heatsink on the shunt. This could throw off it's measurements. However, if the shunt is getting hot, perhaps you need to upgrade the shunt to a larger rated one. I haven't read everything here, but depending on the BMS you're using, you might be able to put two shunts in parallel and make the adjustments needed in software to compensate. Two shunts in parallel will report 1/2 the amps used. So the calibration would need to be double the values read. But I'd say replace with a larger one, less work/hassle/chance of goofup.

Cleaning is my next step. Before removing the shunt to see if there is a difference in temps.

This shunt should be able to handle the power I'm putting through it, it's rated at 1000A and at the absolute max (which isn't very often) I'm putting 300A through it.

 

[Korishan]

Cleaning is my next step. Before removing the shunt to see if there is a difference in temps.

This shunt should be able to handle the power I'm putting through it, it's rated at 1000A and at the absolute max (which isn't very often) I'm putting 300A through it.

Sounds good.

It's possible the shunt could be hot from the connection, then. Which is definitely more desirable than the alternatives. It's cheaper
At 300A, that thing should be cold to the touch. So yeah, I would guess the connection is causing the issue.

Not sure if you saw Petes video when he checked his connections with his Flir, but he found a loose bolt using it. Even by touching, he couldn't tell much of a difference between that connection and the others. But the Flir caught it clearly. I think he got about a 1/4 turn when tightening. So it doesn't take much to cause heat build up through resistance at those amp levels

 

[LPG]

Cleaning is my next step. Before removing the shunt to see if there is a difference in temps.

This shunt should be able to handle the power I'm putting through it, it's rated at 1000A and at the absolute max (which isn't very often) I'm putting 300A through it.

Sounds good.

It's possible the shunt could be hot from the connection, then. Which is definitely more desirable than the alternatives. It's cheaper
At 300A, that thing should be cold to the touch. So yeah, I would guess the connection is causing the issue.

Not sure if you saw Petes video when he checked his connections with his Flir, but he found a loose bolt using it. Even by touching, he couldn't tell much of a difference between that connection and the others. But the Flir caught it clearly. I think he got about a 1/4 turn when tightening. So it doesn't take much to cause heat build up through resistance at those amp levels

Do you have a link to the video?

 

[Korishan]

Not sure if you saw Petes video when he checked his connections with his Flir, but he found a loose bolt using it. Even by touching, he couldn't tell much of a difference between that connection and the others. But the Flir caught it clearly. I think he got about a 1/4 turn when tightening. So it doesn't take much to cause heat build up through resistance at those amp levels

Do you have a link to the video?

I was gonna say no, but I figured I'd scrub his videos and find it for ya:

Timestamp location goes directly to that part. The camera detected 76C on the one connection, and only 50C on the other ones.
(Even tho he says 74, the reading said 76 at one spot)
He states in this video about the crimp not being tight enough. I think there was another video where it was the bolt needed tightening. But still, it was a loose connection "on" the lug itself, whether it was wire or bolt. That's a lotta heat generation from loose connection.

 

[LPG]

Not sure if you saw Petes video when he checked his connections with his Flir, but he found a loose bolt using it. Even by touching, he couldn't tell much of a difference between that connection and the others. But the Flir caught it clearly. I think he got about a 1/4 turn when tightening. So it doesn't take much to cause heat build up through resistance at those amp levels

Do you have a link to the video?

I was gonna say no, but I figured I'd scrub his videos and find it for ya:

Timestamp location goes directly to that part. The camera detected 76C on the one connection, and only 50C on the other ones.
(Even tho he says 74, the reading said 76 at one spot)
He states in this video about the crimp not being tight enough. I think there was another video where it was the bolt needed tightening. But still, it was a loose connection "on" the lug itself, whether it was wire or bolt. That's a lotta heat generation from loose connection.

Thanks for the link. l put the house on bypass and removed the connection to clean it. I did find the bolt wasn't tight enough to my liking so that could have been it, also while I was at it I cleaned all the black off the lug and left it at bare copper, cleaned both surfaces with rubbing alcoholand tightened them backup. While I was at it I also checked my busbar connections and made them a bit tighter since most of the negatives seemed a bit loose.

Here are the results:





The connection to the sensor is now cold as it should be BUT I have found another issue (which funny enough is the same as Pete mentioned in that video). The lug crimp seems to not have been done to par as it is now the hot spot. That's what I get for using the BMS cables the way they came, I'll have to cut off the current lug and re-crimp them properly.

 

[Korishan]

Now you know what to look for

Glad you found out what was going on. Sometimes it takes several steps to get to the root of the problem. It's just a pita to have to keep disassembling/reassembling to find the problem
But progress is made, and your system will be much more efficient because of it.

 

[LPG]

FYI for anyone wanting to recreate my setup. The owner and developer of the EVTV controller I use, has passed away. I'm not sure if anyone is going to continue running the business but at this point I'm considering the controller software to be abadonware with no more updates being available.

If you're going to buy a controller I'd highly recommend first checking if the company is going to stay in business without Jack Rickard or go with the SimpBMS instead.

 

[wifi442]

FYI for anyone wanting to recreate my setup. The owner and developer of the EVTV controller I use, has passed away. I'm not sure if anyone is going to continue running the business but at this point I'm considering the controller software to be abadonware with no more updates being available.

If you're going to buy a controller I'd highly recommend first checking if the company is going to stay in business without Jack Rickard or go with the SimpBMS instead.

Hello LPG just cruised over your thread. Nicely done! I would love to do this to my home some day!I have the same BMS in my motorhome. I am running 5 Modules @ 22v nominal with 4800 watts of solar. The EVTV BMS has been great for the last year that I have used it. I had the same issues with the heat on the shunt as well. I also attributed it to the crimps that came from EVTV. I would take a look at the temps inside the BMS as well. Mine were even hotter in there than at the shunt. I ended up replacing their 2/0 cable with 4/0 cable while I was in there. I also must have ignored the warning a few times about not turningon the BMS while the Laptop USB is plugged in (bypassing the pre charge resistor)as I damaged my contactors (Wouldn't open,kept getting the "Welded" messages). It could have been the heating but i'm guessing it was my error there.Emails to EVTV a few weeks ago were unanswered (not surprising) so I went straight to Gigavac to replace them, that's on me though. Definitelycheck those terminal crimps inside on the contactors!



The community lost a great asset when Jack passed away. He always emailed back promptly during my build.His lengthy videos were always very helpful. Hopefully someone takes over as this is a great product.RIP

 

[Redpacket]

Does that connection point have two holes there like the lower end? would dual cables be an option?

Also check the crimp is good - these need to be done with a high pressure hydraulic crimper.
Wire stands inside should be neat & parallel not crossed or tangled.
You might need to find a lug with a longer "cylinder" where in contact with the cable.
Might also help to have a washer under the bolt to keep the lug flat not "curving up" from bolt pressure?

 

[mobileclimate]

FYI for anyone wanting to recreate my setup. The owner and developer of the EVTV controller I use, has passed away. I'm not sure if anyone is going to continue running the business but at this point I'm considering the controller software to be abadonware with no more updates being available.

If you're going to buy a controller I'd highly recommend first checking if the company is going to stay in business without Jack Rickard or go with the SimpBMS instead.

Hello LPG just cruised over your thread. Nicely done! I would love to do this to my home some day!I have the same BMS in my motorhome. I am running 5 Modules @ 22v nominal with 4800 watts of solar. The EVTV BMS has been great for the last year that I have used it. I had the same issues with the heat on the shunt as well. I also attributed it to the crimps that came from EVTV. I would take a look at the temps inside the BMS as well. Mine were even hotter in there than at the shunt. I ended up replacing their 2/0 cable with 4/0 cable while I was in there. I also must have ignored the warning a few times about not turningon the BMS while the Laptop USB is plugged in (bypassing the pre charge resistor)as I damaged my contactors (Wouldn't open,kept getting the "Welded" messages). It could have been the heating but i'm guessing it was my error there.Emails to EVTV a few weeks ago were unanswered (not surprising) so I went straight to Gigavac to replace them, that's on me though. Definitelycheck those terminal crimps inside on the contactors!



The community lost a great asset when Jack passed away. He always emailed back promptly during my build.His lengthy videos were always very helpful. Hopefully someone takes over as this is a great product.RIP






I did not know the usb was causing the welded contactors thank you for posting that I thought that the contactors were getting welded when the controller shuts off the inverter while loadedI had the same problemwith the welded contacts I finally read a Gigavac manual which stated to cycle the contactor 10 -20 times to break the weld and that has been working for me. Its sad that jack wont get to see FSD because I know he thought that Elon was wasting time with that and itwasn't going happen but he thought wrong

 

[dougbert]

FYI for anyone wanting to recreate my setup. The owner and developer of the EVTV controller I use, has passed away. I'm not sure if anyone is going to continue running the business but at this point I'm considering the controller software to be abadonware with no more updates being available.

If you're going to buy a controller I'd highly recommend first checking if the company is going to stay in business without Jack Rickard or go with the SimpBMS instead.

I have asked for and received some beta firmware from the firmware owner in response to some issues.
he is maintaining it

one issue I see is information flow for existing customer from EVTV is somewhat lacking