Victron is Over-priced Eurotrash. Why would anybody buy Victron over an AiO?

[cdsolar]

Did you have to use any non Victron products, or non listed Victron products to complete your installation?

Even something like their Lynx Distributor doesn't have ETL/UL listing of any sort that I'm aware of, which is a total bummer.

My panels are ETL listed, and I am using MIdnite breakers and boxes and busbars, so I am not using any non UL/ETL products.

 

[Samsonite801]

You download the OS image, put in a microSD card on your desktop then just burn the image to the microSD card and put it in the Pi and boot it then just connect to it via a web browser on your network at http://venus.local and you're on the remote console. It's pretty simple.

raspberrypi install venus image

Victron Energy Unix/Linux OS. Contribute to victronenergy/venus development by creating an account on GitHub.

github.com

After the fresh install, you can also just connect to the RPi Venus OS with Victron Connect over RPi bluetooth, and wham your in, can set IP from there and stuff as well...

 

[sunshine_eggo]

After the fresh install, you can also just connect to the RPi Venus OS with Victron Connect over RPi bluetooth, and wham your in, can set IP from there and stuff as well...

Yep... I just went through this for the first time... and used @Louisvdw / @BradCagle drivers to connect my JBD 4S BMS to it via BT. It's fucking awesome.

Now I have TWO VRMs!

Even did a discharge/charge test!



RPi on wifi and BMS connected via BT from across the room.

 

[cdsolar]

I think part of the Linux issue is the user base doesn’t want a Windows replacement OS. They want Linux to remain distinctly different and as “original” as possible. With innovations like one click installs, it becomes less and less “original”.

There are two basic install methods for Linux:

apt: package manager for Debian-based systems
rpm: package manager for Redhat based systems

It is up to the individual package creator how GUI-enabled the process is.

As for VenusOS, there is 'opkg'. (venusOS is an ' embedded' Linux that doesn't fit into the above pedigrees directly. But there us a third party installer called 'SetupHelper' which knows about a variety of packages specifically built for VenusOS and will install them on request, from the Internet as well as you adding ones it doesn't know about from gitlab/GitHub.

 

[ChrisG]

Yep... I just went through this for the first time... and used @Louisvdw / @BradCagle drivers to connect my JBD 4S BMS to it via BT. It's fucking awesome.

Now I have TWO VRMs!

Even did a discharge/charge test!

View attachment 166989

RPi on wifi and BMS connected via BT from across the room.

So the JBD is reporting SOC to VRM instead of a Victron shunt? Wait...What????

 

[sunshine_eggo]

So the JBD is reporting SOC to VRM instead of a Victron shunt? Wait...What????

Yep.

I have no shunt. I have an RPi and a JBD 4S BMS.

If I had a MPPT and/or inverter, the BMS would also pass charge voltage and current limitations to the GX attached chargers limiting both current and voltage.

That's how BMS comms work.

 

[ChrisG]

Yep.

I have no shunt. I have an RPi and a JBD 4S BMS.

If I had a MPPT and/or inverter, the BMS would also pass charge voltage and current limitations to the GX attached chargers limiting both current and voltage.

That's how BMS comms work.

I have a JBD and tried calibrating SOC using a Victron Shunt (Victron Shunt% ~= to this Voltage in JBD config). Never really was accurate. You trust the SOC of JBD? Need to look up this driver, sounds cool.

 

[sunshine_eggo]

I have a JBD and tried calibrating SOC using a Victron Shunt (Victron Shunt% ~= to this Voltage in JBD config). Never really was accurate.

Neither are. The Victron shunt SoC is much better understood and can tweak how it works. Essentially, you use a preponderance of the evidence for SoC.

You trust the SOC of JBD?

Mostly, but I trust the accuracy of my Batrium BMS over my BMV-702. They're typically ±2-3% from each other, and I sync them up manually every couple weeks.

Need to look up this driver, sounds cool.

Personally, I feel the BT by @BradCagle is experimental. I have discovered that if I try to edit anything in the config file, it pretty much breaks the driver.

I have purchased the UART to USB adapter for the JBD and plan to switch to the @Louisvdw driver once I have it.

The BT driver was to address my delay of gratification issues.

 

[cdsolar]

Yep.

I have no shunt. I have an RPi and a JBD 4S BMS.

If I had a MPPT and/or inverter, the BMS would also pass charge voltage and current limitations to the GX attached chargers limiting both current and voltage.

That's how BMS comms work.

Yes, I also do not have a shunt. Using canbus to connect my Rpi / VenusOS to my EG4 llv2 battery (also UL) this eliminating the need for the shunt.

 

[sunshine_eggo]

Yes, I also do not have a shunt. Using canbus to connect my Rpi / VenusOS to my EG4 llv2 battery (also UL) this eliminating the need for the shunt.

Yep. The only reason I have a BMV on my main system is because I had it before I had the Batrium.

 

[Samsonite801]

Yep... I just went through this for the first time... and used @Louisvdw / @BradCagle drivers to connect my JBD 4S BMS to it via BT. It's fucking awesome.

Now I have TWO VRMs!

Even did a discharge/charge test!

View attachment 166989

RPi on wifi and BMS connected via BT from across the room.

Yes this driver you mention, sounds interesting, where to find such driver download and install instructions?

I will have a Victron shunt on my permanent 48v system, but doesn't hurt to have those 3 JBD BMS's in VRM as well (to read individual BMS's SoC), to cross-compare, I love too much instrumentation (sometimes)...

My present temporary 12v RV system does also have Victron shunt, and 6 JBD BMS's, but I have to log into those separately via Xiaoxiang...

I do already have the USB to UART cables from Overkill here for the JBD's if I wanted to toy with the @Louisvdw driver.

 

[Samsonite801]

Neither are. The Victron shunt SoC is much better understood and can tweak how it works. Essentially, you use a preponderance of the evidence for SoC.



Mostly, but I trust the accuracy of my Batrium BMS over my BMV-702. They're typically ±2-3% from each other, and I sync them up manually every couple weeks.



Personally, I feel the BT by @BradCagle is experimental. I have discovered that if I try to edit anything in the config file, it pretty much breaks the driver.

I have purchased the UART to USB adapter for the JBD and plan to switch to the @Louisvdw driver once I have it.

The BT driver was to address my delay of gratification issues.

Do the wired or the wireless driver solutions allow for multiple JBD BMS's in parallel, to aggregate SoC or even can read the individual SoC numbers separately, or do they only work to read one JBD BMS?

 

[ChrisG]

Neither are. The Victron shunt SoC is much better understood and can tweak how it works. Essentially, you use a preponderance of the evidence for SoC.



Mostly, but I trust the accuracy of my Batrium BMS over my BMV-702. They're typically ±2-3% from each other, and I sync them up manually every couple weeks.



Personally, I feel the BT by @BradCagle is experimental. I have discovered that if I try to edit anything in the config file, it pretty much breaks the driver.

I have purchased the UART to USB adapter for the JBD and plan to switch to the @Louisvdw driver once I have it.

The BT driver was to address my delay of gratification issues.

Thanks, need to look this up. I have a few of the UART to USB adapters I bought from overkill a while ago, used one for a while with Solar Assistant. Need to dust them off and give them a shot with VenusOS. Just never got a warm and fuzzy feeling from JDB SOC no matter how I tweaked the voltage settings in the config.

 

[sunshine_eggo]

Louis' driver:

Victron VenusOS driver for serial connected BMS - JBD / Daly / ANT / JKBMS / Heltec / Renogy / Tian / ECS

I have created a VenusOS driver that can talk to BMS/Batteries that has serial communications (instead of the normal CANbus). RS485/RS232/TTL/UART and Venus 2.80+ have been tested. Currently it works with JBD BMS (LLT Power / Overkill Solar) Daly BMS (Daly Smart BMS / Daly Sinowealth based...

diysolarforum.com

Brad introduced his BT option here:

Victron VenusOS driver for serial connected BMS - JBD / Daly / ANT / JKBMS / Heltec / Renogy / Tian / ECS

Yes if there is a BMS it gets the first piority, but you can then switch that with the Battery Monitor option I have 6 JKBMS.... can I talk to them via bluetooth with the Cerbo device? Or must I use USB converters? And can I talk to all 6 of them? Will they show up like inthe list above?

diysolarforum.com

Louis addresses options for multiple BMS. I believe it requires completely separate drivers by other developers.

 

[Samsonite801]

Thanks, need to look this up. I have a few of the UART to USB adapters I bought from overkill a while ago, used one for a while with Solar Assistant. Need to dust them off and give them a shot with VenusOS. Just never got a warm and fuzzy feeling from JDB SOC no matter how I tweaked the voltage settings in the config.

When I log into my 6 JBD 4s BMS's with Xiaoxiang right now they all show 99%, while my Victron Shunt on the battery gateway shows 100%.

In mid-SoC range, they are about always 1% of each other. Doesn't Xiaoxiang have a calibrate SoC feature in there (I thought I saw something like that)? Also have to set the Ah correct for the cells.


EDIT:
I guess the Xiaoxiang has Calibrate Current, and Calibrate Voltage, not sure how to use those though...

 

[ChrisG]

When I log into my 6 JBD 4s BMS's with Xiaoxiang right now they all show 99%, while my Victron Shunt on the battery gateway shows 100%.

In mid-SoC range, they are about always 1% of each other. Doesn't Xiaoxiang have a calibrate SoC feature in there (I thought I saw something like that), also have to set the Ah correct for the cells.

When I log into my 6 JBD 4s BMS's with Xiaoxiang right now they all show 99%, while my Victron Shunt on the battery gateway shows 100%.

In mid-SoC range, they are about always 1% of each other. Doesn't Xiaoxiang have a calibrate SoC feature in there (I thought I saw something like that), also have to set the Ah correct for the cells.

Hmm. I thought it was based on these values in the config.

 

[Samsonite801]

Hmm. I thought it was based on these values in the config.

Yeah, I'm not like a guru on Xiaoxiang or anything, the only values I really changed on my JBD 4s BMS's is the Ah (mAh) to 280000, Total cycle capacity to 224000, and I turned off the 'Balance on charge only', and mine seem to read fine. I never did even try their calibration features, since it said only do this if you know what you're doing, and I don't know what I'm doing there hehe..

I may have tweaked some of the temperature triggers slightly, but nothing other than the above listed changes yet. I generally try to tweak minimally, and with caution usually (especially when I don't know what the hay I'm doing)..

 

[sunshine_eggo]

1% deviation is awesome. With regular full charges triggering sync criteria, it should stay pretty accurate. My issue is that my big NMC battery never gets to 100% as I'm cycling 20-75% SoC for ludicrous cycle life.

The JBD voltage to SoC numbers are just approximations. They are mostly used only when the BMS loses track of SoC due to power loss, sense cable disconnect, or a new install. In some cases when I've left the battery sit for several months, it will revert to showing a voltage based SoC.

Cycle capacity is not intended to be set to 80% of rated capacity. It is to be set for actual tested capacity. It's ultimately quite silly that they have both listed. It causes more confusion than anything. Cycle capacity is what's used for the SoC calculation, so the BMS will read 0% when you are actually at ~20%. Personally, I want MORE accuracy at lower states of charge rather than less. Last thing I want is for my BMS to show 0% when I know I'm around 20%... then I'll never truly know when I'm at 0%

The shunt capacity needs to be set to the same thing as BMS cycle capacity.

 

[cdsolar]

There are two basic install methods for Linux:

apt: package manager for Debian-based systems
rpm: package manager for Redhat based systems

It is up to the individual package creator how GUI-enabled the process is.

As for VenusOS, there is 'opkg'. (venusOS is an ' embedded' Linux that doesn't fit into the above pedigrees directly. But there us a third party installer called 'SetupHelper' which knows about a variety of packages specifically built for VenusOS and will install them on request, from the Internet as well as you adding ones it doesn't know about from gitlab/GitHub.

And before I forget, here is the link to SetupHelper for VenusOS:

GitHub - kwindrem/SetupHelper: Helper functions to simplify writing setup scripts that modify VenusOs functionality. The package includes automatic reinstallation of the package after a VenusOs update.

Helper functions to simplify writing setup scripts that modify VenusOs functionality. The package includes automatic reinstallation of the package after a VenusOs update. - GitHub - kwindrem/SetupH...

github.com

And here is a video showing how to get it up and running:

 

[Samsonite801]

1% deviation is awesome. With regular full charges triggering sync criteria, it should stay pretty accurate. My issue is that my big NMC battery never gets to 100% as I'm cycling 20-75% SoC for ludicrous cycle life.

The JBD voltage to SoC numbers are just approximations. They are mostly used only when the BMS loses track of SoC due to power loss, sense cable disconnect, or a new install. In some cases when I've left the battery sit for several months, it will revert to showing a voltage based SoC.

Cycle capacity is not intended to be set to 80% of rated capacity. It is to be set for actual tested capacity. It's ultimately quite silly that they have both listed. It causes more confusion than anything. Cycle capacity is what's used for the SoC calculation, so the BMS will read 0% when you are actually at ~20%. Personally, I want MORE accuracy at lower states of charge rather than less. Last thing I want is for my BMS to show 0% when I know I'm around 20%... then I'll never truly know when I'm at 0%

The shunt capacity needs to be set to the same thing as BMS cycle capacity.

Yeah, I don't even remember why I set the JBD total capacity lower, I'm thinking I was just following the original ratio I thought they had in there before, or perhaps I followed some other example seen somewhere.

I'll try setting them the same, honestly I hardly ever log into the JBD BMS's themselves since there's never much need to with the main shunt reading on the BMV-712 (and it is also the sum SoC of the entire bank), I believe I have the BMV set to show me 0% when the batteries are actually at 12% or something like that.

The lowest I've ever seen my BMV SoC get to is about 50% anyways. I generally try to keep it higher since my bottleneck right now is charging capacity (capped at 200a), it takes forever to get back to 100%, before I usually run the AC more than a little bit.