• Have you tried out dark mode?! Scroll to the bottom of any page to find a sun or moon icon to turn dark mode on or off!

diy solar

diy solar

Solar assistant: do you use it?

Do you use solar assistant with your solar system

  • Yes

  • I used to, not anymore

  • No but I plan to use it one day

  • No and I don't see a reason to purchase it

Results are only viewable after voting.
I heard its a good idea to add one by one inverter into the ports.
To get the first one registered before the next etc.
Don't just plunk all the cables in willy-nilly.
 
Hagger said:
Just keep in mind that there are up to 21V (RS232 spec). /15V at Deye as far is I know on the serial buss. Means if you use one of the cheap PLxxx or TLxxxx CH… (usb) converter you will fry them sooner or later if not even the USB port of your Raspis at all. I went all that way and fry serveral prior to buy one of the so called “industrial” converter from Wafeshare on Ali. (Full isolated) They are at 20 bucks or so but worth every penny.

If you run multiple inverters in parallel and use a USB hub for all the connection, keep an eye on the “boot sequence” (power on) of the respective USB ports. That sequence will define the position of your inverter in the user interface in Solar assistance. That tip alone is worth 30 hours you don’t need to spend on debugging for a bug that is no bug… ;-( LOL

View attachment 287672
Click to expand...
Use waveshare exclusively , both for rs485 and rs232, also at work
Never let me down so far
 
Hagger said:
Boot sequence means the physically start up sequence of the individual USB ports on the USB hub.
Click to expand...

I think what you're talking about is a long standing thing with Linux in general, which _should_ be solved by now!

If you have e.g. three identical USB devices (like your Waveshares), then the Linux kernel enumerates them in the order it happens to see them. e.g. /dev/ttyUSB0 ,USB1, USB2 or /dev/ttyACM0, 1, 2

That order depends, to some extent, on randomness if you power them all up at once, which means that the device on the end of USB0 is "whatever I found first" which could easily cause your inverter monitoring to show up in a wrong order.

Or, for example, if the first or middle one is disconnected, cause them to shuffle down one place (0,1,2 doesn't become 0,2 with 1 missing, it becomes 0,2-but-called-1, and no 2).

One solution to this (used on Slackware and other OSs) is udev and "persistent rules", where you can nail down a particular device to be PERMANENTLY USB0,1,2. It's also used to stop embarrassing problems with network interfaces swapping around, removable drives causing a re-assignment of drives.

I don't know if Raspbian/Debian at the heart of SA has this enabled. No doubt similar tools exist under systemd controlled systems.
 
fmeili1 said:
In my opinion, integrating a solar system (via SA) into a smart home system opens up a whole new world of possibilities and automation!

The "Automation" features in SA are very basic compared to automation possibilities when integrating it into a smart home system. If you already using a smart home system and already integrated a lot of typical "things" you have endless possibilities... here are some possible examples:
  • reduce house load if batteries are below a specific SOC (and not enough PV power) by changing the setpoint temperatures of heat pumps OR disconnect an electrical water heater in case of low SOC (via remote relay/contactor) OR stop the pool pump... etc. etc.
  • if using grid as a backup in an off-grid installation, you may adjust the rules depending on a TOU plan, current SOC, weather/solar forecast, etc.
  • you may pre-charge your batteries from grid (if required) before an on-peak TOU time frame begins to not using the grid at on-peak time.
  • you may switch OFF/standby some inverters (if you have more in parallel) in case of lower load time ranges (e.g. at night) to save some idle/self-consume energy of the inverter
  • you may pre-charge the batteries fast (even with grid support) before a storm will arrive (via smart home weather binding).
  • you may transfer all temperatures of all inverters and batteries to the smart home to do e.g. reduce the charging amps of the inverters or to do alerting via messenger to your smart home, etc.
  • ... and hundreds of other ideas...
Click to expand...
I'm using SA with HA via MQTT.
I've implemented several features.
  • Since my inverter is offline, I set up an automation in HA to be my alerting system for grid down (and up). I get notifications on the HA app on my phone.
  • I check for power usage during "Peak" TOU hours. Generally, I should have next to zero grid consumption during peak hours, so this alerts me if something is wrong and I could potentially reduce consumption.
  • I have a "Storm Mode" toggle. Generally, I charge my batteries during Super Off Peak (0.5x) in the middle of the night, and then run off of them during the day. But Off Peak (0.75x) extends all the way until 2PM, so if a storm is coming, I can flip a single toggle in HA to change the TOU settings on the inverter to keep the batteries charged at 100% during everything but Peak to have the fullest batteries in case of a grid failure. I haven't automated this with forecasts yet, but maybe I will in the future.
 
NICE. I look forward to having a dashboard like that somedoy.

Now... do you have a temperature probe in your freezer or do you have a weird climactic thing going on?
 
JaimeZX said:
NICE. I look forward to having a dashboard like that somedoy.

Now... do you have a temperature probe in your freezer or do you have a weird climactic thing going on?
Click to expand...
You guessed it, that's the freezer sensor. I have found that using cheap Accurite wireless sensors and rtl_433 works best for freezers. The batteries last a long time.
 
just bought a SA 2.4ghz, should be showing up tomorrow!
my question is how do i hook up my victron smart shunt to it? i assume VE direct cable, but is it a straight plug in to SA or do i need some sort of converter?
thanks!
 
Hokay...I am now one of the clan. Got it set up today.
After a bit of fiddling got the inverters correctly numbered so I remember who is who (3).
Also set an individual IP instead of just DHCP.

So no more WiFi dongles for me. Security FTW!
The PI5 is ethernet cable connected aswell.
 
I have it running on an Intel NUC (well, Home Assistant anyway). I haven't done much with it yet though, just fooled around with the UI.

My plan is to use ESPHome to bridge all my VE.Direct stuff into it.

I also want to try making my own smart shunts with a cheap non-smart shunt, INA226 module, and an ESP8266. I think you could be all in on a like 300a smart shunt for maybe $20 bucks that way.
 
Lee_Adamson said:
I have it running on an Intel NUC (well, Home Assistant anyway). I haven't done much with it yet though, just fooled around with the UI.

My plan is to use ESPHome to bridge all my VE.Direct stuff into it.

I also want to try making my own smart shunts with a cheap non-smart shunt, INA226 module, and an ESP8266. I think you could be all in on a like 300a smart shunt for maybe $20 bucks that way.
Click to expand...
I went with cerbo gx's to integrate my victron equipment into home assistant. The cergo gx will send everything it knows about to the mqtt server in home assistant making it a breeze to add stuff.
 
I use SA. Their support haven been stellar for the price I paid. They always have a solution.
 
Hagger said:
Boot sequence means the physically start up sequence of the individual USB ports on the USB hub. As same as most of us I was on the impression that it simply doesn’t matter. However, it matters as the first USB port (in such a Hub) will define the first inverter, the 2nd the 2nd inverter. …..

I went with a little more costly hub that has a separate power supply as well as indication lights on each port. You will be wrong when you think that the hub will follow the logical ordert to power on in a sequence from 1-8… . In my case (! Other may different) it power on the 5th port 1st, than the 1st port etc.
if you take a video with your phone you can reply it slowly and see the sequence. Voila ! Your problem is solved.

Plug all cables in , apply a fresh restart of your Raspi and check once again that it remain the same. Once you sure you have the right order of your inverter in SA as well as the subsequencial data via Mqtt in HA or other automation software.

View attachment 287689
Click to expand...

fmeili1 said:
That's a very helpful information for me! Thanks a lot!

Luckily I have an external powered USB HUB (with individual switches and LED's per port) and this port is powered from my existing "blackstart UPS" which I have anyway to black-start my solar system. As a result the external USB HUB will never loose power and as far as I understand it will never loose the order because of this :)
I never got the idea that the power on order of the USB adapters are the key to solve this problem!
Click to expand...
I've tried multiple variants with my external powered USB HUB (with ON/OFF switches for each USB socket) to change the SA order of my AIOs:
  1. USB HUB switch ON order:
    1. click Disconnect in SA
    2. switch on the USB socket for all six AIOs one by one from 1 to 6
    3. click Connect in SA
    4. no success! The old order persists
  2. Shutdown SA (power off)
    1. switch on the USB socket for all six AIOs one by one from 1 to 6
    2. reboot SA
    3. no success! The old order persists
  3. Reorder the USB plugs
    1. using the persistent order in SA and replug the physical AIOs in the order of SA
    2. no success! The old order persists !!!
With my current USB HUB it looks like I can't change the order with the procedure which worked for @Hagger.

It looks like that each USB plug of the RS232-to-USB adapter has a fix address which automatically results in a fixed order in Linux (Raspberry Pi OS).

I'll try to find if I can persist the USB order in Rasperry Pi OS like @SolarUKWM mentioned in his comment above.
 
Digging deeper into this RPi/SA USB order adventure... but still a bit confused.

Ok, logged in via SSH into SA and done the following:

Code: 
$ lsusb
...
Bus 001 Device 023: ID 067b:23a3 Prolific Technology, Inc. ATEN Serial Bridge
Bus 001 Device 024: ID 067b:23a3 Prolific Technology, Inc. ATEN Serial Bridge
Bus 001 Device 025: ID 067b:23a3 Prolific Technology, Inc. ATEN Serial Bridge
Bus 001 Device 026: ID 067b:23a3 Prolific Technology, Inc. ATEN Serial Bridge
Bus 001 Device 027: ID 067b:23a3 Prolific Technology, Inc. ATEN Serial Bridge
Bus 001 Device 028: ID 067b:23a3 Prolific Technology, Inc. ATEN Serial Bridge
Bus 001 Device 011: ID 1a86:7523 QinHeng Electronics CH340 serial converter
...

The first 6 devices (023-028) are the RS232-to-USB adapters for my six AIOs and the last one (Device 011) is the RS485-to-USB adapter connected to all 12 batteries (all in parallel via two interconnected RS485 passive HUB's).

Code: 
$ ls -al /dev/serial/by-id/
lrwxrwxrwx 1 root root  13 Apr 22 13:48 usb-1a86_USB2.0-Ser_-if00-port0 -> ../../ttyUSB5
lrwxrwxrwx 1 root root  13 Apr 22 14:02 usb-Prolific_Technology_Inc._USB-Serial_Controller_A9CAb116L16-if00-port0 -> ../../ttyUSB3
lrwxrwxrwx 1 root root  13 Apr 22 14:02 usb-Prolific_Technology_Inc._USB-Serial_Controller_AKBSb119D15-if00-port0 -> ../../ttyUSB4
lrwxrwxrwx 1 root root  13 Apr 22 14:02 usb-Prolific_Technology_Inc._USB-Serial_Controller_DEBCb119D15-if00-port0 -> ../../ttyUSB1
lrwxrwxrwx 1 root root  13 Apr 22 14:02 usb-Prolific_Technology_Inc._USB-Serial_Controller_EBAQb116L16-if00-port0 -> ../../ttyUSB6
lrwxrwxrwx 1 root root  13 Apr 22 14:02 usb-Prolific_Technology_Inc._USB-Serial_Controller_EIBSb119D15-if00-port0 -> ../../ttyUSB0
lrwxrwxrwx 1 root root  13 Apr 22 14:02 usb-Prolific_Technology_Inc._USB-Serial_Controller_EOCMb119D15-if00-port0 -> ../../ttyUSB2

This shows 6 different "USB addresses" like "A9CAb116L16, etc." (don't know what these addresses are) mapped to the specific tty's. I can't identify the physical USB adapter with theses addresses (there are no labels).

I've tried to get more info via dmesg:
Code: 
$ dmesg | grep tty
[    5.589046] usb 1-1.2.2.2: pl2303 converter now attached to ttyUSB0
[    5.593318] usb 1-1.2.2.3: pl2303 converter now attached to ttyUSB1
[    5.595156] usb 1-1.2.2.1.4: pl2303 converter now attached to ttyUSB2
[    5.623331] usb 1-1.2.2.4: pl2303 converter now attached to ttyUSB3
[    5.650340] usb 1-1.2.2.1.3: pl2303 converter now attached to ttyUSB4
[    6.301978] usb 1-1.2.2.1.1.1: ch341-uart converter now attached to ttyUSB5
[    6.622924] usb 1-1.2.2.1.2: pl2303 converter now attached to ttyUSB6

and the corresponding device tree:
Code: 
$ ls /sys/bus/usb-serial/devices/ -ltrah
total 0
lrwxrwxrwx 1 root root 0 Apr 22 13:48 ttyUSB5 -> ../../../devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb1/1-1/1-1.2/1-1.2.2/1-1.2.2.1/1-1.2.2.1.1/1-1.2.2.1.1.1/1-1.2.2.1.1.1:1.0/ttyUSB5
lrwxrwxrwx 1 root root 0 Apr 22 14:40 ttyUSB6 -> ../../../devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb1/1-1/1-1.2/1-1.2.2/1-1.2.2.1/1-1.2.2.1.2/1-1.2.2.1.2:1.0/ttyUSB6
lrwxrwxrwx 1 root root 0 Apr 22 14:40 ttyUSB4 -> ../../../devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb1/1-1/1-1.2/1-1.2.2/1-1.2.2.1/1-1.2.2.1.3/1-1.2.2.1.3:1.0/ttyUSB4
lrwxrwxrwx 1 root root 0 Apr 22 14:40 ttyUSB3 -> ../../../devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb1/1-1/1-1.2/1-1.2.2/1-1.2.2.1/1-1.2.2.1.4/1-1.2.2.1.4:1.0/ttyUSB3
lrwxrwxrwx 1 root root 0 Apr 22 14:40 ttyUSB2 -> ../../../devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb1/1-1/1-1.2/1-1.2.2/1-1.2.2.2/1-1.2.2.2:1.0/ttyUSB2
lrwxrwxrwx 1 root root 0 Apr 22 14:40 ttyUSB1 -> ../../../devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb1/1-1/1-1.2/1-1.2.2/1-1.2.2.3/1-1.2.2.3:1.0/ttyUSB1
lrwxrwxrwx 1 root root 0 Apr 22 14:40 ttyUSB0 -> ../../../devices/platform/scb/fd500000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/usb1/1-1/1-1.2/1-1.2.2/1-1.2.2.4/1-1.2.2.4:1.0/ttyUSB0

But here is a discrepancy! If you look e.g. to the device "1-1.2.2.4" (last line of the last command) which is mapped to ttyUSB0 in this output is mapped to ttyUSB3 in the dmesg output (previous command line 4) !

I don't understand why the mapping is different and which is the valid one. If I know the correct mapping, I think I could do this manual (persistent) udev mapping via a /etc/udev/rules.d/ file.

Also, even if this will work - how would it be possible to make these changes persistent in SA? As far as I remember, SA will always reset to it's initial image after each reboot?!
 
I've tried different variants to assign the SA inverter number to a specific RS232-USB adapter but without success.

Some findings:

Every time when SA is rebooted, the assignment from a physical RS232-USB adapter to the ttyUSBx changes.

e.g. before reboot:
Code: 
ls -al /dev/serial/by-id/

lrwxrwxrwx 1 root root  13 Apr 23 13:27 usb-Prolific_Technology_Inc._USB-Serial_Controller_A9CAb116L16-if00-port0 -> ../../ttyUSB3
lrwxrwxrwx 1 root root  13 Apr 23 13:27 usb-Prolific_Technology_Inc._USB-Serial_Controller_AKBSb119D15-if00-port0 -> ../../ttyUSB2
lrwxrwxrwx 1 root root  13 Apr 23 13:27 usb-Prolific_Technology_Inc._USB-Serial_Controller_DEBCb119D15-if00-port0 -> ../../ttyUSB5
lrwxrwxrwx 1 root root  13 Apr 23 13:27 usb-Prolific_Technology_Inc._USB-Serial_Controller_EBAQb116L16-if00-port0 -> ../../ttyUSB1
lrwxrwxrwx 1 root root  13 Apr 23 13:27 usb-Prolific_Technology_Inc._USB-Serial_Controller_EIBSb119D15-if00-port0 -> ../../ttyUSB6
lrwxrwxrwx 1 root root  13 Apr 23 13:27 usb-Prolific_Technology_Inc._USB-Serial_Controller_EOCMb119D15-if00-port0 -> ../../ttyUSB4

e.g. after reboot:
Code: 
ls -al /dev/serial/by-id/

lrwxrwxrwx 1 root root  13 Apr 23 02:24 usb-Prolific_Technology_Inc._USB-Serial_Controller_A9CAb116L16-if00-port0 -> ../../ttyUSB6
lrwxrwxrwx 1 root root  13 Apr 23 02:14 usb-Prolific_Technology_Inc._USB-Serial_Controller_AKBSb119D15-if00-port0 -> ../../ttyUSB2
lrwxrwxrwx 1 root root  13 Apr 23 02:20 usb-Prolific_Technology_Inc._USB-Serial_Controller_DEBCb119D15-if00-port0 -> ../../ttyUSB5
lrwxrwxrwx 1 root root  13 Apr 23 02:15 usb-Prolific_Technology_Inc._USB-Serial_Controller_EBAQb116L16-if00-port0 -> ../../ttyUSB3
lrwxrwxrwx 1 root root  13 Apr 23 02:17 usb-Prolific_Technology_Inc._USB-Serial_Controller_EIBSb119D15-if00-port0 -> ../../ttyUSB4
lrwxrwxrwx 1 root root  13 Apr 23 02:10 usb-Prolific_Technology_Inc._USB-Serial_Controller_EOCMb119D15-if00-port0 -> ../../ttyUSB1

But the order how SA numbers the inverters from #1-#6 has nothing to do with the assigned ttyUSBx number. e.g. ttyUSB1 is not assigned to #1 and so on. I found no hint how SA is mapping the inverter # to an USB adapter.

I've added the following udev rule:
Code: 
sudo vi /etc/udev/rules.d/60-usb-SA.rules

SUBSYSTEMS=="usb", KERNEL=="ttyUSB*", ATTRS{idVendor}=="067b", ATTRS{idProduct}=="23a3", ATTRS{serial}=="A9CAb116L16", SYMLINK+="ttyL1A"
SUBSYSTEMS=="usb", KERNEL=="ttyUSB*", ATTRS{idVendor}=="067b", ATTRS{idProduct}=="23a3", ATTRS{serial}=="AKBSb119D15", SYMLINK+="ttyL1B"
SUBSYSTEMS=="usb", KERNEL=="ttyUSB*", ATTRS{idVendor}=="067b", ATTRS{idProduct}=="23a3", ATTRS{serial}=="EOCMb119D15", SYMLINK+="ttyL1C"
SUBSYSTEMS=="usb", KERNEL=="ttyUSB*", ATTRS{idVendor}=="067b", ATTRS{idProduct}=="23a3", ATTRS{serial}=="DEBCb119D15", SYMLINK+="ttyL2A"
SUBSYSTEMS=="usb", KERNEL=="ttyUSB*", ATTRS{idVendor}=="067b", ATTRS{idProduct}=="23a3", ATTRS{serial}=="EIBSb119D15", SYMLINK+="ttyL2B"
SUBSYSTEMS=="usb", KERNEL=="ttyUSB*", ATTRS{idVendor}=="067b", ATTRS{idProduct}=="23a3", ATTRS{serial}=="EBAQb116L16", SYMLINK+="ttyL2C"
After creating this file, the following command make the changes active and SA needs to be rebooted.
Code: 
sudo udevadm control --reload-rules && sudo udevadm trigger

I have 3x AIOs in parallel for L1 and 3x for L2 in split phase setup and would like to see L1A, L1B, L1C as SA inverter #1-#3 and the L2A, L2B, L2C as SA inverter #4-#6. The above rules create symlinks with the names ttyL1A to ttyL2C.

Btw. The only way I found to identify the physical inverter is with the AIO serial numbers which are showing in the bottom line of the SA inverter view:
1745418429061.png

The new tty names show up in the /dev/ folder with the link to the correct physical USB adapter - this also works after reboot and it assigns the tty's correct even if the kernel standard ttyUSBx numbers are changed! This shows, that the udev rules are working in general to assign an USB adapter serial number to a specific tty device name!

But SA does not show the new serial ports (ttyL1A-ttyL2C) in it's connection configuration to let me choose them, just the usual ttyUSB1-ttyUSB6 names are showing.

I found that the /dev/ folder contains the new tty's with the correct symlinks but the /dev/serial/by-id/ and /dev/serial/by-path/ does NOT contain these new tty's.
e.g.
[CODE}
ls -al /dev/serial/by-path/

lrwxrwxrwx 1 root root 13 Apr 23 13:48 platform-fd500000.pcie-pci-0000:01:00.0-usb-0:1.2.2.1.2:1.0-port0 -> ../../ttyUSB3
lrwxrwxrwx 1 root root 13 Apr 23 13:48 platform-fd500000.pcie-pci-0000:01:00.0-usb-0:1.2.2.1.3:1.0-port0 -> ../../ttyUSB5
lrwxrwxrwx 1 root root 13 Apr 23 13:48 platform-fd500000.pcie-pci-0000:01:00.0-usb-0:1.2.2.1.4:1.0-port0 -> ../../ttyUSB6
lrwxrwxrwx 1 root root 13 Apr 23 13:48 platform-fd500000.pcie-pci-0000:01:00.0-usb-0:1.2.2.2:1.0-port0 -> ../../ttyUSB1
lrwxrwxrwx 1 root root 13 Apr 23 13:48 platform-fd500000.pcie-pci-0000:01:00.0-usb-0:1.2.2.3:1.0-port0 -> ../../ttyUSB2
lrwxrwxrwx 1 root root 13 Apr 23 13:48 platform-fd500000.pcie-pci-0000:01:00.0-usb-0:1.2.2.4:1.0-port0 -> ../../ttyUSB4[/CODE]

It looks like that SA does only show all serial devices which are visible in the /dev/serial/ subfolders but the udev rule does not create links in this folders, just in /dev/.

I don't know what to do to let SA showing this new tty devices in it's connection dialog. Maybe @SolarUKWM or @Hagger have some ideas?

But even if I could select the new ttyL1A-ttyL2C devices in the SA connection dialog (which would then always be assigned to the correct physical USB adapter due to the Udev rules mentioned above), I would still have no idea in which order SA would display them as inverters #1-6.

It's really annoying that SA never addressed this problem to assign a fix order of inverters if more than one inverter is used with SA - many people asked for a solution in SA since many years :(
 
fmeili1 said:
I've added the following udev rule:
Click to expand...
First thought .... this looks ok, but instead of making up a new (probably unsupported?) /dev/ttyL1A rule -- just call them ttyUSB0 through ttyUSB6 ?

That way they will probably show up in SA through having standard names, AND by fixed down by serial-number-to-USBx mappings?

I think SA is looking only for /dev/ttyUSBx and /dev/ttyACMx items, e.g. and would be surprised by your nonstandard naming convention!)
 
SolarUKWM said:
First thought .... this looks ok, but instead of making up a new (probably unsupported?) /dev/ttyL1A rule -- just call them ttyUSB0 through ttyUSB6 ?
Click to expand...
I thought the udev rules are only able to add new symlinks for existing ttyUSBx devices... how do I define a udev rule to not create new symlinks for existing ttyUSBx devices but instead changing the kernel used ttyUSBx names itself? I didn't find an example to do this.
SolarUKWM said:
That way they will probably show up in SA through having standard names, AND by fixed down by serial-number-to-USBx mappings?

I think SA is looking only for /dev/ttyUSBx and /dev/ttyACMx items, e.g. and would be surprised by your nonstandard naming convention!)
Click to expand...
This may be the reason.
 
fmeili1 said:
I thought the udev rules are only able to add new symlinks for existing ttyUSBx devices
Click to expand...
As I understand it: When a new device turns up, the kernel looks to see if udev rules have any suggestions, if not, it will give it the first default name for the class of thing.

If you've made a rule that USB device 0123:4567, with serial number 012345678 should be called "/dev/ttyUSB7" and you plug that device in, it will get called 7. If you plug a similar, but un-ruled, device in, you'll get /dev/ttyUSB0 -- the first default.

This means that if you have a comprehensive set of rules for your set of devices, they will always stay put. Even if some are disconnected/missing at boot, or removed and re-inserted after boot. Even if the kernel spots them in a wacky order, they will populate where you want them to.

It's possible to mess up the kernel by removing and re-adding a device, and finding it walks from USB0 to USB1 ... with USB0 being "No such device" and USB1 being "this is your device!". Until you disconnect and reconnect, and goes back to USB0.

Edit: I see what you mean about paucity of examples of fixing USB0 rather than creating a symlink to "my new nonstandard /dev/ttyWhizbang". Hmmm. Try /dev/ttyUSBA ... B ... C ... if SA's scanning is loose enough it might let you see them as they start with USB. On the other hand, they might have been really tight and said only recognise /dev/ttyUSB[0-9][0-9] ...
 
Last edited:
Just use a USB hub. The order will be the order of the physical ports on the hub. If you want to change which inverter goes to a certain number just move the usb male connectors on the hub. Simple ! This works this way on my system.
 
EPicTony said:
Just use a USB hub.
Click to expand...
@fmeili1 is likely already using one, given that this is a Raspberry Pi and there isn't enough USB real estate to connect 6 or 7 devices to it.

Experiences with USB hubs varies, the whole point of udev having rules for things is to solve this exact problem. Glad it's stable and working for you (maybe you have udev rules that are helping you out?) but every so often you discover that the order of USB devices isn't what you thought it was.

Same goes for hard drives, network adaptors, when your /dev/sda goes missing and your /dev/sdb moves down one slot, or you eth0 and eth1 get in a muddle, all sorts of fun begins.
 
EPicTony said:
Just use a USB hub. The order will be the order of the physical ports on the hub. If you want to change which inverter goes to a certain number just move the usb male connectors on the hub. Simple ! This works this way on my system.
Click to expand...
I've tried this a couple of times without any success! It seems like not all USB Hubs behave the same (I remember that one member had the same problem with his USB Hub, but others are working like this).

Because this is not working with my hub (for whatever reason), I want to assign a specific RS232-USB adapter (with it's specific serial number) to a specific ttyUSBx number via udev rules in the Raspberry Pi OS.
 
Just some more info if it helps PI5: USB1 is the blue down to the left, USB2 is blue on top, USB3 is the bottom right, USB4 is the top right.
What I did: config - removed the USB/inverter setup and made it NONE. Saved.
Replugged as to the above where I wanted to put the inverter 1,2 and 3.
Rebooted the PI5.
Then went to the config and added the 3 USBs at the same time under inverters. Saved.
Rebooted again.
Works.

If anyone from SA reads this: an option to rename/move the number of the inverter LATER would be great.
Or at least write in the pdf describing the ports on the unit with the correct number.
Makes it easier.
I for one want the S/N on the inverters to correspond with "1" "2" and "3" onscreen.
Otherwize its a jumble.
 
You must log in or register to reply here.

Similar threads

R
Does Solar Assistant have cell historical data?
Replies
17
Views
722
gary_lankford
G
K
Solar Assistant and 6000XPs in Parallel
Replies
8
Views
482
TM48
T
MattMan119
When you know you need Grid .. How's the best way to get and use it?
Replies
15
Views
591
fnnwizard
F
M
Solar Assistant assistant
2 3
Replies
67
Views
5K
Crowz
Crowz
B
Commission and register LG RESU 16H battery (SolarEdge Solar system)
Replies
0
Views
59
bostondan
B

diy solar

diy solar
Back
Top