bdring
Problem: 'VFD Queue Full' Message.
Wiki Search Terms
I get this message once in a while, resulting in the VFD/Spindle not spinning up. But it doesnt stop the process, resulting in a crash.
Why is there no alarm set when this message is shown?
Im using UGS for my controller.
Controller Board
Xprov5 Controller
Machine Description
Bulkman3D Ultimate Bee 1000x1000 with high current steppers, Xprov5 controller and VFD controlled spindle
Input Circuits
Configuration file
Not needed for now I think
Startup Messages
Not needed for now I think
User Interface Software
UGS
What happened?
Crassh after starting job due to missing spindle rotation. UGS reports VFD queue full message in console. This has happened before, but not very often, and it's been a long time since last. Main problem is that the process doesnt stop, but crashes the machine.
GCode File
No response
Other Information
No response
I will find it when I get home from work.. So Config file and Startup message? Gcode file isn't relevant, as it has nothing to do with the file.. It can be any file.
Here is the missing documentation.
Let me state, that I know that the link is missing the resistors to terminate the connection, but that is not the problem im trying to address.
The problem is that Fluidnc doesn't set an alarm when this message is coming, so that the process continues. Nothing else. I will get the termination done..
We ask for complete information because problems are nearly always context-dependent. Seeing all the information lets us form a mental picture of the full situation. We need to see the forest and the trees.
This is the new startup text with $ss written in the UGS console.. Didn't give much from what I see..
We ask for complete information because problems are nearly always context-dependent. Seeing all the information lets us form a mental picture of the full situation. We need to see the forest and the trees.
Yes I fully understand.. Just needed to understand what you needed :)
Try using the terminal in the web installer.
https://installer.fluidnc.com/
The start messages should look like this.
http://wiki.fluidnc.com/en/support/troubleshooting_config_files#check-the-startup-messages
UGS appears to be suppressing the result from $ss. Try sending $ss from FluidTerm.
Sorry, my message collided with Bart's. Either way will work - installer.fluidnc.com or FluidTerm.
We also need to know exactly what you mean by "crashes the machine". It could mean that the bit slammed into the bed, or that things just stopped working with no messages, or that the firmware issued a bunch of software crash messages, or ... Please be very specific and detailed about what you saw at the time of the failure.
This is the startup text from FluidNC console.
I will ask why UGS supress the command
If you are not using any of the web services, please send $Wifi/Mode=off . It is currently using AP mode which is okay for initial setup, but can cause problems in production runs, since the ESP32 can be bombarded with irrelevant network traffic.
If you are using the FluidNC web services, use STA mode by setting $sta/ssid and $sta/password to connect to your wifi network.
I will do this later, as this is not my main concern in this ticket :) But thank you for the suggestion.
The machined tried to mill out wood with a stopped spindle. It didnt turn on the spindle as it should. Thats basically it
This is the message in got from Joacim at ugs:
Without knowing the details about what messages you are getting from the controller, I think you should raise this with the FluidNC developers. It sounds serious if the controller looses the ability to control the VFD, so the appropriate message from the controller should be an alarm and not just a "MSG: I think we just lost control of the VFD".
It is not quite as simple as alarming on queue full. By the time that the queue fills up, the driver has already attempted to send ten other commands. The fact that they are still in the queue means that something has already gone badly wrong. For a reliable fix we need to understand that earlier sequence. We might need to issue an alarm way before the queue fills up. If you send $message/level=debug, FluidNC will report more details about the interaction with the spindle.
Another thing to try is to switch to the "Generic Modbus VFD" driver with model=Huanyang. That driver has better diagnostics than the direct Huanyang driver, especially when $message/level=debug and the section contains debug:3. You would need this in the config file, instead of the huanyang: section
ModbusVFD:
uart_num: 1
modbus_id: 1
model: Huanyang
debug: 3
tool_num: 0
speed_map: 0=0% 0=25% 6000=25% 24000=100%
Trying the vfd settings I'm no longer able to start and stop spindle from console with m3/m5
Normally ugs will clear the queue in the controller and stop the machine like when I click stop in the interface or estop button.
Are there log messages? With $message/level=debug and debug: 3, M3 and M5 should result in reports. This is on a machine with ModbusVFD:/model: Huanyang but no VFD is connected
INFO: ModbusVFD Spindle Tx:gpio.15 Rx:gpio.16 RTS:gpio.14 Baud:9600
INFO: RS485 Tx: 01 01 03 05 00 00 2C 4F
DBG: ModbusVFD: setState:5 SpindleSpeed:0
DBG: Synced speed. Requested:0 current:0
INFO: Flood coolant gpio.12
INFO: Mist coolant gpio.4
INFO: Probe gpio.32:low
DBG: Pos Limit 0
DBG: Pos Limit 0
DBG: Pos Limit 0
DBG: Pos Limit 0
DBG: ModbusVFD: setState:5 SpindleSpeed:0
DBG: Synced speed. Requested:0 current:0
INFO: RS485 Tx: 01 01 03 05 00 00 2C 4F
INFO: RS485 Rx:
INFO: RS485 No response
INFO: RS485 Tx: 01 01 03 05 00 00 2C 4F
INFO: RS485 Rx:
INFO: RS485 No response
...
> m3 s7000
ok
DBG: Synced speed. Requested:0 current:0
INFO: RS485 Tx: 01 01 03 05 00 00 2C 4F
INFO: RS485 Rx:
INFO: RS485 No response
This is what it shows when trying to run it. So it has connection, but doesn't spin up the spindle
The two images look identical to me. In particular, both say "ModbusVFD Spindle"
The two images look identical to me. In particular, both say "ModbusVFD Spindle"
Sorry about that, wrong picture.. I have updated the post with the correct picture.
I do not see any DBG messages in the log. Please send $message/level=debug
Oki, changing back to the ModbusVFD I get this now, so it looks like that driver is not compatible with my VFD?
[MSG:INFO: ModbusVFD Spindle Tx:gpio.4 Rx:gpio.25 RTS:NO_PIN Baud:9600] [MSG:INFO: Mist coolant gpio.21] [MSG:INFO: Probe gpio.22:low] [MSG:DBG: All Limit 0] [MSG:DBG: All Limit 0] [MSG:DBG: All Limit 0] [MSG:DBG: All Limit 0] [MSG:DBG: feed_hold_pin 0] [MSG:DBG: macro0_pin 0] [MSG:DBG: macro1_pin 0] [MSG:DBG: ModbusVFD: setState:5 SpindleSpeed:0]
Grbl 3.9 [FluidNC v3.9.7 (wifi) '$' for help] [MSG:INFO: VFD RS485 Unresponsive] [MSG:ERR: Critical VFD RS485 Unresponsive] [MSG:INFO: ALARM: Spindle Control] ALARM:10