Sensorless Homing on Expansion board



  • Hi, I see that sensorless homing is currently not supported on expansion boards.
    I currently a working on a project with >6 axes requiring Sensorless Homing. Currently I am just running the axes into hard stops, but this is pretty inelegant. How many releases will it be before sensorless homing is supported on expansion boards?


  • administrators

    I expect to include stall detect homing for motors attached to expansion board in release 3.3.



  • Thank you! That's really useful to know!



  • @dc42 said in Sensorless Homing on Expansion board:

    I expect to include stall detect homing for motors attached to expansion board in release 3.3.

    Any indication on when 3.3 is due for release?


  • Moderator

    @CaLviNx I guess this is the post you wanted a reply on. RRF 3.2 beta1 is imminent (next few days), but David hasn't been able to do all the things for 3.2 (such as sensorless homing on expansion board) that are on his work list. So this is pushed to 3.3. I'd guess 2 to 3 months before 3.3 is out.

    Ian



  • @droftarts

    my main problem is not being able to tune a heater via the tool board......


  • administrators

    Have you seen the manual procedure at https://duet3d.dozuki.com/Wiki/Tuning_the_heater_temperature_control#Section_Setting_the_model_parameters_manually?

    Alternatively, if it's a standard setup that you have used when attached to the main board, or that someone else is using (e.g. E3D Hemera with standard heater and silicone sock), you can copy those values, then make any manual adjustments needed to handle overshoot or oscillation.


  • Moderator

    @CaLviNx said in Sensorless Homing on Expansion board:

    my main problem is not being able to tune a heater via the tool board......

    Oh, I didn't see another thread about that. I expect that's on his work list, but don't know when it's coming.

    Ian



  • @dc42 said in Sensorless Homing on Expansion board:

    Have you seen the manual procedure at https://duet3d.dozuki.com/Wiki/Tuning_the_heater_temperature_control#Section_Setting_the_model_parameters_manually?

    Alternatively, if it's a standard setup that you have used when attached to the main board, or that someone else is using (e.g. E3D Hemera with standard heater and silicone sock), you can copy those values, then make any manual adjustments needed to handle overshoot or oscillation.

    Thanks i have seen that, and i already did some tuning

    And while I appreciate the "work-around" the logical thing to do would be just to be able to tune a heater on the tool board.


  • administrators

    @CaLviNx said in Sensorless Homing on Expansion board:

    And while I appreciate the "work-around" the logical thing to do would be just to be able to tune a heater on the tool board.

    I agree. However, the existing heater tuning mechanism doesn't measure the dead time very well. Rather than write code to extend the existing method to tool and expansion boards and then throw that code away, I want to improve the heater tuning algorithm first.



  • @CaLviNx Tuning remote heaters keeps getting pushed back - it's been over year for me so I've given up hope that it will ever happen.

    Another work around is to extend the cables or make up up extension cables for the heater cartridge and thermistor so that you can plug them into the main board. Then the work flow becomes :

    Power down - disconnect from the tool/expansion board - connect to the main board using the extension/longer cables - change config.g to suit the new pin assignment - power up - tune the heater - enter the new PID values - change the pin assignments back in config.g back - power down - disconnect from the main board - re-connect to the tool/expansion board - and finally power up again.

    It's a mild PITA to do that once, it's a huge PITA if you are trying to develop a new hot end and have to do that every couple of weeks or so for every design iteration but that's what we are stuck with.



  • @deckingman

    I just saw your video on you-tube about your experiences and you were (quite rightly so) pretty pissed to say the least.

    It is somewhat disappointing that hardware was released in a non fully functioning state due to a software issue.

    Im not re-wiring just to tune a hot end, so if that means waiting until the software has caught up with the hardware so be it.

    And that means the project wont complete which for me will mean the pushing back of already planned future projects which also means it will delay purchasing further duet equipment, I already keep some spares, I see no point in buying equipment for other projects if the software cant allow it to be used easily.

    As I was going to convert my 7 home machines to Duet-3 (purposefully for the 1LC tool-boards) and then convert the 21 machines I have at work to the same.


  • administrators

    There is a perfectly adequate manual tuning procedure documented on the wiki.


  • administrators

    PS- several things have been pushed back. The pandemic has resulted in a shortage of TMC2660 driver chips, which is why Duet 2 and DueX boards have been on allocation for several weeks, and will remain so for a while. One of our responses was to bring forward the development of the Duet 3 Mini, and that has been taking up most of my time over the last few months.



  • @dc42 said in Sensorless Homing on Expansion board:

    PS- several things have been pushed back. The pandemic has resulted in a shortage of TMC2660 driver chips, which is why Duet 2 and DueX boards have been on allocation for several weeks, and will remain so for a while. One of our responses was to bring forward the development of the Duet 3 Mini, and that has been taking up most of my time over the last few months.

    So you have decided to prioritize an "up-coming" product instead of fixing a known issue on an already released product because you "feel" that manual tuning is perfectly "adequate" ok then.


  • administrators

    @CaLviNx said in Sensorless Homing on Expansion board:

    So you have decided to prioritize an "up-coming" product instead of fixing a known issue on an already released product because you "feel" that manual tuning is perfectly "adequate" ok then.

    We can't maintain the volume of the Duet 2 products that our customers want to buy, so we need to offer them an alternative. That's why we've prioritised a new product development over porting the useful but less-than-perfect existing auto tune facility to expansion boards, which is work that would be thrown away when we switch to the new heater tuning algorithm.

    I have two Hemera tools on my tool changer, driven by tool boards. I tuned one of them manually, then copied the settings to the other one.

    Development of Duet 3 Mini is nearly complete, so I am starting to get back to other work, in particular RRF bug fixes and the 3.2beta1 release.


Log in to reply