On Thu, 27 Dec 2012, Viesturs L?cis wrote: > Date: Thu, 27 Dec 2012 23:36:11 +0200 > From: "[UTF-8] Viesturs L?cis" <viesturs.la...@gmail.com> > Reply-To: "Enhanced Machine Controller (EMC)" > <emc-users@lists.sourceforge.net> > To: "Enhanced Machine Controller (EMC)" <emc-users@lists.sourceforge.net> > Subject: Re: [Emc-users] Pid saturated, was: Servo error > > 2012/12/27 Peter C. Wallace <p...@mesanet.com>: >> >> Just thinking that noise on the HAL signals might cause >> flakey commutation initialization so might be a good thing to halscope >> > > Ok, here is a result: > http://picpaste.com/hall-1xe8rtyI.png > Seems totally fine for me. > How can I check consistency of hall signals - if the signals do change > every 60 degrees? > I could try to finetune the feedback loop to smooth out encoder > velocity and then try to scope hall signals again and see, but I would > like to find out, if there is any other way to skin the cat. > > > I also tried bldc cfg=qi to find out, if I can make it work without > hall sensors. > > I tried the path suggested by Andy: > 1) started up LinuxCNC > 2) in HalShow ran: > setp hm2_5i23.0.encoder.00.index-enable 1 > 3) watched "counts" and "rawcounts" pins > 4) turned motor by hand until "counts" pin was reset to 0 and > "index-enable" went false; > 5) in HalShow I ran: > setp hm2_5i23.0.3pwmgen.00.A-value 0.9 > setp hm2_5i23.0.3pwmgen.00.B-value -0.9 > setp hm2_5i23.0.3pwmgen.00.C-value -0.9
This is very unlikely to do what you think it will (assuming default +-1.0 PWM ranges this will be 90% of full scale voltage so the 7I39 will current limit and you will not get anywhere near the current ratios you expect) I would try plain old Q mode (with maybe an init value of .1 or .2) > 6) enabled pwmgen with F2; > 7) wrote down "counts" pin value of -103 and saved as: > setp bldc.0.encoder-offset -103 > 8) connected bldc.0.init and pwmgen-enable to estop loop so that motor > is homed, before motion is enabled to avoid triggering following > error; > > 9) for testing purposes I set PID parameters to following values: > P=0.1 > Deadband = 0.001 > all the remaining values are set to 0 > > 10) if I try to jog that joint, pid immediately gets saturated > 11) if I comment out the encoder offset line like this: > # setp bldc.0.encoder-offset -103 > then I can jog that joint > > What puzzles me is that in step 11 bldc still manages to home rotor - > init-done pin is true. > > -- > Viesturs > > If you can't fix it, you don't own it. > http://www.ifixit.com/Manifesto > > ------------------------------------------------------------------------------ > Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > MVPs and experts. ON SALE this month only -- learn more at: > http://p.sf.net/sfu/learnmore_122712 > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users > Peter Wallace Mesa Electronics (\__/) (='.'=) This is Bunny. Copy and paste bunny into your (")_(") signature to help him gain world domination. ------------------------------------------------------------------------------ Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft MVPs and experts. ON SALE this month only -- learn more at: http://p.sf.net/sfu/learnmore_122712 _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
