wskpefqp Posted March 6, 2012 Share Posted March 6, 2012 What command is sent to the controller when I click the Enable motor #1 button in the User Interface of Pewin Tuning Pro? Is this command supposed to close the servo loop for the motor? if so, is that a J command ? Is there any other command can enable the motor, what about O command? I know it is an open loop command, but I noticed that when I sent this command to the controller, the amplifier enables. What’s the difference between the O and J command? My answer : both command will cause the amplifier enable and motor move, but J command will keep the amplifier enable until you issue a K command , O command only enable an amplifier for a very short time. Is that correct? Link to comment Share on other sites More sharing options...
Brian Posted March 6, 2012 Share Posted March 6, 2012 I believe that the command sent for that is "#1J/". Both an "O" and a "J/"(or other J commands, and other commands like "A") will enable the motor and keep them enabled until they are killed by user command or by some error. The "J" commands will also close the position/velocity loop(if it doesn't have a phasing error). Link to comment Share on other sites More sharing options...
wskpefqp Posted March 6, 2012 Author Share Posted March 6, 2012 so the only difference is whether they close the servo loop or not, right? Link to comment Share on other sites More sharing options...
Brian Posted March 6, 2012 Share Posted March 6, 2012 Yes, I believe so. Link to comment Share on other sites More sharing options...
Sina.Sattari Posted March 6, 2012 Share Posted March 6, 2012 wskpefqp, As Brian has pointed out the "#nJ/" and "#nO0" command, where n is the motor number, will cause the amplifier to be enabled. The difference is in open loop command, if the PMAC is controlling the current loop and commutation, the calculations required for both of them are enabled. If PMAC is not closing the current loop and not performing the commutation, then it purely enables the amplifier with 0 torque/velocity command output. Also, as Brian has mentioned, the open loop will not check for phase search error bit. In contrast, "#nJ/" command not only checks for phase search error bit (having a wrong phase angle in closed loop mode can result in positive feedback and runaway), but also it will close the position/velocity PID loop with current position as commanded position input to servo loop. Regards, Link to comment Share on other sites More sharing options...
wskpefqp Posted March 7, 2012 Author Share Posted March 7, 2012 thank you, sina. what are the differences if the PMAC is commutating the motor and not closing the current loop? Also, can I use an O command to make a motor move ? For instance , #1 O10, is intended to add 10% voltage of what I169 has specified to motor 1, the amplifier will respond to that voltage and output phase current to the motor windings, thus cause motor move . Is that possible before a phasing reference has been successfully established? I believe so, because controller will not check for phase search error in open loop mode. Link to comment Share on other sites More sharing options...
Sina.Sattari Posted March 7, 2012 Share Posted March 7, 2012 wskpefqp, Yes. Open loop will not check for a successful phase reference and will output on DACA and DACB voltage commands resembling currents on phases U and V of the motor such that it would generate the commanded percentage, in your example 10%, of Ix69 quadrature current in motor, had the phase reference been correct. Please keep in mind that if the phase reference is set wrong it can cause runaway situation or it can demagnetize your motor magnets. Regards, Link to comment Share on other sites More sharing options...
wskpefqp Posted March 7, 2012 Author Share Posted March 7, 2012 No moves are allowed before the controller phase the motor successfully, though we can force the motor to move, but it's dangerous, right ? How could I know if my motor has been phased right or not ? One way come into my mind is from the Phase-reference error bit, but there is no guarantee on that, that’s what the manual tells us. You all may be very familiar with the Tool provided by Deltatau Called “Turbo/Umac Setup Pro” , It can help us to configure the controller, Step1: choose you controller type, Accessory options,; Step 2: Set Servo IC frequency ,…. When we come to Step 5, it will allow you to make a move so as to set the commutation parameters Ixx70, Ixx71, In this process , the motor will locked in six distinct positions, after doing that the program will popup a window showing the counts number the controller has detected during the move, if the six counts number is consistent and approximate to your motor and encoder specifications, you will be led to do the open loop test, this test will set the phasing search routine parameters Ixx73 and Ixx74. if I fail the commutation test (let’s call it, what happened when I do this is the six counts number are not even) and force this test to pass by changing the motor parameter (length of pole pair), then change it back in the popup I variable confirmation window, after I finish doing the open loop test and the velocity curve is correct, am I able to issue a $ command to close the servo loop. Link to comment Share on other sites More sharing options...
Sina.Sattari Posted March 7, 2012 Share Posted March 7, 2012 If I recall correctly, you had a stepper method phase search setting (Ix80=6). In this case, PMAC checks the commutation angle movement during the movement from 30 degree to 0 degree phase angle and if it doesn't detect at least 22.5 degree of motion, it will set the phase search error bit high. This will give you an indication that the phasing was not successful. Stepper method phasing is usually very reliable as long as Ix73 and Ix74 are setup properly and the stepping is not harmful to your mechanism. Link to comment Share on other sites More sharing options...
wskpefqp Posted March 8, 2012 Author Share Posted March 8, 2012 As is stated in the manual, when doing the commutation test, the motor should lock in six distinct positions and move in one direction, but when I do that, I found that after each step, the motor was not locked, I saw it swaying there (move back and forth, gently though), and there is no distinct positions, it will not stop swaying until the commutation test ends , then we got a very uneven counts number. The reason behind it, in my opinion, is the air-film, which provides little friction, what’s more, our marble rail is kind of oblique due to carriage gravity, every time I open the air valve, the carriage is float and tends to stay in one position where I believe is the lowest position along the marble rail. Regarding to the count error problem, I tend to believe there is interference which cause the encoder interpolator making illegal transitions. Link to comment Share on other sites More sharing options...
wskpefqp Posted March 8, 2012 Author Share Posted March 8, 2012 Sina, based on your analysis, I think there’s one thing I can do. I will monitor the phase search error status bit when doing the commutation test next time. If the controller detect an angular value less than 22.5 degrees, that bit will be set. If this is confirmed, what should I do next? Link to comment Share on other sites More sharing options...
Sina.Sattari Posted March 8, 2012 Share Posted March 8, 2012 You have to either: 1. Redo the phasing process 2. Improve your phase search parameters 3. It would be far better to use some sort of absolute position (even hall sensors) on your motor, so you can use that as your phase reference. Link to comment Share on other sites More sharing options...
Recommended Posts