dzrong Posted March 27, 2014 Share Posted March 27, 2014 Hello,Everyone The PMAC of my customer used is 32 aixs total,and it's used for 2 years,the first year it works well.But... The problem is the ACC24E2As in the slave can't be controlled sometimes,but the ACC36Es and 65E in the slave works well in the sametime. The ACC24E2As can't be controlled means the feedback of encoders totally wrong and the ENA can't be closed through Pewin32 for disable the AMPFILERS. At the beginning,the problem appear ones,but after that it disappeared,but after few days,it appears about every 20 minites after every power cycle,and now became 5 minites. One reason I think mybe the jump E16: I check the manual of ACC24A,it said that,the Jump E16 must be close to pin1 to 2 for used with Macro CPU,but Why I can't find it in the board of ACC24E2A. Who can give me some suggestion? Rong from China Link to comment Share on other sites More sharing options...
dzrong Posted March 27, 2014 Author Share Posted March 27, 2014 Is't the backplane's problem? But why it works well before. Or it's the communication problem of MACRO,but why the ACC-65E and ACC-36E work well when the problem happenning? Link to comment Share on other sites More sharing options...
Sina.Sattari Posted March 27, 2014 Share Posted March 27, 2014 try these changes: I6890=$10 //Master Station Enable I6940=$10 //Master Station Enable I6990=$10 //Master Station Enable I72=$0 // No servo node flag transfer I73=$0 // No servo node flag transfer Link to comment Share on other sites More sharing options...
dzrong Posted March 28, 2014 Author Share Posted March 28, 2014 Thank you for being such a big help. It works well now. Link to comment Share on other sites More sharing options...
Sina.Sattari Posted March 28, 2014 Share Posted March 28, 2014 I'm glad it worked for you. Link to comment Share on other sites More sharing options...
dzrong Posted March 29, 2014 Author Share Posted March 29, 2014 Hello, Sina It's a bad news. It works well only one day. The Encoder feedback unstable on the next day. Axis 17~20 ENC count changed to huge data itself after one hour,but customer do nothing only power on the PMAC. What test can be done to find the problem? Link to comment Share on other sites More sharing options...
Sina.Sattari Posted March 31, 2014 Share Posted March 31, 2014 dzrong, What does your initialization PLC do as far as clearing ring errors and initializing the MACRO ring parameters upon power up/reset? On the other hand, based upon your explanation, it seems that the problem can be either in the MACRO UMAC's backplane or the MACRO16 CPU. Its there any chance for you to replace the backplane and/or the MACRO16 CPU on this machine? Link to comment Share on other sites More sharing options...
dzrong Posted March 31, 2014 Author Share Posted March 31, 2014 Hello, Sina Initialization PLC for Macro? I never heard that.I'm new one in Macro! Is Macro needed to be initialized? All the operation steps are below: (1) Issue $$$*** to the ultralight (2) Download Ultralight.pmc (3) SAV and then $$$ to the ultralight Now we can talk to the Macro 16 station at node 0 (if sw1 set to 7). (4) MS$$$***15, Re-initialize the MACRO16 system to recognize all cards in system MS SAVE15 MS $$$15 (5) Download Macro32CPUsetup.pmc to activate all nodes to be used (6) Issue MSSAV15 and then MS$$$15. Now system has all nodes activated and can accept the next download file (7) SAV $$$ (8)Download Macro32CPUsetup.pmc (9)MS SAVE15 MS $$$15 I think it's a MACRO CLOCK problem by setting MACRO. Of course my customer can replace that,but some days are needed,because the hardwares not at them elbow.Ultralight.pmc Link to comment Share on other sites More sharing options...
dzrong Posted March 31, 2014 Author Share Posted March 31, 2014 By the way,I6890 seems can't be set to $10,it's always $50 after we set it to $10. Why I6940/I6990 can't be set to $0,because we only have two MACRO GATE in ACC-5E. And as the manual talk,it seems if i6890 was be set to $90 will not wrong. Link to comment Share on other sites More sharing options...
steve.milici Posted March 31, 2014 Share Posted March 31, 2014 The clocks used for MACRO are set by I6800 and I6801 on the ACC-5E and by Mi992 and Mi997 on the macro station. I don't see these in your files. Where are these set? If both are left at their default settings this should be ok though. I6890 shows $50 because the "Sync Node Packet Received" status bit (value $40) will get set. This should be set to $90 but it may read back as $D0 because of the status bit just mentioned. Link to comment Share on other sites More sharing options...
naderafsh Posted March 4, 2016 Share Posted March 4, 2016 Hello Sina, We have very similar symptoms and major issues at the Australian Synchrotron. There have been a long history of "changed positions" and "unwanted motions" across two different beamlines which use the same setting. I now think, these are most likely due to corrupt raw encoder readbacks from a MACRO ring. Two MACRO rings are setup, each with a "Turbo PMAC2 PCI Ultralite" card (on the same PC). MicroMech stations are each connected to 2x16 MicroMech stepper drivers. In one instance I have recorded 4 encoder counts changed at the same time, two of them exactly doubled. This happened while I was uploading a backup from the Ultralite. I am attaching a backup I got minutes before this happened (1514), and the second backup which with the jumps happened while it was being uploaded (1517). Are these symptoms common? Any information helping us with diagnostics is very much appreciated. Best regards, Nader AfsharSR05ID01IOC01_pmac1_2016-03-02-1517.zipSR05ID01IOC01_pmac1_2016-03-02-1514.zip Link to comment Share on other sites More sharing options...
steve.milici Posted March 4, 2016 Share Posted March 4, 2016 This is not something common. Where do you see fluctuations in feedback? What is the MACRO hardware is at the other end? Typically you would follow the register map back to the source of the feedback to identify where the issue begins (controller ECT, controller MACRO node, MACRO station node, MACRO station ECT, MACRO station feedback hardware register). What were the sent configuration files intended to show? They were done with different upload schemes (multi-file vs single file) so there is no easy way to compare them. Link to comment Share on other sites More sharing options...
naderafsh Posted March 10, 2016 Share Posted March 10, 2016 This is not something common. Where do you see fluctuations in feedback? What is the MACRO hardware is at the other end? Typically you would follow the register map back to the source of the feedback to identify where the issue begins (controller ECT, controller MACRO node, MACRO station node, MACRO station ECT, MACRO station feedback hardware register). What were the sent configuration files intended to show? They were done with different upload schemes (multi-file vs single file) so there is no easy way to compare them. Thanks for the reply. The configuration is like this: Link to comment Share on other sites More sharing options...
Recommended Posts