Unit101

Here is example logic to scan a ethercat network to see if all devices are present and working as expected on the network. Note now it only works with Acontis type Master. It does scan all device types including those that don't have COE (dumb IO). I hope that soon etherlab version of PP firmware will have a command added that is now in Acontis version. It uses the command "EcatRegReadWrite()" Thanks, Ecat-Diagnostic-Scan.pdf

Yes that is my concern. We need tests/data on limits of ECAT, can it handle high accuracy applications ? I doubt ever do it in Torque mode, but in Position mode it really depends on how well the drive can control the load... so if the drive can close the loop and hold with very little jitter, maybe work. I ask GredD at Omron if any specs or tests done with new Omron 1S Ecat drive and how accurate it can hold a give position. What do you consider high accuracy ? or good enough for your applications ? Another great point to push Managers af ODT/Omron factory with: - IPC with MACRO - IPC in UBUS rack .... no need for PCI with this... similar to the past PC104 with Turbo in a Umac rack

Hi Jeff, I've been talking quite a bit to managers about this, both at DT and Omron HQ. One idea Steve Jones and I had... - you have said that Power PMAC with Enet will not give you the same level of through-put thru the buffer as before using the Bin-Rot and PCI bus - I suspect you are correct. - perhaps a upgrade path for these applications would be to move to the new Power PMAC IPC, this is a I7 Windows PC with Power PMAC embedded. If I'm thinking right on this now you have nothing but Memory transfer rates to get your Gcode to the Buffer. This eliminates the Ethernet or PCI bus. -however this also limits you to ECAT to drives/IO which may not give us the preformance needed at the drive ? -so I value your opinion on this ?? thanks, mike

Curt Wilson has said we may be able to bring these back. Post your requests here... it's all about supply and demand.

RESOLVED !!! Ok I was on the right track. #1 forget about the my.ini it is not being used, not sure why but it is ignored by the MySQL service startup configured by IDE - you can delete this file and it runs exactly the same - the MySQL ignores ANY Settings in this file, including the location of Dbase files - perhaps it is read by the IDE ? #2 Issue is related to the size and type of HDD that IDE is installed on - some GBP or SSD, RAID ... larger drives >1.5TB will show this issue - it is related to sector size of formatted drive #3 FIX is well documented by other MySQL users - it is to tell MySQL to use this when writing the files (logs) innodb_flush_method=normal - most users simply put this i the my.ini (or cnf linux) and it is fixed - but as #2 states the my.ini is here ignored #4 How to FIX - go to regedt32 - HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\MySQL - edit the Image Path (startup cmd) - add cmd line switch to tell MySQL to use above "flush method" - here is what it should look like: "C:\Program Files (x86)\MySQL\MySQL Server 5.0\bin\mysqld-nt.exe" "--innodb_flush_method=normal --defaults-file=C:\Program Files (x86)\MySQL\MySQL Server 5.0\my.ini" MySQL - notice I added the --innodb_flush_method=normal cmd line switch - now this feature is used on startup and it works fine I suspect this will NOT impact other systems... so would like to test on a typical system to see if still works fine... but I'm not a SQL expert... just weathered a bit.

RESOLVED !!! Ok I was on the right track. #1 forget about the my.ini it is not being used, not sure why but it is ignored by the MySQL service startup configured by IDE - you can delete this file and it runs exactly the same - the MySQL ignores ANY Settings in this file, including the location of Dbase files - perhaps it is read by the IDE ? #2 Issue is related to the size and type of HDD that IDE is installed on

Continue to work on this issue. To make it work fine all I need to do is delete file: C:\Program Files (x86)\MySQL\MySQL Server 5.0\data\ib_logfile0 Then start the MySQL service. It starts fine and IDE seems happy with the database, Intellisense works... Tried putting the MySQL in a different directory, no joy. Also seems to ignore the my.ini settings related to database location, if I change this and move the data dir it still works in same local dir as always. In fact you can modify any of the my.ini settings and seems to have no effect when starting the service or process. The fix is supposed to be add this line to my.ini innodb_flush_method=normal But I don't believe the ini is having any effect on startup... it is simply using some defaults. Again if interested here is error log on startup of MySQL: 161226 6:51:18 InnoDB: Operating system error number 87 in a file operation. InnoDB: Some operating system error numbers are described at InnoDB: http://dev.mysql.com/doc/refman/5.0/en/operating-system-error-codes.html InnoDB: File name .\ib_logfile0 InnoDB: File operation call: 'aio read'. InnoDB: Cannot continue operation. Here is same LOG if I Delete the ib_logfile0 FIRST, it starts up fine 161226 6:54:12 InnoDB: Log file .\ib_logfile0 did not exist: new to be created InnoDB: Setting log file .\ib_logfile0 size to 5 MB InnoDB: Database physically writes the file full: wait... InnoDB: Error: all log files must be created at the same time. InnoDB: All log files must be created also in database creation. InnoDB: If you want bigger or smaller log files, shut down the InnoDB: database and make sure there were no errors in shutdown. InnoDB: Then delete the existing log files. Edit the .cnf file InnoDB: and start the database again. 161226 6:54:12 [Note] C:\Program Files (x86)\MySQL\MySQL Server 5.0\bin\mysqld-nt.exe: ready for connections. Version: '5.0.41-community-nt' socket: '' port: 3306 MySQL Community Edition (GPL) I'm convinced the issue is that I'm using a 2TB SSD and that MySQL is NOT using the my.ini settings... it reads it... but ignores the settings Thanks

Problem: New install of IDE (v2.2.0.39) on new PC with fresh Win 10 Pro x64 OS. At end of install it says Database failed and prompts you to run the IDE troubleshooter. Message Dialog like this: "Power PMAC IDE Installation Varification Report" MySQL databases are no created. NOTE: Please go to "IDE Troubleshooting".... In IDE Troubleshooter you run button "IDE Troubleshooting" it says database repaired. However it is NOT. When you run IDE in Output window after starting you get a message like this: Database Error: Unable to connect to any of the specified MySQL hosts... What is Wrong: MySQL service is NOT starting and running. Go to Computer Services - Select MySQL in list - select Start. It will fail with a Error 1067. Open \data\compName.err and you see this : InnoDB: Operating system error number 87 in a file operation. File name .\ib_logfile0 File operation call: 'aio read'. When MySQL starts it can't write to the ib_logfile0 - this seems to be a issue with how large the drive is and it sectors being 4096 instead of 512 in size - so if you are running on a large >1.5TB or so this is a problem FIX: - only temporary fix ... if you manually DELETE the \data\ib_logfile0 then you can start the MySQL no problem - now start IDE and it is working with it's DBASE files fine HOWEVER: - next time you start system MySQL will FAIL to Start Again Permanent fix: - widely reported that a new switch is available in 5.6 > that lets you allow for this, adding a line to my.ini will fix this issue - but it appears we are using a older version for MySQL 5.0 so it is not supported Can we simply manually install a newer MySQL and have the IDE use this ? I don't see why not as long as the service is started and pointing to the correct dbase files... I will give this a try. Thanks for any other suggestions or help. mike

Thanks for the clarification, now makes sense what I saw on this new system.

Update on Win10 issues with IDE after Update: - also had problems with opening a project based on the extension .PowerPmacSuite_sln - breaks the link to IDE, no big deal - I updated my IDE to latest release .107 - simply ran the install - DID NOT Uninstall v .40 - it saw that I had .40 and updated, installed new version perfectly - All is working fine at this point - So the Win10 1511 update created a whole new Windows Directory in my case and it moved the old Win dir to "Windows.old" directory - it did not get the 2 PPMAC460Compile files and move them to the new Win10 dir - I suspect Windows does not expect those files to be be in that directory so ignores them.

I noticed with new Motion Machine that it came with a Dongel. Can I get some clarification on what this does and how to use it ? Is this the ECAT license Dongel ? Does it matter if plugged in before or after power up of unit ? What if it is removed while unit is working with drives ? Can the Dongel be moved from Pmac to Pmac and the license files move as well ? How can I know that the Dongel is working and how many ECAT devices the license is good for ? When Looking at the ECAT setup it has the same number of ECAT devices (4servo + IO) with or without the Dongel plugged in, does this make sense ? I assume the manuals will soon (if not already) be updated with dongle details. Any other tips on how to use or NOT use the Dongel ? Thanks,

Today 04/05/16 Windows 10 did a update on my machine. New Build is 10586.71 Build 1511 Win 10 Pro - After this update IDE will not do a build download. Error in IDE Build/Download is: C:\DTau_Training\Demos-Power\PowerClipperD129-NC\PPNC_PmacSource\PPCNC_ProjectSource\PPCNC_ProjectSource.ppproj(191,5): Error MSB4062: The "PPMAC460CompileTask.CompileTask" task could not be loaded from the assembly C:\Windows\Microsoft.NET\Framework\v3.5\PPMAC460CompileTask.dll. Could not load file or assembly 'file:///C:\Windows\Microsoft.NET\Framework\v3.5\PPMAC460CompileTask.dll' or one of its dependencies. The system cannot find the file specified. Confirm that the declaration is correct, and that the assembly and all its dependencies are available. - Reason is that 2 files are NOT copied to the NEW version of Windows installed after this update. - To Fix this: - go to new directory created by update C:\Windows.old\WINDOWS\Microsoft.NET\Framework\v3.5 - copy 2 files PPMAC460CompileProcess.dll, PPMAC460CompileTask.dll - copy these to C:\Windows\Microsoft.NET\Framework\v3.5 - Now that these files are in correct place of NEW Win10 the IDE will Build OK - Close and restart the IDE after doing this. - Post here if this works for you OR if it does NOT...

Actually Power NC16 will work with Turbo AND Power PMAC hardware. With the Runtime you get 2 complete projects, 1 for Turbo PMAC and other for Power PMAC. However you will have to do some customization based on your hardware to fully implement the NC system. Such as proper initialization in PLC1, Proper system power up and RESET, custom Homing of your machine perhaps. But the code is ready to load a NC file (G-code) and send to the PMAC to run.

Hello, Basically your HMI will need to load a NC file (list of G,Mcodes) and assure the format is compatible with the currently loaded PMAC G/M Code subroutines which are found in Programs 1000, 1001. Once the NC file is in tact and ready your HMI will need to download this to the PMAC into a Motion Program buffer. Then tell the PMAC to execute the program. While it is executing you can monitor the progress line by line (Coord[n].Nsync) and likely on your HMI show some list of the NC file and increment a line indicator to show the progress while executing. Look to the Power PMAC software and users manuals for details on loading a motion program and executing it on the PMAC, these also cover using NC type G-codes. With that said the Power PMAC NC14 software has 100's of hours in development associated with the HMI and corresponding PMAC files (function.plc...), so it is well beyond this forum to explain "how to do it". The NC16 program is well worth it's cost and saves you these 100's of hours trying to do it as well. Also the NC14 (now released version is NC16) is 100% configurable allowing to to create whole new panels and change the entire interface to suit your custom application, but still using the engine underneath to do the hard parts. I would encourage you to use the NC16 as a engine and spend your valuable customizing programming time on modifying this engine instead of trying to duplicate it. However you can surely develop your own NC HMI, this is not so difficult just can take some time both in development and testing, depending on how many options it supports, there are lots of details along the way, so keep asking questions.

Resetting firewall is no big deal... after that you will get the prompts to allow the IDE to work. Seems you accomplished the same or similar by elevating the privileges. Glad it is working... congrats and thanks for getting back.

Try this... I have Win10 as well and have seen similar issue, not sure if same: http://forums.deltatau.com/showthread.php?tid=1991&highlight=zombie Also be sure running application as Admin.

To reset your comms with the PMAC I do this: HERE call pcomm function to close the connection public static void CloseDevice() { try { if (DeviceNum > -1) Pmac.Close(DeviceNum); Online = false; } catch { Online = false; } } Here go to windows and kill the pcommserver process string processName = "PcommServer"; foreach (Process proc in Process.GetProcessesByName(processName)) { proc.Kill(); } Let me know if you need more help with this... I have written a lot of code that will detect when the PMAC connection is lost and make attempts to restore.

This has been seen by 2 different users recently. Running Windows 10 x64 (release) with PeWin 4.4.0.0 Connected to any Turbo PMAC using Ethernet. When issuing a "$$$***" command it is processed but PeWin loses connection to PMAC and will not restore after the reset is done. Message seen in the Results list below looks like this: Device 1-> Error 0x0, SockFlush initial time stamp,....,Error Code: 10060, Description: WSAETIMEOUT socket error Now you can go to Setup - Select the Ethernet PMAC - Ping but the PMAC is not recognized and will not connect to PeWin. Seems you must now Power cycle the PMAC to get it to connect. Solution that has worked: Setup - General Setup - Select - Select Eth PMAC in List - Properties - General now check the box [x] Multihost enable What I "think" is going on is simply a timeout issue, not sure why Win10 acts different ? But the Multihost "I think" sets the timeout to 30 seconds. So not sure why it works, but seems to fix the issue. If it works for you or you see different please post here with details.

I've had this same issue several times, also when using Win7 x64. For me the fix every time is to: Control Panel - Programs - Repair PeWin application Has worked several times for me... I suspect it corrects the real issue CP is describing.

When trying to build and download, it does NOT work. It thinks the download is already done but it is NOT. Problem is with Windows Firewall and your setup. Has to do with Public networks and allow some features of the app (ide) to work. I don't know what problem is but here is the fix. Go to Command Prompt and use this command to reset your firewall settings: "netsh advfirewall reset" Now run the IDE and you will get some messages from Windows about firewall. Be sure to select Public networks and allow. Now it should work. FTS

Swapping Direction when Feedback is a Serial Encoder ? -I don't know how to do this with ALL the different types of serial encoders. -I have not yet tested the below ideas but based on Turbo Brick testing it should work fine as setup below. What I have found is that some allow you to swap the direction sense with wiring. For example the Sick-Stegmann SSI encoder has a wire to Pin which can be connected to GND to reverse the counting sense of the feedback. -If the Serial encoder does not allow you to invert its direction sense then I believe the best way to handle this is simply by swapping 2 motor leads UVW then doing the PWM setup based on this new direction. -If the Serial Encoder allows you to invert its counting sense then the parameters can be used to change this. Disregard the Gate3.Chan[j].EncCtrl parameter since it does not apply to a serial encoder and instead swap the direction sense at the encoder itself (such as with SSI type encoder). -Next you would adjust these parameters to reverse the command sense to motor. Motor[x].PhaseOffset: negate, so if now 683 make this -683 Motor[x].PwmSf: negate, so if now 15237 make this -15237 -For power on phasing use same parameters: -this may not apply to all serial encoder power on phasing Motor[x].AbsPhasePosOffset: 2048 - Current value, so if now 613, make this 1435 Motor[x].AbsPhasePosSf: negate, so if now 170.6667 make this -170.6667 If you do some testing and find different results or a better way, please post it here.

Recently after update/install of Windows 10 build 10240 the IDE would not do a Build/Download. It would stop with Error MSB4062 and show below complete message: "C:\3_TCIP-Server-RockwellParsing\TCPIP-RockCmdInt-PPmac\TCPIP Server.ppproj(144,5): Error MSB4062: The "PPMAC460CompileTask.CompileTask" task could not be loaded from the assembly C:\Windows\Microsoft.NET\Framework\v3.5\PPMAC460CompileTask.dll. Could not load file or assembly 'file:///C:\Windows\Microsoft.NET\Framework\v3.5\PPMAC460CompileTask.dll' or one of its dependencies. The system cannot find the file specified. Confirm that the declaration is correct, and that the assembly and all its dependencies are available." Appears that something IDE uses related to .NET 3.5 was lost? Fix was to re-run the Patch for the IDE 2-0-3-40. After re-running the patch all was fine.

How to reverse the direction sense of a PWM controlled motor on Power Brick. Here we assume the motor is typical 3 phase brushless with Quadrature encoder and using Halls for phasing. After successfully running setup: - you can power up - phase - home - jog motor ok - be sure you have the phasing setup ok first Now we want the motor direction sense to be opposite of what it is now: - E.G. now the motor turns CW with a Jog+ command we want it to go CCW Below the needed parameters to change: Gate3.Chan[j].EncCtrl: invert sense, so if now 7 make this 3 Motor[x].PhaseOffset: negate, so if now 683 make this -683 Motor[x].PwmSf: negate, so if now 15237 make this -15237 Motor[x].AbsPhasePosOffset: 2048 - Current value, so if now 613, make this 1435 Motor[x].AbsPhasePosSf: negate, so if now 170.6667 make this -170.6667 Now save, reset - phase the motor ($) and it the direction sense will reversed.

can you email me a config file for the system with just the cpu and single acc24 card... I will download to my demo box and see if I get same results... thanks, mike mesposito@deltatau.com

Hello, In ACC24E3 Manual: Wiring to: J1/TB1, J2/TB2 Encoder Connectors - the GPInput comes in on Pin1 of the 15pin D-sub. It is labeled "CHT1" So it's state can be read as the Halls T Flag Input for the given channel. -Also for Hardware, look in manual for T Flag hardware wiring HARDWARE: J1/TB1, J2/TB2 Encoder Connectors - this is considered a Hall sensor type input, along with encoders so it is 5v Excerpt from manual pg 17... NOTE the T Flag Input... The U, V, W, and T flag inputs, usually used for Hall-style commutation sensors and digital-output temperature sensors, are 5V CMOS inputs with internal pull-up resistors. These flags share pins with the serial encoder inputs and the pulse-and-direction outputs; the user must configure in software which use the pins have. If the Hall commutation sensors are used for ongoing phase or servo position (not just power-on phase angle), the U, V, and W signals must be wired into the A+, B+, and C+ inputs on the connector. SOFTWARE see PP SRM: You can get to the State of the Flag input here: Gate3[0].Chan[1].T=1 Excerpt from PPSRM manual: Gate3.Chan[j].T Description: T flag input value Range: 0 .. 1 Units: Boolean Gate3.Chan[j].T contains the present logical state of the IC’s T flag input for the channel, latched on the most recent servo interrupt. A value of 1 corresponds to a high voltage into the IC; a value of 0 corresponds to a low voltage into the IC. On most Delta Tau hardware, the buffering external to the IC is non-inverting, so this applies to the voltages into the controller as well. Let us know if need more.