Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Acontis vs Etherlab
#1
In reading the manual I find this:
If Sys.EcatType is 0 at power-on/reset, the IgH Etherlab stack is used. The high-priority tasks of this stack execute in the processor’s “kernel space”, so execute efficiently and with lower latencies. This permits higher network update frequencies. It must be used if “local” motors are to be controlled as well. However, this stack has a smaller feature set and more limited interactive setup software.

If Sys.EcatType is 1 at power-on/reset, the Acontis EtherCAT stack is used. This stack supports a broader array of EtherCAT peripheral devices, and has more extensive setup support through the “EC-Engineer” development software. However, it executes in the processor “user space” and so cannot support as high network update frequencies as the Etherlab stack.
.
Our system consists of mostly "Local" motors due to performance constraints, yet we forced by the latest IDE and the ARM to use Acontis. Since Acontis is lower performance is there a strong reason why support for etherlab has been abandoned? In addition, the machine IO for our machines is mutable, that is we can add or alter attached options at boot time but Acontis support for that type of operations is limited/non-existant. The only saving grace I can see to Acontis is that it does not crash the IO when executing saves or opening/closing the IDE.
Reply
#2
Unfortunately, with the Etherlab stack, there is no prospect for any improvements or extensions going forward, in setup or operation, so we are just maintaining it for existing applications.

The new Quad Core ARM CPU has dramatically better performance for the Acontis stack, especially if you isolate it on its own core. We recently benchmarked reliable performance of 32 EtherCAT axes at an 8 kHz update rate.
Reply
#3
(09-02-2020, 09:46 AM)curtwilson Wrote: The new Quad Core ARM CPU has dramatically better performance for the Acontis stack, especially if you isolate it on its own core. We recently benchmarked reliable performance of 32 EtherCAT axes at an 8 kHz update rate.

The implications of this are interesting.
Are you able to comment on available drive capabilities/limitations to operate in current mode at this update rate?
Will there be ODT drive products to create a 'single source' solution based on higher speed eCat?
Reply
#4
Omron's 1S EtherCAT drives can operate at an 8 kHz update rate. If you want to command a large number of drives at this frequency, you may want to reduce the number of active PDOs per drive from the default to keep the transmission traffic reasonable.
Reply
#5
(09-08-2020, 12:11 PM)curtwilson Wrote: Omron's 1S EtherCAT drives can operate at an 8 kHz update rate. If you want to command a large number of drives at this frequency, you may want to reduce the number of active PDOs per drive from the default to keep the transmission traffic reasonable.
Hi Curt, Let me describe my issues. On a 460 with Wago I/O can add modules (with the proper power cycle of course) for options and the stored project boots and runs as expected. I can then, under plc control at runtime, configure these additional I/O modules and subsequently use the optional functionality. On the Quad core I have a running/saved project and when I add additional I/O modules the project will no longer boot and run. additionally, turning the machine off and removing the added I/O modules does not correct the problem. The project no longer boots and runs. It is necessary to do a $$$*** and build, download, and save the project all over. The downside of this is that every single option and/or combination of options is going to require its own project. From a revision control perspective, this is untenable. is there any way around this?
Reply
#6
Jeff, are you using Acontis or Etherlabs EtherCAT on your quad core?
I feel like you could be making some assumptions about how Acontis works based on Etherlabs.

If Acontis, you may want to take a look at the hot connect groups described starting on page 182 of the current IDE manual. They basically allow you to set up two versions of your ECAT network and switch between them.
Reply
#7
Is it possble to control “local” motors with ARM LS1021A CPU if the Acontis EtherCAT stack is used at the same time?
If not ,what problems will happen? I had done some test, and it seemes that can work at an 1 kHz update rate.
Reply
#8
(09-10-2020, 04:45 PM)Eric Hotchkiss Wrote: Jeff, are you using Acontis or Etherlabs EtherCAT on your quad core?
I feel like you could be making some assumptions about how Acontis works based on Etherlabs.

If Acontis, you may want to take a look at the hot connect groups described starting on page 182 of the current IDE manual. They basically allow you to set up two versions of your ECAT network and switch between them.

On a limited basis this may work, especially for one-off builders and retrofitters, but from the OEM perspective where there are many combinations of I/O driven by customer specified options, this is not feasible given the current state of Acontis. As such, and given the unavailability of Etherlab on the Quad Core. We have come to believe that ModBus TCP is the most viable alternative. Moving forward we would of course order our Quads without EtherCat I/O, but for now can the EtherCat ethernet port be repurposed as a Modbus TCP master? I understand there is a Modbus project with some documentation, but is there any additional support such as existed for Turbo?
Reply
#9
(11-01-2020, 02:44 AM)dzrong Wrote: Is it possble to control “local” motors with ARM LS1021A CPU if the Acontis EtherCAT stack is used at the same time?
If not ,what problems will happen? I had done some test, and it seemes that can work at an 1 kHz update rate.

Yes this is possible. I believe the line in the Software Reference Manual stating otherwise was an error left over from a decision made in early ARM firmware that has been reversed for firmware 2.4 and above.

(11-02-2020, 07:10 AM)JeffLowe Wrote: On a limited basis this may work, especially for one-off builders and retrofitters, but from the OEM perspective where there are many combinations of I/O driven by customer specified options, this is not feasible given the current state of Acontis. As such, and given the unavailability of Etherlab on the Quad Core. We have come to believe that ModBus TCP is the most viable alternative. Moving forward we would of course order our Quads without EtherCat I/O, but for now can the EtherCat ethernet port be repurposed as a Modbus TCP master? I understand there is a Modbus project with some documentation, but is there any additional support such as existed for Turbo?
We have an App Note for changing the EtherCAT port to Ethernet, let me verify it works the same on quad core and then upload it.

There is a section on modbus setup starting on slide 140 of the following presentation.
http://forums.deltatau.com/filedepot/dow...202016.pdf
Reply
#10
Just to share, A probable solution has been identified for these types of multiple configurations an OEM will face without requiring each and every configuration to have its own project. The IDE creates all the EtherCat enumerations in the Configuration folder named eni.xml. After build and download this file is read at the first setting of Ecat[0].enable=1; For projects with multiple possible EtherCat configurations, each configuration can be built up and generate a separate eni.xml file for each. These can be given appropriate names and saved in the Documentation folder so they are downloaded with the project and available at runtime. On startup the configuration is determined, likely from fsave/fload variables and the appropriate saved configuration file can then be copied over the Configuration/eni.xml file.

(Thank you BigCat)
Reply
#11
Quote:Moving forward we would of course order our Quads without EtherCat I/O, but for now can the EtherCat ethernet port be repurposed as a Modbus TCP master?
We should be able to turn the EtherCAT port into an extra Ethernet port. Are you still looking for this? I know you're looking at a solution involving swapping out ESI files.
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)