[Benchmarking Motion Machine and CK3E with EtherCAT]

Hi mbalentine!

 

Henry Bausley from ODT has sent me some more infos you also might be interested into:

 

"You can see from the numbers... [in the table below] even though the i7 is a better performer the CK3E does very well with the etherlab stack it was originally designed for.

The CK3E is not available for sale with the etherlab stack. The UMAC version of the ARM will have the etherlab stack present."

[Benchmarking Motion Machine and CK3E with EtherCAT]

Hi Unit101 and others!

 

Curt Wilson has sent me an answer:

 

There are two reasons for the difference.

 

First, the Motion Machine uses an i7 processor, which is much faster than the ARM embedded processor used in the CK3E. (It is also much more expensive and much more power hungry.)

 

Second, the Motion Machine uses the stripped down Etherlabs EtherCAT stack software, which is very efficient at the cyclic communications, but lacks many of the features – both in setup and execution – of the Acontis stack used in the CK3E.

 

Thanks for sharing this excellent real world data !

I will watch for experts reply... I suspect the difference is that the MM has a more powerful cpu....

[IDE 3x Task-Manager fails on CK3E]

Hi!

 

The Tasks-Manager - Tasks of IDE 2.2.1.25 worked fine on CK3E.

The Tasks-Manager - Tasks of IDE 3.0.1.0 on CK3E does not show any infos!

See attachments.

 

Best Regards, Anton

[Benchmarking Motion Machine and CK3E with EtherCAT]

Hi there!

I am testing one and the same Application (w EtherCAT: 22 axis, IOs) on 2 different Power PMAC HW:

 

1. Motion Machine

Motion Core LX86

1097.554 MHz

dual core

Kernel Version 3.14.28-xenomai-2.6.4

FW 2.1.0.27

Total mem 3922 MB

Free mem 3260 MB

 

2. CK3E

uPower PMAC

arm.LS1021A

1000 MHz

dual core

Kernel Version 3.14.28-ipipe

FW 2.2.0.37

Total mem 1007MB

Free mem 556 MB

 

Benchmarking (CPU-usage):

Motion Machine: FG 10.7%, BG 21.8% Yield to OS 167.6%

CK3E: FG 38.7%, BG 71.4% Yield to OS 89.9%

 

How come there is such big difference between both CPU-types?

Is there any official benchmarking for Delta Taus different Power PMAC types?

 

Regards, Anton

[Joining loops directly - cascading]

Hi!

 

I am cascading 2 motors using motor #1 for the outer loop (position), #21 for the inner loop (velocity,current - direct PWM).

If cascading using method "Joining the Loops through Position Following Function" it all works fine. But using method "Joining loops directly" (exactly after Users Manual p. 338) the net desired position of motor #21 is not overtaking its ActiveMasterPos-value at all (see http://forums.deltatau.com/showthread.php?tid=2087&highlight=activemasterpos)

 

System: UMAC Power PMAC ARM 2 Core (LS1021A) CPU Board ASSY 604112-10

FW: 2.2.0.37

 

Any idea?

Is there somebody with good results cascading 2 motors joining loops directly?!

 

Best Regards,

Anton