[etherlab-users] Configuring DC for a axis control

[Andreas Stewering-Bone]

Hello Ezio,

Depends a little bit on the slave.

I know Beckhoff AX-systems. there you must have  a latency <=2%. So I 
think your system is not the right way to do it.

If you have a bad sync behavior you will have no chance to compensate 
this by other parameter.

Greatings

Andreas


Am 28.01.2011 11:04, schrieb Ezio DelBono:
> Dear Andreas,
>
> I know that. In fact in the past i tried to write some wrapping 
> functions to avoid ioctl system call but now I'm wondering if the 
> jitter is still acceptable without doing any hack to the master.
> How much should be the maximum and minimum latency of the ethercat 
> frame so that the DC mechanism can work properly?
> Must be the 10% of the period time? In my case I have the 20%.
> Can I compensate it optimizing the "shift time"? My idea is to bring 
> the frame arrival in between two Sync0.
> Does it make sense?
>
> Thank you very much,
> Ezio
>
> On 28/01/2011 09:51, Andreas Stewering-Bone wrote:
>> Hello Ezio,
>>
>> Are you really sure, you use the master under lxrt with realtime 
>> capabilities.
>>
>> There is no lxrt or rtdm interface for the master until now.
>>
>> Greatings
>>
>> Andreas
>>
>> Am 28.01.2011 09:19, schrieb Ezio DelBono:
>>> Hello all,
>>> I am using the ethercat master version 1.5 to control some brushless 
>>> motor drives with Dc support.
>>> What my application does, is to read every 1ms the actual position 
>>> (0x6064) and send the target velocity (0x60ff) to the drive and read 
>>> the status word and write the control word as well.
>>>
>>> My application is executed by RTAI-Lxrt in user space with a period 
>>> of 1ms and a jitter of 200us (due to the character device). I am 
>>> quite sure that the dc is running well because I monitor the SYNC0 
>>> generated by the first slave that is the reference clock.
>>>
>>> Anyway I haven't understood how I have to configure the 'shift time' 
>>> in my application because the problem is:
>>> Sometime (every 2-3 seconds) the actual position is not refreshed 
>>> meaning that I read the same actual position in the next cycle. This 
>>> occurs when my application period has the worst jitter. How can I 
>>> set the shift time to compensate it? Or this is not the right solution?
>>>
>>> Kind regard,
>>> Ezio DelBono
>>>
>>> _______________________________________________
>>> etherlab-users mailing list
>>> etherlab-users at etherlab.org
>>> http://lists.etherlab.org/mailman/listinfo/etherlab-users
>>
>>
>
> _______________________________________________
> etherlab-users mailing list
> etherlab-users at etherlab.org
> http://lists.etherlab.org/mailman/listinfo/etherlab-users


-- 

-----------------------------------------------------------------------

Dipl.-Ing. Andreas Stewering-Bone           Amtsgericht Essen HRB 11500
Ingenieurgemeinschaft IgH                   USt-Id.-Nr.: DE 174 626 722
Gesellschaft für Ingenieurleistungen mbH    Geschäftsführung:
Heinz-Bäcker-Str. 34                        Dr.-Ing. S. Rotthäuser
D-45356 Essen                               Dr.-Ing. T. Finke
Tel.: +49 201 / 360-14-15                   Dr.-Ing. W. Hagemeister
Fax.: +49 201 / 360-14-14                   Tel.: +49 201 / 360-14-0
Andreas.Stewering-Bone at igh-essen.com        http://www.igh-essen.com

-----------------------------------------------------------------------