Experimental Physics and
Industrial Control System

[email protected] · Mon, 27 Sep 2010 11:54:25 -0700

On 09:43 Mon 27 Sep     , [email protected] wrote:
> However, you need asynMotor support for your motor controller. By the
> sound of it you are trying to control the motor directly, so this isn't
> much help.

Yes, I am not using asynMotor. Maybe I should have but I didn't.
And now that I have a full blown motor driver, hard to change as I may lose 
expertise and worse many functionalities.

For example, I have 2 types of synchronous move.
1/ Program the hardware and finally ask the hardware to move. That works great.
2/ Compute everything in the software and send a long atomic command.

For each physical motor I have:
* Backlash
* Prelash (to limit deadband when a motor change direction)
* Last direction of move (to see if prelash should be applied)
* Encoder position
* Reference position (internal to motor controller)
* Move in step
* Mode: relative move, absolute move, no move

> If your previous post is related (which I assume it is), note that:
>  * you need a recent version of EPICS (R3.14.11) to support "long
> strings as char waveforms"

So far so good. 

>  * You need to be putting the data into a record field that is large
> enough to take the data. Hence, the stringOut record is still limited by
> the 40 character hard size of the field definition. If you are using
> stream device you can use a waveform record with a FTVL of CHAR or
> UCHAR. You can also use the asynRecord.

Are you saying my strings should be waveform records?
Do you have a sample db file?
BTW, I went back to the epics manuals... I cannot find any mention of this.
I conclude it is expert knowledge..

Thanks,
--
E

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System