Re: Delta Tau Power Brick LV-IMS driver

[Torsten Bögershausen <torsten.bogershausen@esss.se>]

The problem is the PROBLEM bit, as far as I can see.
Which exact version of the motorRecord are you using ?

The code on latest master looks like this:

    /* Get motor-now-moving indicator. */
    if (ls_active == true || msta.Bits.RA_DONE || msta.Bits.RA_PROBLEM)
    {
        pmr->movn = 0;
        if (ls_active == true || msta.Bits.RA_PROBLEM)
        {
            clear_buttons(pmr);
            if (msta.Bits.RA_PROBLEM)
            {
                pmr->stop = 1;
                MARK(M_STOP);
            }
        }
    }
    else
        pmr->movn = 1;

And this was changed here:

commit 303a9208e368ef07839afff67bac69e429af7332
Author: Ron Sluiter <rsluiter@users.noreply.github.com>
Date:   Fri May 23 15:39:57 2014 +0000

    - Set "stop" field true if driver returns RA_PROBLEM true.
    - Increment version to 6.9

So my suggestion would be to
a) revert that commit in your motor code base
  (git revert 303a9208e368ef07839afff67bac69e429af7332)
b) Remove the line in the model 3 driver that sets the PROBLEM bit

HTH
/Torsten



On 06/07/18 17:13, Mark Davis wrote:
> Working with some new equipment here and having some problems getting things 
> working correctly and hoping someone can give me some tips on what to look for.
>
> We have a couple of new Delta Tau Power Brick LV-IMS controllers that I need to 
> get working ASAP.
>
> Here is where I am at the moment:
>     - I have an EPICS 3.14.12.2 IOC running the following:
>         - The powerPMAC driver from 
> http://www.observatorysciences.co.uk/deltatau_downloads.php,
>         - The ssh2 library as described in the manual for the driver
>         - Version 6-9 of the motor module
>         - Version 4-23 of the asyn module
>         - The records included in the example/test IOC config included with the 
> driver (which includes a motor record)
>     - The latest PowerMAC IDE application from Delta Tau running on a Windows 7 
> computer
>     - A FESTO stepper motor (Model EMMS-ST-57-S-SEB-G2), which includes a brake 
> and an encoder
>     - The appropriate power supply from Delta Tau to provide the power for the 
> controller and (through it) the motor
>     - A very basic configuration/program running on the Delta Tau (one which I 
> was told is basically a copy of one of the examples provided by Delta Tau) 
> which includes definitions for 8 open-loop axes (where the reported position is 
> based on the # of steps) and 8 read-only virtual axes that report the position 
> based on the encoder counts.
>
> Both the IOC and the IDE application can connect to the Delta Tau, and I can 
> open a command-line shell to it via SSH (the controller runs some version of 
> Linux)
>
> If the IOC is not running, when I use the IDE application to jog the motor it 
> will move until it reaches the the specified position.
>
> When the IOC is running, if I change the setpoint for the motor record, the 
> motor moves a bit in the correct direction but quickly stops.  This is 
> presumably because the PROBLEM bit in the motor record's MSTA field is always 
> set so the motor record is in the major STATE alarm state, causing the record 
> logic to stop the motor movement.
>
> The brief description of the PROBLEM bit simply says "driver stopped polling, 
> or hardware problem".  But fields in the motor record are frequently updating, 
> showing any changes in the position of the motor, so I wouldn't think it would 
> be a polling problem.  And the fact that everything appears to work fine when 
> using the IDE application would seem to imply that there are no hardware problems.
>
> I will be digging through the code trying to trace this back to some 
> identifiable cause, but if anyone has any experience with this controller and 
> has some tips on what I might check or should be looking for, that would be 
> much appreciated.
>
> Mark Davis
> NSCL/FRIB
>
>
>     - The ssh2 library
>
>
> I have this driver installed on a Linux box running EPICS 3.14.12.2 and (as 
> explained in the doco for this driver) installed the libssh2 library it needs 
> to open an SSH connection to the controller.
>
> I know almost nothing about the controller itself, but was provided with a very 
> basic "project" with some code that I was told is mostly a copy of some example 
> from Delta Tau.
>
> I have