Re: second relative motion stuck in a do loop

[Kevin Peterson via Tech-talk <tech-talk at aps.anl.gov>]

Phi,

I'm not able to reproduce the problem of the motors staying at 1 on 
subsequent 1-unit relative moves.
I do see errors in the positions after the 2nd move following a reset of 
motor positions to zero.  Here is the camonitor output for three 
relative 2-unit moves:
$ camonitor kmp3:test:m1.VAL kmp3:m{1,2,3}.VAL
kmp3:test:m1.VAL               2023-11-14 09:27:07.187920 0
kmp3:m1.VAL                    2023-11-14 09:27:07.122334 0
kmp3:m2.VAL                    2023-11-14 09:27:07.155012 0
kmp3:m3.VAL                    2023-11-14 09:27:07.274250 0

kmp3:m1.VAL                    2023-11-14 09:27:41.662214 2
kmp3:m2.VAL                    2023-11-14 09:27:41.662305 2
kmp3:m3.VAL                    2023-11-14 09:27:41.662371 2
kmp3:test:m1.VAL               2023-11-14 09:27:41.662412 2

kmp3:m1.VAL                    2023-11-14 09:27:50.711933 6
kmp3:m2.VAL                    2023-11-14 09:27:50.712022 6
kmp3:m3.VAL                    2023-11-14 09:27:50.712092 6
kmp3:test:m1.VAL               2023-11-14 09:27:50.712132 4
kmp3:m1.VAL                    2023-11-14 09:27:53.170613 4.21
kmp3:m2.VAL                    2023-11-14 09:27:53.170687 4.21
kmp3:m3.VAL                    2023-11-14 09:27:53.170748 4.21

kmp3:m1.VAL                    2023-11-14 09:28:05.077614 10.21
kmp3:m2.VAL                    2023-11-14 09:28:05.077647 10.21
kmp3:m3.VAL                    2023-11-14 09:28:05.077673 10.21
kmp3:test:m1.VAL               2023-11-14 09:28:05.077689 6
kmp3:m1.VAL                    2023-11-14 09:28:07.302760 6.21
kmp3:m2.VAL                    2023-11-14 09:28:07.302867 6.21
kmp3:m3.VAL                    2023-11-14 09:28:07.302960 6.21

Note: kmp3:test:m1 is my soft motor and kmp3:m{1,2,3} are my "real" motors.

I suspect the error that is introduced is much smaller for you because 
the galil support using a much higher moving poll frequency that my 
simulated motors.
Changing the $(P)-F fanout so that it outputs to the real motors' VAL 
fields instead of their RLV fields remains the simplest solution.  This 
is what the same series of 2-unit relative moves looks like after that fix:
$ camonitor kmp3:test:m1.VAL kmp3:m{1,2,3}.VAL
kmp3:test:m1.VAL               2023-11-14 09:21:57.513328 0
kmp3:m1.VAL                    2023-11-14 09:21:28.188527 0
kmp3:m2.VAL                    2023-11-14 09:21:28.232332 0
kmp3:m3.VAL                    2023-11-14 09:21:28.264741 0

kmp3:m1.VAL                    2023-11-14 09:22:08.629501 2
kmp3:m2.VAL                    2023-11-14 09:22:08.629531 2
kmp3:m3.VAL                    2023-11-14 09:22:08.629542 2
kmp3:test:m1.VAL               2023-11-14 09:22:08.629561 2

kmp3:m1.VAL                    2023-11-14 09:22:14.873323 4
kmp3:m2.VAL                    2023-11-14 09:22:14.873386 4
kmp3:m3.VAL                    2023-11-14 09:22:14.873428 4
kmp3:test:m1.VAL               2023-11-14 09:22:14.873488 4

kmp3:m1.VAL                    2023-11-14 09:22:26.848583 6
kmp3:m2.VAL                    2023-11-14 09:22:26.848654 6
kmp3:m3.VAL                    2023-11-14 09:22:26.848695 6
kmp3:test:m1.VAL               2023-11-14 09:22:26.848722 6

If you really need the real motors to do relative moves, you'll need to 
add records in between the soft motor and the fanout to convert the 
absolute position, which is written to the PV in soft motor's OUT field, 
into a relative displacement.  I don't have good suggestions how to 
implement that.
Kevin


On 11/13/23 16:46, Tran, Phi Dung wrote:
> Hi Kevin,
>
> I am using my Soft Channel $(P) to move my three real motors by relative 
> motions.
> I set the field(RTRY,0) in $(P).  Initially, I set all three real motor 
> positions or xxx.VAL to zero.
> Next, I caput $(P)  1 for relative motion for the real motors.  All real 
> motors moved to 1+0 from 0 for relative move, xxx.RLV.
> _However, all subsequent caput $(P) 1 will not cause the real motors to 
> move from 1.  They stayed at 1 instead of 1+1?_
> If I set all three real motor positions Xxxx.VAL to zero again.
> First caput $(P) 2.  All three real motors moved relatively to 2+0 from 0.
> _Second caput $(P) 2.  All three real motors moved relatively to 
> 2.000,2.001,2.002 not 4,4,4._
> Third caput $(P) 2.  All three real motors moved relatively to 
> 4.000,4.001,4.002.
> All subsequent caput $(P) 2, correctly moved the motors to the 
> additional 2 units relative to the current positions.
> Please advice.
> Thank you,
> Phi