JAWS: enforce minimum and maximum gap values

[FreddieAkeroyd]

As a user on SXD and elsewhere it would be useful to have a minimum and maximum gap setting enforced in the jaws to avoid clashes. You can set DRVH and DRVL fields on HGAP:SP and VGAP:SP but these only work if you set the gap PV directly, if you move an individual motor blade then you may end up with HGAP / VGAP readback going less these DRVH and DRVL values and causing a clash.

• decide if separate PVs are needed for min/max, or just look at existing DRVH and DRVL on gap SP and enforce these more generally. mingap == maxgap means do not enforce
• checking SPs in the jaws db will not catch setting values directly on motors. Y
• You could try and trap at the motor SP level (monitor motor .VAL in the jaws.db) or monitor the realtime readback (HGAP / VGAP). In either case you want to write to .STOP on the motor. Monitoring HGAP/VGAP probably needs less additional logic as the gap maths has been done for you, however monitoring motor SP will give more immediate action. It also depends on what a scientists wants to happen when they give too small a gap e.g. reject move (will happen if monitor SP) v move to smallest/largest allowed (monitor gaps live readback). Move to smallest/largest is consistent with drvh/drvl so may be best?

acceptance cirteria

• HGAP and VGAP cannot fall outside specified values

[Chsudeepta]

It seems that right now it does something similar as requested in the ticket. Either the HGAP:SP or the JE:SP or JW:SP values can be set (similarly for VGAP:SP, JN:SP and JS:SP) and limits are reset. So, if either of JE:SP or JW:SP are set a value, the HGAP:SP is recalculated to the gap between these values, and if HGAP:SP is given a value, both JE:SP and JW:SP are recalculated by taking the mid-point of the two values and separate them from this point by half the value of the HGAP. Then the values are restricted within these ranges.
However, I see that there is a possibility to set the VGAP or HGAP to negative, either by setting them to a negative value or by setting the JW:SP > JE:SP (similarly, when JS:SP > JN:SP). Is this the case that we need to restrict?

[FreddieAkeroyd]

However, I see that there is a possibility to set the VGAP or HGAP to negative, either by setting them to a negative value or by setting the JW:SP > JE:SP (similarly, when JS:SP > JN:SP). Is this the case that we need to restrict?

Yes, we need to allow specifying a minimum VGAP / HGAP so it cannot become too small (negative being an extreme case to too small).

[FreddieAkeroyd]

Add DRVL field to $(P)$(SLIT)$(GAP):SP which can be set by a macro $(MINGAP=0) so you can specify MINGAP in the configuration. This would cover setting gaps directly, to cover separate JN/JS we would need extra calc logic in $(P)$(SLIT)$(S1):SP and $(P)$(SLIT)$(S2):SP to enforce mingap between then

[FreddieAkeroyd]

So $(P)$(SLIT)$(S1):SP could update a new calcout record that enforces a min gap before writing to _SPALT.A etc. The other approach would be to try and update DRVL/DRVH of $(P)$(SLIT)$(S1):SP which would mean tracking the centre of the jaws, which may be a bit messier to do