fb_epid: I/O Intr not valid (no DSET)

[prjemian]

from prjemian/epics-docker#43

after iocInit: PV: xxx:epid1:outpv scanAdd: I/O Intr not valid (no DSET)

The code in the fb_epid.db database has this:

grecord(sseq, "$(P):outpv") {
        field(DESC, "follow OUT PV name")
        field(SCAN, "I/O Intr")
        field(DOL1, "$(P):out.OUTN")
        field(DLY1, "0")
        field(LNK1, "$(P):resume.INBN")
}

The SCAN field is triggering the message above.

[timmmooney]

This is the expected result of setting the SCAN field of a sseq record to "I/O Intr". I/O Intr scanning requires device support that has implemented code described here: https://epics.anl.gov/base/R3-14/12-docs/AppDevGuide/node18.html#12357. The sseq record has no device support that I know of. Tim Mooney ***@***.***) (630)252-5417 Beamline Controls Group (www.aps.anl.gov) Advanced Photon Source, Argonne National Lab

…

________________________________ From: Pete R Jemian ***@***.***> Sent: Thursday, May 13, 2021 3:58 PM To: epics-modules/optics ***@***.***> Cc: Subscribed ***@***.***> Subject: [epics-modules/optics] fb_epid: I/O Intr not valid (no DSET) (#10) from prjemian/epics-docker#43<prjemian/epics-docker#43> after iocInit: PV: xxx:epid1:outpv scanAdd: I/O Intr not valid (no DSET) The code in the fb_epid.db<https://github.com/epics-modules/optics/blob/master/opticsApp/Db/fb_epid.db> database has this: grecord(sseq, "$(P):outpv") { field(DESC, "follow OUT PV name") field(SCAN, "I/O Intr") field(DOL1, "$(P):out.OUTN") field(DLY1, "0") field(LNK1, "$(P):resume.INBN") } The SCAN field is triggering the message above. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub<#10>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ABT6PF7BMRQYR5BUNKIMETTTNQ4NTANCNFSM443HTLTA>.

[MarkRivers]

Note that the swait record in that database also have SCAN=I/O Intr. Is that also an error?

How did this database ever work?

Should these lines all be deleted so that it defaults to SCAN=PASSIVE?

[keenanlang]

From the comments in the code, it appears that the intended function would be to set

$(P):outpv.SCAN = PASSIVE

and

$(P):out.FLNK = "$(P):outpv PP"

[prjemian]

I could try to test this in my simulation IOC by modifying the database locally.

[prjemian]

Looks as if this fb_epid support is ripe for an overhaul. We could compare with current use at 33ID.

[prjemian]

Could be tested by constructing a temperature controller simulation. Here's an example of that using an swait record (as setup for bluesky):

        calcs.calc8.reset()  # remove any prior configuration
        self.description.put("temperature")
        self.report_dmov_changes.put(report_dmov_changes)
        self.previous_value_pv.put(self.readback.pvname)
        if setpoint is not None:
            self.setpoint.put(setpoint)
            self.readback.put(setpoint)
        self.noise.put(noise)
        self.max_change.put(max_change)
        self.tolerance.put(tol)
        self.scanning_rate.put(rate)  # 1 second
        self.calculation.put("A+max(-D,min(D,(B-A)))+C*(RNDM-0.5)")

[prjemian]

Also, there's no button for fb_epid.adl on the XXX green button screen. $(P)epid1

[prjemian]

Could be tested by constructing a temperature controller simulation.

Indeed: such simulation is noted in the database:

optics/opticsApp/Db/fb_epid.db

Lines 236 to 270 in fdf5bc3

	#--------------------------------------------
	# simulate a temperature sensor and heater
	#----------------------
	# A : minimum "temperature" allowed
	# B : cooling rate parameter
	# C : heater power
	# D : output of PID loop
	# E : heater relay closes when D > E
	# F : current "temperature"
	#----------------------
	# PID loop terms:
	# $(P):in.INAN = "$(P):sim"
	# $(P):out.OUTN = "$(P):sim.D"
	# $(P):enable.INAN = "$(P):on.VAL"
	# $(P).KP = "0.01"
	# $(P).KI = "0.1"
	# $(P).I = "0.0"
	# $(P).KD = "0.0"
	# $(P).DRVL = "0.0"
	# $(P).DRVH = "1.0"
	# $(P).FMOD = "PID"
	#----------------------
	grecord(swait, "$(P):sim") {
	field(DESC, "EPID Temperature Controller simulator")
	field(PREC, "5")
	field(SCAN, ".1 second")
	field(CALC, "max(A,F*(1-B)+C*D*G)")
	field(A, "0.1") # minimum "temperature"
	field(B, "0.05") # cooling rate parameter
	field(C, "30") # heater power
	field(D, "0") # driven by epid loop
	field(E, "0.8") # relay close threshold
	field(INFN, "$(P):sim.VAL")
	field(INGN, "$(P).FBON")# is feedback ON?
	}

[prjemian]

Time for an overhaul of the fb_epid support.