Project

General

Profile

Feature #18631

IPM State does not change on I33

Added by Denton Morris over 1 year ago. Updated 7 months ago.

Status:
Closed
Priority:
Normal
Start date:
12/28/2017
Due date:
% Done:

100%

Estimated time:
Spent time:
Duration:

Description

During the activation of a spec the state of the IPM never changes on I33 (upper right hand corner). The state properly changes when observed on a parameter page or the datalogger.

GxSA_Fast_Time_Plot.gif (6.04 KB) GxSA_Fast_Time_Plot.gif Elliott McCrory, 12/28/2017 11:41 AM
2017-12-28.11.53.01.138.png (8.12 KB) 2017-12-28.11.53.01.138.png Denton Morris, 12/28/2017 11:53 AM
CaptureActSpec.JPG (74.2 KB) CaptureActSpec.JPG David Slimmer, 01/02/2018 10:03 AM
2018-01-05.08.34.37.192.png (4.77 KB) 2018-01-05.08.34.37.192.png Correct IPM state reading Denton Morris, 01/05/2018 08:35 AM
2018-01-05.08.34.30.927.png (9.84 KB) 2018-01-05.08.34.30.927.png Broken IPM state reading Denton Morris, 01/05/2018 08:35 AM

History

#1 Updated by Elliott McCrory over 1 year ago

When I tried this, the value of "F.E. Active Spec" successfully changed on the application page (using MI H 104). There was a delay of about 2 seconds from when the 1Hz FTP of this device (I:4HMAS) changed and when the data on the application changed. This could account for what you observed.

#2 Updated by Denton Morris over 1 year ago

Elliott McCrory wrote:

When I tried this, the value of "F.E. Active Spec" successfully changed on the application page (using MI H 104). There was a delay of about 2 seconds from when the 1Hz FTP of this device (I:4HMAS) changed and when the data on the application changed. This could account for what you observed.

That may be but I should have mentioned that the IPM state always reads zero, when it's most often in state 4 (data acquired) for extended periods of time following a successful acquisition of data.

Side note: It looks like the state definitions have been deleted? State numbers are still reported but viewing via D88 shows no state info.

#3 Updated by Elliott McCrory over 1 year ago

Is the readback from the state device of the IPM not working? For example, running Spec 11 on "MI H 104" and plotting the spec number and the IPM status versus time, attached.

#4 Updated by Denton Morris over 1 year ago

Elliott McCrory wrote:

Is the readback from the state device of the IPM not working? For example, running Spec 11 on "MI H 104" and plotting the spec number and the IPM status versus time, attached.

I plotted the state V:4RHIPM and it updates fine. See attached. The state parameter also updates on the parameter page at a 1Hz rate just fine.

#5 Updated by David Slimmer over 1 year ago

  • File CaptureActSpec.JPG CaptureActSpec.JPG added
  • Tracker changed from Bug to Feature
  • Status changed from New to Work in progress
  • Assignee changed from David Slimmer to Elliott McCrory
  • % Done changed from 30 to 60

The active spec device R:4HMAS is set to a non-zero value from the time the spec is activated to the time the spec completes acquisition and analysis. Typically this takes under 3 secs, so this device must be read almost immediately after the spec is activated for the device to contain the value of interest.

#6 Updated by Elliott McCrory over 1 year ago

Trying to fix this now.

#7 Updated by Elliott McCrory over 1 year ago

  • Status changed from Work in progress to Resolved

#8 Updated by Denton Morris over 1 year ago

Why is this marked resolved? It is still broken. It still reads zero 100% of the time on I33 (it's real reading is "4" 99% of the time because it's in a state of completion with "data ready".)

Also the state definitions are no longer available via ACNET states facility?

#9 Updated by Elliott McCrory over 1 year ago

  • % Done changed from 60 to 100

This issue has been fixed and is waiting for go-ahead to "mecca --commit" from MI folks.

In general, I delay doing a commit until I get approval from the "customer"

#10 Updated by Denton Morris over 1 year ago

Elliott McCrory wrote:

This issue has been fixed and is waiting for go-ahead to "mecca --commit" from MI folks.

In general, I delay doing a commit until I get approval from the "customer"

Have you tested it on the Z page? The current Z build (2017-1-5 08:55) still does not work. If you think it's fixed please put the working version on the Z index until it's committed to I33.

#11 Updated by Elliott McCrory 7 months ago

  • Status changed from Feedback to Closed


Also available in: Atom PDF