[Dsg-hallb_magnets] cernox sensors 325K error
Tyler Lemon
tlemon at jlab.org
Fri Feb 10 10:42:54 EST 2017
Is there some advantage to running the start-up algorithm for a Cerenox over using a fixed excitation voltage? It seems to me that using a fixed excitation voltage would be best also, as Rama has mentioned that all other sensors use a fixed voltage/current (below in blue).
As Brian said, it would be easy enough to swap 9870 modules to see if the errors still occur.
Renuka,
Ruben mentioned that you may have a spare 9870 cRIO module, where would the spare be stored?
Thanks,
Tyler
From: "Ramakrishna Bachimanchi" <bachiman at jlab.org>
To: "Tyler Lemon" <tlemon at jlab.org>
Cc: "Nicholas Sandoval" <sandoval at jlab.org>, "Brian Eng" <beng at jlab.org>
Sent: Friday, February 10, 2017 10:03:07 AM
Subject: Re: [Dsg-hallb_magnets] solution to Hall B Magnet LV Excitation Chassis Cerenoxes reading 325 K
Hi Tyler,
this is not the issue with the chassis or cRIO. It's the issue with the
RS232 module that is being used for communicating with LV chassis. There
is intermittent data loss in the RS232 module and it is causing the
readback to go to 0 which will show resistance as 0 and the temperature
is 325K. It happens with other sensors too. That's the reason Nick looks
for 3 consecutive samples before acting on it. Difference between all
the other sensors and cernox is, all the other sensors have fixed
excitation voltage/current. That's the reason they come back by the next
read cycle. Cernox algorithm sets the excitation based on the readback
and once it goes to 325K, it never recovers. By running the start up
algorithm in the case of 325K readback, it will fix this issue. Also, we
didn't notice the data drop issue, when we were testing with only one
chassis on the bench.
Please let me know, if you need more information
Thanks,
Rama
From: "Brian Eng" <beng at jlab.org>
To: "Tyler Lemon" <tlemon at jlab.org>
Cc: "dsg-hallb magnets" <dsg-hallb_magnets at jlab.org>, "Dave Kashy" <kashy at jlab.org>
Sent: Friday, February 10, 2017 10:27:22 AM
Subject: Re: [Dsg-hallb_magnets] cernox sensors 325K error
Since the raw values are the same it isn't the software conversion, so at least that's been eliminated as possible cause.
1. I have no idea how to debug this, but we have seen a failure of a chassis before. Rama would know better on this one.
2. This would be pretty easy to test: just swap the 9870 modules on the cRIO. Since the failures have only been on chassis 1 & 4 they're both on the first module. If the failures follow the module problem found and if it is still chassis 1 & 4 then it is probably upstream of the cRIO (chassis itself, cabling, etc).
3. While this is a possibility I think it's pretty unlikely, we already know from testing that the cRIO can recover from having a sensor unplugged and plugged back in, so even if it was an intermittent short it shouldn't cause it to get stuck, it isn't a complete sensor/cable failure since restarting the cRIO recovers the system.
While it wasn't approved last time the Cernox sensors were misbehaving (partly because it stopped happening) I'm still of the opinion the code should just use a fixed voltage (using only V1 & I1) instead of power (but still use V1/V2 & I1/I2 to calculate the temperature) since this is basically what the startup algorithm does anyway, not to mention that's how the sensors were calibrated from the factory.
> On Feb 10, 2017, at 8:28 AM, Tyler Lemon <tlemon at jlab.org> wrote:
>
> Hello,
>
> The LabVIEW shows that the raw hexadecimal value for V1 and V2 from the LV Chassis are the same. Since December, the 325K error has happened seven times. Five of seven have all been Coil B Cerenoxes in LV Chassis 4; remaining two have been in VCL Cerenoxes in Chassis 1.
>
> This above could mean any of the below:
> • Something wrong in LV Chassis: five of seven sensors with 325K error were in same Chassis. Since we are reading back the same raw hexadecimal value for V1 and V2 when the error occurs, it could be a problem internal to the LV Chassis.
> • cRIO module is bad: Since the majority of the 325K errors are from the same LV Chassis, the port on the cRIO module could be bad and causing incorrect information to be sent/received.
> • Sensor cable is bad: The 325K error will also happen if the cable is broken or shorted
> In the mean time, the LV cRIO has been reset and I have also moved the latching indicators to show whether what is written to the LV Chassis by LabVIEW is the same when the 325K error occurs.
>
> Best regards,
> Tyler
>
> From: "Ruben Fair" <rfair at jlab.org>
> To: "David Kashy" <kashy at jlab.org>
> Cc: "Tyler Lemon" <tlemon at jlab.org>, "dsg-hallb magnets" <dsg-hallb_magnets at jlab.org>
> Sent: Friday, February 10, 2017 7:55:00 AM
> Subject: Re: [Dsg-hallb_magnets] cernox sensors 325K error
>
> Ok - let's see what Tyler learns from the LV chassis output.
>
> Ruben
>
> From: "David Kashy" <kashy at jlab.org>
> To: "Ruben Fair" <rfair at jlab.org>
> Cc: "Tyler Lemon" <tlemon at jlab.org>, "dsg-hallb magnets" <dsg-hallb_magnets at jlab.org>
> Sent: Friday, February 10, 2017 7:49:11 AM
> Subject: Re: [Dsg-hallb_magnets] cernox sensors 325K error
>
> We have a 325 on coil B now
>
>
>
> ----- Original Message -----
> From: "Ruben Fair" <rfair at jlab.org>
> To: "Tyler Lemon" <tlemon at jlab.org>
> Cc: "dsg-hallb magnets" <dsg-hallb_magnets at jlab.org>, "kashy" <kashy at jlab.org>
> Sent: Thursday, February 9, 2017 4:50:28 PM
> Subject: Re: [Dsg-hallb_magnets] cernox sensors 325K error
>
> Ok Tyler - sounds like a reasonable approach.
>
> Regards
>
> Ruben
>
>
>
> From: "Tyler Lemon" <tlemon at jlab.org>
> To: "dsg-hallb magnets" <dsg-hallb_magnets at jlab.org>, "David Kashy" <kashy at jlab.org>
> Sent: Thursday, February 9, 2017 4:48:11 PM
> Subject: [Dsg-hallb_magnets] cernox sensors 325K error
>
> Hello Ruben and Dave Kashy,
>
> Before implementing the LabVIEW that will automatically reinitialize a Cerenox sensor if it reads 325K, a version will be deployed to look at the raw hexadecimal text output from the LV Chassis. As Peter mentioned below, this would tell us if the 325K error is caused by the LV Chassis or some conversion done in LabVIEW.
>
> Since we cannot force the 325K error, we will have to wait for the error to occur to compare the raw LV Chassis output.
>
> Best regards,
> Tyler
>
>
>
> From: "Amrit Yegneswaran" <yeg at jlab.org>
> To: "dsg-hallb magnets" <dsg-hallb_magnets at jlab.org>
> Sent: Thursday, February 9, 2017 2:44:31 PM
> Subject: Re: [Dsg-hallb_magnets] cernox sensors
>
> thanks peter and tyler.
>
>
> From: "Peter Bonneau" <bonneau at jlab.org>
> To: "Amrit Yegneswaran" <yeg at jlab.org>
> Sent: Thursday, February 9, 2017 2:16:49 PM
> Subject: cernox sensors
>
> Hi Amrit,
>
> I looked at the LV cRio code with Tyler.
>
> They don't seem to know if the excitation chassis is the problem or the
> conversions, etc in the code.
>
> The data is sent to cRio as text via serial communications then
> immediately changed to a number for the calculations & lookup table.
>
> We've seen a lot of errors in serial communications in other
> instrumentation.
>
> I suggested to Tyler that we latch the received text numbers V1 and V2
> when this error occurs before the conversion to numbers.
>
> That will tell us if the excitation chassis is not sending or we're not
> receiving the raw data correctly.
>
> Thanks,
>
> Peter
>
> _______________________________________________
> Dsg-hallb_magnets mailing list
> Dsg-hallb_magnets at jlab.org
> https://mailman.jlab.org/mailman/listinfo/dsg-hallb_magnets
>
> _______________________________________________
> Dsg-hallb_magnets mailing list
> Dsg-hallb_magnets at jlab.org
> https://mailman.jlab.org/mailman/listinfo/dsg-hallb_magnets
>
> _______________________________________________
> Dsg-hallb_magnets mailing list
> Dsg-hallb_magnets at jlab.org
> https://mailman.jlab.org/mailman/listinfo/dsg-hallb_magnets
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/dsg-hallb_magnets/attachments/20170210/4f686ac1/attachment-0001.html>
More information about the Dsg-hallb_magnets
mailing list