sPHENIX discussion of the superconducting solenoid

[Kin Yip <kinyip AT bnl.gov>]

Hi,

1.   I've placed the signed new work permit for this low-field test in a folder that contains the generic work permit, which is on the table next to the concrete
      structure/wall surrounding the sPHENIX Magnet.   Inside, it also contains the UNOFFICIAL copy of the two travelers.

      Ray Ceruti holds the OFFICIAL travelers which are meant to be filled out.

2.  The Cryo group has been doing the intial pressure test/leak checking in which they are probably tightening up the tap here or there ....
     They'll invite Mike Gaffney  to come to witness later today or tomorrow.

3.  You may read the email communication below.   Piyush told me that he found the temperatures to be too high after using the constants (in the quadratic equation that I
     had provided the other day) such as ~310 K.
     I suddenly vaguely remembered what Paul Giannotti told me that Yi Zhang had similar problem.  After I checked with her, she told me that the 1 mA current
     source was off by ~1%.    Piyush measured it and indeed it's 1.01 mA (1%  too high).   Putting this correction in, now 2 sensors TT02/TT03 show ~292K whereas
     TT01 still shows ~303 or something.    We did check the voltages (and thus resistances) by measuring directly.

     Apparently, some other temperature sensors that the Cryo group uses (near the coil) also have shown very similar temperatures.   We're puzzled by the 10 K
     difference and in T. Tallerico and P. Joshi's discussion, it seems that the long cable may have capacitance (inductance) which affect the voltage measurements
     (and thus the resistance


So hopefully, if the pressure test/leak check can finish tomorrow, we can do the hypot/impulse test tomorrow.

Kin

On 02/04/2016 11:47 AM, Kin Yip wrote:

Hi Yi,

Thanks a lot !   I tried 295 ohm in one case and I'd get 310.318 K.  Just 0.9% (ie. 0.991 => 292.345 ohm), I'll get 299.408 K.

This definitely sounds a plausible and workable explanation.   I'll try to find Piyush to see whether we can exactly what you have said.

Thanks again ! 

Kin

On 02/04/2016 11:02 AM, Zhang, Yi wrote:

I had the same problem when reading CLTS-2B sensors when we first started. All the CLTS-2B temperatures were too high.

 

Then I found out, after struggled for a long time, that the 1mA excitation current are not exactly 1mA, see highlighted box in the attached signal list. Instead of exactly 1mA, a typical constant current source has a ~0.9% error. This small error causes about 10 degree K in temperature readings.

 

So the solution is to measure your excitation current with an accurate meter (we have one if you need to borrow), then use the measured actual current as a multiplier to your resistance in the “resistance to temperature” curve you created as shown below:

 

I’m writing this email in a hurry(troubleshooting at sPhenix) without going into any details of the calculation. So call me at 4967 or email me if you have any further questions.

Thanks,

Yi

 

From: Yip, Kin
Sent: Thursday, February 04, 2016 9:01 AM
To: Zhang, Yi; Giannotti, Paul
Cc: Joshi, Piyush
Subject: TT01/TT02/TT03 needs correction ?

 

Dear Zhang Yi,

 

This is Kin Yip and I’ve got a question about the sPHENIX Magnet

 

Paul Giannotti has helped connect the 3 temperature signals TT01/TT02/TT03 which are supposed to be used for temperature correction to the

strain gauge.    As you may read below, I found some calibration results (3 points) in the file : "2.4 Gauges & Instrument Connections.pdf"

of our BaBar server :

\\bnl.gov\c-adnas\babar\Volume 2 DESCRIPTION OF COMPONENTS/2.4 GAUGES AND INSTRUMENT CONNECTIONS,

I put it in snapshot6.png (as attached).   I do a quadratic fit for those 3 points and obtained the quadratic equation coefficients.  If I put back

the resistances in those 3 points, I get the exact temperature (within 0.0001 or something like that).   { The quadratic equation

 

But yesterday when our colleague Piyush plugged in my coefficients and measure the constant, he found temperatures : 314K, 309K, 308K

which are 41 C, 36 C and 35 C.  They seem too high/wrong.

 

Assuming the resistances are all measured correctly and coefficients are put in correctly,   is there any special corrections that we may need to do/apply ?

 

I ask you because Paul told me a few days ago that you’ve met something similar and you somehow knew how to correct for it ….

 

Can you please comment ?

 

Kin

 

From: Joshi, Piyush
Sent: Wednesday, February 03, 2016 8:48 PM
To: Yip, Kin <kinyip AT bnl.gov>
Subject: RE: [Sphenix-magnet-l] ohm to temperature (K)

 

Kin,

Today I plugged in the constants “a, b, c” in the quadratic equation for the three sensors and exited it with 1mA constant current. From the read back voltages,  My program calculated :  314K, 309K, 308K which are way off from what I believe should be around 295K which is ambient temp in 912. May be they will get better at colder temp..

Piyush

From: sphenix-magnet-l-bounces AT lists.bnl.gov [mailto:sphenix-magnet-l-bounces AT lists.bnl.gov] On Behalf Of Kin Yip
Sent: Tuesday, February 02, 2016 7:25 PM
To: sphenix-magnet-l AT lists.bnl.gov
Subject: [Sphenix-magnet-l] ohm to temperature (K)

 

Hi Piyush,

Have you figured out how to get the temperature from ohm or voltage ?    The ohm or voltage should be the same thing if you have "constant
current" (especially high-precision).

Paul Giannotti told me to look at 2.4.   So, in \\bnl.gov\c-adnas\babar\Volume 2 DESCRIPTION OF COMPONENTS/2.4 GAUGES AND INSTRUMENT CONNECTIONS,

there is  the "2.4 Gauges & Instrument Connections.pdf"

There are 369 pages with many nos. and curves.   I'm not sure whether the latter stuff are relevant (as they refer to various instrument).

But on p.7-11,  they have 3 resistance and 3 temperatures (temperatures are 4.2 K to 77.4 K to 295 K).

I took a snapshot6.png.    Since there are 3 points, I can do a quadratic fit which almost looks linear (but not quite).   I show you the
fit on TT01.gif for TT01.

So, temperature (K) = a + b*ohm + c*ohm^2 and the following are the coefficients from my fit :

                a                b                c
TT01: -1017.35 , 5.05676 , -0.00187896

TT02: -916.28 , 4.22621 , -0.000191854

TT03: -868.795 , 3.88358 , 0.000384459

I've assumed the resistances  are in ohm.  


I hope this helps.

Kin

Attachment: pngqtTn8ew2YW.png
Description: PNG image