Jumps in hi res data Aug10

07aug10

05aug10:Test data on telescope
Running the pdev diagnostics
Summary of problem
Resolution


Terminology
  • hi-res mode.
    • When the digital low pass filter in the fpga is enabled. used whenever the bandwidth is reduced from 172Mhz.
  • "sp":
    • code in fpga that does the spectral computations: (dlpf,pfb,fft, power,accumulation).
  • pncode:
    • pseudo noise code generated in the fpga. This data is then sent directly out of the fpga. It does not pass thru the "sp" fpga code
  • fpga,spectrometer,fileserver:
    • The mock spectrometer consists of a mock box (fpga, powerpc (ppc) spectrometer) connected to a file server
    • The fpga contains the pncode and sp code. It inputs the data, computes the spectra and outputs to the ppc (which runs in the mock box.
    • spectrometer: this usually refers to the ppc code running on the mock box (the prun program). It inputs data from the fpga and sends it to the file server.
    • the file server connects to the mock box via gigabit ethernet.
  • pnet,prun,psrv Programs:
    • pnet runs on a fileserver and controls all of the mock spectrometers in use.
    • prun runs on the power pc in the mock box. It controls the fpga. Pnet talks to prun.
    • psrv runs on the files server. It gets the data from prun and either verifies the data or writes it to disc.
    • normally refers to the power pc portion of the mock box. The fpga sends data to spectrometer which passes it on to the file server:

aug10 spectral channel jumps (top)


    a2463 had jumps (dropouts and increases) in spectral channels during their xband observations. Their setup was:
    The problems were found in the xband data:

05aug10 test data on telescope. (top)

    Test data was taken on 05aug10 using xband. The system configuration was similar to a2463 with the following changes:
The plots show 1 second spectra for b1s1g1 with jumps in polA (.ps) (.pdf):


Running the pdev diagnostics:  (top)

    The pdev diagnostics were run to try and track down the problem.
    Data was taken on :

06aug10 results: (top)


06aug10 Some of the testing
testing
bw
integration
status
all groups
172
1 sec
no errors
group 0:14bands
86
1 sec
ok
grp 1:
b0s0g1
b1s0g1

b2s1g1
b3s1g1
b4s1g1
b6sxg1
86
1sec
ok
b0s1g1
86
1sec
b0s0g0: Err with b0s1g1 byte 661754 got 1d exp 9d
b0s0g0: Err with b0s1g1 byte 663275 got 08 exp 0a
b0s0g0: Err with b0s1g1 byte 664003 got 0e exp 0f
b0s0g0: Err with b0s1g1 byte 667514 got 06 exp 86
b0s0g0: Err with b0s1g1 byte 688707 got 0c exp 0e
b0s0g0: Err with b0s1g1 byte 690139 got 0e exp 0f
b0s0g0: Err with b0s1g1 byte 692707 got 08 exp 0a

b2s0g1


fails
b3s0g1


fails but not very often. need to stop,restart
then failed a bunch
b1sxg1


no failure yet
b1s1g1
86
1
b0s0g0: Err with b5s1g1 byte 136515 got 07 exp 06
b0s0g0: Err with b5s1g1 byte 136851 got 0d exp 0c
b0s0g0: Err with b5s1g1 byte 137219 got 07 exp 06
b0s0g0: Err with b5s1g1 byte 137955 got 0b exp 0a
b0s0g0: Err with b5s1g1 byte 139395 got 07 exp 06

b4s0g1
b1s1g1


failed
b0s1g1


failed
b5s1g1


failed

b1s0g1
86

b0s0g0: Err with b1s0g1 byte 10059 got 0e exp 0c
only once

09aug10 test results:  (top)

    Different configuration permutations were tried on 09aug10:

Used
dec
Ts_noise
hr_shft
fcnt
Notes

g0,g1
none
-
172Mhz
21000
no errors

g0,g1
4
65535
11
5250
many errors. Looks like never got started correctly

g0sXg0
4
65535
11
5250
runs ok. As long as only 1 box, dec 4 seems to always start ok.

g0
2
32767
12
10500
10mb no errors

g1
2
32767
12
10500
4.9mb then b1s1g1 too many errors. psrv running on pdevs8

gs0g0,g1
2
32767
12
10500
3.5 mb then b2s0g1 too many errors. psrv running on pdevs1




summary of problems  (top)


Looking at data on the sky: 

Test pattern data

both



Resolution (top)

    The problem goes away when the clock frequency is reduced from 172.032 Mhz to somewhere around 160 to 165 Mhz. It must be thtat the fpga is having trouble at the higher clock frequency.

<- page up
home_~phil