what is the exact exposure pattern for TESS?

[davidwhogg]

Hey @zkbt : I am simulating a TESS lightcurve.

TESS takes data in 2-second exposures. What is the exact number I should replace "2.0000" with? Also, what is the dead time (if any) between the 2-second exposures?

I pay beer for answers.

[zkbt]

The time between adjacent sub-exposures will be 2.000000 seconds. According to Ed Morgan on the TESS team, the exposure time is defined as two pulses of 1 pulse-per second spacecraft clock. This clock isn't guaranteed to be exact and isn't synchronized with GPS, so a separate onboard "High Accuracy Clock" will be used to track any long-term drift between its exposure timestamps and real time systems. These drifts should be small compared to the barycenter correction, but are worth mentioning.

The CCD's are being read out as frame transfer arrays. The time to transfer pixels into the frame store will be something like 5 milliseconds. (This number will depend on the final properties of the flight detector and electronics, but this is the latest I've heard.) During this 5 milliseconds, the light from every star will be smeared out over its entire column, and this smear will be subtracted off in calibration. Unless something really crazy happens to a very bright star during that 5 milliseconds, this effectively means the cadence will be 1.995 seconds of exposure followed by 0.005 seconds of dead time.

Hope this helps! Will send a response on the Hogg 2.0 cosmic ray filter when I get the chance. It's pushing the memory limits, so will have to be really good to make it on board.

Cheers, Z

————————————

Zachory K. Berta-Thompson Torres Exoplanet Fellow MIT Kavli Institute 77 Massachusetts Ave., 37-673 Cambridge, MA 02139 phone: 617-253-5084 web: space.mit.edu/~zkbt

——————————__——

On Mon, Apr 20, 2015 at 12:11 PM, David W. Hogg notifications@github.com wrote:

Hey @zkbt https://github.com/zkbt : I am simulating a TESS lightcurve.

TESS takes data in 2-second exposures. What is the exact number I should replace "2.0000" with? Also, what is the dead time (if any) between the 2-second exposures?

I pay beer for answers.

— Reply to this email directly or view it on GitHub https://github.com/davidwhogg/HoneyComb/issues/25.

[zkbt]

Oops -- a factor of 10 correction to what I just said. The frame transfer time could be as large as 50 milliseconds (but will hopefully be significantly lower). In that worst case scenario, the exposure time would be 1.95 seconds and the dead time would be 0.05 seconds.

————————————

Zachory K. Berta-Thompson Torres Exoplanet Fellow MIT Kavli Institute 77 Massachusetts Ave., 37-673 Cambridge, MA 02139 phone: 617-253-5084 web: space.mit.edu/~zkbt

——————————__——

On Mon, Apr 20, 2015 at 5:57 PM, Zach Berta-Thompson zkbt@mit.edu wrote:

The time between adjacent sub-exposures will be 2.000000 seconds. According to Ed Morgan on the TESS team, the exposure time is defined as two pulses of 1 pulse-per second spacecraft clock. This clock isn't guaranteed to be exact and isn't synchronized with GPS, so a separate onboard "High Accuracy Clock" will be used to track any long-term drift between its exposure timestamps and real time systems. These drifts should be small compared to the barycenter correction, but are worth mentioning.

The CCD's are being read out as frame transfer arrays. The time to transfer pixels into the frame store will be something like 5 milliseconds. (This number will depend on the final properties of the flight detector and electronics, but this is the latest I've heard.) During this 5 milliseconds, the light from every star will be smeared out over its entire column, and this smear will be subtracted off in calibration. Unless something really crazy happens to a very bright star during that 5 milliseconds, this effectively means the cadence will be 1.995 seconds of exposure followed by 0.005 seconds of dead time.

Hope this helps! Will send a response on the Hogg 2.0 cosmic ray filter when I get the chance. It's pushing the memory limits, so will have to be really good to make it on board.

Cheers, Z

————————————

Zachory K. Berta-Thompson Torres Exoplanet Fellow MIT Kavli Institute 77 Massachusetts Ave., 37-673 Cambridge, MA 02139 phone: 617-253-5084 web: space.mit.edu/~zkbt

——————————__——

On Mon, Apr 20, 2015 at 12:11 PM, David W. Hogg notifications@github.com wrote:

Hey @zkbt https://github.com/zkbt : I am simulating a TESS lightcurve.

TESS takes data in 2-second exposures. What is the exact number I should replace "2.0000" with? Also, what is the dead time (if any) between the 2-second exposures?

I pay beer for answers.

— Reply to this email directly or view it on GitHub https://github.com/davidwhogg/HoneyComb/issues/25.

[zkbt]

(Realizing you probably prefer the predicted rather than worst case...):

Word just came from Roland Vanderspeck that it looks like current best estimate is 20 milliseconds of frame transfer time.

————————————

Zachory K. Berta-Thompson Torres Exoplanet Fellow MIT Kavli Institute 77 Massachusetts Ave., 37-673 Cambridge, MA 02139 phone: 617-253-5084 web: space.mit.edu/~zkbt

——————————__——

On Mon, Apr 20, 2015 at 6:16 PM, Zach Berta-Thompson zkbt@mit.edu wrote:

Oops -- a factor of 10 correction to what I just said. The frame transfer time could be as large as 50 milliseconds (but will hopefully be significantly lower). In that worst case scenario, the exposure time would be 1.95 seconds and the dead time would be 0.05 seconds.

————————————

Zachory K. Berta-Thompson Torres Exoplanet Fellow MIT Kavli Institute 77 Massachusetts Ave., 37-673 Cambridge, MA 02139 phone: 617-253-5084 web: space.mit.edu/~zkbt

——————————__——

On Mon, Apr 20, 2015 at 5:57 PM, Zach Berta-Thompson zkbt@mit.edu wrote:

The time between adjacent sub-exposures will be 2.000000 seconds. According to Ed Morgan on the TESS team, the exposure time is defined as two pulses of 1 pulse-per second spacecraft clock. This clock isn't guaranteed to be exact and isn't synchronized with GPS, so a separate onboard "High Accuracy Clock" will be used to track any long-term drift between its exposure timestamps and real time systems. These drifts should be small compared to the barycenter correction, but are worth mentioning.

The CCD's are being read out as frame transfer arrays. The time to transfer pixels into the frame store will be something like 5 milliseconds. (This number will depend on the final properties of the flight detector and electronics, but this is the latest I've heard.) During this 5 milliseconds, the light from every star will be smeared out over its entire column, and this smear will be subtracted off in calibration. Unless something really crazy happens to a very bright star during that 5 milliseconds, this effectively means the cadence will be 1.995 seconds of exposure followed by 0.005 seconds of dead time.

Hope this helps! Will send a response on the Hogg 2.0 cosmic ray filter when I get the chance. It's pushing the memory limits, so will have to be really good to make it on board.

Cheers, Z

————————————

Zachory K. Berta-Thompson Torres Exoplanet Fellow MIT Kavli Institute 77 Massachusetts Ave., 37-673 Cambridge, MA 02139 phone: 617-253-5084 web: space.mit.edu/~zkbt

——————————__——

On Mon, Apr 20, 2015 at 12:11 PM, David W. Hogg <notifications@github.com

wrote:

Hey @zkbt https://github.com/zkbt : I am simulating a TESS lightcurve.

TESS takes data in 2-second exposures. What is the exact number I should replace "2.0000" with? Also, what is the dead time (if any) between the 2-second exposures?

I pay beer for answers.

— Reply to this email directly or view it on GitHub https://github.com/davidwhogg/HoneyComb/issues/25.

[davidwhogg]

Thank you very much @zkbt ! This is exactly what I wanted.

[davidwhogg]

I am closing this for now; I am doing things as realistically as I ever intend to (for now).