Improve raster of timings

[ckolbPTB]

No description provided.

[github-actions]

Coverage Report

File	Stmts	Miss	Cover	Missing
/home/runner/.local/lib/python3.12/site-packages/mrseq
_version.py	7	7	0%	2–9
/home/runner/.local/lib/python3.12/site-packages/mrseq/preparations
t2_prep.py	55	1	98%	198
/home/runner/.local/lib/python3.12/site-packages/mrseq/sequences
epi2d_fid.py	128	9	93%	256, 359, 416–421, 433, 439
epi2d_se.py	148	12	92%	204–206, 317, 420, 477–482, 494, 500
grpe_flash_dixon.py	205	18	91%	120, 123, 188, 205–207, 329, 348, 399, 409, 511, 580–585, 597, 603
radial_flash.py	141	11	92%	102, 105, 260, 354, 412–417, 429, 435
spiral_flash.py	120	10	92%	95, 231, 320, 370–375, 387, 393
t1_inv_rec_gre_single_line.py	94	9	90%	163, 264, 307–312, 331, 337
t1_inv_rec_se_single_line.py	107	9	92%	220, 325, 384–389, 408, 414
t1_molli_bssfp.py	142	13	91%	217, 283–285, 300, 384, 438–443, 455, 461
t1_t2_spiral_cmrf.py	149	12	92%	89, 95, 116, 170, 399, 446–451, 465, 471
t1rho_se_single_line.py	103	8	92%	319, 376–381, 400, 406
t2_multi_echo_se_single_line.py	97	9	91%	163, 269, 322–327, 345, 351
t2_t2prep_flash.py	144	15	90%	104, 227, 244, 278–280, 292, 376, 438–443, 455, 461
t2star_multi_echo_flash.py	146	14	90%	114, 233, 280–282, 294, 399, 462–467, 479, 485
/home/runner/.local/lib/python3.12/site-packages/mrseq/utils
EpiReadout.py	333	90	73%	52, 61, 193, 197, 248, 327, 330–341, 463, 516–600, 746–752, 772–817
cartesian_sampling.py	37	2	95%	38, 80
sequence_helper.py	66	4	94%	107, 135–136, 164
spiral_sampling.py	67	1	99%	79
TOTAL	2479	254	90%

Tests	Skipped	Failures	Errors	Time
1530	0 💤	2 ❌	0 🔥	4m 32s ⏱️

[schuenke]

Are the raster timings in the EPI scripts correct already, or do we have to do some adjustments there as well?

[ckolbPTB]

They need to be adapted. I will also add one more case with non-integer raster times.

[ckolbPTB]

There is still a problem with the ADC dead time. For our current system defaults that is very short so it is never a problem. For other systems it can be longer.

The ADC dead time I guess is only important between RF pulses and ADC events but most interpreters will simply throw an error if the time between block border and ADC event is less than the ADC dead time, independent if there is an RF event beforehand.

One option would be to simply ensure that the gy blip gradient durations are always at least 2 * ADC dead time: https://github.com/ckolbPTB/mrseq/blob/e50a366dad901328fefc66c4bcd9e10b471c0416/src/mrseq/utils/EpiReadout.py#L225

@schuenke Can you think of a smarter way of how to solve this?

[ckolbPTB]

Needs pulseq/pypulseq#417