Central line artefact with 2D spin echo

[aTrotier]

Hi !

I am working on a 2D spin echo sequence for the bruker pulseq initiative.

Without crusher around the 180 block pulse I get :

And with the crusher :

The sequence looks like this :

 for i in range(Ny):

        seq.add_block(rf, gz)
        seq.add_block(pp.scale_grad(grad=gy_pre,scale=scale_area[i]), gz_reph)
        seq.add_block(b_delay_TE1)
        #seq.add_block(gx_spoil) #crusher
        seq.add_block(rf_180)
        #seq.add_block(gx_spoil) #crusher
        seq.add_block(b_delay_TE2)
        seq.add_block(gx_pre)
        seq.add_block(gx, adc)
        seq.add_block(gx_pre)
        seq.add_block(b_delay_TE1)

        # spoiler
        seq.add_block(gx_spoil, pp.scale_grad(grad=gy_pre,scale=-scale_area[i]), gz_spoil,lin_inc)
        seq.add_block(b_delay_TR)

Any idea what I am not taking into account ?

PS : I have a TR = 1000 ms for spoiling

[cncastillo]

Is the x-direction up-down in the plots?

[aTrotier]

I don't think so.

I think up-down is the phase encoding direction

[cncastillo]

can you append the .seq file so i can check it out?

[aTrotier]

bruker.zip

you can find :
the jupyter notebook and the .seq + simulation script

[aTrotier]

If I remove the upper-sampling of the phantom I also get that issue :
obj=brain_phantom2D()

[cncastillo]

Part of the problem, at least, is that it is using Pulseq 1.5.0 and we haven't merged #614 yet:

[aTrotier]

Oh my bad.

We should really add a warning for that kind of stuff !