permanent quadrupole

[Philippe Piot]

Hello,
I am simulating a linac with many quadrupole electromagnets but some sections have permanent-magnet quadrupoles (PMQs). In some instance I vary the beamline energy (by loading different sddsbeam and automatically reseting pCentral to the right energy). I was wondering is there is a way to accordingly scale the k1 of the PMQs to reflec the fact that for these magnet the field gradient is constant (so k1 changes as I vary the beam energy). I guess I am after an option that locally (when the beam crosses the PMQ) sets the pcentral to scale appropriately the k1 to keep the PMQ field gradient constant. Right now I do this by hand. Thank you, -- Philippe.

[michael_borland]

Philippe,

It might help to use the KQUAD element, which has the ability to set the gradient based on the pole tip field (B) and bore radius (BORE). If BORE is non-zero, this overrides the K1 parameter.

--Michael

[Philippe Piot]

Michael,
Thank you for your answer. I did try the KQUAD element but I am puzzled by its behaviour. For instance if I have only KQUAD' elements in my beamline I would expect the pCentral not to affect anything in the results while the po associated to an initial "sddsbeam" to change the focusing properties. I observe the inverse: in order to affect the focusing strength of the KQUAD I need to alter the pCentral while changing the initial bunch momentum via po does not affect the focusing. I am attaching simplified files. Thank you very much. -- Philippe.

[michael_borland]

Philippe,

This is the expected behavior. Elegant always expands about the momentum given in the &run_setup namelist. The Po parameter in bunched_beam is only there as a way of defining a beam with a different average momentum.

The more fundamental reason this doesn't work as you expected is that elegant tracks with delta=(p-pCentral)/pCentral as the energy coordinate. So it has to convert B/BORE to K1 using pCentral before it starts tracking.

--Michael

[Philippe Piot]

HI Michael,
Thank you. This makes sense. What I am trying to do is explore chromatic effects in a linac which incorporates electromagnet and permanent quadrupoles. The way I usually perform such a study is to fit my lattice (electromagnet only) for a given pCentral and then track a beam with different off-nominal energies. If I do this, I see an effect from the electromagnet change in focusing strength (the k1 being scaled appropriately) but not from the PMQs. It seems to me that to perform such a study I would have to rescale by hand the PMQs (to make sure the beam see the same field gradient as its energy is changed). Do I understand correctly and/or is there a smarter way to do what I am trying to do? Thank you, -- Philippe.

[michael_borland]

Philippe,

I put together an example that I think does what you want. Please give it a try.

--Michael