Elegant users forum

Michael,

I have a straight beamline (few meters) that we plan to use to measure emittance from our linac. The beam from the linac has a large energy spread (dp/p>~50%). I would like to know how the particles track since they will affect the emittance measurement. Word on the street is that such large momentum offsets will not track accurately in elegant. One idea that is being floated is to bin the beam in energy and then track the smaller bins in ELEGANT.

Is it possible to trakc accurately a beam with such a large momentum spread in elegant??

Ray

Ray,

This should not be a problem if you use KQUAD, KSEXT, and CSBEND elements, which are accurate to all orders in energy deviation. You can also use EDRIFT, which is an exact drift space and is recommended if slopes are large.

If you use QUAD, SEXT, and SBEN, it is only good to 2nd order and you might not get accurate results.

--Michael

Michael,

the manual says that SEXT and QUAD are up to 3rd order.

Anyway, I tried to use both kick elements (KQUAD) and matrix ones (QUAD) and tracking results are very similar (I set momentum spread to 10%).

At which point would you expect inaccurate results using elements implemented as matrices?

Thanks!

Slava

Slava,

It is hard to define just when the third-order quads and sextupoles would prove inadequate. Certainly one must use KQUAD and KSEXT for ring tracking. For single-pass systems, it is probably not necessary.

--Michael