How to set pulse lengths

It almost always works for solid samples with symmetric lines which have short enough Tg s so that the field inhomogeneity can be ignored. 

A variation of this method will work even if, for some 

Pulses which are multiples of n/2 pulses are the easiest to set, but other pulse lengths are occasionally needed, like a 45 degree pulse for Jeener echoes (IV.B.4.), or arbitrary length pulses for variable nutation T experiments (III.D.5.) and the DANTE sequence (II.D.2.). For any submultiple of n/2 pulses, the trick is to repeat the pulse an appropriate number of times so that the combined effect is that of a n/2 pulse or multiples thereof. For example, you could go for 30 degree pulses by having three identical pulses act like a 90 degree pulse. Any odd length pulses must be interpolated after that. 

When studying a new sample, a sample containing a nucleus with which you are not familiar, or a sample about which you have some suspicion, you should try to estimate the S/N so that you will not be taken by total surprise when you get no signal at all. This task is considerably easier for solution samples than solids because the line broadening mechanisms are more predictable and you do not have to worry about losing intensities to satellite transitions as you might for quadru-polar nuclei in solids. The most likely circumstance under 

The main source of information for S/N is an NMR table such as the one by Lee and Anderson (Appendix A) which gives the calculated relative sensitivity per nucleus. These values are calculated with several assumptions (such as T Tg) and represent the maximum signal possible so that the actual signal may not scale according to these values. One of the effects not taken into account in the table is the degradation of the usual NMR coil quality factor as the resonance frequency is increased and coil inductance reduced if the same probe cannot be used at the higher frequency. The net 

---