A new DT campaign (DTE2) is planned at JET in 2020 to minimize the risks of ITER operations. In view of DT operations, a calibration of the JET neutron monitors at 14 MeV neutron energy has been performed using a well calibrated 14 MeV neutron generator (NG) deployed, together with its power supply and control unit, inside the vacuum vessel by the JET remote handling system. The NG was equipped with two calibrated diamond detectors, which continuously monitored its neutron emission rate during the calibration, and activation foils which provided the time integrated yield. Cables embedded in the remote handling boom were used to power the neutron generator, the active detectors and pre-amplifier, and to transport the detectors' signal. The monitoring activation foils were retrieved at the end of each day for decay gamma-ray counting, and replaced by fresh ones. About 76 hours of irradiation, in 9 days, were needed with the neutron generator in 73 different poloidal and toroidal positions in order to calibrate the two neutron yield measuring systems available at JET, the U-235 fission chambers (KN1) and the inner activation system (KN2). The NG neutron emission rates provided by the monitoring detectors were in agreement within 3%. Neutronics calculations have been performed using MCNP code and a detailed model of JET to derive the response of the JET neutron detectors to DT plasma neutrons starting from the response to the NG neutrons, and taking into account the anisotropy of the neutron generator and all the calibration circumstances. These calculations have made use of a very detailed and validated geometrical description of the neutron generator and of the modified. MNCP neutron source subroutine producing neutron energy-angle distribution for the neutrons emitted by the NG. The KN1 calibration factor for a DT plasma has been determined with +/- 4.2%' experimental uncertainty. Corrections due to NG and remote handling effects and the plasma volume effect have been calculated by simulation modelling. The related additional uncertainties are difficult to estimate, however the results of the previous calibration in 2013 have demonstrated that such uncertainties due to modelling are globally <= +/- 3%. It has been found that the difference between KN1 response to DD neutrons and that to DT neutrons is within the uncertainties in the derived responses. KN2 has been calibrated using the Nb-93(n,2n)Nb-92m and Al-27(n,a)Na-24 activation reactions (energy thresholds 10 MeV and 5 MeV respectively). The total uncertainty on the calibration factors is +/- 6% for Nb-93(n,2n)Nb-92m and +/- 8% Al-27(n,a)Na-24 (1 sigma). The calibration factors of the two independent systems KN1 and KN2 will be validated during DT operations. The experience gained and the lessons learnt are presented and discussed in particular with regard to the 14 MeV neutron calibrations in ITER.