Voltage adding pulse transformer
In most cases a pulse transformer has to be necessary for real high level conversion. A second reason to use a transformer is match the impedance level between pulse-forming part and klystron to effect maximum transfer of energy.
The pulse is unidirectional, what implicates a danger to saturate the transformer. To improve the effective induction range a bias current has required, mostly realized in combination with the primary or with extra windings. In both cases an inductor has placed to prevent the bias power supply from pulse influence.
To permit heater power to be supplied to a grounded-anode klystron
the secondary has a bifilar winding or there is a separate transformer.
A difficulty with bifilar secondary windings is that the windings
form a parasitic "delay line" by their distributed stray capacitance and stray
inductance. At the start of the pulse, a standing wave is generated because the
"delay line" is shorted at both ends. This effect must be damped at the HV side
by a RC filter.
This RC damping circuit can be placed on a tap (ca. 1: 10) of the secondary or
on extra windings. It can be used for droop correction too [1] [2].
A sometimes used approach to lower the specification of the switch-components is to use a transformer with several primaries. The primaries can be used in three way’s;
In all cases there is a build in redundancy if there are
more pulse-forming units then necessary for each pulse. Such a transformer
can be called a ‘cycle pulse transformer’.
There are two ways to achieve such a situation;
The transformer exists of a number of primaries with their own core and a common secondary. All the primaries are connected on the same (low)voltage level but functional in series. This means that not used primaries have to be shorten, because they act like current sources. For small pulses and low repetition rates a diode will be sufficient otherwise a switching device is needed. It means a high leakage inductance too. Because of the large leakage induction the rise- and falltime of the generated pulse are rather long. Therefore this transformer type can be used only in long pulse width modulators where the risetime is still small (<10%) in respect to the pulse width [3].
The transformer exists of a number of primaries with a common core and secondary. This transformer has a high step up ratio (e.g. 200:1) and therefore bad risetime properties. There is no special switch necessary to shorten an unused primary. The advantages and disadvantages are about the same as the voltage adding transformer [4].
In both cases solid state components with relative low voltage and high current capabilities can be used. It is to consider or a ‘cycle pulse transformer’ has profit to increase the redundancy and flexibility and lowered the voltage specifications of the pulse-forming part. The, in that case used, pulse forming units can be build with ‘low’ voltage material and components in mass production what decreases costs and increases reliability and manageability. Besides, the use of low voltage results in low stored energy/pfn in case of shortings.
Voltage adding pulse transformer
Primary shortcircuit devices for voltage
adding pulse transformer