Peak current mode control

Another form current-mode control is called hysteretic current-mode control. Here both the peak and the valley currents are controlled. This is obviously better for continuous-mode forward for boost converters. It is somewhat complicated to set-up, but it does offer very fast response times. It is not a very common method of control and its frequency varies. 

If we take the Bode plot of any current mode controlled converter (one that has not yet entered this wide-narrow-wide-narrow state), we will discover an unexplained peaking in the gain plot, at exactly half the switching frequency. This is the source of subharmonic instability. Because, though this point is much past the crossover frequency, it is potentially dangerous because of the fact that if it peaks too much, it can end up intersecting the 0 dB axis again — which we know is one of the prescriptions for full instability. 

The original models for current mode control did not predict this half-switching-frequency peaking (i.e. subharmonic instability). But it has been well known that we need to set a minimum amount of slope compensation — the value of which depends on the slopes of the up-ramp and down-ramp of the inductor current. But the criteria used for setting the precise amount of slope compensation, have been slightly differing. Our approach, outlined later, is based on more recent trends. 

Current mode control In this method, as shown in Fig. 10.79, a second inner control loop compares the peak inductor current with the control voltage, which provides improved open-loop line regulation. All of the problems of the direct duty cycle control method (1) and (2) are corrected with this method. An additional advantage of this method is that the two pole second-order filter is reduced to a single pole (the filter capacitor) first-order filter resulting in simpler compensation networks. There are several current mode control ICs available in the market. 

AVe in this case can be the peak-to-peak voltage of the oscillator ramp if the method of control is voltage-mode, or the maximum peak voltage representing the primary current within the current-mode method of control. The gain can be converted into decibels, which is shown in Equation B.7. 

The mode of application of this method need not be described in this article, since it is thoroughly illustrated in many user s manuals of commercial potentiostats. It should be recalled, however, that when this technique is adopted use must be made of an oscilloscope and a function generator because for an optimum choice of the feed-back amplitude it is necessary that the wave shape of the current peak be strictly controlled. It may be added that the use of the Solartron mod. 1286 electrochemical interface makes it possible also to choose the amount of the positive feed-back numerically. 

The most convenient means of making time-resolved SH measurements on metallic surfaces is to use a cw laser as a continuous monitor of the surface during a transient event. Unfortunately, in the absence of optical enhancements, the signal levels are so low for most electrochemical systems that this route is unattractive. A more viable alternative is to use a cw mode-locked laser which offers the necessary high peak powers and the high repetition rate. The experimental time resolution is typically 12 nsec, which is the time between pulses. A Q-switched Nd YAG provides 30 to 100 msec resolution unless the repetition rate is externally controlled. The electrochemical experiments done to date have involved the application of a fast potential step with the surface response to this perturbation followed by SHG [54, 55,116, 117]. Since the optical technique is instantaneous in nature, one has the potential to obtain a clearer picture than that obtained by the current transient. The experiments have also been applied to multistep processes which are difficult to understand by simple current analysis [54, 117]. 

For MS work, the electron impact (El) mode with automatic gain control (AGC) was used. The electron multiplier voltage for MS/MS was 1450 V, AGC target was 10,000 counts, and filament emission current was 60 pA with the axial modulation amplitude at 4.0 V. The ion trap was held at 200°C and the transfer line at 250°C. The manifold temperature was set at 60°C and the mass spectral scan time across 50-450 m/z was 1.0 s (using 3 microscans). Nonresonant, collision-induced dissociation (CID) was used for MS/MS. The associated parameters for this method were optimized for each individual compound (Table 7.3). The method was divided into ten acquisition time segments so that different ion preparation files could be used to optimize the conditions for the TMS derivatives of the chemically distinct internal standard, phenolic acids, and DIMBOA. Standard samples of both p-coumaric and ferulic acids consisted of trans and cis isomers so that four segments were required to characterize these two acids. The first time segment was a 9 min solvent delay used to protect the electron multiplier from the solvent peak. 

The potential of the peak Ep is indicative of which species is involved. If the reduction (or oxidation) mechanism is diffusion-controlled the concentration of the species controls the Faradaic current. Since differential pulse polarography effectively displays the derivative of this current, theoretically it is the area under the peak which is proportional to the concentration. However, provided the shape of the peak does not change, the height of the peak is also proportional to concentration. The choice between the two modes of measurement will be discussed later. 

Fig. 1. Visualization of the data of the first control sample, measured in positive mode. The top of the figure shows the square root of the Total Ion Current (TIC) background color indicates the intensity of the signal in the plane formed by retention time and m/zaxes. Circles indicate features found by the peak picking the fill colour of these circles indicates the intensity of the features.

---