Area of occupation, and power consumption are two of the most important constrains, when it comes to designing integrated DC to DC converters. In order for these circuits to work, a compensation network is required, that involves use of large resistors and capacitors. These passive elements increase the required area of implementation. The traditional voltage mode compensation also requires high performance amplifiers, with large DC gains and wide bandwidths. The current consumption of the traditional compensation methods is significant, which makes the converter less efficient. This thesis attempts a novel time-based compensation method for a DC to DC boost converter. The time-based compensator replaces the network of passive elements, with time-based structures. The classic PID transfer functions is implemented with the help of a voltage-controlled oscillator that acts as integral control, while voltage-controlled delay lines are used to implement derivative and proportional control. The capability of the time-based compensation strategy is proven through simulations in Spectre, by comparing its transient response to the transient response of a more commonly used voltage mode compensator. The similar transient response of the two compensation strategies proves that the time-based compensator can achieve the same bandwidth and phase margin, as the voltage mode compensator. The size required by the time-based compensator is 12 times less than that of the voltage mode counterpart, when both compensators are used to implement singularities at same position in frequency. The current consumption of the time-based compensator is five times less than that of the voltage mode compensator. These gains in both current consumption and area occupation are significant, especially when designing an integrated circuit, and prove the viability of the time-based compensation method as a worthwhile alternative to the more widely used compensation methods. The relative increase in design complexity of the time-based components is rewarded with gains in area and current consumption.
L’occupazione di area e il consumo di potenza sono due vincoli molto importanti nella progettazione di convertitori di potenza DC-DC. Per poter funzionare, questi circuiti necessitano di un circuito di compensazione composto da capacità e resistenze di grandi dimensioni che impattano fortemente sull’occupazione di area del convertitore. Il circuito di compensazione più comunemente utilizzato (Voltage-mode PID), richiede l’implementazione di un amplificatore errore a larga banda e alto guadagno in DC. Per ottenere queste performance, quest’ultimo componente richiede un alto consumo di corrente con conseguente impatto sull’efficienza del convertitore. Lo scopo di questa tesi è la realizzazione di un circuito di compensazione per un convertitore di tipo boost che utilizzi un approccio time-based. Grazie a questo è possibile sostituire le reti di elementi passivi con strutture di tipo time-based. La funzione di trasferimento del compensatore PID è stata ottenuta utilizzando un oscillatore controllato in tensione, per realizzare il controllo integrale, e delle linee di ritardo controllate in tensione, per l’implementazione del controllo di tipo proporzionale e derivativo. L’efficacia della struttura time-based è stata verificata in Spectre confrontando i le risposte in transitorio della struttura proposta con quelle di un boost di tipo voltage-mode. I risultati del confronto dimostrano che l’architettura di tipo time-based è in grado di raggiungere le stesse performance in termini di banda e margine di fase della controparte voltage-mode con un’occupazione di area 12 volte inferiore. Il consumo di potenza dell’architettura time-based è cinque volte inferiore rispetto al voltage-mode. I vantaggi, in termini di occupazione di area e consumo, dell’architettura proposta sono significativi, e giustificano la validità dell’approccio di tipo time-based come alternativa al più utilizzato voltage-mode, soprattutto per convertitori integrati. L’aumento della complessità del sistema di controllo time-based è ricompensato da un guadagno in termini di area e consumo di corrente.
Time-based-compensation method for a DC-DC boost converter
GEORGIEVSKI, GORJAN
2018/2019
Abstract
Area of occupation, and power consumption are two of the most important constrains, when it comes to designing integrated DC to DC converters. In order for these circuits to work, a compensation network is required, that involves use of large resistors and capacitors. These passive elements increase the required area of implementation. The traditional voltage mode compensation also requires high performance amplifiers, with large DC gains and wide bandwidths. The current consumption of the traditional compensation methods is significant, which makes the converter less efficient. This thesis attempts a novel time-based compensation method for a DC to DC boost converter. The time-based compensator replaces the network of passive elements, with time-based structures. The classic PID transfer functions is implemented with the help of a voltage-controlled oscillator that acts as integral control, while voltage-controlled delay lines are used to implement derivative and proportional control. The capability of the time-based compensation strategy is proven through simulations in Spectre, by comparing its transient response to the transient response of a more commonly used voltage mode compensator. The similar transient response of the two compensation strategies proves that the time-based compensator can achieve the same bandwidth and phase margin, as the voltage mode compensator. The size required by the time-based compensator is 12 times less than that of the voltage mode counterpart, when both compensators are used to implement singularities at same position in frequency. The current consumption of the time-based compensator is five times less than that of the voltage mode compensator. These gains in both current consumption and area occupation are significant, especially when designing an integrated circuit, and prove the viability of the time-based compensation method as a worthwhile alternative to the more widely used compensation methods. The relative increase in design complexity of the time-based components is rewarded with gains in area and current consumption.| File | Dimensione | Formato | |
|---|---|---|---|
|
2019_07_Georgievski.pdf
non accessibile
Descrizione: Thesis text
Dimensione
3.29 MB
Formato
Adobe PDF
|
3.29 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/148591