1. Field
The present invention relates in general to bias generator circuits, and more specifically to bias generators with minimized susceptibility to environmental and manufacturing changes.
2. Background
A bias generator provides a bias voltage to a device such as transistor to allow the device to operate in a preferred region of the device operational characteristics. In many applications, the selection of the relative sizes of the transistors used in the bias generator and the load device is critical to maintain the operational characteristics within an acceptable region over variances due to temperature and manufacturing. For example, bias generators are often used to set a voltage at an input of a low noise amplifier (LNA) transistor where small changes in operational characteristics result in increased noise and non-linear input to output relationships. Since the devices used to implement bias generators are susceptible to temperature and manufacturing process variations, conventional biasing schemes attempt to minimize the effects of temperature and process fluctuations. In an attempt to maximize the performance of the biased device, conventional bias generates consume significant amounts of current relative to the current used by the biased device. Performance of bias generators typically suffers when the devices in the bias generator are mismatched from the biased device. A mismatch between devices causes time-independent random variations in physical characteristics of identically designed devices. Typical characteristics that may be different between mismatched devices include device dimensions, threshold voltage, and mobility. Performance of a bias generator is typically improved by selecting a biasing device similar to the biased device. Unfortunately, conventional biasing schemes typically require a significant tradeoff between current draw and relative device size between the biased device and the devices in the bias generator.
Therefore, there is a need for a bias generator with minimized current consumption and maximized performance.
In accordance with the exemplary embodiment, a bias generator comprises a first transistor and a second transistor having a control port connected to a control port of the first transistor and to an input port of the second transistor, where a second current through the second transistor is greater than a first current through the first transistor. The current through the bias generator is minimized by providing the different currents through the transistors having a similar size.
In the exemplary embodiment, current through a bias generator is minimized by providing different currents through transistors having a similar size. As compared to conventional bias generators where equal current are forced through devices of different sizes, the overall current consumption of the exemplary bias generator is less. The word “exemplary” is used herein to mean “serving as an example, instance, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.
The bias generator 100 includes at least two transistors 104, 106 that provide a bias voltage (Vbias) 130 to the biased device 102. Each of the transistors 104, 106 has an input port 108, 114, an output port 110, 116, and a control port 112, 118 where the voltage at the control port 112, 118 determines the current flow through the transistor 104, 106 from the input port 108, 114 to the output port 110, 116. Where the transistor is a field effect transistor (FET), the input port 108, 114, output port 110, 116 and control port 112, 118 are the drain, source and gate of the FET, respectively. Where the transistor is a bipolar junction transistor (BJT), the input port 108, 114, output port 110, 116 and control port 112, 118 are the collector, emitter and base of the BJT, respectively. Those skilled in the art will readily apply the teachings herein to known techniques in order to utilize any of numerous three terminal devices to implement the bias generator.
In the exemplary embodiment, the first control port 112 of the first transistor 104 is connected to the second control port 118 of the second transistor 106. The common node formed at the two control ports 112, 118 is connectable to the biased device 102 to provide the bias voltage 120 (Vbias). The first output port 110 of the first transistor 104 is connected to ground through a reference load (R) 118. A first current source 120 supplies a first current (I1) 124 to the first transistor 104 from a power supply at a voltage (VDD) 128. The first current (I1) 124 flows from the input port through the transistor to the output port and through the reference load 118. A second current source supplies a second current (I2) to the second transistor 106. Although other devices may be used in some circumstances for implementing the current sources 120, 122, the first current source 120 and the second current source 122 are field effect transistors (FETs) in the exemplary embodiment.
In order to minimize the total current consumption while minimizing the variation of the bias voltage over temperature and manufacturing variations, the second current 122 source provides a second current (I2) that is greater than the first current (I1) and the difference in size between the first transistor 104 and the second transistor 106 is minimized. In the exemplary embodiment, the first transistor 104 and the second transistor 106 are selected to have the same size. As discussed below, the total current through the bias generator 100 is minimized while the performance is maximized.
A field effect transistor (FET) is typically fabricated using any of numerous doping techniques to create a channel in a substrate. The channel may be formed with one or more elements often referred to as “fingers”. The operational characteristics of the FET depend on the aspect ratio of the fingers and the number of fingers where the aspect ratio is the ratio of the width (Wf) to the length (Lf) of the finger. The size (M) of the FET is the aspect ratio of each finger (Wf/Lf) time the number of fingers (Nf). Therefore, the size of the first transistor and the second transistor can be expressed as follows:
where M1 is the size of the first transistor, M2 is the size of the second transistor, Wf is the width of each finger, Lf is the length of each finger, and Nf is the total number of fingers.
In the exemplary embodiment, the first current source 120 includes a third FET 202 where the gate is connected to the source and the second current source 122 includes a fourth FET 204. The first current (I1) 124 and the second current (I2) 126 are determined, at least in part, by the sizes (M3, M4) of the third FET 202 and the fourth FET 204. In the exemplary embodiment, the size of the fourth FET (M4) is selected to be approximately A times the size (M3) of the third FET 202 in order that the second current (I2) 126 be about A times the first current (I1) 124. Accordingly, for the exemplary embodiment, the following relationships apply.
where R is the resistance of the reference load 118, μn is the mobility of the FETs, Cox is the capacitance per area of the FETs, and Nf is the number of fingers of the second transistor 106. Accordingly, the current is proportional to A, the ratio of the M4 to M3.
Comparing the conventional bias generator 300 to the exemplary bias generator 100, it can be seen that the total current in the exemplary bias generator 100 is less than the total current in the conventional bias generator 300 for biasing the same biased device. The current savings is easily observed when applying values in the following example.
For the example, the biased device 102 has a channel width (Wbiased) equal to 500 μm resulting in a bias current (Ibiased) through the biased device 102 equal to 10 mA. For the comparison, scaling factors A and B are both equal to 4. In both the conventional bias generator 300 and the exemplary bias generator 100, the sizes (M2Conv) (M2EX) are ten times smaller than the biased device 102 and, therefore, the channel widths (WM2Conv) (WM2EX) of the second transistors 304, 106 are selected to be ten times smaller than the biased device width. Accordingly, WM2Conv=WM2EX=500 μm/10=50 μm. The current (I2) through the second transistors 106, 304 is equal to 10 mA/10=1 mA. Since the current through the first transistor 302 in the conventional bias generator 300 it the same as the current through the second transistor 304, the current through the first transistor 302 is equal to 1 mA and the total current through the bias generator 300 is equal to 2 mA. The size (M1Conv) of the first transistor 304, however, is four times smaller than the size (M2Conv) of the second transistor 304 resulting in a channel width (WM2Conv) equal to 12.5 μm. In the exemplary bias generator 100 discussed with reference to
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.