Floating Point DSCs Lift Performance of Solar Inverters

Jun 1, 2007 12:00 PM
By David Morrison, Editor, Power Electronics Technology

Click here for the enhanced PDF version of this article including diagrams and/or equations.

Three new digital signal controllers (DSCs) from Texas Instruments promise to boost system performance, while simplifying software development for solar-power inverters and various types of industrial motor drives. Described as the industry's first floating-point DSCs, the TMS320F2833x are 32-bit controllers that carry out 300 million floating-point instructions per second at 150 MHz. That performance represents, on average, a 50% increase in operations versus 32-bit fixed-point DSCs running at the same clock speed.

In the power inverters used to convert energy from photovoltaic panels, the higher performance afforded by the floating-point DSCs will allow the inverters to operate with higher efficiency. These inverters typically use digital loop control in their dc boost and dc-ac converter stages. The use of floating-point DSCs will make it possible to run more efficient control algorithms in both of these stages. At the same time, the DSCs can implement additional system features.

The floating-point DSCs also can boost performance in other applications, such as in the inverter stages in UPSs. And in factory automation, the DSCs can provide both wider dynamic range and finer control to robotic equipment and CNC machines using ac induction motors. For industrial applications using ac servo motors, the floating-point DSCs enable more efficient servo drives.

In addition to improving performance, the floating-point DSCs ease software development. Typically, software developers create algorithms in a floating-point environment for validation, and then convert the code to run on fixed-point devices through a process of scaling, saturation and adjustment of numerical resolution. That conversion process, which can take weeks or even months, is eliminated with the new DSCs.

Though system designers have known about the advantages of floating-point operation in industrial applications, they were previously hindered by the high cost of implementation. They could employ floating-point digital signal processors, but these chips lacked the required peripherals such as a-d converters, PWMs and memory. Consequently, floating-point operation required expensive multichip solutions, which typically could not be justified by performance requirements.

However, as performance requirements in industrial applications have grown, so too has the demand for cost-effective floating-point DSCs. TI responded to that demand in developing the three F2833x devices, which are distinguished from one another mainly by the amounts of on-chip memory they offer.

Designed to facilitate single-chip control, the F2833x offer a high level of system integration (see the figure and table). Features include a 12-bit, 16-channel A-D converter operating at 12.5 Msamples/sec and up to 18 PWM channels including six with high (150-ps) resolution (HRPWM). Communication interfaces include CAN, I2C, UART, SPI and TI's Multi-channel Buffered Serial Ports (McBSPs).

Samples of the F28335, F28334, and F28332 devices will be available in September 2007 and will be fully AEC Q-100 qualified for automotive applications. For more information, see www.ti.com/f2833x.

Table. Floating-point controller specifications.

TMS320	MHz	Flash KB	RAM KB	PWM/HRPWM	CAP/QEP	SCI	Pricing in lots of 1000
F28335	150	512	68	18/6	6/2	3	$19.95
F28334	150	256	68	16/6	4/2	3	$17.65
F28332	100	128	52	16/4	4/2	2	$13.30