English
Language : 

HD6417705 Datasheet, PDF (582/739 Pages) Renesas Technology Corp – Renesas 32-Bit RISC Microcomputer SuperH RISC engine Family/SH7700 Series
21.5 Interrupts and DMAC Transfer Request
The A/D converter generates an interrupt (ADI) at the end of A/D conversion. The ADI interrupt
request is enabled when ADF in ADCSR is set to 1and the ADIE bit in ADCSR is set to 1 after
A/D conversion. The ADI interrupt can activate the direct memory access controller (DMAC) by
setting the ADIE and DMASL bits to 1. Continuous conversion without loads of software is
enabled by reading data that has been converted by the ADI interrupt with DMAC.
When activating DMAC by ADI, the ADF bit in ADCSR is automatically cleared to 0 at DMAC
data transfer.
Table 21.5 A/D Converter Interrupt Source
Name
ADI
Interrupt source
A/D conversion end
Interrupt flag
ADF
DMAC activation
Yes
21.6 Definitions of A/D Conversion Accuracy
The following shows the definitions of A/D conversion accuracy. In the figure, the 10 bits of the
A/D converter have been simplified to 3 bits.
• Resolution
Digital output code number of the A/D converter
• Quantization error
Intrinsic error of the A/D converter and is expressed as 1/2 LSB (figure 21.3)
• Offset error
Deviation between analog input voltage and ideal A/D conversion characteristics when the
digital output value changes from the minimum (zero voltage) 0000000000 (H'00; 000 in
figure 21.3) to 0000000001 (H'01; 001 in figure 21.3) (figure 21.4)
• Full-scale error
Deviation between analog input voltage and ideal A/D conversion characteristics when the
digital output value changes from the 1111111110 (H'3EF; 110 in figure 21.3) to the maximum
1111111111 (H'3FF; 111 in figure 21.3) (figure 21.4).
• Nonlinearity error
Deviation between analog input voltage and ideal A/D conversion characteristics between zero
voltage and full-scale voltage (figure 21.4). Note that it does not include offset, full-scale, or
quantization error.
• Absolute accuracy
Deviation between analog and digital input values. Note that it includes offset, full-scale,
quantization, or nonlinearity error.
Rev. 2.00, 09/03, page 536 of 690