A／D量化,非线性,微分非线性误差对非同步,准同步采样算法误差影响 …

本文提出了具有量化误差、非线性误差和微分非线性误差的Ａ／Ｄ转换器的数据学模型,在该数学模型的基础上,仿真分析了非同步采样法、准同步采样递推算法各种情况下的误差,包括功率测量中电压线性、电流线性、相位变化误差及电压测量的误差。由于分析中的条件更近接应用中的实际情况,因此,仿真分析的结论为测量仪器设计中的误差估计提供了重要依据。     

A/D量化、非线性、微分非线性误差对非同步、准同步采样算法误差影响的分析

本文提出了具有量化误差、非线性误差和微分非线性误差的A/D转换器的数学模型,在该数学模型的基础上,仿真分析了非同步采样法、准同步采样递推算法各种情况下的误差,包括功率测量中电压线性、电流线性、相位变化误差及电压测量的误差.由于分析中的条件更接近应用中的实际情况,因而,仿真分析的结论为测量仪器设计中的误差估计提供了重要依据.     

基于混合采样方法的交流功率转换器

提出了一种混合采样的交流功率转换器 ,叙述了该种采样功率测量的原理 ,推导了功率测量的算法和功率测量误差。通过计算机仿真 ,分析了量化误差对混合采样功率测量方法的影响。最后给出了采用 12 bitA D和D A转换器、频率范围为 4 5～ 6 5Hz时的功率测量实验数据 ,结果表明 :混合采样功率转换器的准确度为 3× 10 - 4,比数字采样方法的测量误差有明显的改善     

大量程触针杠杆式轮廓仪的非线性误差分析和补偿

介绍了一种基于光栅干涉测量原理的大量程、高精度触针杠杆式表面形貌传感器,并对传感器在大量程测量时的非线性误差进行了分析。测量时轮廓仪在水平和垂直方向上都会由于测杆绕固定支点的转动而产生非线性误差,为此提出了一种对水平和垂直方向上非线性误差进行校正的新方法。通过使用大量程、高精度的一维垂直微位移工作台和标准楔形斜块进行误差校正。最后分别给出校正前后误差曲线,并进行对比。通过分析可知,此方法可以有效减小测量时的非线性误差,大大提高轮廓仪在进行大量程测量时的精度,从而实现大量程、高精度测量。     

自准直仪非线性误差补偿方法的研究

自准直仪角度测量方法基于正切原理,测量灵敏度会随着角度变化而变化,测量原理存在非线性误差。本文针对在自准直仪设计的系统时将测量原理进行线性化,并对非线性误差进行了补偿的方法提出了解决和研究的方法。     

大范围二维纳米位移台的控制及非线性校准的实验研究

搭建了基于激光干涉仪测量原理的三轴微位移测量系统,对大范围二维纳米位移台的控制及非线性校准进行实验研究。介绍了双频激光干涉仪测量系统的构成;编写了纳米位移台的控制程序和激光干涉仪数据采集程序;阐述了位移台的非线性校准方法,并通过实验对比了多项式三阶拟合和三阶分段拟合的差别,验证了校准方法的准确性。实验表明:使用三阶分段拟合的校准方法效果更好,校准前,x轴的最大非线性误差为4.052μm,y轴的最大非线性误差为2.927μm;校准后,x轴的最大非线性误差为15nm,y轴的最大非线性误差为17nm,仅为原始非线性误差的1%。     

真有效值交直流转换器测量非正弦电压的误差

真有效值交直流转换器不仅可以高精度地测量失真的正弦电压,而且还可以测量非正弦电压。对于实际使用的有效值转换器都有一定的幅值范围和频率范围,因此在测量非正弦电压时,除了由于被测电压含有丰富的高频谐波而产生高频误差外,还要在低频产生附加的误差,以及频响附加误差,运算放大器有限的转换速率产生的误差等。下面以交流耦合(这是电     

基于模数转换器的非线性校正

本文讨论了基于A/D转换器反馈电路的非线性变换原理,分析了铂电阻温度测量的非线性校正问题并确定了优化设计参数,简化了电路设计,有效提高了铂电阻测量温度的准确度。     

机床主轴回转精度测试实验平台设计

本文首先利用双向测量法测量原理,通过A/D转换器、由计算机控制对主轴回转误差进行数据采集、预处理与转换。以快速傅里叶变换为基础对回转误差信号进行功率谱分析,并提出以快速傅里叶变换和逆变换的方法对一次偏心分量进行消除。该系统运用于ZHS-1型多功能转台主轴回转精度的实际测量,实现了主轴回转误差的在线测量和实时分析处理。     

偏振分光镜旋转角度误差的确定

偏振分光镜存在旋转角度误差是引起激光外差干涉非线性误差的来源之一.提出了一种确定激光外差干涉偏振分光镜旋转角度误差的方法.通过分析偏振分光镜旋转角度误差在其它条件下对激光外差干涉非线性误差的影响,得出偏振分光镜存在的旋转角度误差将极大地增加非线性误差的一次谐波,但改变二次谐波很小.对光电接收器输出信号进行频谱分析,分离出激光外差干涉非线性误差的一次谐波和二次谐波,通过测量非线性误差二次谐波相对测量信号的大小,通过应用非线性误差二次谐波幅度比值与偏振分光镜旋转角度误差之间的模型,确定偏振分光镜旋转角度误差.实验结果表明,应用该方法能够得到偏振分光镜的旋转角度误差,从而为调整偏振分光镜提供了实验依据.     

Single-chip interpolating time counter with 200-ps resolution and43-s range

In this paper, we present a design and test results of the interpolating time counter implemented on a single field programmable gate array (FPGA) chip. The counter contains two 6-bit time-to-digital converters (TDCs), each having 200-ps resolution (LSB) within 10 ns range, and the 32-bit, 100-MHz real-time counter, which is also used for frequency measurement. The utilization of the logic cells on the FPGA chip is 93%. The software correction of the TDC's nonlinearity errors resulted in lowering the random error of the counter to 0.65 LSB or 129 ps (RMS)     

基于正弦失真测量的A/D动态有效位数评价

针对A/D动态有效位数的测量,提出了基于正弦波采样序列失真度的测量方法,给出了相应的计算公式。不用执行复杂的正弦波拟合运算即可实现A/D转换器动态有效位数的测量校准。通过一组仿真实验, 使A/D位数在3bits到24bits范围内变动,以及变动每周期采集数据点数等情况下,将所述方法与正弦波曲线拟合法测量动态有效位数进行了比较研究,获得结论是两者结果一致性良好,正弦波拟合方法的适应性更宽,而所述方法的过程更加简洁。在数据采集系统上的实测实验验证了所述方法的正确性和切实可行性。     

A 12-Mc, Nine-Bit Digital Function Generator

This paper describes some developed equipment found very useful in evaluating circuits that require digital inputs. For example, D-A converters are generally tested by using their associated A-D converter as a test-signal source. Since the A-D converter itself introduces errors, the measurement of analog input and output indicates over-all performance but not individual converter performance. The digital function generator allows separate evaluation of D-A converters by synthesizing useful test functions from discrete samples. Comparison of the known input with the output gives an accurate performance check. The present equipment generates up to sixteen different samples, represented by nine-bit words in parallel form, at a word rate of 12 Mc. A discussion of useful test functions for determining D-A converter linearity and transient response will include discussion of digital sine waves, ramps, and step functions. Although D-A converter evaluation is stressed, the equipment is also useful for other tasks.     

An enhanced method for characterizing successive approximationconverters

A strategy for characterizing successive approximation, analog-to-digital converters with quad current switch architectures is described. It permits detailed parameter specification (accuracy and linearity) without recourse to testing every transition voltage. The results show an encouraging reduction in the residual error from 0.144 step (from a currently used test) to 0.048 step with the proposed test for the specific case of 12-b converters. To achieve this result, only 2.6% of the transition voltages require measurement. The percentage of codes to be tested reduces dramatically as the number of bits of the converter increases. The evolution of the proposed test is investigated by examining the cause and effect of superposition errors     

Code probability distributions of A/D converters with random inputnoise

The specific architecture of an A/D converter influences the code probability distributions that result from random input noise. In particular, the output codes of successive approximation A/D converters have a spiked distribution, and its variance is half that of the corresponding input noise. In addition, the distribution has a small bias. These and other related results are derived, and are qualitatively supported by measurement data on a real 16-bit A/D converter     

A Technique for Measuring the Equivalent RMS Input Noise of A/D Converters

A simple accurate technique is described for measuring the equivalent rms input noise of A/D converters. Noise can typically be measured with 10-percent accuracy in 1 s, and the method has successfully been applied to converters with up to 16 bits of resolution. The measurements are made at input voltages corresponding to the test converter's decision levels, where the effects of noise are most pronounced. A feedback loop incorporating the unit under test locates and locks onto these levels. The method utilizes a theoretical relationship between the input noise and an expected number of counts derived digitally from the feedback loop response. A low-noise wide-band operational amplifier is the only critical component required.     

Linearization and stabilization of the conversion characteristics of a thermoelectric thermometer

An analysis is given of the condition of a digital thermometer utilizing thermoelectric converters (thermocouples). A method is established for digitally correcting the errors due to nonlinearity and instability of the calibration characteristic of a thermocouple by providing additional heating and cooling of its working end by passing calibrated electric current pulses through it. A microprocessor is used to process the effects on the thermocouple of the additional current, without dismantling it from its operating location, in order to correct the errors of the thermoelectric thermometer. Translated from Izmeritel'naya Tekhnika, No. 5, pp. 42–46, May, 1996.     

A new method for estimating the aperture uncertainty of A/Dconverters

Aperture uncertainty measurements are strongly influenced by quantization, converter nonlinearity and test setup noise. The proposed method solves the difficulties arising from quantization, and nonlinearity. The method estimates the noise distribution function by fine adjustment of the input signal offset. The contribution of jitter-induced voltage noise is separated from additive noise in two different ways. Finally, the aperture uncertainty, assumed independent of the frequency, is estimated by varying the signal frequency. Experimental results obtained on 8 and 10 bit converters are discussed     

Built-in self-test design of current-mode algorithmicanalog-to-digital converters

Analog MOS circuits are becoming increasingly sophisticated in terms of checking and correcting themselves. Self-correcting, self-compensating, or self-calibrating techniques has been employed in analog-to-digital (A/D) converters to eliminate errors caused by offset and low frequency noise, and to cancel the error effect. However, the self-compensating/calibration techniques may no longer work properly in the presence of faulty switching elements. This paper presents the fault behaviors and test generation of a current-mode algorithmic A/D converter, where the single stuck-at faults in the switching elements of the converter are assumes. The converter requires only two test currents to achieve a full testability. Due to the simplicity of generating test currents and the expected outputs, a simple built-in self-test (BIST) structure is proposed. Two extra pins for test enable signal and error indicator are needed. Results show that the full self-testability of the BIST structure is achieved with a low pin/hardware overhead, and the use of expensive test equipment is not necessary