A multi-digital channel EPA calibration method

Abstract

The present invention relates to the technical field of EPA calibration, and specifically discloses an EPA calibration method for multiple digital channels, comprising the following steps: S1: obtaining the functional relationship between the voltage change over time during the signal rise process when each digital channel of the first calibration board is driven; S2: obtaining the judgment parameter based on the functional relationship and generating the coordinate point; determining whether the digital channel is abnormal based on the coordinate point; S3: determining the first delay data, and calibrating the comparison edge according to the maximum first delay data; determining the second calibration board, obtaining the second delay data, and calibrating the driving edge according to the minimum second delay data; eliminating the path time delay RTD according to the minimum second delay data. The present invention can quickly and automatically complete the EPA calibration of multiple digital channels, which is very convenient for the periodic maintenance of the production line.

Description

Technical Field

[0001] The invention relates to the technical field of EPA calibration, and in particular to an EPA calibration method for multiple digital channels. Background Art

[0002] EPA (Edge Placement Accuracy) in the field of ATE (Automatic Test Equipment) refers to the accuracy of signal edge placement when the equipment is performing high-speed digital testing. With the continuous improvement of electronic equipment performance, especially the improvement of input / output (I / O) speed and electrical characteristics, the requirements for ATE's EPA are getting higher and higher. This is because in high-speed interface testing, even very small timing errors may lead to inaccurate test results.

[0003] The edge here generally refers to the edge (i.e., moment) of driving or comparison, which is reflected in the digital logic as the edge of the logic jump. In order to place these two edges accurately, the general calibration method is to drive or compare multiple channels simultaneously in the control logic: for the driving edge, the difference in the signal edge moment generated by multiple channels needs to be within the required range; for the comparison edge, the difference in the same edge moment scanned by multiple channels needs to be within the required range. In addition, since signal propagation will cause time delays due to the routing path (such as cables, PCB routing), the path time delay RTD (Route Time Delay) needs to be eliminated so that the difference in the driving and comparison edge moments can be within the required range.

[0004] There are two obvious difficulties in EPA calibration: one is the high accuracy requirement, which is generally in the tens to hundreds of picoseconds. Therefore, it is necessary to purchase expensive instruments or develop special calibration equipment for calibration, which is costly. The second is the large number of channels. Because the board channel density in digital test equipment is relatively high, such as the common number of channels is 64 or 128, most existing technical solutions use 2 to 4 channel comparisons, which are repeated many times to obtain the relative delay of all channel driving and comparison, thereby obtaining the compensation time of all channel driving and comparison. However, due to the small number of channels for a single comparison, it is necessary to repeat many times, and the calibration probe may need to be moved after each comparison. The calibration takes a long time, which is not conducive to product maintenance. Summary of the invention

[0005] The object of the present invention is to provide a multi-digital channel EPA calibration method to solve the following technical problems:

[0006] Due to the high accuracy requirements, it is necessary to purchase expensive instruments or develop special calibration equipment for calibration, which is costly. The number of single comparison channels is small, so it needs to be repeated many times, and the calibration probe may need to be moved after each comparison. The calibration takes a long time, which is not conducive to product maintenance.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A multi-digital channel EPA calibration method comprises the following steps:

[0009] S1: Determine a first calibration board, and obtain a functional relationship f(t) of voltage variation over time during a signal rise process of each digital channel of the first calibration board when the digital channels are driven, and the wiring lengths from the endpoints to the connection points of the digital channels are the same;

[0010] S2: based on the functional relationship f(t), the judgment parameters are obtained, the judgment parameters are the rise time, the average rise speed and the maximum rise speed, and the mean value F1 of the rise time, the mean value F2 of the average rise speed and the mean value F3 of the maximum rise speed after m times of driving are calculated, and the coordinate points (F1, F2, F3) are generated;

[0011] Determine whether there is an abnormality in the digital channel according to the coordinate points, and if there is no abnormality in the digital channel, execute step S3;

[0012] S3: driving a preset comparison channel and a digital channel to obtain first delay data, and calibrating a comparison edge of the digital channel based on the first delay data;

[0013] Determine a second calibration board, connect two adjacent digital channels on the second calibration board so that the wiring length between the two adjacent digital channels is the same, determine second delay data according to the delay when the two connected digital channels are driven, and calibrate the digital channel driving edge based on the second delay data;

[0014] The path time delay is eliminated based on the second delay data.

[0015] As a further solution of the present invention: in step S2, the process of determining whether there is an abnormality in the digital channel specifically includes:

[0016] Step 1: With the coordinate point as the center, construct a circle with a radius of a preset value R, determine the coordinate point a that is closest to the origin, use it as the first coordinate point A, remove the coordinate points in the circle AY corresponding to the coordinate point A, repeat the above steps until all the coordinate points are removed, and record the number of removed circles C1;

[0017] Step 2: Determine the coordinate point b that is closest to the coordinate point a, use it as the first coordinate point B, remove the coordinate points in the circle BY corresponding to the coordinate point B, until all the coordinate points are removed, and record the number of removed circles C2;

[0018] Step 3: Repeat steps 1 and 2 until the distance between the first coordinate point and the coordinate point a is greater than the radius R, and generate the quantity set and C jh = (C1, C2, …, C k ), C k Indicates the number of circles removed when determining the first coordinate point for the kth time;

[0019] Determine the minimum quantity C min =min(C jh ), the minimum number C min The corresponding removed circle is used as the target circle, the center coordinates of the target circle are determined and used as the target coordinates, and it is determined whether the corresponding digital channel has an abnormality based on the target coordinates.

[0020] As a further solution of the present invention, the process of determining whether there is an abnormality in the digital channel further comprises the following steps:

[0021] Calculate the judgment value P=η*(FYX2*FYX3) / FYX1, where FYX1, FYX2, and FYX3 represent the rise time, average rise speed, and maximum rise speed corresponding to the target coordinates, respectively. When the judgment value is less than or equal to the preset judgment value threshold, it is determined that the corresponding digital channel is abnormal.

[0022] When the judgment value is greater than the judgment value threshold, it is determined that there is no abnormality in the corresponding digital channel.

[0023] As a further solution of the present invention: when there are two or more of the same number and it is the minimum number, perform the following steps:

[0024] Mark the same and minimum number as the candidate number, take the removed circle corresponding to the candidate number as the candidate circle, determine the candidate circle with the largest coordinate point density, and take the corresponding candidate number as the minimum number.

[0025] As a further solution of the present invention: the process of determining the density of the coordinate points of the circle to be selected specifically includes:

[0026] Calculate the density of the coordinate point G = g / (πR 2 ), g represents the number of coordinate points in the circle to be selected.

[0027] As a further solution of the present invention: in step S2, the process of obtaining the judgment parameter specifically includes:

[0028] Calculate average ascent speed , where [t sta , t end ] represents the domain of the functional relation f(t);

[0029] Determine the tangent slope of each point on the graph of the functional relationship f(t), and take the maximum value thereof as the maximum rising speed.

[0030] As a further solution of the present invention: in the process of calculating the mean F1 of the rise time, the mean F2 of the average rise speed and the mean F3 of the maximum rise speed after m times of driving, when the difference between the judgment parameter obtained by driving the digital channel at a certain time and the corresponding mean is greater than a preset difference threshold, the judgment parameter obtained by driving the digital channel at that time is removed and the mean is calculated again.

[0031] As a further solution of the present invention: in the step S3, the comparison edge and the driving edge of the digital channel are calibrated to eliminate the path time delay, which specifically includes:

[0032] The i-th digital channel is used as a comparison channel, and the remaining digital channels are driven to obtain the first delay data Di, which represents the difference between the rise / fall time of the remaining digital channels when driven and the rise / fall time of the comparison channel, and generates the first delay data set D JH = (D1, D2, …, D n ), n represents the total number of digital channels, and the maximum first delay data Dmax=max(D JH ), the comparison edge of the jth digital channel increases by Dmax-D j Delay to achieve calibration of digital channel comparison edge;

[0033] Determine a second calibration board, connect two adjacent digital channels on the second calibration board, and make sure that the wiring length between the two adjacent digital channels is the same, compare the delay when the x-th digital channel is driven with the digital channel x' connected to the x-th digital channel, and obtain the second delay data DS x The second delay data represents the difference between the rise / fall time when the xth digital channel is driven and the rise / fall time when the digital channel x' is driven, and generates a second delay data set DS JH = (DS1, DS2, …, DS n ), get the minimum second delay data DSmin=min(DS JH ), the driving edge of the yth digital channel is increased by DS y -DSmin delay to calibrate the driving edge of the digital channel;

[0034] Let the path time delay RTD = DSmin, add it to the delay of all digital channel comparison edges to eliminate the path time delay RTD.

[0035] The beneficial effects of the present invention are as follows: in this solution, by judging whether there is an abnormality in the digital channel of the first calibration board, incorrect calibration caused by the abnormality of the digital channel is prevented; in this solution, when a number of channels (such as 64 or 128) are driven in the same comparison, even if they have not been calibrated, due to the existence of the above-mentioned application premise (the routing lengths from the end points to the connection points of the digital channels of the first calibration board are the same), a steep rising / falling edge can be generated, and the comparison edge can be calibrated by using the edge; and since there are a large number of channels to be driven, removing any channel will not affect the timing and steepness of the generated edge; after calibrating and compensating the comparison edge of each channel, the calibrated comparison edge can be used to calibrate the driving edge by short-connecting the channels in pairs, and the path time delay RTD (Route Time Delay) from driving to comparison can be obtained. Delay); then calibrate and compensate the driver, and eliminate the path time delay RTD; this solution makes full use of the characteristics of the multi-digital channel board itself, and only requires two PCB boards (i.e., the first calibration board and the second calibration board), without the need to purchase an oscilloscope or develop dedicated calibration equipment, which reduces the high cost of calibration equipment compared to the existing technology. Since the two PCBs require almost no additional maintenance, the maintenance cost during use is also greatly reduced; at the same time, by installing the two calibration boards in succession and running the calibration program, the EPA calibration of multiple digital channels can be completed quickly and automatically, which is very convenient for the periodic maintenance of the production line. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The present invention will be further described below in conjunction with the accompanying drawings.

[0037] Figure 1 It is a flow chart of a multi-digital channel EPA calibration method of the present invention;

[0038] Figure 2 is a schematic diagram of the wiring of the first layer on the first calibration board of the present invention;

[0039] Figure 3 is a schematic diagram of the wiring of the second layer on the first calibration board of the present invention;

[0040] Figure 4 is a schematic diagram of the wiring of the third layer on the first calibration board of the present invention;

[0041] Figure 5 is a schematic diagram of the wiring of the fourth layer on the first calibration board of the present invention;

[0042] Figure 6It is a schematic diagram of digital channel connection of the second calibration board of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] See also Figure 1 As shown, the present invention is a multi-digital channel EPA calibration method, comprising the following steps:

[0045] S1: Determine a first calibration board, and obtain a functional relationship f(t) of voltage variation over time during a signal rise process of each digital channel of the first calibration board when the digital channels are driven, and the wiring lengths from the endpoints to the connection points of the digital channels are the same;

[0046] S2: based on the functional relationship f(t), the judgment parameters are obtained, the judgment parameters are the rise time, the average rise speed and the maximum rise speed, and the mean value F1 of the rise time, the mean value F2 of the average rise speed and the mean value F3 of the maximum rise speed after m times of driving are calculated, and the coordinate points (F1, F2, F3) are generated;

[0047] In a preferred embodiment of the present invention, the process of obtaining the judgment parameter specifically includes:

[0048] Calculate average ascent speed , where [t sta , t end ] represents the domain of the functional relation f(t);

[0049] Determine the tangent slope of each point on the graph of the functional relationship f(t), and take the maximum value thereof as the maximum rising speed.

[0050] In another preferred embodiment of the present invention, in the process of calculating the mean F1 of the rise time, the mean F2 of the average rise speed and the mean F3 of the maximum rise speed after m times of driving, when the difference between the judgment parameter obtained by driving the digital channel at a certain time and the corresponding mean is greater than a preset difference threshold, the judgment parameter obtained by driving the digital channel at that time is removed and the mean is calculated again.

[0051] Determine whether there is an abnormality in the digital channel according to the coordinate points, and if there is no abnormality in the digital channel, execute step S3;

[0052] It is worth noting that when the digital channel is abnormal, a prompt message is sent to report an error;

[0053] In a preferred embodiment of the present invention, the process of determining whether a digital channel is abnormal specifically includes:

[0054] Step 1: With the coordinate point as the center, construct a circle with a radius of a preset value R, determine the coordinate point a that is closest to the origin, use it as the first coordinate point A, remove the coordinate points in the circle AY corresponding to the coordinate point A, repeat the above steps until all the coordinate points are removed, and record the number of removed circles C1;

[0055] Step 2: Determine the coordinate point b that is closest to the coordinate point a, use it as the first coordinate point B, remove the coordinate points in the circle BY corresponding to the coordinate point B, until all the coordinate points are removed, and record the number of removed circles C2;

[0056] Step 3: Repeat steps 1 and 2 until the distance between the first coordinate point and the coordinate point a is greater than the radius R, and generate the quantity set and C jh = (C1, C2, …, C k ), C k Indicates the number of circles removed when determining the first coordinate point for the kth time;

[0057] Determine the minimum quantity C min =min(C jh ), the minimum number C min The corresponding removed circle is used as the target circle, the center coordinates of the target circle are determined and used as the target coordinates, and it is determined whether the corresponding digital channel has an abnormality based on the target coordinates.

[0058] It is worth noting that, firstly, the function relationship f(t) is obtained to convert the original signal into a standardized mathematical model, which is convenient for subsequent parameter calculation. The key performance parameters such as rise time, average rise speed and maximum rise speed are extracted through the function relationship f(t). These parameters reflect the electrical characteristics of the channel, such as signal transmission delay and speed change. This ensures the physical meaning of the test, is directly linked to the channel performance, and ensures the physical validity of the analysis basis. After that, based on these performance parameters, the average value of multiple driving experiments is calculated to reduce random errors, eliminate accidental fluctuations in the test, enhance the stability and credibility of the data, and finally obtain the three-dimensional coordinate points. By constructing these coordinate points in three-dimensional space, the scheme converts the dynamic behavior of the digital channel into the distribution problem of spatial points, which is convenient for subsequent cluster analysis. Then, in the minimum circle search process, the scheme selects the circle with the least number of removed points and regards it as the target circle. The minimum coverage principle is used to ensure that the selected target circle is representative of normal data. The coordinates of the center of the target circle therefore become the standard reference point for the performance of the digital channel and are used to calculate the judgment value to determine whether the digital channel has an abnormality.

[0059] In a preferred situation of this embodiment, the process of determining whether there is an abnormality in the digital channel further includes the following steps:

[0060] Calculate the judgment value P=η*(FYX2*FYX3) / FYX1, where FYX1, FYX2, and FYX3 represent the rise time, average rise speed, and maximum rise speed corresponding to the target coordinates, respectively. When the judgment value is less than or equal to the preset judgment value threshold, it is determined that the corresponding digital channel is abnormal.

[0061] When the judgment value is greater than the judgment value threshold, it is determined that there is no abnormality in the corresponding digital channel.

[0062] It can be understood that the determination method of the second calibration board also refers to the above-mentioned step of determining whether there is an abnormality in the digital channel on the first calibration board.

[0063] It should be noted that when there are two or more quantities that are the same and are the minimum quantity, the following steps are performed:

[0064] Mark the same and minimum number as the candidate number, take the removed circle corresponding to the candidate number as the candidate circle, determine the candidate circle with the largest coordinate point density, and take the corresponding candidate number as the minimum number.

[0065] It is worth noting that the process of determining the density of the coordinate points of the circle to be selected specifically includes:

[0066] Calculate the density of the coordinate point G = g / (πR 2 ), g represents the number of coordinate points in the circle to be selected.

[0067] S3: driving a preset comparison channel and a digital channel to obtain first delay data, and calibrating a comparison edge of the digital channel based on the first delay data;

[0068] Determine a second calibration board, connect two adjacent digital channels on the second calibration board so that the wiring length between the two adjacent digital channels is the same, determine second delay data according to the delay when the two connected digital channels are driven, and calibrate the digital channel driving edge based on the second delay data;

[0069] Eliminating the path time delay based on the second delay data;

[0070] In a preferred embodiment of the present invention, the process of calibrating the comparison edge and the driving edge of the digital channel and eliminating the path time delay specifically includes:

[0071] The i-th digital channel is used as a comparison channel, and the remaining digital channels are driven to obtain the first delay data Di, which represents the difference between the rise / fall time of the remaining digital channels when driven and the rise / fall time of the comparison channel, and generates the first delay data set D JH = (D1, D2, …, D n ), n represents the total number of digital channels, and the maximum first delay data Dmax=max(D JH ), the comparison edge of the jth digital channel increases by Dmax-D j Delay to achieve calibration of digital channel comparison edge;

[0072] Determine a second calibration board, connect two adjacent digital channels on the second calibration board, and make sure that the wiring length between the two adjacent digital channels is the same, compare the delay when the x-th digital channel is driven with the digital channel x' connected to the x-th digital channel, and obtain the second delay data DS x The second delay data represents the difference between the rise / fall time when the xth digital channel is driven and the rise / fall time when the digital channel x' is driven, and generates a second delay data set DS JH = (DS1, DS2, …, DS n ), get the minimum second delay data DSmin=min(DS JH ), the driving edge of the yth digital channel is increased by DS y -DSmin delay to calibrate the driving edge of the digital channel;

[0073] Let the path time delay RTD = DSmin, add it to the delay of all digital channel comparison edges to eliminate the path time delay RTD;

[0074] It is worth noting that when the above method is implemented on the first calibration board, multiple (eg, 64, 128) digital channels are connected to one point, ie, the connection point, using strictly equal-length traces.

[0075] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A multi-digital channel EPA calibration method, characterized in that: The following steps are involved: S1: Determine a first calibration board, and obtain a functional relationship f(t) of voltage variation over time during a signal rise process of each digital channel of the first calibration board when the digital channel is driven, and the wiring lengths from the endpoints of the digital channels to the connection points are the same, and the number of the connection points is 1; S2: based on the functional relationship f(t), the judgment parameters are obtained, the judgment parameters are the rise time, the average rise speed and the maximum rise speed, and the mean value F1 of the rise time, the mean value F2 of the average rise speed and the mean value F3 of the maximum rise speed after m times of driving are calculated, and the coordinate points (F1, F2, F3) are generated; Determine whether there is an abnormality in the digital channel according to the coordinate points, and if there is no abnormality in the digital channel, execute step S3; S3: driving a preset comparison channel and a digital channel to obtain first delay data, and calibrating a comparison edge of the digital channel based on the first delay data; Determine a second calibration board, connect two adjacent digital channels on the second calibration board so that the wiring length between the two adjacent digital channels is the same, determine second delay data according to the delay when the two connected digital channels are driven, and calibrate the digital channel driving edge based on the second delay data; Eliminating the path time delay based on the second delay data; The process of calibrating the comparison edge and driving edge of the digital channel and eliminating the path time delay includes: The i-th digital channel is used as a comparison channel, and the remaining digital channels are driven to obtain the first delay data Di, which represents the difference between the rise / fall time of the remaining digital channels when driven and the rise / fall time of the comparison channel, and generates the first delay data set D JH = (D1, D2, …, D n ), n represents the total number of digital channels, and the maximum first delay data Dmax=max(D JH ), the comparison edge of the jth digital channel increases by Dmax-D j Delay to achieve calibration of digital channel comparison edge; Determine a second calibration board, connect two adjacent digital channels on the second calibration board, and make sure that the wiring length between the two adjacent digital channels is the same, compare the delay when the x-th digital channel is driven with the digital channel x' connected to the x-th digital channel, and obtain the second delay data DS x The second delay data represents the difference between the rise / fall time when the xth digital channel is driven and the rise / fall time when the digital channel x' is driven, and generates a second delay data set DS JH = (DS1, DS2, …, DS n ), get the minimum second delay data DSmin=min(DS JH ), the driving edge of the yth digital channel is increased by DS y -DSmin delay to calibrate the driving edge of the digital channel; Let the path time delay RTD = DSmin, add it to the delay of all digital channel comparison edges to eliminate the path time delay RTD.

2. The EPA calibration method for multiple digital channels according to claim 1, characterized in that: In step S2, the process of determining whether there is an abnormality in the digital channel specifically includes: Step 1: With the coordinate point as the center, construct a circle with a radius of a preset value R, determine the coordinate point a that is closest to the origin, use it as the first coordinate point A, remove the coordinate points in the circle AY corresponding to the coordinate point A, repeat the above steps until all the coordinate points are removed, and record the number of removed circles C1; Step 2: Determine the coordinate point b that is closest to the coordinate point a, use it as the first coordinate point B, remove the coordinate points in the circle BY corresponding to the coordinate point B, until all the coordinate points are removed, and record the number of removed circles C2; Step 3: Repeat steps 1 and 2 until the distance between the first coordinate point and the coordinate point a is greater than the radius R, and generate the quantity set and C jh = (C1, C2, …, C k ), C k Indicates the number of circles removed when determining the first coordinate point for the kth time; Determine the minimum quantity C min =min(C jh ), the minimum number C min The corresponding removed circle is used as the target circle, the center coordinates of the target circle are determined and used as the target coordinates, and it is determined whether the corresponding digital channel has an abnormality based on the target coordinates.

3. The EPA calibration method for multiple digital channels according to claim 2, characterized in that: The process of determining whether there is an abnormality in the digital channel also includes the following steps: Calculate the judgment value P=η*(FYX2*FYX3) / FYX1, where FYX1, FYX2, and FYX3 represent the rise time, average rise speed, and maximum rise speed corresponding to the target coordinates, respectively. When the judgment value is less than or equal to the preset judgment value threshold, it is determined that the corresponding digital channel is abnormal. When the judgment value is greater than the judgment value threshold, it is determined that there is no abnormality in the corresponding digital channel.

4. The EPA calibration method for multiple digital channels according to claim 2, characterized in that: When there are two or more quantities that are the same and are the minimum quantity, perform the following steps: Mark the same and minimum number as the candidate number, take the removed circle corresponding to the candidate number as the candidate circle, determine the candidate circle with the largest coordinate point density, and take the corresponding candidate number as the minimum number.

5. The EPA calibration method for multiple digital channels according to claim 4, characterized in that: The process of determining the density of the coordinate points of the circle to be selected specifically includes: Calculate the density of the coordinate point G = g / (πR 2 ), g represents the number of coordinate points in the circle to be selected.

6. The EPA calibration method for multiple digital channels according to claim 1, characterized in that: In step S2, the process of obtaining the judgment parameter specifically includes: Calculate average ascent speed , where [t sta , t end ] represents the domain of the functional relation f(t); Determine the tangent slope of each point on the graph of the functional relationship f(t), and take the maximum value thereof as the maximum rising speed.

7. The EPA calibration method for multiple digital channels according to claim 1, characterized in that: In the step S2, in the process of calculating the mean F1 of the rise time, the mean F2 of the average rise speed and the mean F3 of the maximum rise speed after m times of driving, when the difference between the judgment parameter obtained by driving the digital channel at a certain time and the corresponding mean is greater than the preset difference threshold, the judgment parameter obtained by driving the digital channel at that time is removed and the mean is calculated again.

Images