A synchronous calibration method based on multi-device control

Abstract

The present invention relates to the technical field of equipment synchronization calibration, and specifically discloses a synchronization calibration method based on multi-device control, including: designing a calibration board, and counting the devices connected to the calibration board and requiring synchronization calibration; controlling any two devices to communicate, obtaining the time delay generated in the process of a signal from a sending device to a receiving device during the communication process, and generating a matrix T after traversing all combinations; operating the matrix T to obtain a matrix C; calculating the average of each row of the matrix C, and obtaining the maximum value therein; and using the difference between the maximum value and the average corresponding to the device as the time delay that the device needs to compensate. The present invention can achieve synchronization of device sending / receiving signals based on the time point of the device sending / receiving signals.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment synchronization calibration, and in particular to a synchronization calibration method based on multi-equipment control. Background Art

[0002] ATE test machine, full name Automatic Test Equipment, is a device used for functional testing of semiconductor chips. It is composed of a large number of test functions and is controlled by a computer for testing.

[0003] Because the ATE machine is composed of a single board, each board has a certain number of test channels. Since the number of test channels of the customer's chip may be greater than the number of test channels of a single board, or the test channels of the chip use the site positions of different boards, there may be such a situation: a chip is tested by the test channels of more than two boards. Therefore, the ATE test machine must ensure that all sub-board test channels of the entire system remain synchronized, otherwise it will affect the timing of the chip test. Due to SMT differences and different distances between boards, the PIPE times of different boards are not consistent, so the boards of the ATE test machine need to be synchronized and calibrated.

[0004] The existing solution is based on the synchronization calibration of the controller on the board. Although it can achieve relatively accurate synchronization between boards, it still has the following disadvantages:

[0005] 1. From the controller sending out control instructions to the final signal output and then to the user interface, it needs to go through dedicated conversion chips, PCB traces, connectors, cables, spring pins, etc. These signal paths cannot be guaranteed to be completely consistent during mass production, which is reflected in different time delays. Existing solutions cannot detect these delays, nor can they be calibrated or compensated;

[0006] 2. The boards in the ATE test machine generally have controllers, and the controllers have calibration / compensation functions; but some devices outside the test machine environment do not have controllers, or the controllers do not have the function of synchronous calibration between boards. Summary of the invention

[0007] The purpose of the present invention is to provide a synchronous calibration method based on multi-device control to solve the above technical problems.

[0008] The purpose of the present invention can be achieved through the following technical solutions:

[0009] A synchronous calibration method based on multi-device control comprises the following steps:

[0010] Step S1: designing a calibration board and counting the devices connected to the calibration board and requiring synchronous calibration;

[0011] Step S2: Control any two devices to communicate, obtain the time delay generated when the signal reaches the receiving device from the sending device during the communication process, and generate a matrix T after traversing all combinations;

[0012] Step S3: Get the transposed matrix T of matrix T T , operate on the matrix T to obtain the matrix TT T , and the matrix TT T Convert to get matrix ; For the matrix Supplement the diagonal to get the matrix C;

[0013] Step S4: Calculate the average of each row of the matrix C and obtain the maximum value therein; use the difference between the maximum value and the average corresponding to the device as the time delay that the device needs to compensate.

[0014] As a further solution of the present invention: in the calibration board, the device is connected to the common point through two relays, and the connection between the device and the calibration board is achieved through the on-off function of the relay.

[0015] As a further solution of the present invention: in the step S2, the specific method of generating the matrix T is as follows:

[0016] The number of devices connected to the calibration board and performing synchronous calibration is set to n;

[0017] Mark each device as D1, D2, ..., Dn; and set the time for each device to send / receive a signal as T1, T2, ..., Tn;

[0018] Choose any two devices Di and Dj, command Di to send and Dj to receive, and get the time delay t ij ; Command Dj to send Di to receive, and get the time delay t ji ; i, j are positive integers, i∈[1, n], j∈[1, n] and i≠j;

[0019] After traversing all combinations of Di and Dj, we get the matrix T:

[0020] .

[0021] As a further solution of the present invention: in step S3, the specific method of obtaining the matrix C is as follows:

[0022] Operate on matrix T to get matrix TT T :

[0023] ;

[0024] According to t ij -t ji =2(T i -T j ) for the matrix TT T Convert to get matrix :

[0025] ;

[0026] Pair Matrix Diagonal supplement Ti-Ti=0 to get the matrix C:

[0027] .

[0028] As a further solution of the present invention: in step S4, the specific method of calculating the time delay that the device needs to compensate includes:

[0029] Calculate the average of each row of matrix C, and we get:

[0030] ;

[0031] ;

[0032] ;

[0033] Take the maximum value ;

[0034] Then the time delay that device Di needs to compensate is ti=C max -C i .

[0035] As a further solution of the present invention: when obtaining the time delay t ij or ji During the process, only the relays connecting devices Di and Dj to the common point are closed.

[0036] As a further solution of the present invention: obtaining the time delay t ij The steps specifically include:

[0037] When the controller sends a synchronization instruction to the devices Di and Dj at the same time, the time when the devices Di and Dj receive the synchronization instruction is set to TCi and TCj respectively;

[0038] After receiving the synchronization instruction, the devices Di and Dj start to send / receive signals at times Ti and Tj respectively;

[0039] The time when device Dj actually receives the signal sent by Di is Tj';

[0040] Then the time delay t ij =Tj'-Tj.

[0041] As a further solution of the present invention: the time for the signal to be transmitted from Di to Dj is Tj'-Ti.

[0042] The beneficial effects of the present invention are as follows: compared with the prior art, from the controller sending out a control instruction to the final signal output and then to the user interface, it is necessary to go through a dedicated conversion chip, PCB routing, connectors, cables, spring pins, etc., and these delay problems cannot be detected; the present invention is based on the time point when the device sends / receives the signal, and can realize the synchronization of the device sending / receiving the signal, eliminating the differences caused by the device and routing delays on the internal link of the device, and can also be applied to devices without a controller or the controller does not support the corresponding calibration scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The present invention will be further described below in conjunction with the accompanying drawings.

[0044] Figure 1 It is a schematic diagram of communication time nodes of devices D0 and D1 in one embodiment of the present invention;

[0045] Figure 2 It is a schematic diagram of the structure of a calibration board in another embodiment of the present invention. DETAILED DESCRIPTION

[0046] 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.

[0047] The present invention is a synchronous calibration method based on multi-device control, which can be applied to the production, maintenance and use of ATE, and can also be applied to the maintenance and use of several clock homologous devices.

[0048] The following steps are involved:

[0049] Step S1: designing a calibration board and counting the devices connected to the calibration board and requiring synchronous calibration;

[0050] Step S2: Control any two devices to communicate, obtain the time delay generated when the signal reaches the receiving device from the sending device during the communication process, and generate a matrix T after traversing all combinations;

[0051] Step S3: Get the transposed matrix T of matrix T T, operate on the matrix T to obtain the matrix TT T , and the matrix TT T Convert to get matrix ; For the matrix Supplement the diagonal to get the matrix C;

[0052] Step S4: Calculate the average of each row of the matrix C and obtain the maximum value therein; use the difference between the maximum value and the average corresponding to the device as the time delay that the device needs to compensate.

[0053] like Figure 1 As shown, the two devices are D0 and D1, which are controlled by the same clock. The controller is at a certain moment ( Figure 1 The origin of the number axis in the figure) sends synchronization instructions to devices D0 and D1 at the same time. Assume that the time when the two devices receive the synchronization instructions from the controller is TC0 and TC1. TC0 and TC1 represent the time required for the synchronization instructions to be transmitted from the controller to devices D0 and D1 respectively. Since the clocks are of the same source and the transmission link remains unchanged, the relative time delays of TC0 and TC1 (TC1-TC0) remain unchanged; T0 and T1 represent the time when the two devices send / receive signals respectively, that is, the reaction time of the device from receiving the instruction to executing the action. Since the internal structure of the device remains unchanged, it can be considered that TC0-T0 and TC1-T1 remain unchanged. The specific values ​​depend on the device itself. From the above two points, it can be seen that the relative time delay T0-T1 of T0 and T1 remains unchanged.

[0054] And, in Figure 1 In the example, the first axis represents the time record of the signal sent by D0 and received by D1, where t 01 It represents the time interval from when device D1 starts to perform the receiving action to when it actually receives the signal, that is, the time delay; td represents the time interval from when the signal is sent by device D0 to when it is transmitted to device D1, that is, the signal transmission time; the second number axis represents the time record in the usage scenario where the signal is sent by D1 and received by D0. The principle is the same as above and will not be repeated.

[0055] Use a board or cable to connect two devices. The controller sends synchronization instructions to devices D0 and D1 at the same time, and instructs D0 to send and D1 to receive. Then D0 and D1 receive the synchronization instructions at TC0 and TC1 respectively. D0 sends a signal at T0, and D1 starts receiving timing at T1. The signal passes through the board or cable with a delay td and is received by D1 at T1'. The time delay t can be measured. 01 ; command D1 to send and D0 to receive at the same time, the time delay can be measured by the same principle. Considering that no matter which device sends or receives, the signal transmission time on the board or cable will not change, that is, td, the following relationship can be obtained:

[0056] t 01 -t 10 =2(T0-T1);

[0057] Then we calculate the time t1 as follows and compensate it to the D1 device to achieve synchronization of the two devices:

[0058] t1=T0-T1=0.5(t 01 -t 10 ).

[0059] Also, it should be noted here that in the process of calculating the compensation time t1, why T0 and T1 are not directly substituted for calculation, but the time delay t 01 and t 10 The calculation is performed because, in actual use, the actions corresponding to T0 and T1 occur inside the device and are difficult to measure, resulting in greater difficulty in obtaining and lower accuracy, making it impossible to achieve synchronous calibration.

[0060] In another preferred embodiment of the present invention, different from the previous embodiment in which there are only two devices, in the case of multiple devices, if you want to achieve synchronous calibration between multiple devices, you need to perform expansion on the original basis, as shown below:

[0061] Design a calibration board that can realize the interconnection between any two devices, and in order to ensure the signal quality, when the two devices are connected, the connection with other devices needs to be disconnected. Figure 2 The figure shows an implementation of the calibration board, which can achieve synchronization and calibration of up to 20 devices. Each device is connected to the common point through two relays, and the above functions are realized through the on-off control of the relays. At the same time, the relay near the common point should be as close to the common point as possible to reduce the length of the stub line when disconnected to ensure signal quality.

[0062] Assume that the number of devices to be synchronized and calibrated is n, then the time when each device sends / receives the signal is T1~Tn (T1~Tn here is only used for the derivation below. The specific time cannot be measured in actual use and is not substituted into the formula for calculation). Select devices Di and Dj, command Di to send and Dj to receive. To ensure signal quality, only close the relay connecting Di and Dj to the common point, and get the time delay t ij , then command Dj to send Di to receive, and get the time delay t ji , when Di and Dj traverse all combinations, the following matrix can be obtained:

[0063] ;

[0064] Perform matrix operations:

[0065] ;

[0066] According to the description in the inventive principle, it can be obtained that:

[0067] t ij -t ji =2(T i -T j );

[0068] Then for the matrix TT T Calculation yields:

[0069] ;

[0070] Supplementing the above matrix diagonal with 0=Ti-Ti, we can get the following matrix

[0071] ;

[0072] Calculate the average of each row of matrix C, and we can get

[0073]

[0074] Take the maximum value

[0075] ;

[0076] Assume that the corresponding device is Dq, and the time when it sends / receives the signal is Tq. According to the description in the inventive principle, the time delay that each device needs to compensate is:

[0077]

[0078] By compensating the above time to each device, synchronous calibration of multiple devices can be achieved.

[0079] 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 synchronous calibration method based on multi-device control, characterized in that: The following steps are involved: Step S1: designing a calibration board and counting the devices connected to the calibration board and requiring synchronous calibration; Step S2: Control any two devices to communicate, obtain the time delay generated when the signal reaches the receiving device from the sending device during the communication process, and generate a matrix T after traversing all combinations; Step S3: Get the transposed matrix T of matrix T T , operate on the matrix T to obtain the matrix TT T , and the matrix TT T Convert to get matrix ; For the matrix Supplement the diagonal to get the matrix C; Step S4: Calculate the average of each row of the matrix C and obtain the maximum value; The difference between the maximum value and the average value corresponding to the device is taken as the time delay that the device needs to compensate.

2. A synchronous calibration method based on multi-device control according to claim 1, characterized in that: In the calibration board, the device is connected to the common point through two relays, and the connection between the device and the calibration board is achieved through the on-off function of the relays.

3. The synchronous calibration method based on multi-device control according to claim 1, characterized in that: In step S2, the specific method of generating the matrix T is as follows: The number of devices connected to the calibration board and performing synchronous calibration is set to n; Mark each device as D1, D2, ..., Dn; and set the time for each device to send / receive a signal as T1, T2, ..., Tn; Choose any two devices Di and Dj, command Di to send and Dj to receive, and get the time delay t ij ; Command Dj to send and Di to receive, and get the time delay t ji ; i, j are positive integers, i∈[1,n], j∈[1,n] and i≠j; After traversing all combinations of Di and Dj, we get the matrix T: 。 4. The synchronous calibration method based on multi-device control according to claim 3, characterized in that: In step S3, the specific method of obtaining the matrix C is as follows: Operate on matrix T to get matrix TT T : ； According to t ij -t ji =2(T i -T j ) for the matrix TT T Convert to get matrix : ； Pair Matrix Diagonal supplement Ti-Ti=0 to get the matrix C: 。 5. A synchronous calibration method based on multi-device control according to claim 4, characterized in that: In step S4, the specific method of calculating the time delay that the device needs to compensate includes: Calculate the average of each row of matrix C, and we get: ； ； ； Take the maximum value ; Then the time delay that device Di needs to compensate is ti=C max -C i .

6. The synchronous calibration method based on multi-device control according to claim 2, characterized in that: In obtaining the time delay t ij or ji During the process, only the relays connecting devices Di and Dj to the common point are closed.

7. The synchronous calibration method based on multi-device control according to claim 3, characterized in that: Get the time delay t ij The steps specifically include: When the controller sends a synchronization instruction to the devices Di and Dj at the same time, the time when the devices Di and Dj receive the synchronization instruction is set to TCi and TCj respectively; After receiving the synchronization instruction, the devices Di and Dj start to send / receive signals at times Ti and Tj respectively; The time when device Dj actually receives the signal sent by Di is Tj'; Then the time delay t ij =Tj'-Tj.

8. A synchronous calibration method based on multi-device control according to claim 7, characterized in that: The time it takes for a signal to be transmitted from Di to Dj is Tj'-Ti.

Images