Chip performance optimization method based on single-chip programming voltage adjustment

Abstract

The invention discloses a chip performance optimization method based on single-chip programming voltage adjustment, and relates to the technical field of chip performance optimizing.The method comprises the steps that trimming is started, an initial parameter set is determined firstly, and a to-be-programmed object is selected; performing programming operation on the programming object by using the initial parameter; if half or more of the test units in the programming object are successfully programmed, determining the final value of the parameter, and writing and fixing the calculated final value into the ROM Fuse of the chip; if no half or more test units in the programming object are successfully programmed, whether the upper limit of the cycle index is reached or not is judged; if the upper limit of the cycle index is not reached, the parameter value is increased, and programming is carried out again; and if the upper limit of the cycle index is reached, the chip is judged to be failed in trimming. According to the invention, parameter adjustment is based on a single chip, and adaptive parameter adjustment is carried out by taking the chip as a unit, so that the adaptive to performance differences of different chips is met.

Description

Technical Field

[0001] The invention relates to the technical field of chip performance optimization, and in particular to a chip performance optimization method based on single-chip programming voltage adjustment. Background Art

[0002] With the continuous development of technology, single chips are increasingly used in various electronic devices. In order to improve the performance and reliability of single chips, the programming parameters of the chip need to be accurately adjusted. The programming voltage is one of the key parameters that affect the performance of a single chip. A reasonable programming voltage can improve programming efficiency, reduce bit error rate, and extend chip life.

[0003] The programming voltage adjustment method is usually manually adjusted based on empirical data and experimental results. This method is not only time-consuming and laborious, but also difficult to ensure that each chip can achieve optimal performance. In addition, due to process fluctuations during the production process, different batches of chips may require different programming voltages to achieve optimal performance.

[0004] Traditional methods rely on manual experience and experimental data. The adjustment process is complex and time-consuming and cannot quickly adapt to large-scale production needs. Moreover, manual adjustment can hardly guarantee that each chip can achieve optimal performance and performance fluctuations are prone to occur. Different batches of chips may require different programming voltages, and traditional methods cannot flexibly respond to these changes. Existing methods lack systematic adjustment processes and calculation methods, making it difficult to achieve automation and standardization. Summary of the invention

[0005] The purpose of the present invention is to provide a chip performance optimization method based on single-chip programming voltage adjustment to solve the problems raised in the prior art. To achieve the above purpose, the present invention provides the following technical solution: a chip performance optimization method based on single-chip programming voltage adjustment, the method comprising the steps of:

[0006] S1. Start adjustment, first determine the initial parameter set, set other related parameters, and select the object to be programmed;

[0007] S2, performing programming operation on the programming object using the initial parameters, and after programming is completed, determining whether half or more of the test cells in the programming object are successfully programmed;

[0008] S3. If half or more of the test cells in the programming object are successfully programmed, the chip is judged to be successfully trimmed; the parameter is calculated according to the specification requirements to determine the final value of the parameter, and the calculated final value is written and fixed into the ROM Fuse of the chip to complete the trimming process;

[0009] S4, if half or more of the test cells in the programming object are not successfully programmed, determine whether the upper limit of the number of cycles is reached; if half or more of the test cells in the programming object are not successfully programmed and the upper limit of the number of cycles is not reached, increase the parameter value, return to step S2, and re-perform the programming and determination operations;

[0010] S5. If half or more of the test cells in the programming object are not successfully programmed and the upper limit of the number of cycles is reached, the chip is determined to have failed trimming.

[0011] According to the above scheme, the calculation of the new parameter value includes the steps of:

[0012] S2-1, perform Erase processing on the programming object to ensure that the programming object is in the programming state; and determine the initial parameters, maximum number of cycles and step value;

[0013] S2-2, programming the programming object using the parameters, and recording the number of programming cycles for each page, and excluding abnormal pages;

[0014] S2-3, calculating the average number of programming cycles for all normal pages;

[0015] S2-4, comparing the calculated average number of programming cycles with a preset target number of programming cycles to obtain a deviation value;

[0016] S2-5, set the correction coefficient to help correct the deviation between the target value and the actual value, and calculate the actual deviation value based on the deviation value;

[0017] S2-6. Calculate the final value of the parameter.

[0018] According to the above scheme, each page of the chip will undergo multiple programming cycles; during the programming cycle, if some pages show abnormal programming speed, these abnormal pages need to be eliminated, and the abnormal pages include data of abnormally fast programming speed of the page or abnormally slow programming speed of the page. The abnormal pages are not included in the cumulative value to avoid affecting the accuracy of the average value; abnormally fast or slow pages are usually inconsistent with the programming behavior of most pages due to process deviations or environmental influences, so these abnormal values ​​should be excluded.

[0019] According to the above scheme, the average number of programming cycles is as follows:

[0020] AL=L A / P A ;

[0021] Where AL represents the average number of programming cycles; L A Expressed as the cumulative number of cycles; P AExpressed as cumulative number of pages;

[0022] The deviation value is as follows:

[0023] P=AL-L M ;

[0024] Where P represents the deviation value; L M It is expressed as the target number of cycles;

[0025] The actual deviation value is expressed as follows:

[0026] P C =P×B;

[0027] Among them, P C It is represented by the actual deviation value, which needs to be rounded up to ensure that the final adjusted parameter value can meet the strict programming requirements; B is represented by the correction coefficient; the calculated deviation value has a large deviation and needs to be corrected by the correction coefficient B to improve the accuracy of the calculation; the correction coefficient B is an empirical value obtained based on historical data and a large number of test results;

[0028] The final value is as follows:

[0029] V n =V c +(P C ×b);

[0030] Among them, V n Expressed as final value; V c represents the current parameter value; b represents the step value. The calculated final value V n The adjustment parameters for the next programming will be written and fixed into the ROM Fuse of the chip;

[0031] According to the above solution, the upward adjustment of the parameter value includes: upward adjustment of the parameter value, the formula is as follows:

[0032] V n =V n +b;

[0033] Among them, V x Represents the adjusted parameter value; V n a represents the current parameter value; b represents the step value;

[0034] When adjusting, you need to select different step values ​​to adjust the parameter value according to the adjustment time requirements, and balance the accuracy and time. If higher adjustment accuracy is required, adjust the parameter value with a smaller step value. If less adjustment time is required, adjust the parameter value with a larger step value.

[0035] When both accuracy and reduced adjustment time are required, the parameter value needs to be adjusted according to the larger step value and the smaller step value. The formula is as follows:

[0036] V x =V n +M×(b B -b S );

[0037] Among them, V x Represents the adjusted parameter value; V n represents the current parameter value; M represents the target value; b B Indicates a larger step value; b S Expressed as smaller step values.

[0038] According to the above scheme, in the first round of adjustment, a smaller step value is selected for adjustment to ensure a higher adjustment accuracy;

[0039] In actual application, choose a larger step value for adjustment to ensure less adjustment time;

[0040] As the adjustment process progresses, the step value gradually transitions from a smaller step value to a larger step value to reduce the adjustment time. At this time, the parameter value needs to be adjusted. The formula is as follows:

[0041] V x2 =V n +(M×k)×(b B -b S );

[0042] Among them, V x2 Represents the parameter value after transition; V n represents the current parameter value; M represents the target value; k represents the compensation coefficient; b B Indicates a larger step value; b S Expressed as a smaller step value;

[0043] The transition process needs to be corrected by a compensation coefficient k to ensure that the adjusted parameters do not deviate too far from the target value when changing from a small step to a large step;

[0044] By accumulating a large amount of data, the adjusted compensation coefficient is confirmed to provide correction values ​​for different chip models or production batches; the common compensation coefficient range is x1 to x5.

[0045] According to the above scheme, during the adjustment, each step must calculate the final adjustment value based on the result, and verify the calculation through the algorithm to ensure the accuracy and consistency of the final parameters;

[0046] When new parameters are adjusted, use new programming objects and new parameters to perform adjustment tests to verify the accuracy of the new parameters. If the adjusted new parameters perform well in the actual programming test and the deviation of the target cycle number is within ±0.5 cycles, the adjustment result is considered successful.

[0047] After the new parameters are adjusted and tested, check whether the new parameters can meet the target cycle number deviation within ±0.5 cycles to ensure that the final application can meet the deviation requirements.

[0048] According to the above scheme, for circuit designs without a given number of cycles, a single-step PulsebyPulse programming method is used to calculate the number of pulses, and the obtained number of pulses represents the number of cycles.

[0049] According to the above scheme, the single-step Pulse by Pulse programming designs an initial programming voltage and pulse frequency according to the expected programming effect of the chip, applies pulses to the chip step by step according to the programming object until the chip responds to the expected programming effect, and records the change of the chip state after each pulse is applied;

[0050] The optimal programming conditions are obtained by continuously adjusting the programming voltage and pulse frequency, and after completing a programming operation, the total number of pulses required for this programming operation is counted; based on the feedback mechanism, optimization is performed so that after each pulse is applied, the state of the chip gradually approaches the desired programming effect, reducing the number of unnecessary pulses, thereby improving programming efficiency and stability;

[0051] The number of cycles is obtained by gradually recording and analyzing the number of pulses; the changing pattern of the pulse number is associated with the characteristics of the chip response, so as to infer the minimum number of pulse cycles required to achieve a specific programming effect; after multiple experiments and analyses, a standardized pulse number range can be determined to guide subsequent programming operations and ensure that programming tasks can be completed in the best way under different chips and programming conditions.

[0052] According to the above scheme, the chip determined to have failed the adjustment is judged as scrap or downgraded.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] 1. The present invention adjusts parameters based on a single chip, and performs adaptive parameter adjustment based on the chip, which can meet the needs of adapting to production line fluctuations including performance differences of different chips;

[0055] 2. The chip adaptive adjustment of the present invention does not require frequent code upgrade and maintenance for testing, and can achieve adjustment of a single chip, thus reducing the loss of yield caused by a single parameter;

[0056] 3. The present invention realizes fine monitoring of the production line and hierarchical management of key parameters of the chip by monitoring the adaptive adjustment parameters of the chip. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A schematic flow chart of a chip performance optimization method based on single-chip programming voltage adjustment according to the present invention;

[0058] Figure 2 The present invention is a flow chart of the steps of parameter calculation of a chip performance optimization method based on single-chip programming voltage adjustment. DETAILED DESCRIPTION

[0059] 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.

[0060] Example: Figure 1-Figure 2 As shown, the present invention provides a technical solution, a chip performance optimization method based on single-chip programming voltage adjustment, to determine the key programming voltage VPGM;

[0061] To start tuning, first determine the initial parameter set, set other related parameters, and select the object to be programmed;

[0062] The object to be programmed is: 128 Blocks × 64 Pages; the initial parameter set includes: initial programming voltage: VPGM = 13V; maximum loop count max Loop count = 15; step value dVPGM = 0.25V;

[0063] Erase the selected 128Block×64Pages, and use the initial parameter VPGM=13V to program the programming object, record the number of programming cycles for each page, and exclude abnormal pages; the abnormal pages include data with abnormally fast programming speed of the page or abnormally slow programming speed of the page, and the abnormal pages are not included in the cumulative value to avoid affecting the accuracy of the average value; abnormally fast or slow pages are usually inconsistent with the programming behavior of most pages due to process deviations or environmental influences, so these abnormal values ​​should be excluded.

[0064] For example, the programming results are as follows: Page 1: Loop = 10; Page 2: Loop = 11; Page 3: Loop = 12; Page 4: Loop = 2 (abnormally fast, excluded); Page 5: Loop = 30 (abnormally slow, excluded); Page 6: Loop = 10; Page 7: Loop = 9; Page 8: Loop = 11;

[0065] Cumulative number of cycles: L A =10+11+12+10+9+11=63; cumulative number of pages P A =6;

[0066] Calculate the average number of programming cycles using the following formula:

[0067] AL=L A / P A ;

[0068] Where AL represents the average number of programming cycles; L A Expressed as the cumulative number of cycles; P A Expressed as cumulative number of pages;

[0069] Average programming cycle number AL = 63 / 6 = 10.5;

[0070] Calculate the deviation value using the following formula:

[0071] P=AL-L M ;

[0072] Where P represents the deviation value; L M It is expressed as the target number of cycles;

[0073] Target number of cycles: L M =8; Deviation value: P = AL-L M =10.5-8=2.5;

[0074] Calculate the actual deviation value, the formula is as follows:

[0075] P C =P×B;

[0076] Among them, P C It is expressed as the actual deviation value; B is expressed as the correction coefficient; it is determined based on a large amount of data accumulation and historical data to improve the accuracy of the calculation; the common compensation coefficient range is x1 to x5;

[0077] Correction coefficient: B = 2; actual deviation value: P C =P×B=2.5×2=5; the actual deviation value needs to be rounded up;

[0078] Calculate the final value, the formula is as follows:

[0079] V n =V c +(P C ×dVPGM);

[0080] Among them, V n Expressed as final value; V c Indicates the current parameter value; dVPGM indicates the step value;

[0081] Current parameter value: VPGM = 13V; step value: dVPGM = 0.25V; final value V n =V c +(P C ×dVPGM)=14.25V;

[0082] During the adjustment, each step must calculate the final adjustment value based on the results, and verify the calculation through the algorithm to ensure the accuracy and consistency of the final parameters;

[0083] After adjusting the programming voltage or other parameters, a verification test needs to be performed on another test area to ensure that all pages can be stably programmed under the new programming parameters and the number of programming cycles is close to the target value; at this time, it is necessary to ensure that the new programming voltage is within the target range and that there are no abnormalities in the entire programming process;

[0084] Select new programming object: 64Block×64Pages;

[0085] Use the new parameter V n =14.25V for programming;

[0086] The new programming results are as follows: Page 1: Loop = 8; Page 2: Loop = 8; Page 3: Loop = 7; Page 4: Loop = 8; Page 5: Loop = 8; Page 6: Loop = 7; Page 7: Loop = 8; Page 8: Loop = 8;

[0087] Cumulative number of cycles: L A =8+8+7+8+8+7+8+8=64; cumulative number of pages P A =8;

[0088] Calculate the average number of programming cycles using the following formula:

[0089] AL=L A / P A ;

[0090] Where AL represents the average number of programming cycles; L AExpressed as the cumulative number of cycles; P A Expressed as cumulative number of pages;

[0091] Average programming cycle number AL = 64 / 8 = 8;

[0092] Calculate the deviation value using the following formula:

[0093] P=AL-L M ;

[0094] Where P represents the deviation value; L M It is expressed as the target number of cycles;

[0095] Target number of cycles: L M =8; Deviation value: P = AL-L M =8-8=0; meet the deviation within 0.5 cycles; V n =14.25V is written into the ROM Fuse of the chip.

[0096] If the verification fails, continue to increase the parameter value and repeat the above steps;

[0097] The upward adjustment parameter value gradually transitions from a smaller step value dVPGM=0.25 to a larger step value dVPGM=0.5 as the adjustment process proceeds, so as to reduce the adjustment time. At this time, the parameter value needs to be adjusted, and the formula is as follows:

[0098] V x2 =V n +(M×k)×(b B -b S );

[0099] Among them, V x2 Represents the parameter value after transition; V n represents the current parameter value; M represents the target value; k represents the compensation coefficient; b B Indicates a larger step value; b S Expressed as a smaller step value;

[0100] The transition process needs to be corrected by a compensation coefficient k to ensure that the adjusted parameters do not deviate too far from the target value when changing from a small step to a large step;

[0101] Through the accumulation of a large amount of data, the compensation coefficients are confirmed and adjusted to provide correction values ​​for different chip models or production batches; the common compensation coefficient range is x1 to x5;

[0102] Target value: M = 8; New parameter value after upward adjustment: V x2 =V n +(M×k)×(b B -bS )=14.25+8×0.25=14.25+2=16.25V;

[0103] When new parameters are adjusted, use new programming objects and new parameters to perform adjustment tests to verify the accuracy of the new parameters. If the adjusted new parameters perform well in the actual programming test and the deviation of the target cycle number is within ±0.5 cycles, the adjustment result is considered successful.

[0104] After the new parameters are adjusted and tested, check whether the new parameters can meet the target cycle number deviation within ±0.5 cycles to ensure that the final application can meet the deviation requirements.

[0105] The present invention provides a technical solution, a method for optimizing a key erase voltage VERA;

[0106] To start tuning, first determine the initial parameter set, set other related parameters, and select the object to be programmed;

[0107] The object to be programmed is: 128 Blocks; the initial parameter set includes: initial programming voltage: VERA = 13V; minimum loop count min Loop count = 2; maximum loop count max Loop count = 8; step value dVERA = 0.25V;

[0108] Erase the selected 128Blocks, and use the initial parameter VERA=13V to program the programming object, record the number of programming cycles for each page, and exclude abnormal pages; the abnormal pages include data with abnormally fast programming speeds or abnormally slow programming speeds, and the abnormal pages are not included in the cumulative value to avoid affecting the accuracy of the average value; abnormally fast or slow pages are usually inconsistent with the programming behavior of most pages due to process deviations or environmental influences, so these abnormal values ​​should be excluded.

[0109] For example, the programming results are as follows: Block 1: Loop = 3; Block 2: Loop = 2; Block 3: Loop = 5; Block 4: Loop = 1 (abnormally fast, excluded); Block 5: Loop = 30 (abnormally slow, excluded); Block 6: Loop = 5; Block 7: Loop = 4;

[0110] Cumulative number of cycles: L A =3+2+5+5+4=19; cumulative number of blocks P A =5;

[0111] Calculate the average number of programming cycles using the following formula:

[0112] AL=L A / P A ;

[0113] Where AL represents the average number of programming cycles; L A Expressed as the cumulative number of cycles; P A Expressed as cumulative number of pages;

[0114] Average programming cycle number AL = 19 / 5 = 3.8;

[0115] Calculate the deviation value using the following formula:

[0116] P=AL-L M ;

[0117] Where P represents the deviation value; L M It is expressed as the target number of cycles;

[0118] Target number of cycles: L M =8; Deviation value: P = AL-L M =3.8-4=-0.2;

[0119] Calculate the actual deviation value, the formula is as follows:

[0120] P C =P×B;

[0121] Among them, P C It is expressed as the actual deviation value; B is expressed as the correction coefficient; it is determined based on a large amount of data accumulation and historical data to improve the accuracy of calculation;

[0122] Correction coefficient: B = 3; actual deviation value: P C =P×B=-0.2×3=-0.6;

[0123] Calculate the final value, the formula is as follows:

[0124] V n =V c +(P C ×dVERA);

[0125] Among them, V n Expressed as final value; V c Indicates the current parameter value; dVERA indicates the step value;

[0126] Current parameter value: VERA = 13V; step value: dVERA = 0.25V; final value V n =V c +(P C ×dVPGM)=12.7V;

[0127] During the adjustment, each step must calculate the final adjustment value based on the results, and verify the calculation through the algorithm to ensure the accuracy and consistency of the final parameters;

[0128] After adjusting the programming voltage or other parameters, a verification test needs to be performed on another test area to ensure that all pages can be stably programmed under the new programming parameters and the number of programming cycles is close to the target value; at this time, it is necessary to ensure that the new programming voltage is within the target range and that there are no abnormalities in the entire programming process;

[0129] Select the new programming object: 64Blocks;

[0130] Use the new parameter V n =12.7V for programming;

[0131] If the new parameters after adjustment perform well in the actual programming test and the deviation of the target cycle number is within ±0.5 cycles, the adjustment result is considered successful.

[0132] If the verification is successful, it is determined that the adjustment is successful, and V n =12.7V is written into the ROM Fuse of the chip; if the verification fails, continue to increase the parameter value until the requirement is met;

[0133] If half or more of the test cells in the programming object are not successfully programmed and the upper limit of the number of cycles is reached, the chip is judged to have failed trimming; the chip judged to have failed trimming is judged to be scrapped or downgraded.

[0134] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A chip performance optimization method based on single-chip programming voltage adjustment, characterized in that: The method comprises the steps of: S1. Start adjustment, first determine the initial parameter set, set other related parameters, and select the object to be programmed; S2, performing programming operation on the programming object using the initial parameters, and after programming is completed, determining whether half or more of the test cells in the programming object are successfully programmed; S3. If half or more of the test cells in the programming object are successfully programmed, the chip is judged to be successfully trimmed; the parameter is calculated according to the specification requirements to determine the final value of the parameter, and the calculated final value is written and fixed into the ROM Fuse of the chip to complete the trimming process; S4. If half or more of the test cells in the programming object are not successfully programmed, determine whether the upper limit of the number of cycles is reached; If half or more of the test cells in the programming object are not successfully programmed and the upper limit of the number of cycles is not reached, the parameter value is adjusted upward, and the process returns to step S2 to perform programming and judgment operations again; S5. If half or more of the test cells in the programming object are not successfully programmed and the upper limit of the number of cycles is reached, the chip is determined to have failed trimming.

2. The chip performance optimization method based on single-chip programming voltage adjustment according to claim 1, characterized in that: The calculation of the new parameter value comprises the steps of: S2-1, perform Erase processing on the programming object to ensure that the programming object is in the programming state; and determine the initial parameters, maximum number of cycles and step value; S2-2, programming the programming object using the parameters, and recording the number of programming cycles for each page, and excluding abnormal pages; S2-3, calculating the average number of programming cycles for all normal pages; S2-4, comparing the calculated average number of programming cycles with a preset target number of programming cycles to obtain a deviation value; S2-5, set the correction coefficient to help correct the deviation between the target value and the actual value, and calculate the actual deviation value based on the deviation value; S2-6. Calculate the final value of the parameter.

3. The chip performance optimization method based on single-chip programming voltage adjustment according to claim 2, characterized in that: The abnormal page includes data of a page with an abnormally fast programming speed or a page with an abnormally slow programming speed. The abnormal page is not included in the cumulative value to avoid affecting the accuracy of the average value.

4. The chip performance optimization method based on single-chip programming voltage adjustment according to claim 2, characterized in that: The average number of programming cycles is as follows: AL=L A / P A 4 Where AL represents the average number of programming cycles; L A Expressed as the cumulative number of cycles; P A Expressed as cumulative number of pages; The deviation value is as follows: P=AL-L M 4 Where P represents the deviation value; L M It is expressed as the target number of cycles; The actual deviation value is expressed as follows: P C =P×B; Among them, P C It is represented as the actual deviation value; B is represented as the correction coefficient; The final value is as follows: V n =V c +(P C ×b); Among them, V n Expressed as final value; V c a represents the current parameter value; b represents the step value.

5. The chip performance optimization method based on single-chip programming voltage adjustment according to claim 1, characterized in that: The parameter value of the upward adjustment includes: Adjust the parameter value upwards, the formula is as follows: V n =V n +b; Among them, V x Represents the adjusted parameter value; V n a represents the current parameter value; b represents the step value; When adjusting, you need to select different step values ​​to adjust the parameter value according to the adjustment time requirements, and balance the accuracy and time. If higher adjustment accuracy is required, adjust the parameter value with a smaller step value. If less adjustment time is required, adjust the parameter value with a larger step value. When both accuracy and reduced adjustment time are required, the parameter value needs to be adjusted according to the larger step value and the smaller step value. The formula is as follows: V x =V n +M×(b B -b S ); Among them, V x Represents the adjusted parameter value; V n represents the current parameter value; M represents the target value; b B Indicates a larger step value; b S Expressed as smaller step values.

6. The chip performance optimization method based on single-chip programming voltage adjustment according to claim 5, characterized in that: In the first round of adjustment, select a smaller step value to ensure higher adjustment accuracy; In actual application, choose a larger step value for adjustment to ensure less adjustment time; As the adjustment process progresses, the step value gradually transitions from a smaller step value to a larger step value to reduce the adjustment time. At this time, the parameter value needs to be adjusted. The formula is as follows: V x2 =V n +(M×k)×(b B -b S ); Among them, V x2 Represents the parameter value after transition; V n represents the current parameter value; M represents the target value; k represents the compensation coefficient; b B Indicates a larger step value; b S Expressed as a smaller step value; By accumulating a large amount of data, the compensation coefficients for adjustment are confirmed and correction amounts are provided for different chip models or production batches.

7. The chip performance optimization method based on single-chip programming voltage adjustment according to claim 6, characterized in that: During the adjustment, each step must calculate the final adjustment value based on the results, and verify the calculation through the algorithm to ensure the accuracy and consistency of the final parameters; When new parameters are adjusted, use new programming objects and new parameters to perform adjustment tests to verify the accuracy of the new parameters; After the new parameters are adjusted and tested, check whether the new parameters can meet the target cycle number deviation within ±0.5 cycles to ensure that the final application can meet the deviation requirements.

8. The chip performance optimization method based on single-chip programming voltage adjustment according to claim 1, characterized in that: For circuit designs without a given number of cycles, the single-step Pulse by Pulse programming method is used to calculate the number of pulses, and the number of pulses obtained represents the number of cycles.

9. The chip performance optimization method based on single-chip programming voltage adjustment according to claim 7, characterized in that: The single-step Pulse by Pulse programming designs an initial programming voltage and pulse frequency according to the expected programming effect of the chip, and gradually applies pulses to the chip according to the programming object until the chip responds to the expected programming effect; and records the change of the chip state after each pulse is applied; By continuously adjusting the programming voltage and pulse frequency, the optimal programming conditions are obtained, and after completing a programming operation, the total number of pulses required for this programming operation is counted; The cycle number is obtained by gradually recording and analyzing the number of pulses.

10. The chip performance optimization method based on single-chip programming voltage adjustment according to claim 1, characterized in that: Chips that are judged to have failed adjustment are judged as scrap or downgraded.

Images