Home >
Bitcoin Mining: In-depth analysis of the Antminer S19 Pro

Bitcoin Mining: In-depth analysis of the Antminer S19 Pro

The Antminer S19 Pro has been available for several months. It is believed that the miners in the industry have a specific understanding of this miner's basic parameters and operating conditions.

Today, let's take you to the mining machine to have a glimpse of the internal structure of the Antminer S19 Pro and to explain some common undesirable phenomena and treatment methods of the computing board.

Antminer S19 Pro.png

The Antminer S19 Pro machine is mainly composed of 3 computing boards, one control board, APW12 power supply, and four cooling fans, as shown in the figure below:

The Antminer S19 Pro computing board is composed of 114 chips, divided into 38 groups (domains), each group is formed of 3 ICs; the working voltage of the S19 Pro chip is 0.32V; the 38th, 37, 36, 35, 34, 33, and 32 groups (7 groups in total) The 20V output from the booster circuit Q9 supplies the LDO and outputs 1.8V. The 31st group-the first group provides 1.8V from the VDD 12.6V via the LDO, and 0.32V reduces the voltage for each backward domain.

All 0.8V is provided by the 1.8V of this domain through the LDO output, as shown in the figure below:

All right. In fact, after testing, under the officially recommended environmental conditions, the performance of S19 Pro's energy consumption ratio, stability, and other operating conditions are excellent. If the arithmetic board is defective, the most common phenomena include reporting 0, reporting the number, low computing power, etc. We can deal with these situations like this——

1. When the test fixture LCDs "ASIC NG: (0)", first measure the total voltage of the domain and the boost circuit 20V is normal, and then use the short-circuit probe to test the RO test point between the first and second chips with 1V8 Click the short circuit, and then run the program to find the chip. Looking at the serial port log, if 0 chips are still found at this time, it will be one of the following situations:

(1) Use a multimeter to measure whether the 1V8 and 0V8 test points voltages are 1.8V or 0.8V. If not, it may be that the 1.8V or 0.8V LDO circuit of the domain is abnormal, or the two ASIC chips of this domain are not. Solder well, most of them are caused by the short circuit of 0.8V, 1.8V patch filter capacitors (measure the resistance of the patch filter capacitors related to the front and back of the PCBA);

(2) Check whether the circuit of U2&U3&U4 is abnormal, such as resistance welding, etc.;

(3) Use a multimeter to measure the resistance of R232 or R233 to see if it is within 1 ohm and the reading will not jump randomly. If not, please replace these two resistors;

(4) Check whether the pins of the first chip are not soldered well (it has been repaired and found that the pins are tinned from the side, but when the chip is removed, it is found that the pins are not stained with tin at all).

2. If one chip can be found in step 1, the first and the previous circuits are sound. Use a similar method to check the following chips. For example, short-circuit the 1V8 test point and RO test point between the 38th and 39th chips. If the log can find 38 chips, then the first 38 chips are fine; if you still find 0 chips, check the 1V8 first Whether it is normal or not, there is a problem with the chip after 38. Continue to investigate with dichotomy until the problematic chip is found. Assuming that there is a problem with the Nth chip when the 1V8 and RO between the N-1th and Nth chips are short-circuited, N-1 chips can be found, but the 1V8 and RO between the Nth and N+1th chips can be found. When there is a short circuit, the chip cannot be found.

3. When the LCD of the test fixture displays "ASIC 113: (Report 113)", it means that the arithmetic board can detect 114 chips at 115200 baud rates, but only 113 chips are found at 12M baud rate, and one chip is at 12M. The baud rate was not found.

Repair method: Using the dichotomy method, short-circuit the 1V8 test point and the RO test point between the 38th and 39th chips through the short-circuit probe. If the log can find 38 chips, there is no problem with the first 38 chips; if there is a short circuit 47, and the log reports 46, indicating that the 47th chip cannot be detected, and there is no problem with the visual inspection. Generally, the 47th chip can be replaced.

4. When the LCD of the test fixture displays "ASIC NG: (X, a certain chip)," there are two situations:

(1) In The first case; the test time is the same as the OK board (usually the value of X will not change each time you test) (the test time refers to the time from pressing the start test button to the LCD "ASIC NG: (X) "The time of this result).

This situation is likely to be caused by the abnormal resistance welding of the front and rear CLK, CI, and BO series of the Xth chip, so focus on these six resistors. The small probability is caused by abnormal welding of the following pins of the three chips of X-1, X, and X+1:

(2) The second case; the test time is almost twice as long as the OK board (sometimes the value of X will change every time you test, and sometimes X=0); at this time, the log usually has the following information (the red numbers are not necessarily It is 13, depending on which seat the fixture is connected to); During the test, assume that the domain voltage of all the fields in front of the abnormal position is almost less than 0.3V, and the field voltage of the back fields are almost all higher than 0.34V.

This situation is caused by the chip not being soldered well. Usually, 1.8V, 0.8V, RTX, CLK are not soldered well. It is recommended to directly measure the domain voltage to locate which domain is the problem. The 1V8 and RO short-circuit method used in step 1 can also locate the abnormal position.

(3) The current maintenance experience found that most of the signals are micro-short circuits (0-several hundred ohms) caused by the small resistance of the chip pins. You can first use a hot air gun to add welding and see if it is OK.

5. And if you encounter a low computing power of a single board, here is a convenient and quick way to locate and repair it to share with everyone that is, log in to the IP through the Putty software to observe whether the domain is working voltage and NONCE return of this board are normal, and according to Putty's log information Carry out repairs.

The specific operations are as follows:

(1) Open the IP of the machine with the problem and click OPEN

(2) Enter the username, password, and test command to see the NONCE return status and the status of the voltage domain. If the NONCE and domain voltage are abnormal, you can perform measurement and maintenance based on the printed abnormal chip.

If you want to know more about maintenance instructions for Antminer S19 Pro and other models, please sign up for the maintenance training courses of our Ant Training Academy!