Pages:
Author

Topic: Troubleshooting and repair guide for AvalonMiner models 721, 741, 761, 821 & 841 - page 10. (Read 23873 times)

legendary
Activity: 2422
Merit: 1706
Electrical engineer. Mining since 2014.
Troubleshooting and repair guide for AvalonMiner models 721, 741, 761, 821 & 841

Ongoing repair process for an AvalonMiner 821 unit. Click for a higher resolution image.

1) General information, most common tools needed
2) Troubleshooting
3) Tear down
4) Conclusion


Last updated in: March 14, 2019.

Nicely formatted version (v. May 14, 2018) is available for download here (Mega.nz cloud service):
https://mega.nz/#!o10yxaIC!yAODmwOLd8Fwqp3Wi-7fz54w05sdTzjOlPWtnubRIn8

1) General information, most common tools needed
(top)
This is an official troubleshooting and repair guide for AvalonMiner 7 and AvalonMiner 8 series.
All rights to fix or improve this document are reserved to Canaan Creative.

In this guide you will see how to disassemble and separate all miner components and it will help
you to understand how to replace a faulty miner component like for example a broken cooling fan or faulty PCB board.

You will also learn how to do the fault diagnosis to your miner in a case of
a miner malfunction and how to find out which miner component is most likely not working correctly.

This guide is made using an AvalonMiner 841 unit, but same principles apply
with other compatible models mentioned in this guide as well.

Most common tools needed:
  • Phillips headed screwdriver (it is good to have both PH1 and PH2 size screwdrivers / tips)
  • Digital multimeter
  • Static electricity protection (ESD mat or antistatic wrist strap for example)
  • If you don’t have anything for static electricity protection, please at least ground yourself by touching a grounded object to discharge the static electricity before touching a PCB board
  • Good quality thermal paste
  • Pure isopropyl alcohol for cleaning the old thermal paste
  • Soft cloth or soft paper for cleaning the old thermal paste


Phillips head screwdriver. Click for a higher resolution image.


Digital multimeter.


ESD mat and antistatic wrist strap.


Thermal paste.


Isopropyl alcohol.

These instructions apply for following AvalonMiner 7 and 8 series models:
  • AvalonMiner 721
  • AvalonMiner 741
  • AvalonMiner 761 (with a built-in power supply unit)
  • AvalonMiner 821
  • AvalonMiner 841


AvalonMiner 721 technical specifications.


AvalonMiner 741 technical specifications.


AvalonMiner 761 technical specifications.


AvalonMiner 821 technical specifications.


AvalonMiner 841 technical specifications.


2) Troubleshooting
(top)
In this chapter you will see the common methods for the troubleshooting process and you will also see example cases about the most common issues.

The goal is to help understand how to determine which miner component is causing the miner operation malfunction.

How to indicate your miner’s physical location
You can use the indicator LED to help you locate the miner.
Just simply press the indicator LED switch button to change the state from ’LED OFF’ to ’LED ON’.
You can find the indicator LED switch button location by following these steps:
  • 1. Go the controller web configuration page by using your web browser and login.
  • 2. Click ’Advanced Version’ to access the advanced user features in the OpenWrt firmware.
  • 3. Click ’CGMiner Status’ to see status for all the devices connected to the controller.
  • 4. Choose the AvalonMiner unit which you would like to locate and click ’LED OFF’ button located in the Indicator field.
    Now you should see that status change into ’LED ON’.
    Further observations about this situation are shown in two figures below.


Normal miner operation before pressing the indicator LED button - the indicator LED color is blue. Click for a higher resolution image.


After pressing the indicator LED button the miner in question changes the indicator LED color to white.

Video: How to Identify your Avalonminer for Troubleshooting
https://www.youtube.com/watch?v=vNMW0QtX5Bk



How to locate and open the CGMiner API Log
These following instructions will tell you how to locate and open the CGMiner API Log.

By reading the log text you are able to see lots of technical information related to the miner performance.

Step 1.
Use your web browser and access to the controller web configuration page.

Web configuration page, login screen.

Step 2.
Click ’Advanced Version’ to access all user features.

Dashboard view.

Step 3.
You are now able to see all the features available.
Now you need to click ’CGMiner API Log’ to access the log text.

System overview.

Step 4.
Click ’Debug Switch’ and after that you are ready to read and examine the CGMiner API Log.

CGMiner API Log.


Information about indicator LED colors and CGMiner API Log error codes
During the troubleshooting it is best to find and use all the information available, so that the repair is a quick and straight forward process.

The miner is able to report about specific issue with different LED colors and error codes.
In these following tables you can see a description for each LED color and error code.

Please remember to click 'Debug Switch' button before reading and examining the CGMiner API Log.

Video: Avalonminer 741 and 841 - LED Color Guide

https://www.youtube.com/watch?v=P59HPZnIk7s

Table 1. AvalonMiner LED colors.


Table 2. PMU board LED colors.


Table 3. AUC3 Converter LED colors.



One of the most frequent errors is the socket connect error.
Socket connect failed => Connection refused

This error can happen because of numerous different things, but here is a list showing few of them.
•  Miner is still in the startup process
•  Incorrect pool settings
•  Wrong OpenWrt firmware version
•  Communication error.

Table 4. CGMiner API Log error codes.



How to read the abbreviations and values found in the CGMiner API Log

In this following table you can see the CGMiner API Log thoroughly explained, row by row.
This provides you a lot of help in the troubleshooting process, as you are able to determine possible issues based on the values seen in the CGMiner API Log. The information is divided into three parts, part 1-3.

Table 5. List of codes found in the CGMiner API Log. Part 1.


Table 6. List of codes found in the CGMiner API Log. Part 2.


Table 7. List of codes found in the CGMiner API Log. Part 3.


How to check the power supply unit health
If you want to know if the power supply is passing any voltage to the PMU board,
you can measure the +12V rail output voltage to find out.

According to the ATX specification, the measured voltage should be from 11.4 to 12.6 VDC.
Please also take into account the rated input voltage for your AvalonMiner.

The most safe area for the input voltage is 12.0 to 12.5 VDC.

Video: How to check your AvalonMiner power supply
https://www.youtube.com/watch?v=1Ty3D5vuj_I

For measuring the voltage you need a multimeter.
You can easily measure the voltage by applying one multimeter test lead to the +12V (yellow) and one multimeter test lead to the COM (black).


6-pin PCI-E power connector pinout. Click for a higher resolution image.

First you need to select the right measuring mode by using the selection knob in the middle of the multimeter.
Make sure that your test leads are connected correctly to the multimeter for voltage measurement setup.


Selecting the right measuring mode for the 12 VDC voltage measurement. Click for a higher resolution image.

You can use any safe methods, tools or tape you need to hold the power connector still
so that it is easy for you to hold the test leads in position.


Measuring the voltage. Click for a higher resolution image.


Cooling fan
If your miner won’t start operating, one of the first things to check if the cooling fan is working.
You can check it by doing both visual inspection and looking at the miner status page.

If you find a miner with fan reading at 0 rpm or close to it, then it is quite clear that you need to replace that miner’s cooling fan.

Please also see if there are any indicator LED colors or log error codes related to a cooling fan issue.


Fan speed 0 rpm at the miner status page.


Fan speed 0 rpm at the CGMiner API Log page.

If the cooling fan failure occurs during the warranty period, you can request a replacement cooling fan from Canaan Creative.

If your miner is out of warranty, here is a list of few compatible replacement cooling fan models:
  • Cooler Master A12038-60BB-4RP-F1
  • Delta Electronics QFR1212GHE
  • Delta Electronics QFR1212GHEXUM
  • Nidec V12E12BS1B5-07
  • Sanyo Denki 9GV1212P1J01
  • Shengjiu SG121238BS
  • Sunon PSD1212PMBXA-F-PWM

AvalonMiner Controller / Raspberry Pi
If you are not able to access the controller unit at all and you have double-checked that your setup is ok,
then it is possible that your controller unit is faulty and needs to be replaced with a new one.

Please also make sure that the 5 VDC wall wart is in ok condition and that
it is able to provide at least 5 VDC 2.5 A power for the controller unit.

Here is the list of compatible controller units:
  • AvalonMiner Controller (RPi 3 based device)
  • Raspberry Pi 1 Model B
  • Raspberry Pi 2 Model B
  • Raspberry Pi 3 Model B



AUC3 Converter

If you are able to access the controller unit successfully, but you don’t see any miner devices connected, it is possible that there is a faulty AUC3 converter or a faulty control board (MM board).

You can look at the LED colors and log error codes and try to figure out the issue that way.

You can try a spare AUC3 converter if you have another one.

You can easily troubleshoot if it is the AUC3 which is faulty, if you have six miners or more in your setup,
because this means that you have at least two AUC3 converters in your setup.

You can swap those two AUC3 converters and try to see if the issue moves to the another group of miners. If it does, you should be able to know which one is the faulty AUC3 converter.

Never buy unshielded USB cables for connecting AUCs to the controller. Keep a multimeter around and check if the resistance of shielded USB cable you want to use is lower than 1ohm. If possible, check all connections, power should be around 0.1ohm, data (two pins in the middle) should be lower than or around 0.2ohm. Unfortunately, there is no way to find out if the cable is actually shielded or not unless you want to cut the actual cable and see the inside. A lot of USB cables out there just short the shield to the ground to make it impossible to know if it is actually shielded or not. Almost all cheap USB cables are not shielded properly.

There was a user who wasted a week trying to find out the reason behind random disconnects as he used two 3 meter USB cables from Anker to connect two of AUCs to the controller.

It is also important to remember check that your USB cable is not for charging only.

Control board (MM board)
If you are able to access the controller unit successfully, but you don’t see a miner device connected, it is possible that there is a faulty AUC3 converter or a faulty control board (MM board).

Another example of possible control board malfunction is that if there is a huge number hash boards down suddenly.

One good hint is that you have for example five miners connected to one AUC3 converter, and you are only able to connect four of them, it indicates that there might be issues with one miner control board.

You can look at the LED colors and log error codes and try to figure out the issue that way.

You can try using a spare control board (MM board) for the miner in question if you have another one.

PMU board
PMU board stands for Power Management Unit board.

If you have one or more hash boards down and without power, it is a good idea to figure out how is the PMU board condition.
You can do this by looking at the CGMiner API Log.

If you any values zero or values having an odd number, it is possible that you have a faulty PMU board.
When you look at the log, please pay attention to Vi and Vo values (voltage in and voltage out).

One PMU board has two power terminals, so there is a total number of four values visible in the log.
There are four digits presenting the measured input voltage and output voltage. For example value 1199 means 11.99 VDC.


Both PMU boards are working normally.


The second PMU board is not working at all.

In this next image you can see how to identify which PMU board is the first one and which is the second.


PMU board identification.

Hash board
To find out if you have a hash board issue, you can look at two things in the CGMiner API log.
  • Works (MW0 – MW3)
  • Chip temperatures (PVT_T0 – PVT_T3).

If there is one board with no works, then it is clearly a faulty one.
Your miner might also have issues with constant rebooting because one of the ASIC chips overheats.

In this case you can first try to clean the old thermal paste and then apply new thermal paste.
If new thermal paste does no good, you need a new hash board.


Number of works shown in the log. This miner is working normally.


Every single ASIC chip temperature shown in the log.

In this next image you can see how to identify which ID number is for which hash board.


Hash board identification.


3) Tear down
(top)
You only need one tool for the tear down process and it is a screwdriver.
Make sure you have a good quality Phillips head screwdriver and right sized,
to avoid damaging the small bolts.

Please also be careful not to use too much force which can potentially damage your
screwdriver, bolt or the thread.

Please pay attention using the right type of a bolt when assembling it all back together.

These instructions show you how to separate each miner component,
so that you are able to replace a faulty part in question with a new spare part.

Video: Avalonminer 741 Disassembly Guide
https://www.youtube.com/watch?v=1W_BO1PimlA

Step 1.
Remove the steel tube case.
There are four small bolts in both sides of the miner, so a total number of eight bolts.

Removing the steel tube case. Click for a higher resolution image.


Steel tube case removal completed.

Step 2.
Remove the cooling fan. There are four screws holding the cooling fan shroud against the frame.

Removing the cooling fan shroud.


Unplug the 4-pin fan cable connector and remove the cooling fan.

Unplugging the 4-pin fan cable connector.


Removing the cooling fan.

Video: Avalonminer 741 and 841 Fan Replacement Tutorial
https://www.youtube.com/watch?v=-ltqGavbzd0


Step 3.
Unplug the 4-pin fan extension cable from the control board.

Unplugging the 4-pin fan extension cable from the control board.

Step 4.
Remove the control board (MM board).
There are two small bolts to be removed, after which you can gently detach the control board which is connected to both two PMU boards (Power Management Unit).

Removing the control board.


Removing the control board.

Step 5.
Remove the hexagon pattern steel grill.
There are four small bolts holding it against the frame.

Removing the hexagon pattern steel grill.

Step 6.
Remove both two frame parts.
There are six small bolts holding one frame parts against the heat sinks, so a total number of twelve bolts.

Removing the frame parts.

Step 7.
Remove both two power management unit boards (PMU board).
There are six bolts per PMU board and four of them

Removing the PMU board.

Step 8.
Remove the hash board.
There are three small bolts and six big bolts holding one hash board against the heat sinks.

Two hash boards with one shared large heat sink and two invidual heat sinks.

Remove the old thermal paste if needed, if it is hardened or damaged in any different way. You should do this by using pure isopropyl alcohol and a soft cloth or paper. You can also use a cotton swab if needed.

Removing the old thermal paste.

After the surfaces are cleaned, you can apply new thermal paste. Do not use your finger or anything else that causes impurities to mix with the thermal paste when applying the new thermal paste.

Applying new thermal paste.

Now we have gone through how to tear down the miner,
how to separate all the miner components
and how to replace each invidual miner component.

Assembling it all back together is essentially doing this same process backwards.



4) Conclusion
(top)
These instructions should be able to help you with most of the issues that might occur,
even though AvalonMiner failure rate in field is close to 1%, which is a low number compared to the competitors.

But please remember that if you need help with an miner issue, you can always also do these two things:
  • Submit a support ticket at Canaan Creative website
  • Join the discussion thread at Bitcointalk.org forum to get comments and instructions on your issue.

Any questions, suggestions or comments are welcome to be posted in this Bitcointalk.org forum thread.

Canaan Creative: http://www.canaan.io


Canaan Creative support ticket form at www.canaan.io . Click to show a higher resolution image.
Pages:
Jump to: