My dead avalon 741 story starting 9/21/2017

[philipma1957]

Okay 1 avalon 741 from   my July 21st order Shipped  July 27  delivered August 3?

died today  after mining  for 48 days.

I sent a support ticket to canaan.io today

This thread will track how long a repair takes.

Any and all  insulting anti Chinese posts will be deleted.

So  lets  start out with my troubleshooting.

I had  a rasp pi  with a chain of 3x 721's

I had a rasp pi with   1x 741

first trick  was switching    the controller connecting each one .

If the problem  follows the controller it means  the controller or a wire is bad.

the good chain  of 3  works  with both controllers.
the bad 741  does not work with either controller

then  put in a known good psu  the     bad 741  does not work.

so  all cables good
controller good
psu good.


next step try  one board from the dead gear  does not work
next step try  the other board from the dead gear does not work.

so  both boards may be dead

or more likely the internal daughter board died.

At this point  I can not do more testing as all my 741's are under warranty.

If I could test  I would use the  hash boards   on the good unit   with the internal daughter board  from the dead unit.

I think  it would not work as I suspect the daughter board died.

I want to do a daily  chronology of this  and show  how good canaan.io is with dead gear under warranty.


Sent this pm to steve of canaan.io

Hi Steve,

I filed a ticket at canaan.io  for my order ##6....    July 21, 2017   Completed


Most likely it is a dead interface board or daughter board in side the miner.

maybe two dead hashboards more likely the daughter board.

And not to complain  but  my order on august 6th   was supposed to be free shipping  and I was charged 51 usd to have it shipped.

I know you are understaffed and over worked but I have to ask for service on my miner as  it is under warranty.

as for the over charge on the shipping if you can refund fine if not let me know.

the ticket number is 98872:

ID_______Status____Date_______________________Product___________   Department____Order
#9....   New.........   September 21, 2017 10:54 pm.....AvalonMiner 741........Support.............#64035

philipma1957 4 hours ago
The miner from my july 21st order # 6.... no longer mines.  I think the daughter board has failed.

I will follow up once a day

[1714541752]

1714541752

[1714541752]

1714541752

[1714541752]

1714541752

[1714541752]

1714541752

[fanatic26]

Have you tried doing an MM update on the bad unit? Maybe there was corruption on the controller board. I havent had to deal with that on the A7 but on the A6 I was able to revive unresponsive units pretty regularly with an MM flash. Even if the controller doesnt 'see' the unit, the mm upgrade process would complete on the A6.

[NotFuzzyWarm]

Is the AUC dongle and miner at least seen in CGMiner Status?
Think what is reported under "Avalon Devices" is the AUC dongle itself and right under that is "Avalon Devices Status" reporting each miner.

[philipma1957]

understand I have 3 setups.


Rasp pi 3    with a 3 chain   721

Rasp pi 3  with a 1 chain   good 741

Rasp pi 3 with  a 1 chain   dead 741


so  I take dead 741  and move it to the good 741 rasp pi 3  using all the cables from the  good 741  the dead one stays dead.

at the same time  I move good 741 to the dead 741 rasp pi cables   and the good 741 works.

that test show the problem is in the avalon 741 not the gear it attaches to.


I do same test with the 3 chain rasp pi 3 and the dead unit stay dead.

so the dead unit does not work on 3 independent   setups.

most likely the unit has the problem  not the setups.

I do have a fourth  rasp pi  I will try that one.

I also have another avalon 741 coming on tues  I could try that dongle.

but I have had  a lot of experience testing gear like this.  I am 95% sure the internal daughter board is dead.

[fanatic26]

I understand you have 3 setups, I understand that you think the problem is the daughter board.

My suggestion was a possible fix for a problematic daughter board.

The MM firmware is the firmware that runs the daughter board. It can get corrupt without killing the entire board which is where my suggestion comes into play. None of the steps you have mentioned attempted to fix the daughter board after you decided that it was the problem.

If you dont want to try it thats fine, I was just offering my experience as I have managed over 8000 Avalon devices in the past 3 years.

[philipma1957]

I understand you have 3 setups, I understand that you think the problem is the daughter board.

My suggestion was a possible fix for a problematic daughter board.

The MM firmware is the firmware that runs the daughter board. It can get corrupt without killing the entire board which is where my suggestion comes into play. None of the steps you have mentioned attempted to fix the daughter board after you decided that it was the problem.

If you dont want to try it thats fine, I was just offering my experience as I have managed over 8000 Avalon devices in the past 3 years.

my bad I misread your fix..
so when I click on mmupgrade

it wants a mcs file  is the link on canaan.io

MM741.mcs

I now have mm upgrading

I now have mm upgrade finish

I show this firmware

7411706-3162860  

and the website reads that

version 7411706-3162860 is newest.

still dead.  neither board fires up

[fanatic26]

Interesting. It looks like the daughter board is working. The DNA is the serial number off of that little board so the controller is talking to the unit itself. It is seeing at least some information from the hashing boards as it is reporting temperature and fan speed. Whats strange is there are no real error codes that would clue you into what is going on with the boards as you would expect with a hardware failure.

[philipma1957]

Interesting. It looks like the daughter board is working. The DNA is the serial number off of that little board so the controller is talking to the unit itself. It is seeing at least some information from the hashing boards as it is reporting temperature and fan speed. Whats strange is there are no real error codes that would clue you into what is going on with the boards as you would expect with a hardware failure.

yeah  it is weird .  now I could chain it to the working unit
fans will fire it will turn the led blue then the led goes to white fans slow.
it will come and go on the interface.
when fan fires and the led goes blue  it is on the interface then it turns down.

this was something the spondoolies would do we used to need to point a blow dryer at them for a few minutes to get them to start.

I got a reply from steve of canaan I am hoping for a part or something to fix it.

[NotFuzzyWarm]

Interesting. It looks like the daughter board is working. The DNA is the serial number off of that little board so the controller is talking to the unit itself. It is seeing at least some information from the hashing boards as it is reporting temperature and fan speed. Whats strange is there are no real error codes that would clue you into what is going on with the boards as you would expect with a hardware failure.

That ^^ is why I asked what CGMiner Status was showing: Was the AUC seen (yes) and more to the point was the miners control board talking at all (again yes).

Anything different in the Kernel Log reported by a good vs dead miner?

[philipma1957]

That ^^ is why I asked what CGMiner Status was showing: Was the AUC seen (yes) and more to the point was the miners control board talking at all (again yes).

Anything different in the Kernel Log reported by a good vs dead miner?

they look pretty much the same.  I will screen shot later.

One time the board actually  hashed   about 130gh  and submitted a share..

Years ago this happened with  both the s-5  and the sp20  on occasion.

I got an email from  support and from steve of canaan.io  maybe it could be fixed quickly.

[TheYankeesWin!]

Any updates phil?

[philipma1957]

They said they will send a replacement unit.

I have a dhl number from them.

I will post when it arrives.

[philipma1957]

only 48 days  , maybe it's not a good products.

Hard to tell.

I have 3 x   avalon 721's  mining since December zero issues.

I have had  100 avalon's   I killed 1 due to using a high voltage  power  of 12.8 volts.  I think it was an avalon six around March of 2016.

My second 741  has been running well.

So I need to figure what  died on the  dead one.

So 2 dead units  across  4 styles and 100 units  is very good , but I am  a bit nervous  that the 741's  have an issue.

To soon to tell. By next week I will be up to 3x 741's and hopefully it was an outlier not a sign of things to come.

[fanatic26]

I have managed a fair number of A741s and the overall failure rate after 60 days including DOA is under 2%. This is far and away the best I have seen from an asic hardware company.

[philipma1957]

I have managed a fair number of A741s and the overall failure rate after 60 days including DOA is under 2%. This is far and away the best I have seen from an asic hardware company.

I have good success with avalon better the my s-9's for sure .

the replacement came it works and a third unit came it works.
x
I have 3x 721's and 3x 741's at this time.

[Steamtyme]

Hey I saw your post in the Dragonmint thread. Did you have a 2nd 741 fail? Or was that before you started this thread?

What went wrong? If you know

[cromsps]

I'm sorry to re-open this topic, but I saw that you have, or at least had :-) , some of 741
I bought 2, downloaded the img for nicehash, put in a PI 2 B, configure the Lan and the nicehash pools, and everything is working.

But every time the machine "solves" on share, it restarts CGminerStatus...
its all working correctly, but on CGminerStatus I can only see Accepted = 1, after finishing this, it starts from zero again... Do you have any issue like this ?
for example,

Starting this quote,

Code:

-> 5m 36s  - 6775.91  - 1 - 0 - 12 - 1,755,789

Ending this quote,

Code:

-> 3m 48s  - 6649.32 - 0 - 0 - 1   174,731

[Steamtyme]

Hey there,

You are best to ask this here in the Troubleshooting and repair guide for AvalonMiner models 721, 741, 761, 821 & 841 , Which it looks like you were the last person to post in regarding a different issue.


In an attempt to assist, I don't use Nicehash so I was unaware there was a specific firmware needed. Are you sure you used the version for PI 2 b, and not one of the other 2 options, Also double check it's the 741 version?

When you do post could you try and add links to images for the miner status page and post the Api log using the code feature like you did in this post. The more information the better.

Also did you try letting them run normally at first pointed elsewhere? This would help in figuring out if it is a firmware issue or maybe an issue with the controller/controller power supply.

Edit:

This is also the 7 series hardware thread you might be able to find some info there, as it was active during the time that firmware was released, definitely worth a serch.

[cromsps]

[...]

Hello,

Ok, thanks :-)