HELP PLEASE -Lost ALL my miners after internet outage?

[BBitcoin92]

Hi

I noticed seven of my remote s9 miners were off online and when i looked up there was an internet fault in the area that lasted 14 hurs outage. (Virgin Media) and never had problems before)

I have just been to location and some were just ticking over/not mining, some where totally dead so i checked main 13a fuses and they had blown?

I replaced them/tried new leads and had 3 go BANG when the PSU's blew?

I now have 2 miners that i am running, but both are just ticking over and getting tuner failed, there is not any voltage being displayed on monitir screen when it tries to tune either?

Plus sometimes i get this error message -'Hashchain refused to start I2C error', sometimes i do not, plus its random on 2 miners.

All are running BrainsOS+ and are a mixture of cards that i know brains does not like......But i updated them to Brains a few weeks ago and all has seen fine since this outage.

I have brought 5 miners to my location on my network now and i'm fetting the same results??

Any ideas?

I'm running one now at mine at its TUNER FAILED but it did not even display any hashrate/voltage/frequency/asics???

Don't knwo where to start here tbh... But i know Brains does not like mixing control boards/cards and this is what these miners are....

I have older model control bards here and i bet the one i'm thinking that if i fit an old board with old Bitmain FW it would boot up...

Anyone help please....

Thanks

[BBitcoin92]

Just tried each card individually and disabled cards not used and getting -

'Hashchain refused to start I2C error' ?

This miner has ran for several months if not a over 1 year and at least 5 weeks on new BrainsOS

[BBitcoin92]

Just plugged a second miner in at my home network (where i also run several S9s) and again it just revs up/down.

Tuner status running.

No hashrate/voltage/freq/asic?

On Tuner Status on both cards it states 'testing status or something' but how can it  be testing cards when above shows there is no voltage/freq/asics?

Anyhow it has just stopped and this one gives error -'

Hashchain refused to start Enumeration: no chips detected on the current chain'

I have had this before and solved it simply by fitting an older S9 card with the old autotune 2017 02 2019 FW, i replaced it and it ran fine.

I'm thinking i may need to revert ALL seven boards back to the old FW but ive tried in the past and u can't just flash it???

This is a right mess atm and BrainsOS is not helping at all.....

[BitMaxz]

Why don't you start editing your thread? Why keep consecutive posting? It's prohibited. You likely keep bumping your thread.

Anyway, try to flash it back to stock firmware since you said that you already had experienced it before and solved your case after reverting it to stock firmware.

If you had problem contacting Braiins support I'd suggest try to contact them on this thread below

- https://bitcointalk.org/index.php?topic=5412808.0

[philipma1957]

Well are they?
s-9
s-15
s-17
s-19

and are you fully dead on all units?

it does seem that is the case.

[BBitcoin92]

@BitMaxz Sorry will edit threads in future.



@philipma1957 'I noticed seven of my remote s9 miners '

I've been trying the hash boards with various cards and FW, with the older fW sometimes very rarely it will recognise a card for a few seconds and ghow a few Gh, then that drops off.

I have then tried known working cards from my miners at home and all is ok.

So i can only determine that all 14 cards have somehow gone faulty?  (2 cards/7 miners).

Its odd that BrainsOS recognises the cards, shows testing performance profile shows, but it does not display any hashrate/voltage/freq/asic?

So how can it be testing if its not sending any voltage (or displaying any voltage)?

When tuner then fails i get these messages -

'Hashchain refused to start
Enumeration: no chips detected on the current chain'

UPDATE 2

Miner is now stable lol -
                                                                                                                asics
6   1.181MH/s   161.4GH/s     8.945 V   0.00 °C   0.00 °C   708 MHz   2
8   0.000H/s           484.3GH/s     8.945 V   0.00 °C   0.00 °C   708 MHz   6

1.142MH/s   645.7GH/s   8.945 V   0.00 °C   0.00 °C   708 MHz   8

Fans @ 100%  

Or similar to -
'Hashchain refused to start
Hashchip: number of responses 27 of read_register(reg=0x18) doesn't match chip count 33: reply 0x36 missing'

Has anyone heard of this before, is there anything else i could try here or have i somehow broken all 14 cards?

There is no surge protection on the supplies, but ive had MCB's, and plus fuses blow many times without damage to all boards?

Im at a loss here.

Thanks for looking in guys.

edit -Just had 2  cards actually show and a voltage of 9.42v and one card had 400mhz and dropping, plus one board showed 2 asics, the other 1 asic.

As i'm watching and 'tuner' is running the one board is 40mhz and dropping to 0.....

[DaveF]

Are you sure it was an internet fault due to your ISP and not a power issue in the area?


In our old location we had times where the power was on but WAY undervoltage and it was causing all kinds of issues for equipment, some printers were even destroyed.
I would think the power supplies and miners would be tolerant but if the power supplies were trying to keep the voltage steady and the actual line voltage was bouncing as they were trying to fix something I could see something bad happening.

The issue we had was it was just over the cutoff for the generator and UPS units and only 1 leg so it turned into a real disaster.

-Dave

[BitMaxz]

Seems you already tried almost all available firmware for s9 miner. If all doesn't fix the issue then there might be issue on power supply you need some extra PSU to test or if not then hashboard has some shorted components that leads into this issue.

Can you tell us how much amp that fuse can handle? If it just 5amp fuse then it would blew.

I also want to suggest to check the hashboard you can follow this guide below it needs multitester to check components and test points.

- https://www.zeusbtc.com/manuals/Antminer-S9-Hash-Board-Repair-Guide.asp

[BBitcoin92]

Seems you already tried almost all available firmware for s9 miner. If all doesn't fix the issue then there might be issue on power supply you need some extra PSU to test or if not then hashboard has some shorted components that leads into this issue.

Can you tell us how much amp that fuse can handle? If it just 5amp fuse then it would blew.

I also want to suggest to check the hashboard you can follow this guide below it needs multitester to check components and test points.

- https://www.zeusbtc.com/manuals/Antminer-S9-Hash-Board-Repair-Guide.asp

That's way to much work for me...

I just wish there was a way to downgrade from Brains to the old Bitmain FW, can't do it no matter how i try as some  2 cards work on the older 2017/2018/2019 FW?

There must be a way, i stupidly presumed that when i upgraded to brains that my SSH locked boards that i had to flash to NAND with SD card removed the lock, hence allowing you to downgrade if needed.

I can't find/see a way to remove Brains?

[BitMaxz]

I can't find/see a way to remove Brains?

So you mean the SD card flashing with stock firmware from Bitmain can't able to replace your current firmware? Or you might just did the wrong way?

If the SD card do not work then try the guide from Braiins documentation. Here check this link below

- FLASHING A FACTORY FIRMWARE

[NotATether]

I also speculate that it could be a power or firmware issue you're dealing with - there's no way an internet outage could do that kind of thing to devices. I live near a factory with heavy equipment (not ASICs but very high-watt industrial machines nevertheless) that gets brief power cuts almost daily. Yes the internet goes down as well but neither event seems to break anything.

[Artemis3]

That's way to much work for me...

I just wish there was a way to downgrade from Brains to the old Bitmain FW, can't do it no matter how i try as some  2 cards work on the older 2017/2018/2019 FW?

There must be a way, i stupidly presumed that when i upgraded to brains that my SSH locked boards that i had to flash to NAND with SD card removed the lock, hence allowing you to downgrade if needed.

I can't find/see a way to remove Brains?

That's what bos-toolbox uninstall does, it puts factory fw from 2018 (which has ssh unlocked). sd recovery also works with the S9 family only.

From what you describe, you have a hardware issue. A power fluctuation could easily damage your PSU causing that error.

Various things to check/try, in no specific order:

- Make sure the device is connected to ground at all times (case and psu).
- Check all settings are default (Delete any value, then Save & Apply).
- Try reducing power limit. Wait at least 12 hours for Tuner Status to show Stable.
- Check the cables are well connected, try swapping or replacing them, tighten the power rail screws.
- Disable the hashboard in Configuration > Performance > Override Global Hash Chain about 6 hours.
- Try with another power supply unit, problems with the PSU can also manifest as HW errors.
- Clean the device carefully, if you use liquids you should wait a few days to dry fully.

If the problem persists, it probably has a physical issue that needs repair.

I seldom read other threads (this forum is too large), so you should have listened to what BitMaxz said.

[hZti]

Since so many devices are affected, I would suspect that the there was some kind of voltage peak or similar at your location that damaged your miners. What you now need to find out is what parts are damaged.

[BBitcoin92]

I can't find/see a way to remove Brains?

So you mean the SD card flashing with stock firmware from Bitmain can't able to replace your current firmware? Or you might just did the wrong way?

If the SD card do not work then try the guide from Braiins documentation. Here check this link below

- FLASHING A FACTORY FIRMWARE

My thinking that the older FW i have (you label it stock -pre brains) is a .tar file and not an image file like Brains.

Brains therefore i can easily flash via SD card, the .tar older FW requires command line or some other process to flash (i'm talking pre BrainsOS 2017 FW etc, the old type Bitmain Fw)

I honestly don't understand how to do it via the 'commands', i presume i use command prompt and manually enter details as it describes, plus i somehow create an image of the .tar file to insert in this command line?

Brains FW already is an image file, so it's easily created to put on an SD card.

Since so many devices are affected, I would suspect that the there was some kind of voltage peak or similar at your location that damaged your miners. What you now need to find out is what parts are damaged.

It was water damage by the looks, i can now see it and the last two miners i opened that are nearest the cold air input tip garage they are in are were wet, we did have a big storm, power outages etc and a fine mist of water ha entered.

They have sat as they are running for years until this, hence my first thinking was some sort of spike in current/network etc as it was all over the news/social media of big outages locally to me.

That's way to much work for me...

I just wish there was a way to downgrade from Brains to the old Bitmain FW, can't do it no matter how i try as some  2 cards work on the older 2017/2018/2019 FW?

There must be a way, i stupidly presumed that when i upgraded to brains that my SSH locked boards that i had to flash to NAND with SD card removed the lock, hence allowing you to downgrade if needed.

I can't find/see a way to remove Brains?

That's what bos-toolbox uninstall does, it puts factory fw from 2018 (which has ssh unlocked). sd recovery also works with the S9 family only.

From what you describe, you have a hardware issue. A power fluctuation could easily damage your PSU causing that error.

Various things to check/try, in no specific order:

- Make sure the device is connected to ground at all times (case and psu).
- Check all settings are default (Delete any value, then Save & Apply).
- Try reducing power limit. Wait at least 12 hours for Tuner Status to show Stable.
- Check the cables are well connected, try swapping or replacing them, tighten the power rail screws.
- Disable the hashboard in Configuration > Performance > Override Global Hash Chain about 6 hours.
- Try with another power supply unit, problems with the PSU can also manifest as HW errors.
- Clean the device carefully, if you use liquids you should wait a few days to dry fully.

If the problem persists, it probably has a physical issue that needs repair.

I seldom read other threads (this forum is too large), so you should have listened to what BitMaxz said.

Issue was water damage by the looks.

I definitely need to look inti bos-toolbox, i did download it and did find it somewhat confusing tbh, then after reading on here some where saying it's main use is for large mining farms, mass flashing/control of 100's of miners etc and people were doing it simply via SD card, so i went with that and found the process simple and i only have several miner.

Cheers

[moderator's note: consecutive posts merged]

[Artemis3]

It can do operations to a single miner, you simply use the IP address and that's it. For batch operations you just use a simple text file with a list of IP addresses, one per line.