Topic: DragonMint T1 16TH/S halongmining.com - page 2. (Read 87880 times)

I'll update the link

Here: https://drive.google.com/file/d/1h5wITp761yeM0r-UXXfAQ2KHpU-uk-zU/view?usp=sharing

Hi! Can you update de link? I wanna try if flash that firmware could help to revive 2 dead hashboards ( all red lights are on but still not recognizing).

Thank you for checking the fan pin outs. Can you tell me the model, maker and specs of the fans you currently have installed.

The only ones that seem to work for me in either of my Dragonmint T1 miners is the GH12038B.

I can put these in either of the miners and they work without a problem they also work with the antminers I have. The problem is none of the other fans I have will work in the T1's.

I checked my unit and the fan power lines are the same color order as the Antminer S9. So at least on the versions of the T1 I have seen it's not the fans that are different.

Weird. As for the other person with the hashing, my recommendation is don't push it: T1's can vary in speeds and the best course of action is to run it at default or slight economy and just let it make bits of coin.

Is there a way to overclock? I need a more specific and simple tutorial. my unit varies a lot, and stays at 13 ths instead of 16. even in performance mode.

That would be insane. Do they have a department that literally flips around the wires on the fans just to screw with people? Or did they have a drunk order 20,000 fans the wrong way and they're stuck with it?

Regardless I'd be happy to check tomorrow. I never really noticed to be honest.

I think so. they are really into miners.

I think he needs to flip yellow and blue wires.

red and black are power.

yellow and blue are Pwm and rpm gauge.

IIRC inno flips them. which means dragon flips them.

Do they have a different pinout in the 4-pin fan connector compared to the usual?

Hm. That's a little odd. I have swapped in fans on these with no issue, can you see the model number of your fans? If not I can trundle out and check.

Personal thought: I wouldn't run T1's faster than default speed. They generate a lot of heat and I have come across a *lot* of dead boards. Either they had a bad run of heat sinks or you really should not overclock them too much.

I just purchased 2 of the (Delta QFR1212GHE) fans and installed them on the unit.

The issue persists that both fans do not spin at speed and the front fan stops multiple times. I tested the fans in an antminer S9 and they work without issue. I swapped out the OEM fans from another dragonminer T1 and they spin at speed and work just fine, where as the replacement (Delta QFR1212GHE) do not.

Any thoughts on what I may be missing?

Wiziwig

@wiziwig
This fan model is suitable and quite widely available: Delta QFR1212GHE

Can anyone assist with this?

I purchased a couple used Dragonmint T1 miners and the fans are dead. Does anyone know what fans reliably work in these miners as replacements?

I have tried several 2.7amp fans however all the fans do when connected is spin at a low speed and randomly speed up for a few seconds but never come to the correct speed of the original GH12038B models.

(I pulled the GH12038B fans from a working unit and verified those fans function correctly and the miners hash.
I have not been able to locate the specs for the GH12038B fans on the manufactures site.)

I bought some cheap Dragonmint T1 w/ G19 controll boards.
Wanted to try Braiins on them, since I have it on my other miners.
Please note, that if anything wrong happens, Dragonmint dont have SD card to help.
Use this guide at own risk!!!
Something can happen under installation like power failure, internet failure osv that can make the install fail.

Used ssh from Linux.
SSH root:dr4g0nhash Web is admin:dragonadmin

If you want to install Braiins use thise comands from a new folder.

#Update the repositories and install dependencies
$ sudo apt update && sudo apt install python3 python3-virtualenv virtualenv

#Download and extract the firmware package
#Dragonmint
$ wget -c https://feeds.braiins-os.org/19.06/braiins-os_dm1-g29_ssh_2019-06-05-0-0de55997.tar.gz -O - | tar -xz

#Change the directory to the unpacked firmware folder
$ cd ./braiins-os_dm1-g29_ssh_2019-06-05-0-0de55997

#Create a virtual environment and activate it
$ virtualenv --python=/usr/bin/python3 .env && source .env/bin/activate

#Install the required Python packages
$ python3 -m pip install -r requirements.txt


Install Braiins OS using SSH package

#Change the directory to the unpacked firmware folder (if not already in the firmware folder)
$ cd ./braiins-os_dm1-g29_ssh_2019-06-05-0-0de55997

#Activate the virtual environment (if it is not already activated)
$ source .env/bin/activate

#Run the script to install Braiins OS
$ python3 upgrade2bos.py IP_ADDRESS
(Change "IP_ADDRESS" to miners IP adress)


Done   BTC

If you want to go back to original firmware:

#Using previously created backup to flash back to original firmware if you dont want to use Braiins
$ cd braiins-os_dm1-g29_ssh_2019-06-05-0-0de55997 && source .env/bin/activate
$ python3 restore2factory.py backup/NAME_OF_BACKUP_FOLDER/ IP_ADDRESS

#Comand you use if you want to backup
$ python3 restore2factory.py backup/BACKUP_ID_DATE/ IP_ADDRESS)----

SSH root:dr4g0nhash Web is admin:dragonadmin

Github guide:
https://github.com/KoreNBeatZ/braiins-os/blob/master/docs/index.md

Please note, that if anything wrong happens, Dragonmint dont have SD card to help.
Use this guide at own risk!!!
Something can happen under installation like power failure, internet failure osv that can make the install fail.

Maybe try a different pool? Just an idea. Best of luck!  

i change t1 setting
But Hash Rate can't increase
my power is 1800watt
i use 2500 watt 12volt server power
i set
factory
factory+
factory++
performance --
performance -
performance
performance +
performance ++

the hashrate cannot increase from 15.8th/s  in 24hour

What could be the reason?

You can overclock/underclock just by moving the performance slider.

hi
i have dragon t1
what i do to overclock???
must change framware?

Still working on them, some higher priority projects piled in. The thing that is really annoying is the clock isn't running until the controller commands it so it's a pain to try and find the bad chip. Need to think about that and come up with a way to quick-test the board like one can do with ants.

Resurrecting a dead thread: I got one of these Dragonminers in for repair, well actually it is a pile of T1 boards that are shot. Doesn't seem like there is much documentation on *how* to troubleshoot them so off we go....

I'll probably open a separate thread but for starters:
1) Don't get the latest firmware: This one has the Innosillicon firmware on it and while it is pretty there is no way to log into the console (dr4g0nhash does not work) and worse yet the logs cannot be seen from the web console. UG!

2) These are finicky as fuck in terms of power. I have one working board that oddly enough does different things with different power supplies:
On a Bitmain 1600 watt power supply the board will start up, pull up to 200 watts of power (measured at the wall) and will then shut down to 0 watts, then up to 200, then down. Ok, board is bad.

Except it is not: If I use one of my old Corsair CX750M supplies it will come up, climb to 550 watts and start hashing. Of course it overloads the small PCIe wires on the Corsair and the voltage at the miner has dropped to 11.2 volts, but it does work.

Tried my bench power supplies: HP 750 watt server supplies. Same symptom as the Bitmain one. Note both of these supplies work perfectly on my Antimer S9 test units for weeks on end. And the voltage never sags below 12.0 volts.

Even odder: If I plug two PCIe plugs in from the Corsair it works. If it's working and I plug the third wire into the HP supply it will shut down.

Even odder: If I plug in two PCIe plugs from the Corsair and one from another CX500 it will still work.

Even *odder*: If I plug in one PCIe from the Corsair 750, one from the 500, and one from the HP supply it *will* work and current will be balanced between the three supplies.

I'm wondering if maybe these boards are highly sensitive to some sort of ripple in the DC supply. It's not the DC voltage, the Corsairs sag like old boobs but it's something else in the newer and much better supplies.

Oh well. Tuning is done and board is stable at 4.0th in max economy mode, 60c temps.

The other dead boards do a somewhat similar thing: They come up, pull up to 60 watts, then crash to zero. I used my IR camera on them outside the box and I think I see which chips are dead. So I may be able to fix those.

Edit: After running the "tuning" it's at 380 watts for the hashing board for 4.1th output. Not too bad, also it will now run with the 750 watt HP supply. Something was going weird during the auto-tuning....

Oh I'm running the t1_20190311_114300 version. Very pretty but stupid.

Edit: Fuzzing found the logs at /logs. Nice they left it in, wonder what else is on there.

C

---

Hi, why does the device work this way?? It's been a few days.

T1-16th:

4.90
5.24
4.60