Dragon 1TH blade directly to RPi

[Rusalka68]

A guy @ work gave us his old Dragon Miner 1Th.  It was not working and he did not want to mess with it after going thru several ps.

I was finally able to find a working image for the rpi and got 2 of the blades running.  Actually, we can only get 2 blades @ a time running.  On 3 blades it crashes.

So, need to find another controller card or does anyone know how to wire a blade directly to the RPi??

[notlist3d]

A guy @ work gave us his old Dragon Miner 1Th.  It was not working and he did not want to mess with it after going thru several ps.

I was finally able to find a working image for the rpi and got 2 of the blades running.  Actually, we can only get 2 blades @ a time running.  On 3 blades it crashes.

So, need to find another controller card or does anyone know how to wire a blade directly to the RPi??

I don't think you will be able to go directly to the RPI.  From memory there was at least one board they connect to before going to RPI inside on the A2's.

What PSU does it have in it now?   Have you tried another PSU that can do around 1100 watt.    I think that would be my first try, or if you don't have another one so big you can try to switch cables and run of the ones that are not running.   

It is possible for A2 blades to go bad aswell.  They ran very hot and some were ran in hard conditions.  So there is no guarantee you will get more working.

[Rusalka68]

We have tested all the blades by running them 2 @ a time.  We have used different PS with no luck, so that is why we are looking to see if we can connect them directly to the rpi.

[notlist3d]

We have tested all the blades by running them 2 @ a time.  We have used different PS with no luck, so that is why we are looking to see if we can connect them directly to the rpi.

So running 2 at a time all the blades work?   Or 2 still have problems when using just 2?   

How many watt PSU did you use to power 2 at a time?   

You can't connect directly to RPI there is at least one board you have to go between that I remember off top of head.  You won't be able to skip it.  You will need to find out the problem.

[MCHouston]

You should have this board in the link wired to your boards.

https://bitcointalk.org/index.php?topic=485497.0

[Rusalka68]

Currently we have a 750Watt to run two of the blades and a separate 400W PS to run the 3rd blade.

This is what we get when we do:

ASC Name ID Enabled Status Temperature MHS av MHS 5s Accepted Rejected Hardware Errors Utility Last Share Pool Last Share Time Total MH Diff1 Work Difficulty
0 BA1 0 Y Alive   257,916.16 255842.13 54 0 0 11.88/m 0 05:19:34 70,334,384.4393 54 13,824.00 0.00 256.00 -3s 0.00% 0.00% 273
1 BA1 1 Y Alive   257,916.16 256260.56 58 0 0 12.76/m 0 05:19:35 70,334,384.4393 58 14,848.00 0.00 256.00 -2s 0.00% 0.00% 273
2 BA1 2 Y Alive   2,154,362.18 2128816.91 0 0 0 0.00/m None Never 587,731,914.7192 0 0.00 0.00 0.00 -283s 0.00% 0.00% 273
Total:           2,670,194.5 2640919.6 112 0 0 24.64/m     728,400,683.5978 112 

Any help figuring out what is wrong with the third blade would be appreciated.

MCHouston, were you able to find a spare controller board?

[Rusalka68]

We have tested all the blades by running them 2 @ a time.  We have used different PS with no luck, so that is why we are looking to see if we can connect them directly to the rpi.

So running 2 at a time all the blades work?   Or 2 still have problems when using just 2?   

How many watt PSU did you use to power 2 at a time?   

You can't connect directly to RPI there is at least one board you have to go between that I remember off top of head.  You won't be able to skip it.  You will need to find out the problem.

When running just 2 of the 4 blades @ a time, blades hash normally, regardless of which two are running.  Its only when we try to run more than two of the blades that it acts up.  See prior post.

[notlist3d]

We have tested all the blades by running them 2 @ a time.  We have used different PS with no luck, so that is why we are looking to see if we can connect them directly to the rpi.

So running 2 at a time all the blades work?   Or 2 still have problems when using just 2?  

How many watt PSU did you use to power 2 at a time?  

You can't connect directly to RPI there is at least one board you have to go between that I remember off top of head.  You won't be able to skip it.  You will need to find out the problem.

When running just 2 of the 4 blades @ a time, blades hash normally, regardless of which two are running.  Its only when we try to run more than two of the blades that it acts up.  See prior post.

A lot of the Dragon info is on old pages at this point.   I would try new or different PSU and make sure 1100 watts or more, they say 1000 watt ... but you need more for it to really run good from my experience.  After reading again I see your using a 400 watt are you using it on two or just one?  I have a feeling your issue is related to PSU's and probley low quality PSU's.  What are the brand/models and watts on the two PSU's your using?

If that does not pan out I would look for other firmware.  There is multiple out there might have to ask for one use on a different SD card so you don't lose current firmware (You might actually back it up as-well never bad to have backup).

[pinhead666]

Currently we have a 750Watt to run two of the blades and a separate 400W PS to run the 3rd blade.

This is what we get when we do:

ASC Name ID Enabled Status Temperature MHS av MHS 5s Accepted Rejected Hardware Errors Utility Last Share Pool Last Share Time Total MH Diff1 Work Difficulty
0 BA1 0 Y Alive   257,916.16 255842.13 54 0 0 11.88/m 0 05:19:34 70,334,384.4393 54 13,824.00 0.00 256.00 -3s 0.00% 0.00% 273
1 BA1 1 Y Alive   257,916.16 256260.56 58 0 0 12.76/m 0 05:19:35 70,334,384.4393 58 14,848.00 0.00 256.00 -2s 0.00% 0.00% 273
2 BA1 2 Y Alive   2,154,362.18 2128816.91 0 0 0 0.00/m None Never 587,731,914.7192 0 0.00 0.00 0.00 -283s 0.00% 0.00% 273
Total:           2,670,194.5 2640919.6 112 0 0 24.64/m     728,400,683.5978 112  

Any help figuring out what is wrong with the third blade would be appreciated.

MCHouston, were you able to find a spare controller board?

I think it's pretty clearly said here. 8 boards suck exact 2000w from the wall, i have measured that. So one board is 250w. If underclocked that's less. My sw allowed to lower voltages, for each blade separately.

[notlist3d]

Currently we have a 750Watt to run two of the blades and a separate 400W PS to run the 3rd blade.

This is what we get when we do:

ASC Name ID Enabled Status Temperature MHS av MHS 5s Accepted Rejected Hardware Errors Utility Last Share Pool Last Share Time Total MH Diff1 Work Difficulty
0 BA1 0 Y Alive   257,916.16 255842.13 54 0 0 11.88/m 0 05:19:34 70,334,384.4393 54 13,824.00 0.00 256.00 -3s 0.00% 0.00% 273
1 BA1 1 Y Alive   257,916.16 256260.56 58 0 0 12.76/m 0 05:19:35 70,334,384.4393 58 14,848.00 0.00 256.00 -2s 0.00% 0.00% 273
2 BA1 2 Y Alive   2,154,362.18 2128816.91 0 0 0 0.00/m None Never 587,731,914.7192 0 0.00 0.00 0.00 -283s 0.00% 0.00% 273
Total:           2,670,194.5 2640919.6 112 0 0 24.64/m     728,400,683.5978 112  

Any help figuring out what is wrong with the third blade would be appreciated.

MCHouston, were you able to find a spare controller board?

I think it's pretty clearly said here. 8 boards suck exact 2000w from the wall, i have measured that. So one board is 250w. If underclocked that's less. My sw allowed to lower voltages, for each blade separately.

8 boards?  I think most are the 4 board A1 1T.   There were some with 6 on early on A1's.  But most were the 4 though as it was much cheaper to product.

But I would agree on firmware some did have a lot more ability to adjust then others.  At this point we don't know which one he has.  The original firmware's were pretty basic for most part.

[pinhead666]

No, I had not 1T, I had (and have one still) but 2TH miners. Same shit,little bigger. Same hash-boards anyway. 2TH version has 2 1000W psu's. All still running without problems.

[Rusalka68]

first off, thank you all for helping me on this!

1.  I just took another good look @ the controller card and I would say that its the 1.5Th model (has connection for 2 PSU and 6 serial inputs)

2.  We have swapped several PS's but currently have 2 blades on a CX750 and the 3rd blade is on a 400W Dell server PS.

3.  Just identified one of the serial ports on the controller as bad, when ever a blade is plugged into it, the system will crash within 15 minutes.

4.  Just to make sure I'm not "chasing my tail", but I should be able to run just 3 blades, correct?  Or do I need to run them in pairs?  I was also told that it did not matter what serial ports i used (since this one has 6).

5.  Wareck made me Minera build to try.  Had been using the default version I downloaded from manufactures site.  I will try the Minera build later today.

[notlist3d]

first off, thank you all for helping me on this!

1.  I just took another good look @ the controller card and I would say that its the 1.5Th model (has connection for 2 PSU and 6 serial inputs)

2.  We have swapped several PS's but currently have 2 blades on a CX750 and the 3rd blade is on a 400W Dell server PS.

3.  Just identified one of the serial ports on the controller as bad, when ever a blade is plugged into it, the system will crash within 15 minutes.

4.  Just to make sure I'm not "chasing my tail", but I should be able to run just 3 blades, correct?  Or do I need to run them in pairs?  I was also told that it did not matter what serial ports i used (since this one has 6).

5.  Wareck made me Minera build to try.  Had been using the default version I downloaded from manufactures site.  I will try the Minera build later today.

If it's a 1.5T from lketc you will know it it was roughly 50 pounds to pick up... they but a ton of material in heatsink.   I think whoever you got it from though would have marked 1.5T if it really is to get extra money.

But yes you can run with any number of blades 1,2,3,4 (more on some).  But it makes no difference on odd or even number.  You can do 3 blades only.