[ANN] Spondoolies-Tech - carrier grade, data center ready mining rigs

[philipma1957]

Hi,

Yesterday I ordered a mini farm (#4521), and didn't recibe the tracking code, when you will ship it?

thanks

they are closed on Fri. and Sat.


  they are open from Sun. to Thur.

[1714165766]

1714165766

[1714165766]

1714165766

[1714165766]

1714165766

[goxed]

Yes, I am sorry.
I also have pending pull request from Luke which is probably good, but I do not have the time to make the integration (with all the release cycle, and testing on all the miner flavors). Especially in the web, which is not my strongest side, I am reluctant to change things
But developers are welcome to do custom builds, and develop their own flavors if they feel it is interesting for them.

Zvi, there's no reason to say "I'm sorry" In my opinion the code repositories should be used interactively.

Some of the guys here are addicted enough to even code stuff for you (web or not), but unfortunately they can't integrate their stuff in the mainstream.
That's why 2112 suggested a step back from day to day issues and look back in with a bigger perspective. You have a unique position being on the team since the first SP-T chip was being created. Since SP35 is almost out of stock, with the new chips (if any) one can try a new approach. It will be a pity to use the same (or slightly modified) chip optimization model since the SP10 units

From a personal perspective I know it's hard to break routines and take new challenges when there is a multitude of small issues to be solved. However, those issues will always be there. Sometimes I'd like to code stuff instead of driving my team to do the coding Particularly when the team is small and busy. But they need guidance and planning while they do what they do best: code. They can't be efficient when they focus on both current issues and future planning. Someone needs to do the planning for them It's the "management" job security paradigm

With the continuous increase in the number of SP-T models, just testing new firmware versions for all situations out there will eventually stress your team more than it can cope with. If I were you, I'd drop support for older models (I'm still happy with my SP10 running FW 1.5.8 and I'm not looking to upgrade it any longer). That will let you focus on the new ones. Also, try an "agile" way of running things. Let your guys focus on the improvements with the largest impact and with minimum work-hours .. the R&D supporting the firmware using the future line of chips should be one of those.

By the way, no one needs a weekly FW update. It can be bi-weekly or even monthly. There's no time to upgrade and monitor the units (particularly in large numbers) for improvement, that quick. What it really boils out of it, is to sort out priorities, communicating the sort result to the team. And one can't do that properly when (s)he's always busy with small stuff.

I believe that the fw logic is very close to optimum. Despite some things said here i have mathematical background, and i took part in both asic interface definition and implementation.

i use greedy algorithm for finding the optimal working point  not because it is simplest but because i believe it is best for this task. And i did some simulations and tests before reaching that conclusion.
I see no point to rewrite any of the existing fws since i dont believe it will give higher performance by more then a couple of %. I would be happy to be proven wrong by someone from this forum )

How about letting us have direct control over clocks & voltages?!

Yes, that would be excellent for those of us who what to maximize GH/Watt for each voltage regulator / bank.

[mavericklm]

The simplest way is to give speed and have the wattage reported, or the other way around. no voltage!
But is more complicated to put it in soft&hardware

[GenTarkin]

The simplest way is to give speed and have the wattage reported, or the other way around. no voltage!
But is more complicated to put it in soft&hardware

Dont know what u are saying exactly...

[GenTarkin]

If not direct clock / voltage control. I would at least like to set a max voltage for each 'board', not just one overall voltage.

[mavericklm]

My idea is that we as end users, we care less about voltage and more about hash rate and wattage.

Hash and wattage should be lock together, change one and the other one should changes accordingly by itself. and no voltage controlor or just in ''advanced option''

example: i want 1kw consume on my sp20, hash will go to 1.55gh! i want 1gh from my sp20, consume will be reported at 500w.....

my idea comes from the fact that if i set the max voltage at .7 i simply have no idea what the wattage is! without an wattage meter
and this will be more user friendly

[innergy]

Bought my first SP30, second hand from eBay and with - FW: 2.6.1, start V: 0.655, max V: 0.75, max P: 1290W and 80% fan

I can't get more than 2800GH/s stable (two ASIC chips auto disabled - Loop [5]: 15, 17)

It looks like the device can't set the right voltage. No matter what voltage I set, it is fluctuating between 625-635, the PSU powers fluctuating between 700-800W and the temps are about 30 °C / 53,5 °C.
Is there anything I can do about that?

Code:

 2.6.1 Uptime:3498 | FPGA ver:62 | BIST in 1 
-----BOARD-0----- 
PSU[EMERSON1200]: 768->(688w/688w)[688 688 684] (->746w[729 727 746]) (lim=1290) 0c 1491GH cooling:0/0x0 
-----BOARD-1----- 
PSU[EMERSON1200]: 688->(652w/652w)[648 652 648] (->672w[654 653 672]) (lim=1290) 0c 1343GH cooling:0/0x0 

LOOP[0] ON TO:0 (w:220)
 0: DC2DC/1/:[vlt1:628 vlt2:632(DCl:794 Tl:794 Ul:749) 38W 60A 49c] ASIC:[ 85c (125c) 500hz(BL: 500) 75 (E:193) F:0 L:0]
 1: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 39W 61A 47c] ASIC:[ 85c (125c) 500hz(BL: 500) 85 (E:193) F:0 L:0]
 2: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 36W 57A 44c] ASIC:[ 85c (125c) 470hz(BL: 470) 60 (E:193) F:0 L:0]
LOOP[1] ON TO:0 (w:255)
 3: DC2DC/1/:[vlt1:628 vlt2:632(DCl:794 Tl:794 Ul:749) 38W 60A 49c] ASIC:[ 85c (125c) 510hz(BL: 510) 90 (E:192) F:0 L:0]
 4: DC2DC/1/:[vlt1:628 vlt2:632(DCl:794 Tl:794 Ul:749) 37W 59A 55c] ASIC:[ 85c (125c) 490hz(BL: 490) 80 (E:193) F:0 L:0]
 5: DC2DC/1/:[vlt1:628 vlt2:632(DCl:794 Tl:794 Ul:749) 38W 60A 56c] ASIC:[ 85c (125c) 500hz(BL: 500) 85 (E:193) F:0 L:0]
LOOP[2] ON TO:0 (w:262)
 6: DC2DC/1/:[vlt1:628 vlt2:632(DCl:794 Tl:794 Ul:749) 38W 61A 58c] ASIC:[ 85c (125c) 510hz(BL: 510) 89 (E:193) F:0 L:0]
 7: DC2DC/1/:[vlt1:628 vlt2:632(DCl:794 Tl:794 Ul:749) 39W 62A 57c] ASIC:[ 85c (125c) 510hz(BL: 510) 82 (E:193) F:0 L:0]
 8: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 41W 66A 53c] ASIC:[ 85c (125c) 540hz(BL: 540) 91 (E:193) F:0 L:0]
LOOP[3] ON TO:0 (w:250)
 9: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 40W 64A 58c] ASIC:[ 85c (125c) 530hz(BL: 530) 96 (E:192) F:0 L:0]
 10: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 42W 67A 62c] ASIC:[ 85c (125c) 550hz(BL: 550) 75 (E:193) F:0 L:0]
 11: DC2DC/1/:[vlt1:628 vlt2:632(DCl:794 Tl:794 Ul:749) 38W 60A 61c] ASIC:[ 85c (125c) 510hz(BL: 510) 79 (E:193) F:0 L:0]
LOOP[4] ON TO:0 (w:237)
 12: DC2DC/1/:[vlt1:625 vlt2:632(DCl:794 Tl:794 Ul:749) 40W 64A 63c] ASIC:[ 85c (125c) 530hz(BL: 530) 81 (E:192) F:0 L:0]
 13: DC2DC/1/:[vlt1:632 vlt2:632(DCl:794 Tl:794 Ul:749) 42W 67A 65c] ASIC:[ 85c (125c) 550hz(BL: 550) 86 (E:192) F:0 L:0]
 14: DC2DC/1/:[vlt1:628 vlt2:632(DCl:794 Tl:794 Ul:749) 40W 64A 54c] ASIC:[ 85c (125c) 530hz(BL: 530) 70 (E:193) F:0 L:0]
LOOP[5] ON TO:0 (w:69)
 15: disabled (Asic all engines fail BIST)
 16: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 38W 61A 42c] ASIC:[ 85c (125c) 510hz(BL: 510) 69 (E:192) F:0 L:0]
 17: disabled (Asic all engines fail BIST)
LOOP[6] ON TO:0 (w:260)
 18: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 39W 62A 45c] ASIC:[ 85c (125c) 530hz(BL: 530) 90 (E:193) F:0 L:0]
 19: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 40W 63A 50c] ASIC:[ 85c (125c) 530hz(BL: 530) 89 (E:193) F:0 L:0]
 20: DC2DC/1/:[vlt1:628 vlt2:632(DCl:794 Tl:794 Ul:749) 36W 58A 48c] ASIC:[ 85c (125c) 490hz(BL: 490) 81 (E:193) F:0 L:0]
LOOP[7] ON TO:0 (w:269)
 21: DC2DC/1/:[vlt1:628 vlt2:632(DCl:794 Tl:794 Ul:749) 38W 60A 53c] ASIC:[ 85c (125c) 520hz(BL: 520) 79 (E:193) F:0 L:0]
 22: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 43W 68A 54c] ASIC:[ 85c (125c) 560hz(BL: 560) 102 (E:193) F:0 L:0]
 23: DC2DC/1/:[vlt1:628 vlt2:632(DCl:794 Tl:794 Ul:749) 39W 61A 49c] ASIC:[ 85c (125c) 520hz(BL: 520) 88 (E:193) F:0 L:0]
LOOP[8] ON TO:0 (w:283)
 24: DC2DC/1/:[vlt1:625 vlt2:632(DCl:794 Tl:794 Ul:749) 42W 67A 53c] ASIC:[ 85c (125c) 560hz(BL: 560) 89 (E:193) F:0 L:0]
 25: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 40W 63A 58c] ASIC:[ 85c (125c) 530hz(BL: 530) 101 (E:193) F:0 L:0]
 26: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 42W 66A 58c] ASIC:[ 85c (125c) 550hz(BL: 550) 93 (E:193) F:0 L:0]
LOOP[9] ON TO:37 (w:259)
 27: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 41W 65A 61c] ASIC:[ 85c (125c) 550hz(BL: 550) 87 (E:193) F:0 L:0]
 28: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 40W 64A 58c] ASIC:[ 85c (125c) 550hz(BL: 550) 90 (E:192) F:0 L:1]
 29: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 41W 66A 56c] ASIC:[ 85c (125c) 560hz(BL: 560) 82 (E:193) F:0 L:1]

[H:HW:2835Gh (1000),W:1117,L:0,A:28,MMtmp:0 TMP:(31/31)=>=>=>(52/52 , 51/51)]
Pushed 32 jobs , in HW queue 4 jobs (sw:0, hw:0)!
min:55 wins:2364[this/last min:40/33] bist-fail:7979, hw-err:277
leading-zeroes:42 idle promils[s/m]:0/1, rate:2306gh/s asic-count:28 (wins:23+17)
Fan:80, conseq:200
AC2DC BAD: 0 0
R/NR: 3297/0
RTF asics: 0
FET: 0:3 1:3
146 restarted 0 reset 0 reset2 25 fake_wins 0 stuck_bist 0 low_power 0 stuck_pll 72 runtime_dsble 0 purge_queue 0 read_timeouts 0 dc2dc_i2c 159 read_tmout2 169 read_crptn 2 purge_queue3 2336 bad_idle 0 err_murata
Adapter queues: rsp=1, req=20

[klondike_bar]

Bought my first SP30, second hand from eBay and with - FW: 2.6.1, start V: 0.655, max V: 0.75, max P: 1290W and 80% fan

I can't get more than 2800GH/s stable (two ASIC chips auto disabled - Loop [5]: 15, 17)

It looks like the device can't set the right voltage. No matter what voltage I set, it is fluctuating between 625-635, the PSU powers fluctuating between 700-800W and the temps are about 30 °C / 53,5 °C.
Is there anything I can do about that?

Code:

 2.6.1 Uptime:3498 | FPGA ver:62 | BIST in 1 
-----BOARD-0----- 
PSU[EMERSON1200]: 768->(688w/688w)[688 688 684] (->746w[729 727 746]) (lim=1290) 0c 1491GH cooling:0/0x0 
-----BOARD-1----- 
PSU[EMERSON1200]: 688->(652w/652w)[648 652 648] (->672w[654 653 672]) (lim=1290) 0c 1343GH cooling:0/0x0 

LOOP[5] ON TO:0 (w:69)
 15: disabled (Asic all engines fail BIST)
 16: DC2DC/1/:[vlt1:626 vlt2:632(DCl:794 Tl:794 Ul:749) 38W 61A 42c] ASIC:[ 85c (125c) 510hz(BL: 510) 69 (E:192) F:0 L:0]
 17: disabled (Asic all engines fail BIST)
[H:HW:2835Gh (1000),W:1117,L:0,A:28,MMtmp:0 TMP:(31/31)=>=>=>(52/52 , 51/51)]
Pushed 32 jobs , in HW queue 4 jobs (sw:0, hw:0)!
min:55 wins:2364[this/last min:40/33] bist-fail:7979, hw-err:277
leading-zeroes:42 idle promils[s/m]:0/1, rate:2306gh/s asic-count:28 (wins:23+17)
Fan:80, conseq:200
AC2DC BAD: 0 0
R/NR: 3297/0
RTF asics: 0
FET: 0:3 1:3
146 restarted 0 reset 0 reset2 25 fake_wins 0 stuck_bist 0 low_power 0 stuck_pll 72 runtime_dsble 0 purge_queue 0 read_timeouts 0 dc2dc_i2c 159 read_tmout2 169 read_crptn 2 purge_queue3 2336 bad_idle 0 err_murata
Adapter queues: rsp=1, req=20

not sure - theres no obvious reason for it to not push itself higher. Your start and max voltages are quite spread apart though, so it could take several hours to actually ramp up very slowly. It looks like your system uptime is only 1hr

try bumping your start voltage to 700mv - if it does down from there, its a sign of problem. if it stays there or barely rises, the system is likely very slow to adjust upward to nominal speed

[innergy]

..

not sure - theres no obvious reason for it to not push itself higher. Your start and max voltages are quite spread apart though, so it could take several hours to actually ramp up very slowly. It looks like your system uptime is only 1hr

try bumping your start voltage to 700mv - if it does down from there, its a sign of problem. if it stays there or barely rises, the system is likely very slow to adjust upward to nominal speed

OK, I'll try.. how much time it takes for the system to stabilize.. I never mined with ASICs before 

[RoadStress]

..

not sure - theres no obvious reason for it to not push itself higher. Your start and max voltages are quite spread apart though, so it could take several hours to actually ramp up very slowly. It looks like your system uptime is only 1hr

try bumping your start voltage to 700mv - if it does down from there, its a sign of problem. if it stays there or barely rises, the system is likely very slow to adjust upward to nominal speed

OK, I'll try.. how much time it takes for the system to stabilize.. I never mined with ASICs before 

It takes a bit of time to ramp up the speed.

[klondike_bar]

..

not sure - theres no obvious reason for it to not push itself higher. Your start and max voltages are quite spread apart though, so it could take several hours to actually ramp up very slowly. It looks like your system uptime is only 1hr

try bumping your start voltage to 700mv - if it does down from there, its a sign of problem. if it stays there or barely rises, the system is likely very slow to adjust upward to nominal speed

OK, I'll try.. how much time it takes for the system to stabilize.. I never mined with ASICs before 

It takes a bit of time to ramp up the speed.

+1. Theres a reason the settings page suggests keeping your start voltage close to your max voltage. If they are more than 50mV apart it will take a few hours or more to slowly adjust. you should at least see the hashrate stepping up 5-10GH every so often (5 min?)

[Biffa]

..

not sure - theres no obvious reason for it to not push itself higher. Your start and max voltages are quite spread apart though, so it could take several hours to actually ramp up very slowly. It looks like your system uptime is only 1hr

try bumping your start voltage to 700mv - if it does down from there, its a sign of problem. if it stays there or barely rises, the system is likely very slow to adjust upward to nominal speed

OK, I'll try.. how much time it takes for the system to stabilize.. I never mined with ASICs before 

Are you on 110V?

[topminingcontracts]

Did any try extreme undercloking.

If I am correct can get 0.5W/G in SP20 with 600W for a total 1.2 TH

Did any tested lower? Results?

Thanks in advance

JuanG

[Epoch]

Did any try extreme undercloking.

If I am correct can get 0.5W/G in SP20 with 600W for a total 1.2 TH

Did any tested lower? Results?

I haven't seen 600W @ 1.2TH, but I've measured the following; YMMV:

1275GH/s
max PSU: 210W
start: 0.62V
max: 0.67V
fan: auto
power: 730W

1250GH
start: 0.62
max: 0.665
fan: auto
power: 710W

1245GH
start: 0.62
max: 0.66
fan: auto
power: 695W

1200GH
start: 0.62
max: 0.65
fan: auto
power: 640W

1137GH
start: 0.62
max: 0.64
fan: auto
power: 577W

1085GH
start: 0.62
max: 0.63
fan: 20%
power: 534W

[Guy Corem]

Did any try extreme undercloking.

If I am correct can get 0.5W/G in SP20 with 600W for a total 1.2 TH

Did any tested lower? Results?

Thanks in advance

JuanG

Welcome to our thread Juan.

Latest report I got from a customer (privately, by email):

..."The SP20 were slightly more time consuming - anyway I ordered them
only because I had some 1200W PSUs left which I didn't want to let
unused. I figured that it should be possible to underclock the SP20 so
that one 1200W PSU can feed two of them - which is what I got finally.
Two SP20 connected to a 1.2kW gold PSU pull 1.07kW at the wall and
generate ~2.15TH. Even better than 0.5J/GH at wall - very efficient and
in every discipline superior to the only competitor S5."...

The customer has experience with mining gears from all the vendors. He is mining since 2011.

Guy

[goxed]

Did any try extreme undercloking.

If I am correct can get 0.5W/G in SP20 with 600W for a total 1.2 TH

Did any tested lower? Results?

Thanks in advance

JuanG

Welcome to our thread Juan.

Latest report I got from a customer (privately, by email):

..."The SP20 were slightly more time consuming - anyway I ordered them
only because I had some 1200W PSUs left which I didn't want to let
unused. I figured that it should be possible to underclock the SP20 so
that one 1200W PSU can feed two of them - which is what I got finally.
Two SP20 connected to a 1.2kW gold PSU pull 1.07kW at the wall and
generate ~2.15TH. Even better than 0.5J/GH at wall - very efficient and
in every discipline superior to the only competitor S5."...

The customer has experience with mining gears from all the vendors. He is mining since 2011.

Guy

My setup is similar, just that I am using  a) 1K 80Plus Plat and a) 1K 80 Plus Gold PSUs to power 4 SP20s (2 SP20 per PSU).
I have tried to hand-calibrate the voltages so that the power@wall consumed is 1100W for 2 miners connected to one PSU.

Here are the stats from my 4 SP20s.  

SP20 #1, #2 connected to the 1st PSU: Rosewill 80Plus Platinum, 220V Input

1082 GH/s (0.628V), 1098 GH/s (0.63V): power @ wall = 1070W: Wall efficiency = 2180GH/s / 1070Watt =  2.05GH/s / Watt

SP20 #3, #4 connected to the 2nd PSU: Rosewill 80Plus Gold, 220V Input

1100 GH/s (0.63V), 1020 GH/s (0.635V): power @ wall = 1101W: Wall efficiency = 2120GH/s / 1101Watt = 1.92GH/s / Watt
Accounting for 2% loss in efficiency by using a gold PSU vis-a-vis Platinum PSU, the estimated efficiency with plat PSU would be ~2120/1079=1.96GH/s / Watt

I think there's a range of Rockerbox ASIC quality (like GPUs) and the final efficiency also depends a bit on which kind of chips went into that particular SP20.
E.g. two chips in one of my SP20 are really stubborn in ramping up their mining speeds at high-efficiency levels (~0.635V)
But overall it's not a biggie, since the performance loss due to ASIC quality is quite negligible for a home miner like me.

[Guy Corem]

...
I think there's a range of Rockerbox ASIC quality (like GPUs) and the final efficiency also depends on which kind of chips went into that particular SP20.

Correct (http://en.wikipedia.org/wiki/Process_corners)
In addition, we skewed later batches to faster corners. Recent SP20s will be (usually) a bit more efficient.

[innergy]

..

not sure - theres no obvious reason for it to not push itself higher. Your start and max voltages are quite spread apart though, so it could take several hours to actually ramp up very slowly. It looks like your system uptime is only 1hr

try bumping your start voltage to 700mv - if it does down from there, its a sign of problem. if it stays there or barely rises, the system is likely very slow to adjust upward to nominal speed

OK, I'll try.. how much time it takes for the system to stabilize.. I never mined with ASICs before

It takes a bit of time to ramp up the speed.

+1. Theres a reason the settings page suggests keeping your start voltage close to your max voltage. If they are more than 50mV apart it will take a few hours or more to slowly adjust. you should at least see the hashrate stepping up 5-10GH every so often (5 min?)

No. What I see is CGMiner restarting every 2-3min and it restarts often when the voltages are higher. To lower the number of restarts I need to lower the voltages to 655, 660max but it leads to low rate and power from PSUs and after some time -  10-20 minutes the voltage always drops somewhere between 620-630. The average rate for the last 12 hours was 2500, and the graph looks ugly..

Are you on 110V?

No, 220V, EU

[RoadStress]

No. What I see is CGMiner restarting every 2-3min and he restarts often when the voltages are higher. To lower the number of restarts I need to lower the voltages to 655, 660max but it leads to low rate and power from PSUs and after some time -  10-20 minutes the voltage always drops somewhere between 620-630. The average rate for the last 12 hours was 2500, and the graph looks ugly..

Are you on 110V?

No, 220V, EU

Go to Settings tab. What limit do you have for "Restart the miner if it mines below X speed" (or something like that). If your limit is 3000GH and the miner starts hashing at 2500GH then it will always restart. Try to lower that limit to a speed that is below your starting speed.

[innergy]

No. What I see is CGMiner restarting every 2-3min and he restarts often when the voltages are higher. To lower the number of restarts I need to lower the voltages to 655, 660max but it leads to low rate and power from PSUs and after some time -  10-20 minutes the voltage always drops somewhere between 620-630. The average rate for the last 12 hours was 2500, and the graph looks ugly..

Are you on 110V?

No, 220V, EU

Go to Settings tab. What limit do you have for "Restart the miner if it mines below X speed" (or something like that). If your limit is 3000GH and the miner starts hashing at 2500GH then it will always restart. Try to lower that limit to a speed that is below your starting speed.

I made this limit to 1000 yesterday. The CGMiner stays relatively stable only if the voltages are set below, let say .660
The main problem I see is that the device couldn't stabilize the voltage to the given levels and they always drop somewhere around .625
The PSUs power and rate drops too.

Now I have startV: .655, maxV: .665 and CGMiner stable for about half an hour but the ASIC voltages are .625
They should be somewhere between .655 and .665, right?


EDIT: Looks like incident droppings of the ASIC's voltage to 0V causes mining restarting. I thing the voltage regulating circuit not working as it should. Is that repairable somehow?


EDIT2: OK, I think I get that SP30 stable around 3000Gh/s (straight rate line, no restarts) with: startV: .635, maxV: .640

Fan 60%, temps 30 / 60 C, ASICs voltage .637 stable
Board 0 - 800W, 1545Gh
Board 1 - 720W, 1383Gh (two ASIC chips disabled)

Every try to set voltage over .640 leads to crashes and mining restarts..
Just thinking.. is it possible that the two disabled failed chips may cause some short circuits when voltage is higher? Should I try to remove them from the PCB?

Obviously it looks like a hardware problem. What is the practice in that case?
Does Spondoolies-Tech offers replacement parts or repairing service?