ANTMINER S7 is available at bitmaintech.com with 4.86TH/s, 0.25J/GH - page 135.

J4bberwock

hero member

Activity: 840

Merit: 1000

Quote from: klondike_bar on December 20, 2015, 04:24:15 PM

Quote from: RichBC on December 20, 2015, 12:29:46 PM

Quote from: J4bberwock on December 20, 2015, 12:05:34 PM

Quote from: RichBC on December 20, 2015, 12:00:52 PM

Quote from: J4bberwock on December 20, 2015, 11:28:01 AM

ok, so R15 = Rc1 in the datasheet, and C73, C75 would be Cc1,Cc2.
U2 should be a digital pot made by onsemi, then. In the cat51xx family
And that means that we could have some software control over it.
I spotted the 6k8 as being tied to ground.
I'll do a quick test on one board replacing the 6k8.
I noticed that without the controller plugged, the voltage output at the coil was closer to 8v, can't remember exact value.
And, it went higher to 10.2 with the control board plugged

8V would make sense for the sort of voltage you would expect without the shunting effect of the Digital Pot. So this implies that the PIC is being controlled or at least reset by the controller as opposed to it being completely stand alone as I had assumed?

A trick to feed the chips with a voltage so low they won't start (burn the entire board) if the controller doesn't send a signal to tell he is ready with ethernet connected?

Yes I guess it makes it safe. CAT51xx looks spot on against the pinout, not sure how that ties to AG30/ AG6K but guessing it's a 10K pot?
I do not have sufficient info with just the pictures to suss out all the connections to the PIC, in particular any interface to the control board?

Any chance the Batch8 is software controllable? digging around vi SSH (root/admin), found these references:
etc/config/cgminer.conf: "bitmain-voltage" : "0706"
etc/cgminer.conf.factory: "bitmain-voltage" : "0725"

seems strange that they are different, yet still necessary to reference. Anyone know what voltage the B8 actually runs at?

currenty playing with the values to see if they do something

RichBC

hero member

Activity: 588

Merit: 500

Quote from: J4bberwock on December 20, 2015, 05:16:48 PM

Found it:

MCP4017T-503E/L

http://ww1.microchip.com/downloads/en/DeviceDoc/22147a.pdf

Yes perfect, pinout still matches, except this time we have I2C. 127 Positions, Ram Memory. AGNN Code with NN being only the traceability code so explaining why we have 30 & 6K on the 2 boards. Plus it confirms my 50K.

J4bberwock

hero member

Activity: 840

Merit: 1000

Quote from: RichBC on December 20, 2015, 04:36:36 PM

Quote from: RichBC on December 20, 2015, 12:29:46 PM

Quote from: J4bberwock on December 20, 2015, 12:05:34 PM

Quote from: RichBC on December 20, 2015, 12:00:52 PM

Quote from: J4bberwock on December 20, 2015, 11:28:01 AM

ok, so R15 = Rc1 in the datasheet, and C73, C75 would be Cc1,Cc2.

U2 should be a digital pot made by onsemi, then. In the cat51xx family

And that means that we could have some software control over it.

I spotted the 6k8 as being tied to ground.
I'll do a quick test on one board replacing the 6k8.

I noticed that without the controller plugged, the voltage output at the coil was closer to 8v, can't remember exact value.
And, it went higher to 10.2 with the control board plugged

8V would make sense for the sort of voltage you would expect without the shunting effect of the Digital Pot. So this implies that the PIC is being controlled or at least reset by the controller as opposed to it being completely stand alone as I had assumed?

A trick to feed the chips with a voltage so low they won't start (burn the entire board) if the controller doesn't send a signal to tell he is ready with ethernet connected?

Yes I guess it makes it safe. CAT51xx looks spot on against the pinout, not sure how that ties to AG30/ AG6K but guessing it's a 10K pot?

I do not have sufficient info with just the pictures to suss out all the connections to the PIC, in particular any interface to the control board? However several of the connections come to the pads on the board edge probably including programming the PIC. If I had a board I would have a go at reading it in case they have not Read Protected :-)

I have done some maths on the Digital Pot, which has 32 positions, in combination with the 6K8 to ground & series 12K. So assuming it is a 50K Pot we get.

     Supply V   Core V
Min     8.63     0.58
Mid     9.17     0.61
Max     11.87     0.79

Which would be a very sensible range of core voltages, so may be correct?

Rich

Found it:

MCP4017T-503E/L

http://ww1.microchip.com/downloads/en/DeviceDoc/22147a.pdf

RichBC

hero member

Activity: 588

Merit: 500

Quote from: RichBC on December 20, 2015, 12:29:46 PM

Quote from: J4bberwock on December 20, 2015, 12:05:34 PM

Quote from: RichBC on December 20, 2015, 12:00:52 PM

Quote from: J4bberwock on December 20, 2015, 11:28:01 AM

ok, so R15 = Rc1 in the datasheet, and C73, C75 would be Cc1,Cc2.

U2 should be a digital pot made by onsemi, then. In the cat51xx family

And that means that we could have some software control over it.

I spotted the 6k8 as being tied to ground.
I'll do a quick test on one board replacing the 6k8.

I noticed that without the controller plugged, the voltage output at the coil was closer to 8v, can't remember exact value.
And, it went higher to 10.2 with the control board plugged

8V would make sense for the sort of voltage you would expect without the shunting effect of the Digital Pot. So this implies that the PIC is being controlled or at least reset by the controller as opposed to it being completely stand alone as I had assumed?

A trick to feed the chips with a voltage so low they won't start (burn the entire board) if the controller doesn't send a signal to tell he is ready with ethernet connected?

Yes I guess it makes it safe. CAT51xx looks spot on against the pinout, not sure how that ties to AG30/ AG6K but guessing it's a 10K pot?

I do not have sufficient info with just the pictures to suss out all the connections to the PIC, in particular any interface to the control board? However several of the connections come to the pads on the board edge probably including programming the PIC. If I had a board I would have a go at reading it in case they have not Read Protected :-)

I have done some maths on the Digital Pot, which has 32 positions, in combination with the 6K8 to ground & series 12K. So assuming it is a 50K Pot we get.

     Supply V   Core V
Min     8.63     0.58
Mid     9.17     0.61
Max     11.87     0.79

Which would be a very sensible range of core voltages, so may be correct?

Rich

klondike_bar

legendary

Activity: 2128

Merit: 1005

ASIC Wannabe

Quote from: RichBC on December 20, 2015, 12:29:46 PM

Quote from: J4bberwock on December 20, 2015, 12:05:34 PM

Quote from: RichBC on December 20, 2015, 12:00:52 PM

Quote from: J4bberwock on December 20, 2015, 11:28:01 AM

ok, so R15 = Rc1 in the datasheet, and C73, C75 would be Cc1,Cc2.
U2 should be a digital pot made by onsemi, then. In the cat51xx family
And that means that we could have some software control over it.
I spotted the 6k8 as being tied to ground.
I'll do a quick test on one board replacing the 6k8.
I noticed that without the controller plugged, the voltage output at the coil was closer to 8v, can't remember exact value.
And, it went higher to 10.2 with the control board plugged

8V would make sense for the sort of voltage you would expect without the shunting effect of the Digital Pot. So this implies that the PIC is being controlled or at least reset by the controller as opposed to it being completely stand alone as I had assumed?

A trick to feed the chips with a voltage so low they won't start (burn the entire board) if the controller doesn't send a signal to tell he is ready with ethernet connected?

Yes I guess it makes it safe. CAT51xx looks spot on against the pinout, not sure how that ties to AG30/ AG6K but guessing it's a 10K pot?
I do not have sufficient info with just the pictures to suss out all the connections to the PIC, in particular any interface to the control board?

Any chance the Batch8 is software controllable? digging around vi SSH (root/admin), found these references:
etc/config/cgminer.conf: "bitmain-voltage" : "0706"
etc/cgminer.conf.factory: "bitmain-voltage" : "0725"

seems strange that they are different, yet still necessary to reference. Anyone know what voltage the B8 actually runs at?

RichBC

hero member

Activity: 588

Merit: 500

Quote from: J4bberwock on December 20, 2015, 12:05:34 PM

Quote from: RichBC on December 20, 2015, 12:00:52 PM

Quote from: J4bberwock on December 20, 2015, 11:28:01 AM

ok, so R15 = Rc1 in the datasheet, and C73, C75 would be Cc1,Cc2.

U2 should be a digital pot made by onsemi, then. In the cat51xx family

And that means that we could have some software control over it.

I spotted the 6k8 as being tied to ground.
I'll do a quick test on one board replacing the 6k8.

I noticed that without the controller plugged, the voltage output at the coil was closer to 8v, can't remember exact value.
And, it went higher to 10.2 with the control board plugged

8V would make sense for the sort of voltage you would expect without the shunting effect of the Digital Pot. So this implies that the PIC is being controlled or at least reset by the controller as opposed to it being completely stand alone as I had assumed?

A trick to feed the chips with a voltage so low they won't start (burn the entire board) if the controller doesn't send a signal to tell he is ready with ethernet connected?

Yes I guess it makes it safe. CAT51xx looks spot on against the pinout, not sure how that ties to AG30/ AG6K but guessing it's a 10K pot?

I do not have sufficient info with just the pictures to suss out all the connections to the PIC, in particular any interface to the control board? However several of the connections come to the pads on the board edge probably including programming the PIC. If I had a board I would have a go at reading it in case they have not Read Protected :-)

J4bberwock

hero member

Activity: 840

Merit: 1000

Quote from: RichBC on December 20, 2015, 12:00:52 PM

Quote from: J4bberwock on December 20, 2015, 11:28:01 AM

ok, so R15 = Rc1 in the datasheet, and C73, C75 would be Cc1,Cc2.

U2 should be a digital pot made by onsemi, then. In the cat51xx family

And that means that we could have some software control over it.

I spotted the 6k8 as being tied to ground.
I'll do a quick test on one board replacing the 6k8.

I noticed that without the controller plugged, the voltage output at the coil was closer to 8v, can't remember exact value.
And, it went higher to 10.2 with the control board plugged

8V would make sense for the sort of voltage you would expect without the shunting effect of the Digital Pot. So this implies that the PIC is being controlled or at least reset by the controller as opposed to it being completely stand alone as I had assumed?

A trick to feed the chips with a voltage so low they won't start (burn the entire board) if the controller doesn't send a signal to tell he is ready with ethernet connected?

RichBC

hero member

Activity: 588

Merit: 500

Quote from: J4bberwock on December 20, 2015, 11:28:01 AM

ok, so R15 = Rc1 in the datasheet, and C73, C75 would be Cc1,Cc2.

U2 should be a digital pot made by onsemi, then. In the cat51xx family

And that means that we could have some software control over it.

I spotted the 6k8 as being tied to ground.
I'll do a quick test on one board replacing the 6k8.

I noticed that without the controller plugged, the voltage output at the coil was closer to 8v, can't remember exact value.
And, it went higher to 10.2 with the control board plugged

8V would make sense for the sort of voltage you would expect without the shunting effect of the Digital Pot. So this implies that the PIC is being controlled or at least reset by the controller as opposed to it being completely stand alone as I had assumed?

not.you

legendary

Activity: 1726

Merit: 1018

Quote from: scyth33 on December 20, 2015, 06:18:17 AM

"bitmain-use-vil=true" in cgminer.conf. any idea what does this do?

did you make that up?

bitmain-us-evil=true

J4bberwock

hero member

Activity: 840

Merit: 1000

Quote from: RichBC on December 20, 2015, 10:43:27 AM

Quote from: J4bberwock on December 20, 2015, 04:03:45 AM

Quote from: RichBC on December 20, 2015, 01:50:23 AM

Quote from: J4bberwock on December 19, 2015, 07:47:01 AM

Quote from: Subw on December 19, 2015, 07:27:12 AM

B6 backside photo https://imgur.com/sIYNhVX

Ok, so, the "30C" resistor is the 20k reference one.
"
and one of the ~~"123"~~ resistor allows you to output 10.6v.
divided by 15 chips in a string = 0.71v at the chips
Edit:
not "123"(12k), but there should be a 1.2k resistor or close value.

1.1k resistor in place of the 1.2k will give 0.77v at the chips
1.3k resistor will give 0.65v at the chips
18k resistor in place of the 20k should also work for 0.64v at the chips
22k resistor in place of the 20k will give 0.77v at the chips

Edit, I'm aging, took me 20 minutes to figure the over/undervolt.

I am just working from the available pictures, but just to say that I agree with all of the above including the fact that I cannot find the approx 1.2K resistor which is needed from pin 2 on the LM27402 to ground as the bottom part of the potential divider?

I have a possible solution which is that U3 on the other side of the board, which in the pictures I cannot read the ident, is a Digital Pot, perhaps set up from the pads on the edge of the board, which would enable a value & hence the voltage to be set? Bit of a wild guess and could be for something completely different but easy to check if someone has a board out of a system.

Also uncertain about U2 marked AG30, is this an LDO & if so what is it for?

Also note that the Bitmain pictures show a V1 without U3 and that Batch 6 is V2 & Batch 8 a V3 Board

Rich

Changing the 20k for 16k and 22k resistors didn't change anything to the hashing capabilities Sad

All 3 blades are still working at 800MHz.
All those heatsinks aren't helping to trace the circuits. I need to look deeper into this once I have more time available

I have not fully sussed it but am making progress with the better picture. R15, 20K is not the Reference resistor it's part of the compensation network to pin 3 of the Buck with C73 & C75. The reference resistor (fb1) is R9, 82K taking the voltage from the Converter output.The resistor to ground (fb2) is R10, 6K8.

However this is where it get's interesting and I have not fully worked it out. In parallel with R10 to Ground is a series combination of R17, 12K and U2.

Which of course begs the question as to exactly what U2 (AG30 on one board AG6K on the other) My guess is that it's a single channel Digital pot controlled over I2C by the PIC, but more work needed to confirm this.

Rich

ok, so R15 = Rc1 in the datasheet, and C73, C75 would be Cc1,Cc2.

U2 should be a digital pot made by onsemi, then. In the cat51xx family

And that means that we could have some software control over it.

I spotted the 6k8 as being tied to ground.
I'll do a quick test on one board replacing the 6k8.

I noticed that without the controller plugged, the voltage output at the coil was closer to 8v, can't remember exact value.
And, it went higher to 10.2 with the control board plugged

RichBC

hero member

Activity: 588

Merit: 500

Quote from: J4bberwock on December 20, 2015, 04:03:45 AM

Quote from: RichBC on December 20, 2015, 01:50:23 AM

Quote from: J4bberwock on December 19, 2015, 07:47:01 AM

Quote from: Subw on December 19, 2015, 07:27:12 AM

B6 backside photo https://imgur.com/sIYNhVX

Ok, so, the "30C" resistor is the 20k reference one.
"
and one of the ~~"123"~~ resistor allows you to output 10.6v.
divided by 15 chips in a string = 0.71v at the chips
Edit:
not "123"(12k), but there should be a 1.2k resistor or close value.

1.1k resistor in place of the 1.2k will give 0.77v at the chips
1.3k resistor will give 0.65v at the chips
18k resistor in place of the 20k should also work for 0.64v at the chips
22k resistor in place of the 20k will give 0.77v at the chips

Edit, I'm aging, took me 20 minutes to figure the over/undervolt.

I am just working from the available pictures, but just to say that I agree with all of the above including the fact that I cannot find the approx 1.2K resistor which is needed from pin 2 on the LM27402 to ground as the bottom part of the potential divider?

I have a possible solution which is that U3 on the other side of the board, which in the pictures I cannot read the ident, is a Digital Pot, perhaps set up from the pads on the edge of the board, which would enable a value & hence the voltage to be set? Bit of a wild guess and could be for something completely different but easy to check if someone has a board out of a system.

Also uncertain about U2 marked AG30, is this an LDO & if so what is it for?

Also note that the Bitmain pictures show a V1 without U3 and that Batch 6 is V2 & Batch 8 a V3 Board

Rich

Changing the 20k for 16k and 22k resistors didn't change anything to the hashing capabilities Sad

All 3 blades are still working at 800MHz.
All those heatsinks aren't helping to trace the circuits. I need to look deeper into this once I have more time available

I have not fully sussed it but am making progress with the better picture. R15, 20K is not the Reference resistor it's part of the compensation network to pin 3 of the Buck with C73 & C75. The reference resistor (fb1) is R9, 82K taking the voltage from the Converter output.The resistor to ground (fb2) is R10, 6K8.

However this is where it get's interesting and I have not fully worked it out. In parallel with R10 to Ground is a series combination of R17, 12K and U2.

Which of course begs the question as to exactly what U2 (AG30 on one board AG6K on the other) My guess is that it's a single channel Digital pot controlled over I2C by the PIC, but more work needed to confirm this.

Rich

RichBC

hero member

Activity: 588

Merit: 500

Quote from: J4bberwock on December 20, 2015, 08:47:56 AM

Here are the pics I took yesterday.
U3 is almost readable.
http://imgur.com/a/YoNul

I'd bet on pic12-1572

http://ww1.microchip.com/downloads/en/DeviceDoc/40001723D.pdf

Yes good spot, you nailed it...

So we have us a Microcontroller, it must be to do with the voltage control. Just need to work out how it's connected and how they are using it?

Rich

J4bberwock

hero member

Activity: 840

Merit: 1000

Quote from: RichBC on December 20, 2015, 08:11:59 AM

Quote from: J4bberwock on December 20, 2015, 04:03:45 AM

Quote from: RichBC on December 20, 2015, 01:50:23 AM

Quote from: J4bberwock on December 19, 2015, 07:47:01 AM

Quote from: Subw on December 19, 2015, 07:27:12 AM

B6 backside photo https://imgur.com/sIYNhVX

Ok, so, the "30C" resistor is the 20k reference one.
"
and one of the ~~"123"~~ resistor allows you to output 10.6v.
divided by 15 chips in a string = 0.71v at the chips
Edit:
not "123"(12k), but there should be a 1.2k resistor or close value.

1.1k resistor in place of the 1.2k will give 0.77v at the chips
1.3k resistor will give 0.65v at the chips
18k resistor in place of the 20k should also work for 0.64v at the chips
22k resistor in place of the 20k will give 0.77v at the chips

Edit, I'm aging, took me 20 minutes to figure the over/undervolt.

I am just working from the available pictures, but just to say that I agree with all of the above including the fact that I cannot find the approx 1.2K resistor which is needed from pin 2 on the LM27402 to ground as the bottom part of the potential divider?

I have a possible solution which is that U3 on the other side of the board, which in the pictures I cannot read the ident, is a Digital Pot, perhaps set up from the pads on the edge of the board, which would enable a value & hence the voltage to be set? Bit of a wild guess and could be for something completely different but easy to check if someone has a board out of a system.

Also uncertain about U2 marked AG30, is this an LDO & if so what is it for?

Also note that the Bitmain pictures show a V1 without U3 and that Batch 6 is V2 & Batch 8 a V3 Board

Rich

Changing the 20k for 16k and 22k resistors didn't change anything to the hashing capabilities Sad

All 3 blades are still working at 800MHz.
All those heatsinks aren't helping to trace the circuits. I need to look deeper into this once I have more time available

Hmm that's surprising. 20K & 16K would have the core voltage to approx 0.57V & 0.77V, which should have been noticeable. However the key thing rather then the hashing is did you measure the Volts out from the Converter?

I straightened and tidied the pictures as best I could.

Need to identify what u2 / AG30 is and what it's function is? It may be an LDO but these surface mount idents are always tricky.

U3 is the chip that may be a Digital Pot? If any one can read the ident from one that would be great. As you can see this is an addition from the V1 Hash board on the Bitmain product page.

Rich

Here are the pics I took yesterday.
U3 is almost readable.
http://imgur.com/a/YoNul

I'd bet on pic12-1572

http://ww1.microchip.com/downloads/en/DeviceDoc/40001723D.pdf

RichBC

hero member

Activity: 588

Merit: 500

Quote from: J4bberwock on December 20, 2015, 04:03:45 AM

Quote from: RichBC on December 20, 2015, 01:50:23 AM

Quote from: J4bberwock on December 19, 2015, 07:47:01 AM

Quote from: Subw on December 19, 2015, 07:27:12 AM

B6 backside photo https://imgur.com/sIYNhVX

Ok, so, the "30C" resistor is the 20k reference one.
"
and one of the ~~"123"~~ resistor allows you to output 10.6v.
divided by 15 chips in a string = 0.71v at the chips
Edit:
not "123"(12k), but there should be a 1.2k resistor or close value.

1.1k resistor in place of the 1.2k will give 0.77v at the chips
1.3k resistor will give 0.65v at the chips
18k resistor in place of the 20k should also work for 0.64v at the chips
22k resistor in place of the 20k will give 0.77v at the chips

Edit, I'm aging, took me 20 minutes to figure the over/undervolt.

I am just working from the available pictures, but just to say that I agree with all of the above including the fact that I cannot find the approx 1.2K resistor which is needed from pin 2 on the LM27402 to ground as the bottom part of the potential divider?

I have a possible solution which is that U3 on the other side of the board, which in the pictures I cannot read the ident, is a Digital Pot, perhaps set up from the pads on the edge of the board, which would enable a value & hence the voltage to be set? Bit of a wild guess and could be for something completely different but easy to check if someone has a board out of a system.

Also uncertain about U2 marked AG30, is this an LDO & if so what is it for?

Also note that the Bitmain pictures show a V1 without U3 and that Batch 6 is V2 & Batch 8 a V3 Board

Rich

Changing the 20k for 16k and 22k resistors didn't change anything to the hashing capabilities Sad

All 3 blades are still working at 800MHz.
All those heatsinks aren't helping to trace the circuits. I need to look deeper into this once I have more time available

Hmm that's surprising. 20K & 16K would have the core voltage to approx 0.57V & 0.77V, which should have been noticeable. However the key thing rather then the hashing is did you measure the Volts out from the Converter?

I straightened and tidied the pictures as best I could.

Need to identify what u2 / AG30 is and what it's function is? It may be an LDO but these surface mount idents are always tricky.

U3 is the chip that may be a Digital Pot? If any one can read the ident from one that would be great. As you can see this is an addition from the V1 Hash board on the Bitmain product page.

Rich

Sweminer777

hero member

Activity: 754

Merit: 500

1xBit the largest casino

Quote from: J4bberwock on December 20, 2015, 04:03:45 AM

Quote from: RichBC on December 20, 2015, 01:50:23 AM

Quote from: J4bberwock on December 19, 2015, 07:47:01 AM

Quote from: Subw on December 19, 2015, 07:27:12 AM

B6 backside photo https://imgur.com/sIYNhVX

Ok, so, the "30C" resistor is the 20k reference one.
"
and one of the ~~"123"~~ resistor allows you to output 10.6v.
divided by 15 chips in a string = 0.71v at the chips
Edit:
not "123"(12k), but there should be a 1.2k resistor or close value.

1.1k resistor in place of the 1.2k will give 0.77v at the chips
1.3k resistor will give 0.65v at the chips
18k resistor in place of the 20k should also work for 0.64v at the chips
22k resistor in place of the 20k will give 0.77v at the chips

Edit, I'm aging, took me 20 minutes to figure the over/undervolt.

I am just working from the available pictures, but just to say that I agree with all of the above including the fact that I cannot find the approx 1.2K resistor which is needed from pin 2 on the LM27402 to ground as the bottom part of the potential divider?

I have a possible solution which is that U3 on the other side of the board, which in the pictures I cannot read the ident, is a Digital Pot, perhaps set up from the pads on the edge of the board, which would enable a value & hence the voltage to be set? Bit of a wild guess and could be for something completely different but easy to check if someone has a board out of a system.

Also uncertain about U2 marked AG30, is this an LDO & if so what is it for?

Also note that the Bitmain pictures show a V1 without U3 and that Batch 6 is V2 & Batch 8 a V3 Board

Rich

Changing the 20k for 16k and 22k resistors didn't change anything to the hashing capabilities Sad

All 3 blades are still working at 800MHz.
All those heatsinks aren't helping to trace the circuits. I need to look deeper into this once I have more time available

yo man . areyou trying to verlocko about the capacitieS? would be cool if you could get like 10thz-

Btw i wrote to you , i

scyth33

member

Activity: 99

Merit: 10

"bitmain-use-vil=true" in cgminer.conf. any idea what does this do?

J4bberwock

hero member

Activity: 840

Merit: 1000

Quote from: RichBC on December 20, 2015, 01:50:23 AM

Quote from: J4bberwock on December 19, 2015, 07:47:01 AM

Quote from: Subw on December 19, 2015, 07:27:12 AM

B6 backside photo https://imgur.com/sIYNhVX

Ok, so, the "30C" resistor is the 20k reference one.
"
and one of the ~~"123"~~ resistor allows you to output 10.6v.
divided by 15 chips in a string = 0.71v at the chips
Edit:
not "123"(12k), but there should be a 1.2k resistor or close value.

1.1k resistor in place of the 1.2k will give 0.77v at the chips
1.3k resistor will give 0.65v at the chips
18k resistor in place of the 20k should also work for 0.64v at the chips
22k resistor in place of the 20k will give 0.77v at the chips

Edit, I'm aging, took me 20 minutes to figure the over/undervolt.

I am just working from the available pictures, but just to say that I agree with all of the above including the fact that I cannot find the approx 1.2K resistor which is needed from pin 2 on the LM27402 to ground as the bottom part of the potential divider?

I have a possible solution which is that U3 on the other side of the board, which in the pictures I cannot read the ident, is a Digital Pot, perhaps set up from the pads on the edge of the board, which would enable a value & hence the voltage to be set? Bit of a wild guess and could be for something completely different but easy to check if someone has a board out of a system.

Also uncertain about U2 marked AG30, is this an LDO & if so what is it for?

Also note that the Bitmain pictures show a V1 without U3 and that Batch 6 is V2 & Batch 8 a V3 Board

Rich

Changing the 20k for 16k and 22k resistors didn't change anything to the hashing capabilities Sad

All 3 blades are still working at 800MHz.
All those heatsinks aren't helping to trace the circuits. I need to look deeper into this once I have more time available

VirosaGITS

legendary

Activity: 1302

Merit: 1068

Quote from: marvykkio on December 20, 2015, 03:56:03 AM

I do not think is correct, the price difference, should cost the same price of the batch 9, because it is the same hardware, same power, nothing changes, this is only a policy to make more money, and all that we allow ourselves.
cmq I need coupon for batch 8, you have something to sell?

I do not think you understand. Getting a S7 1 month earlier mean you generate 1 BTC in the meantime. So the hardware is worth a bit more.

If you do not understand why the S1 was worth hundreds of dollars a long time ago and is now only worth 40$, then i guess there is not much else i can say to explain it to you.

marvykkio

hero member

Activity: 798

Merit: 1000

I do not think is correct, the price difference, should cost the same price of the batch 9, because it is the same hardware, same power, nothing changes, this is only a policy to make more money, and all that we allow ourselves.
cmq I need coupon for batch 8, you have something to sell?

cryptichermit

member

Activity: 106

Merit: 10

Quote from: notlist3d on December 19, 2015, 09:40:07 PM

Quote from: cryptichermit on December 18, 2015, 05:12:08 PM

Quote from: PPOC on December 18, 2015, 04:27:00 PM

Where are you pointing all that hash?

Usually between westhash and f2pool - whichever one is higher payout at the moment.
I've written scripts to manage the miners and change pools according to highest payout.
Sometimes I'll swing them at a PPLNS pool if I think it's due to have a good day, but mostly leave it auto to PPS pools.

How often does westhash win? It seems like nice/westhash have been below bitcoin payments most of time unless I have missed something.

There are some times that westhash has significant spikes. The spikes are not overly frequent. They seem to occur when new SHA coins are being mined.
Also during difficulty adjustments - like today/yesterday - Westhash seems to hold on to the older rate for several hours as it adjusts down. My connection to Westhash is better than f2pool as well.
My overall hashrate is generally 2-3% higher on Westhash, so I factor in these connectivity variants into the script calculations. Probably mining westhash 30-40%.

Topic: ANTMINER S7 is available at bitmaintech.com with 4.86TH/s, 0.25J/GH - page 135. (Read 528055 times)