Thanks for the reply.  I've traced the problem to no data on pin 13 ribbon from pin 35 on ASIC1 (U2).  Wish I had a pinout on the BM1382.  Level is low, flat, but not ground.  Am seeing 25MHz on TP11.

I've compared signals between the good and bad hash board on the S3+.  The one significant difference is that on the good hash board pin 13 of the ribbon cable has data flowing while the bad hash board does not.  I traced the pin 13 to U2 (the first ASIC in the chain) pin 35.  Now I'm stumped.  Without knowing the criteria for getting data from pin 35 U2, I'm stuck.

That seems to be it.  The bad board generates no heat when powered up.  Dragging out my old Tektronics 475,  with an uncalibrated probe (adjusted with the time VAR on the calibrator at 1khz) I get 27MHz from the crystal's TP11 and putting the VAR back it give me 20MHz so I go to the bad board and there's no oscillation.

-----------------------

Nope, got it wrong.  I had looked for the 25MHz while the bad board was on the bench.  Putting the bad board back into the miner and giving it the ribbon cable then looking again I get a booming signal.  So, that's not it.

Here's an odd bit.  On the S3 boards from memory there are at least a couple of 12v test points that if I remember correctly are common with 12v in.  On this version board I see only a single 12v test point and it doesn't appear to be common with 12v in however I do see voltage on the board.

Opening up the machine to compare with the other board now.

Odd.  The second working board has no continuity between the 12v test point and 12v in nor any pin on the ribbon cable connector.  Firing the board up, hashes well and no voltage showing at the test point so the test point is dead on both boards.

------------------

Dead crystal would explain it.  Changed crystals on USB miners in the past but looking at the two nearby smt resistors, I suspect that if the crystal is heated enough to remove the resistors will blow away.

May I ask what you mean by string design?  The ASICs are in series?  Then the S3+'s are also string design?  I have a dead S3+ beyond warranty I'm trying to fix.

I've tried this board with two S3+ controller boards.  Anyone know if an S3 controller will be harmed or do harm with an S3+ board?

This is an S3_Hash Board_262 V1.1.

https://drive.google.com/file/d/0B_DLmNkCkkVpOS05bkRCblRjVzQ/view?usp=sharing
https://drive.google.com/file/d/0B_DLmNkCkkVpa2NPWmh0VjAwXzg/view?usp=sharing

I tried the image button but get an invalid image type.  Let me know if there's a view problem as I might need to change share settings.

Nice.  Thanks.

I see the ribbon cables will conflict if both board are orientated top up.  Or one could buy two water blocks and only put a single board on each so as to have the distance.

Following Bitmain's suggestion I shut down my second S3+ and tested the bad hash board with the other's controller.  The hash board is still bad but I did just find a difference between the two S3+'s.  The one that has remained good has components on the inside above the heatsink as S3's have.  The S3+ with a failed hash board has no components on the reverse side except the ribbon connector of course.

Good thing I did what they said as dust was accumulating on the internal heatsink of the good S3+ behind the front fan.

The large tank pump has arrived from China but the other two pumps have been moving water with no loss for quite some time so I won't disturb the setup.  My two C1's are running at 37/39/37/38° & 36/40/38/40° so I like water cooling.  These board are the same as the S3+'s if I'm not mistaken but well cooled so capable of higher hashrate?  I'm running my C1's 243.75M/00675v/981GH/s for lower current on the line.  One S3+ is running at 42/39° and the other with a single board is 39°.  The one with the single board had started to heat up so I removed a cheesecloth filter dropping the temps down from mid-50° to the 40°'s but perhaps the damage had already been done.  The next day one of its boards had failed.  The higher temp meant higher current due to increased inefficiency and some damage occurred.  So, I like the water cooling.  I'd like to put water blocks in the S3+'s.

The S3+ has two hashing boards.  The miner runs at 226.76GH/s with only one board in place.  I swap that out for the other board and no hashing.  The miner had gotten up into the 50°'s.  Perhaps an ASIC was going and drawing more power?  I'm guessing that the increased current may have damaged a conducting thru hole or internal trace.  cgminer shows zero hashing and no chain nor board data.  

The miner had been tested with a number of power supplies.  I have plenty.

Running 2 S3+'s on a single cart, one lower than the other but due to power supply lead length had supplies on the first shelf, miner on second and third and supplies on the top.  Had cheesecloth on each intake.  Yesterday one of the S3+'s was running 8-10° hotter than the other.  Remove cheesecloth from both and the temperature on the hotter dropped.  The cart faces a window fan exhausting heat from the room and the cooler S3+ was more directly in line with the window fan pull.  This morning the hashrate dropped according to the pool so I checked my miner's GUI's and found that yesterday's heating S3+ was only showing chain1.  Took the miner out, cleaned dust, put it back and still only chain1 which in this case is the left side while facing the RJ45 and typically that's chain2 in a 2 board configuration.  Both fans run.  Will continue to troubleshoot but what I found so far is that with the top off and checking voltages, with 12.00 volts set on my power supplies and the supplies having digital readouts, I measure 11.85v on the inside hashing board PCIE (this on the one hashing while the non-hashing board shows 12.00).  So, each hashing board getting fed with two 16 gauge wires, one to each PCIE, .15vdc drops in the wire run (about 30").  Will now reverse the controller board....

It's been very humid, rain, present pressure 29.40"Hg.

----------------------

Flipped the controller board around so the hashboards have swapped ribbons and fired up.  Again the same hashboard is hashing and the same dead is dead although fan running.

-----------------------

Took the outside cover off the bad hashboard and cleaned the ASICs well.  Figured that removing the cheesecloth might have put dust in.  Fired it back up and still no hashing.

------------------------

Shut it down, reversed the controller board again but this time didn't connect the ribbon to the working hashboard.  Have power to all four PCIE.  Fan on the good hashboard running full out by its sound.  Fan on the other running normally but cgminer reports 0 hashing no chains.

Will try a reset to default although it's been running at stock 225M S3+ default.

-----------------------------

Reset to 192.168.1.99 using the 90 second reset switch close.  First attempt after reset, both red and green of that pair were out but the green and yellow RJ45 leds were lit, shut down for more than a minute and retried.  The top red of the red/green pair tho first solid, went to fast flashes for a period then off then fast flashes then off then fast flashes then off then on then green lit and the RJ45 leds lit, I connected to 192.168.1.99 and found only chain1.  So, it's time to remove the hashboard from the miner and look closely for a problem.  Voltages to the ASICs on the bad board ranged from .80 to .78 so they are seeing power.

----------------------------

Tried it again.  Hashing board and controller both out of the miner, ASICs cleaned and outside heatsink replaced without compound.  All the ASIC pins are clear.  Fired up, settled at red/green pair red lit bright and green lit dimly.  GUI connects but no hashing going on.

----------------------------

Up and mining on a single hashing board.  Discovered a mistake I was making.  I want both fans running.  Previously on a problem S3 I had interrupted the blue wire and brought it to 12v from the supply to get full fan speed.  Then I put in a toggle to be able to switch from 12v to dictated voltage at will.  So, to get the front fan running at full speed I ran leads to the blue and ground.  Didn't work.  Running power to the red and black worked.  In this instance without a board going  to the black, yellow and red, power to the blue didn't get it done.

soy

-------------

Drat.  The Bitmain Wiki page says:'We also give you a 24 Month Warranty to all our Products.

But then lower down it says 90 days from manufacturer shipping.

When did the S3+ stop shipping?  Are any S3+'s still under warranty?  Come this summer, heat will kill them.  Down south here this one took an 80° day just fine but all the rain recently, a 29.40"Hg humidity reduced cooling sufficiently that 70°'s killed it.  I did have some cheesecloth over the intake fan to reduce dust accumulation which could have been a contributing factor.

This miner ran stock frequency, never overclocked, since day 1 directly here from Bitmain.

I try to ignore the 'don't bother it loses more by not mining' advice I've gotten not infrequently but then I spent years in a proto electronics labs and probably view it differently as a result.

I've sometimes considered trying the cook it in an oven electronics reflow solution.  I have used that approach on a laptop.

Now I see this promising scheme:  http://martindrapeau.tumblr.com/post/116344427765/how-to-fix-iphone-4s-wi-fi-grayed-out-with-a  

Not going to try it right away tho.

Changed my mind when I considered the fan noise.  The room has air exhausting via one window with radiators for C1's and another window with an exhaust fan for the 4 miners just inside.  So a lot of fans.  The room draws air from the rest of my place which causes a draft in the kitchen where the chicks are now sleeping but no worries, a warming pad underneath and a cover over the mesh.  Couldn't hack the warming pad PIC, now they time out at an hour as heat can damage tissue, so whipped up a 55 minute timer and a solid state relay as cycling power to the heating pad resets its timer.

Found a use for the heat.  My hens have been coming up short with eggs so I picked up a couple of sex-linked red hen chicks this morning and the two are in a cage on my kitchen table crying missing their flock.  At night they'll be moved into the miner room.  Nice and warm.

1) ssh in
2) run the command 'screen -r'
3) type P and that will take you to the Pools screen (took multiple P entries to get it to the Pool Management screen)
4) Then one could usually enter pool information but I see on mine the cgminer notification messages scroll down the page, e.g. Invalid Bitmain-voltage

So, this method won't work.

This isn't working.  After 2days it was down to 428GH/s so I rebooted and 9½ hrs later it's down to 424GH/s.  What I notice now is that although Chain2#8 isn't x'd, it's a degree hotter than chain1 but the fan is running slower - chain1/2400/42, chain2/2160/43.  

So, I tried to add a variable speed control to the pull fan and dial in same RPM's.  Not only didn't it work but now I'm not getting the fan to run with a simple SPDT switch between 12v & standard out.  Either my wiring is wrong or I fried something in the fan.  The PWM controller has pos & neg for both IN and Mtr.  I had made neg common for both.  If the PWM pulls motor in below ground, e.g. intending the motor to be isolated from IN, then there may be a problem.

------------------

No, I had put a strain on the fan wires and the connector to the hashboard, under the cover, had partly come free.  Interestingly, when the S3 starts up the ground to the fan is disconnected.  I know this because while the red led is blinking I can flip the switch from 12v (from the supply) to the internal fan speed output, the fan doesn't turn in either position.  When the board starts hashing both fans start and the toggle from dictated speed to full speed works properly.

It may be that the PWM motor controller can work and that the internal plug had been partly separated already but I don't think so as the fan ran full blast ignoring my dialing down the rheostat on the PWM motor control.

After 8 minutes the chain2 fan was at 2220.  I flipped the switch and the fan is now, well it was 4200 but has dropped to 3900.

I stand corrected.  I picked up my Dell N750P-S0 today and it clearly states12VDC 62.4A and 3.3VDC=5.5A.

---------------------

Have it powered up and seeing 12.20vdc!

---------------------

Connecting to the new supply - I could solder here but the lower flat washer makes contact with all three connections:  https://drive.google.com/file/d/0B_DLmNkCkkVpcjc4ZUduYmM3dUE/view?usp=sharing

--------------------

Tested an older miner the other day and it was 368 watts/281GH/s @ 12.0vdc with 2 of my cheap supplies (<$25 ea).

Am running the N750P-S0 to the same miner, 372 watts/285GH/s @12.2vdc ($17.96 w/free shipping).

I tried running the pull fan on a problem S3 at full speed and found that to get it done and still have speed readings I had to cut the blue wire and bring that from the fan to the supply 12v while leaving the hashboard blue wire dead ended.  

I bought a PWM speed controller from China and was going to put the output to the blue wire and have an adjustable fan speed control with normal speed feedback to the GUI.  But, since I wanted to troubleshoot the problem S3 I put it back to stock and haven't attempted the variable speed yet.  I decided I wanted to get some connectors so as not to cut the wires as the fans don't have a lot of extra lead length.  Also, summer is on its way and I may just want full speed on fans so then I might be better off with a simple SPDT toggle on each fan for hashboard dictated speed or full speed.

I wonder why on the S3's there are two different fan speeds.  When I added a pull fan to my S5 I took its power from the 12v supply without concern for variable speed.  The fan cable splitters on ebay would seem good for adding a fan to an S5, again why two fan speeds.  And on an S3 the splitter could be put to the board with faster fan and then have a SPST toggling the fans' blue wire from either the hashboard blue wire output or to supply 12v.  Think I'll order some now, give it a try and then if I'm happy with the option I'll put it on all my miners for the summer.  Though mid-summer I may be shutting down during the day and only running at night, remains to be seen.

S3#2 on which I re-flowed solder at the PCIE connectors and on the inside top of hashboard2/chain2, has been coming up with an x at Chain2#8.  That was one, among others, that had been coming up x before the work.  Now the S3, after running a day, will settle out at 426.xx with that ASIC x'd out.  This has happened a couple of times.  That difference from 441 is just about 1/32nd or what 1 ASIC puts out.

Someone said they had luck putting thermal conductive tape on all ASIC tops.  Noting his work, I took conductive tape from a TO-220 and cut it down, put it on the single ASIC showing an x, and have just put it back up to run.  It remains to be seen if the tape on that one ASIC, giving it a little more pressure than the other ASICs, pressing its underside harder onto the groundplane (if not well soldered under there) and tighter to the heatsink, will keep that ASIC from x'ing out.