[ANN] Bitfury is looking for alpha-testers of first chips! FREE MONEY HERE!

[GandalfG]

Here's a couple of photo's I took while assembling an S-HASH board.

cscape, what kind of solder paste you using for stencil ?

[cscape]

cscape, what kind of solder paste you using for stencil ?

I've tried various kinds, but this is Multicore 96SC LF320 AGS88 lead free (Farnell order code 5091100) and so far I like it the best. The problem I have with some kinds of paste is that they are very runny, falling off the squeegee, and slumping when on the board. This one is nice and firm, but not dry.

[GandalfG]

cscape, what kind of solder paste you using for stencil ?

I've tried various kinds, but this is Multicore 96SC LF320 AGS88 lead free (Farnell order code 5091100) and so far I like it the best. The problem I have with some kinds of paste is that they are very runny, falling off the squeegee, and slumping when on the board. This one is nice and firm, but not dry.

Heh I tok some, testing few and didn't fund best.Temporary  use Koki  (standard S series) but I'm looking for something more stable with longer be used.

http://www.ko-ki.co.jp/product/

[jspielberg]

Hey Guys...

I posted this in the bitfury retail saies threads, but this seems like the bitfury technical thread so I thought I would ask here too.

------
...To get my 25 GH, I received an H-board that is functioning at full capacity (~19GH) and an "EOL" end of line card with is running about 6 to 10 GH.  After looking at the stats.log, it appears that the EOL card is only seeing the first 6 chips.

I took some macro hi-res pictures of the chips 6 and 7 (45 and 46 labelled on the card), where the spi communication I am guessing is broken.

https://i.imgur.com/Iddp2WQ.jpg


https://i.imgur.com/MpXXHdS.jpg


For the life of me, I don't see anything out of the ordinary.

The european thread seems more advanced at getting full speed out of these boards and was wondering if any experienced eyes see something suspicious.
-------

Anyway... the current advice seems to be that reflowing the 6 and 7 position asics would be a good first step.  I am going to pickup a hot air rework station, and was wondering if anyone knew the flow temperature for the solder used on the H-boards (I guess that is directed mostly at c-scape or intron since they seem to have the most inside knowledge of these boards).
Anyone else have any advice for a biochemist/computer geek jumping into the EE world to get 10 additional asics hashing?  

[kaerf]

Hey Guys...

I posted this in the bitfury retail saies threads, but this seems like the bitfury technical thread so I thought I would ask here too.

------
...To get my 25 GH, I received an H-board that is functioning at full capacity (~19GH) and an "EOL" end of line card with is running about 6 to 10 GH.  After looking at the stats.log, it appears that the EOL card is only seeing the first 6 chips.

I took some macro hi-res pictures of the chips 6 and 7 (45 and 46 labelled on the card), where the spi communication I am guessing is broken.

https://i.imgur.com/Iddp2WQ.jpg


https://i.imgur.com/MpXXHdS.jpg


For the life of me, I don't see anything out of the ordinary.

The european thread seems more advanced at getting full speed out of these boards and was wondering if any experienced eyes see something suspicious.
-------

Anyway... the current advice seems to be that reflowing the 6 and 7 position asics would be a good first step.  I am going to pickup a hot air rework station, and was wondering if anyone knew the flow temperature for the solder used on the H-boards (I guess that is directed mostly at c-scape or intron since they seem to have the most inside knowledge of these boards).
Anyone else have any advice for a biochemist/computer geek jumping into the EE world to get 10 additional asics hashing?  

just solder those solder jumpers (SJs) next to the bad chip. this turned my 15GH/s EOL board into a 25GH/s board.

[jspielberg]

Thanks Kaerf -

Bridge all 4 (i.e. SJ16,SJ17,SJ18,SJ54)?

[-Redacted-]

That would take out the 6th chip, which you said is hashing.  You want to start bridging on the next set of jumpers, attempting to bypass chip 7.  Without looking at a scope, no way to tell if it's the 6th or 7th chip that's the problem - you may end up having to bridge them both, but start by trying to bypass chip 7 which would be 55, 19, 30, and 21...

[jspielberg]

Got it... SJ(55,19,20,21).

Looks like I will be investing in a soldering iron this week

Small steps....

[-Redacted-]

Get something reasonably decent...  I've had my eye on this:

http://www.amazon.com/X-TRONIC-MODEL-4010-XTS-Centigrade-ANTI-MAGNETIC/dp/B0053491YO

[jspielberg]

Get something reasonably decent...  I've had my eye on this:

http://www.amazon.com/X-TRONIC-MODEL-4010-XTS-Centigrade-ANTI-MAGNETIC/dp/B0053491YO

Thanks Redacted... I was thinking of getting https://www.sparkfun.com/products/11704 but your link looks to be a better value.

[jimrome]

Get something reasonably decent...  I've had my eye on this:

http://www.amazon.com/X-TRONIC-MODEL-4010-XTS-Centigrade-ANTI-MAGNETIC/dp/B0053491YO

Thanks Redacted... I was thinking of getting https://www.sparkfun.com/products/11704 but your link looks to be a better value.

Go with the Hakko. Get a decent magnifier too, you'll need it.

[-Redacted-]

If you go with the Hakko, get a couple of extra T18-B tips for it.  That standard chisel tip is going to look enormous compared to the size of stuff you'll be working on.   And don't forget some solder and small roll of solder wick...

[jspielberg]

Lol... the shopping list is getting longer... those sparkfun guys are going to love me.

The advice/recommendations are very much appreciated.

[intron]

Thanks Kaerf -

Bridge all 4 (i.e. SJ16,SJ17,SJ18,SJ54)?

Bridging all solder jumpers can fix the SPI chain by taking
out the bad ASIC. But when the ASIC still has functional
output drivers (meaning: the ASIC is not hashing, but the
outputs still drive a logic level on there pins) this shorting
of the solder pads won't help much. The ASIC has to be
removed from the chain completely then. To do that, cut
the traces at two points with a small snap-off  knife and
apply some heat with the tip of a soldering iron. The glue
between the copper and the board will soften and the trace
can be removed easily, without doing any damage to the
board. You even hardly see it when done with care.

Shorting the jumpers also has a little trick. Applying solder
in one go can be tricky. Just apply solder to one pad and
let it cool down. Then apply solder to the other pad and let
it cool too. Repeat a few times until a small lump has 'grown'
on each pad. Then in one swift motion connect both lumps
with some more solder.

intron

[Doff]

Lol... the shopping list is getting longer... those sparkfun guys are going to love me.

The advice/recommendations are very much appreciated.

Hehe since your pictures are good if you wouldn't mind showing what this looks like after you have cut the traces and soldered anything that would be cool.

Just in case I need to do this in the future id like to see how its actually done.

[vs3]

Lol... the shopping list is getting longer... those sparkfun guys are going to love me.

The advice/recommendations are very much appreciated.

Hehe since your pictures are good if you wouldn't mind showing what this looks like after you have cut the traces and soldered anything that would be cool.

Just in case I need to do this in the future id like to see how its actually done.

See also this example: (source: https://bitcointalk.org/index.php?topic=250249.msg3099241#msg3099241)

(although that is a bit of a overkill)

In essence - make a narrow cut (1mm) just enough so that you can visually tell that the traces have been cut apart and that should do it. It is also a good idea to keep the cut relatively small - in case it turns out that this is the wrong chip
(yes - believe me - Murphy's laws were not invented out of boredom!)

Be careful as once you cut the traces it would be relatively easy to grab an end and just peel the entire trace off. That's fine if you're sure you'll never ever need it again (but are you?)

You can also use a small piece of wire (any wire will do) to connect the points. Or make a large slump of solder and bridge them that way. I'd go with a wire though - much easier.

[-Redacted-]

The trace cuts at U44 are not a very good example of what it should look like when you get done....  The cuts near the bottom traces of U43 are much more like what you want to strive for - you can barely tell they've been cut.  Those can be repaired with a solder bridge if necessary, the cuts near U44 not so much..

[cscape]

I usually take an X-Acto knife with a fresh blade and a metal ruler. Cut across the traces. Shift the ruler a tiny bit, and make a parallel cut. Now, using tip of the blade peel away the little strips of copper between the two cuts.

[fpgaminer]

If you go with the Hakko, get a couple of extra T18-B tips for it.  That standard chisel tip is going to look enormous compared to the size of stuff you'll be working on.

Pffft, sissies.  I soldered 0402 with a big honking chisel tip on my Hakko.    Not that I'll ever want to hand solder 0402 again.

But for the love of electronics, please get good solder.  Good solder, good iron, and flux is all a man needs to be happy.

[cscape]

I do everything with 0.8 mm chisel tip. But I agree with fpgaminer. Use good solder, and make sure it's fresh. Don't let the flux evaporate, and if you do then wipe tip clean and start again. My favorite solder wire for rework is Multicore crystal 511.