Hacking a BFL Jalapeno to 32GH and beyond....(???)

[lightfoot]

I can't recall what temps I used... I had to max out my Hakko hot air wand though to get it to go.  It's that @#$#@ lead-free solder, makes it a bitch to work on.  I hate it so much.

Hm. Well, let's go at this from the other end: At what point will these boards literally catch fire? :-)

I was just running 450c out of the unit at max speed for 3 minutes at that non-balled chip. Didn't work, it's on the board but not perfect. Damn damn damn. :-)

Yes, I hate ROHS solder in ways that would make a normal person go "hm". To be honest one way to deal with it on modern boards is to use Chip-quik to get the component off, then clean it up and use leaded solder to put the new parts on.

But what it sounds like is that the problem is not I have burned the chips, but that the damn thing just needs that much heat. One truth here is the more chips I have on the board, the more difficult it is to get the heat on the edges, that might be it as well. Or something like it, but man....

I'm going to re-group for a bit here and think: I tried setting the parameters to very... high... levels and got zilch. I have broken a few of the pads on two chips (including at least one of the critical communication pads), and a third doesn't seem to want to work even when low to the board.

I've also re-read the code, and I don't see anything in the chip init routines that would prevent the chip from firing up due to lower 12 volt voltage, so that's not it. I've tried running without heavy diags, and with disable engine zero on. So it's not a code issue.

Hm. I think I have two options:

1) Pre-heat the board to a solid 300 degrees F to warm it up, then use the wand for 120 seconds at 450c at that low chip. This is about as much as I can throw, and at that level I am running a serious risk of blowing capacitors off the board.

2) Pull the chip, and re-ball it at a much lower temp with 64 lead solder balls. Not much fun, but at least I will be working with a lower temp solder.

In the meantime I need to find another single chip to work on. So far I have knocked three out of commission, this is starting to get a bit expensive. :-) But that's part of learning.

[Inaba]

Yes, you have to heat it up like a BBQ to get the chips off that board.  The first time I stripped the chips off a board it took me about an hour because I couldn't figure out why it wasn't coming off... I finally cranked it up to around 450 and had a very tight nozzle on it and the chip came right off.  I had to have the air right up on the chip, if I put it at "normal" distance, the solder just won't melt... the board has a very efficient thermal plane, so it will start migrating the heat out of the chip rapidly.

Don't worry about catching the board on fire, your wand should not produce enough heat to cause any problems with the board itself.  Just be careful of some of the surrounding components... however, I think the worst case scenario is a few junction temps that max out at 120C, but it's doubtful you'll reach that if you have focused air on the chip itself.

[lightfoot]

Yes, you have to heat it up like a BBQ to get the chips off that board.  The first time I stripped the chips off a board it took me about an hour because I couldn't figure out why it wasn't coming off... I finally cranked it up to around 450 and had a very tight nozzle on it and the chip came right off.  I had to have the air right up on the chip, if I put it at "normal" distance, the solder just won't melt... the board has a very efficient thermal plane, so it will start migrating the heat out of the chip rapidly.

Don't worry about catching the board on fire, your wand should not produce enough heat to cause any problems with the board itself.  Just be careful of some of the surrounding components... however, I think the worst case scenario is a few junction temps that max out at 120C, but it's doubtful you'll reach that if you have focused air on the chip itself.

Ok. Tomorrow I'll give it a "what's the worst that could happen" run up with 300F on the infrared pre-heater and all-out on the chip. And yes, the board has a very impressive amount of thermal conductivity, this is different from the boards of the 90's which were more along the lines of laminated fiberglass. It's pretty easy to overheat those and either delaminate the pad or burn the board beneath it.

Interesting. On the older boards one of the failures of component burnout was that the board itself would get hot and liquify inside, causing electrical shorts through the planes. This might not be a problem here.

In the meantime I went to the Ebay china crap store and bought a bag of lead solder balls. I'll give hand-reballing a shot as well, lead will melt at a way lower temp...

And BTW thanks for the advice and help. This is an experiment, if it turns out these chips are still usable I will be pretty tickled.

C

[lightfoot]

Ok, 2:30am again. Took the board to 300F, and ran the wand to 460c, as high as it would go.

Hit the chip for 2-3 minutes at full temp. Cooled down, tried the chip on the board. Nothing. Tried removing the chip from the board, could not do it with the board on the side. Chip is not getting hot on startup, so it's probably 100% dead.

Oh well. I needed a space-holder there.

Next up is to perhaps find a new chip, and in my spare time see if I can re-ball one of these with a few torn pads. The other one is being sold to a bitcoin collector for .01btc. Till then I'll take a break, anybody got any BFL chips for sale?

C

[lightfoot]

Morning theory: Maybe the problem isn't in these new chips I am putting down, but with one of the existing chips I *have* put down.

Theory:
The BFL system seems to use three lines to control what address a chip has. Each chip socket is wired in a specific binary pattern, the system then selects a chip by putting an address on the bus. Maybe one of the chips I have put down and is working is masking that bus somehow.

Facts:
LEDs on back:
LED, status, engine number, engine ID from BFL
7 , on, engine 1, 0A 000
8 , off, engine 2, 0B 001
9 , off, engine 3, 0C 010
10, on, engine 4, 0D 011
11, on, engine 8, 0H 111
12, on, engine 7, 0G 110
13, on, engine 6, 0F 101
14, off, engine 5, 0E 100

The problem is I don't know how to map sockets to chips. Is this right?
Theory:
Pad, LED, binary, chip in place and working?
U3  engine 1 000 yes
U5  engine 2 001 no
U6  engine 3 010 no, does not work
U8  engine 4 011 yes
U7  engine 8 111 yes
U9  engine 5 101 no, does not work
U13 engine 6 101 yes
U14 engine 7 110 yes

The sockets which don't seem to work are U6 and U9. I have never tried U5

Maybe one of my "working" chips has an open address line and is masking these other chips. Not sure.

Counterpoint: If this was the case, then I doubt a "race" condition would always pick the same processor every time. Either I would see other lights coming on, or I would see a reduction in chips when I add the one that would overlap.

What's bugging me though is that my own reports show chip 5 didn't work, then more heat got it to work "sometimes", then running it at super heat got it online. Maybe I just was too timid about chip 6 there, it did come online for *one* power cycle, but did not repeat. So I never got to see how many cores it had. It was sinking power and generating heat, which is what the other chip I failed on (the one I didn't re-ball and is now probably welded to the board) didn't do.

[Drug5bitz]

I would just like to say I'm enjoying this jalapeno thread as much as any other.

Sad to say I was looking at the chips on eBay prior to reading this. After that I can't find another one.

I can't afford a 100 chips by myself so I think the only thing I can do is coordinate a group buy if anyone would be interested.

I personally want prolly about 20 chips. seeing as I said fuck the upgrade they used lower grade chips I would assume.

Is this a good idea, would anyone want to organize this with me? I think I'm going to go start a thread because I know other peoples projects include these chips as well.

[nexus99]

If there was a good way to get a Jally upgraded to run 6 or 8 chips I would be onboard to mod 2 units.

[lightfoot]

If there was a good way to get a Jally upgraded to run 6 or 8 chips I would be onboard to mod 2 units.

Sure. It's probably one of the better ROI deals out there right now. We have the equipment, one just needs to get chips on them. Problem is I don't have quite the confidence in myself yet to do this for others, or the spares pool to make up for any errors.

Keep watching, and if you think you can do it give it a shot. What's the worst that can happen? :-)

C

[lightfoot]

I would just like to say I'm enjoying this jalapeno thread as much as any other.

Sad to say I was looking at the chips on eBay prior to reading this. After that I can't find another one.

I think people have been reading this thread. Yes, I need another chip myself, might be awhile. On the other hand I helped some people who were trying to get rid of the chips, so I guess I did a good thing overall.

I can't afford a 100 chips by myself so I think the only thing I can do is coordinate a group buy if anyone would be interested.

Indeed, that's a huge order size, especially given the chips have been around for awhile. If the price was in the 10-20 range I'd give it a go.

I personally want prolly about 20 chips. seeing as I said fuck the upgrade they used lower grade chips I would assume.

No, I think what they did was fair; I bought a 4.5gh unit and I got a 5gh unit. They used mediocre chips on mine, but I was still able to get to 7.3 which was not bad at all. It also got our units shipped earlier, as I think they used up all the taller gen 1 chips on the jallys because they were ready to go.

Is this a good idea, would anyone want to organize this with me? I think I'm going to go start a thread because I know other peoples projects include these chips as well.

Please do and feel free to link back from this thread. A lot of people seem to be reading it, and I'd be happy to serve as a clearing point for knowledge and ideas and such.

C

[dentldir]

Thanks for posting your progress lightfoot. I was inspired to give hacking my Jally a shot today.  I was able to get 2 chips from the ebay vendor before he was out of stock.

For what its worth:

1) I upgraded it to a self compiled 1.2.9 via AVR dragon in Atmel Studio 6.1, no problems (using danatacker's settings).
2) My unit was an older unit without heatpipe and older unbadged ASICs which sit higher.
3) It had two chips aligned vertically so I added the two chips horizontally to balance the heatsink load.
4) I used Kester 186 flux that I've used for Xbox 360 repairs in the past.
5) I used 450 c on hot air with a nozzle the same size as the chip, no preheater.
6) Reflowing took a few minutes, watched with a magnifying glass until things looked right on both chips.
7) I used different solder pastes to level the height difference in chips.  Didn't have any pads around.

The result was one good new chip, one unrecognized new chip.  Hashing away at 11.5 Gh/s now. 43 c @ 3.95V.

Stats say that the processors are 3, 5, and 7.  So those for sure are the sockets up, down, left, and right from the center.

I will probably order some more chips and try reflowing all my added chips.  If I come up with any good temp combos, timing, or other insights I'll let you know.

[Drug5bitz]

Chips AHOY!!!!

https://bitcointalk.org/index.php?topic=236103.0

Found this with some help from some great forum members.

[lightfoot]

Thanks for posting your progress lightfoot. I was inspired to give hacking my Jally a shot today.  I was able to get 2 chips from the ebay vendor before he was out of stock.

For what its worth:

1) I upgraded it to a self compiled 1.2.9 via AVR dragon in Atmel Studio 6.1, no problems (using danatacker's settings).
2) My unit was an older unit without heatpipe and older unbadged ASICs which sit higher.
3) It had two chips aligned vertically so I added the two chips horizontally to balance the heatsink load.
4) I used Kester 186 flux that I've used for Xbox 360 repairs in the past.
5) I used 450 c on hot air with a nozzle the same size as the chip, no preheater.
6) Reflowing took a few minutes, watched with a magnifying glass until things looked right on both chips.
7) I used different solder pastes to level the height difference in chips.  Didn't have any pads around.

The result was one good new chip, one unrecognized new chip.  Hashing away at 11.5 Gh/s now. 43 c @ 3.95V.

Stats say that the processors are 3, 5, and 7.  So those for sure are the sockets up, down, left, and right from the center.

I will probably order some more chips and try reflowing all my added chips.  If I come up with any good temp combos, timing, or other insights I'll let you know.

Congratulations! I think the horozontal chips are the easiest to add; you have air around all four corners, best chance for a good reflow. Thanks also for the 450c for several minutes, I'll use hotter temps from now on. Likewise light 3 tells me a lot; mine is off, so I know which ones are at the corners now. Will verify tonight.


If you go past 4 I am going to highly recommend putting heat sinks on your FETs, possibly top and bottom. The little copper ones are working perfectly for me at this time.

What kind of pastes were you using?

[dentldir]

My go to pastes are the Antec Nano Diamond 6 and 7 series.  6 is thin, easy to spread, and works great.  7 is thick, really hard to spread, but works just a touch better than 6 in practice.  It's good for making thicker layers on stuff.  Both are available off the shelf at Staples and Staples takes gift cards bought from Gyft with Bitcoin.

I'll look at getting some heatsinks for the FETs.  I was also thinking about treating the Jalapeno like MrTeal's Chili board and doing a CPU heatsink under the chip area.  Not sure how I could mount that yet, but the sandwiched heatsinks on the Chili's work great.

Since the chips are so hard to reflow on the Jally, a thermal pad and a giant heatsink for the entire bottom might make sense too.  (like the Block Erupter blades).
 

[lightfoot]

My go to pastes are the Antec Nano Diamond 6 and 7 series.  6 is thin, easy to spread, and works great.  7 is thick, really hard to spread, but works just a touch better than 6 in practice.  It's good for making thicker layers on stuff.  Both are available off the shelf at Staples and Staples takes gift cards bought from Gyft with Bitcoin.

I'll look at getting some heatsinks for the FETs.  I was also thinking about treating the Jalapeno like MrTeal's Chili board and doing a CPU heatsink under the chip area.  Not sure how I could mount that yet, but the sandwiched heatsinks on the Chili's work great.

Since the chips are so hard to reflow on the Jally, a thermal pad and a giant heatsink for the entire bottom might make sense too.  (like the Block Erupter blades).
 

Take a look at my pics on the beginning of the thread. I put a big heat sink on the bottom, clears the board's vias through the AL plate, with a layer of heat sink grease. It drops the temps on the board by a LOT; very worth the time to do.

Ah, thermal compound. I'm using heat pads from frozencpu.com, seems to work well and flexible enough I don't worry about crusing the chips. Although as of late I have just been putting a touch of grease on the two original chips; they don't mine very hot, and it puts them and the new chips in prety much the same plane.

C

[Drug5bitz]

Check this out, looking around to help.

http://imgur.com/a/CMazz#0

Every BFL PCB with chip layout.

P.S. I'm looking at water cooling options could you find me dimensions on the chip area (would a 50x50mm block work?) and the hole centers for the heat sink mount.

[eoakland]

Hello LightFoot, 

I have a question for you.  I recently tried to follow your footsteps, and mod a Jally.  I dropped in 2 chips.  When I powered on, it looked like all was going smoothly....Jally was up to 25ghs...and all the sudden it konk'd out.  it still powers on, but just gives errors.  When I try to Jtag it, it will not allow me to.  It indicates volts are 0.0.   I am hoping you may be able to shed some light, if I still have hope or if I am super F'd...LoL


thanks

[lightfoot]

Hello LightFoot, 

I have a question for you.  I recently tried to follow your footsteps, and mod a Jally.  I dropped in 2 chips.  When I powered on, it looked like all was going smoothly....Jally was up to 25ghs...and all the sudden it konk'd out.  it still powers on, but just gives errors.  When I try to Jtag it, it will not allow me to.  It indicates volts are 0.0.   I am hoping you may be able to shed some light, if I still have hope or if I am super F'd...LoL


thanks

Hmmmm. Can you post pics what were the temps how.was it cooled and how did you get 25gh on 4 chips total?

Check the fets and does it smell odd.

C

[eoakland]

speed may be incorrect, there were no work units.  mosfets were uncooled, but it was not on long enough to matter.  the unit currently powers on, but does not detect.  and will not allow Jtag to get into program mode.  I had someone perform the work for me, so I am passing along what I was told.

sadly, no pics. 

just figured I would get your take, since you have adding for a while. 

[lightfoot]

speed may be incorrect, there were no work units.  mosfets were uncooled, but it was not on long enough to matter.  the unit currently powers on, but does not detect.  and will not allow Jtag to get into program mode.  I had someone perform the work for me, so I am passing along what I was told.

sadly, no pics.  

just figured I would get your take, since you have adding for a while.  

Well hm....

So let's think about this: The jally was a (stock code or not-stock code?) unit, two chips were added, it came up, and now does nothing. Let's try some diagnostics:

1) Does the front light flash on power-up and then go solid?
2) Do any lights on the back come on? There should be the power LED by the power plug, then at least two of the eight LEDs along the back between the power plug and the USB plug.

It should light the rear power LED, then flash the front light, then stop flashing the front light and light the appropriate number of rear LEDs from that set of eight. That indicates the system is up and running.

Nothing we did here should affect the JTAG port; have you ever had it programmed before now? In all my cases, I could always erase memories, then reprogram. Yep, checked my notes, even when the thing was dead as a doornail.

Can you get pics of the board without the heat sink on it? Something's odd.

Before I did my first chip, I knew I had to go to 1.2.9 firmware. I tried Tarkin's code stock and it failed my Jally before I even put another chip on. I had to realize that one has to define LITTLE_SINGLE and not SINGLE and compile to get things to work.

If you have the 8 pin programmer on backwards you will get the 0 volt thing. Always a good idea to do stuff one step at a time, never change two things (new programming and new chips) at once if you can help it.

I'd say have the person check the LEDs and report back, and make absolutely sure pin 1 on the dragon JTAG is on the red wire side of the 10 pin connector, and the jally's pin 1 is on the red wire side of the 10 pin connector. On mine the dragon has to be upside down in order for the cable not to have a 180 twist in it. If this person is going straight he (or she) is doing it wrong.

If you're still sunk, drop me a line and maybe I can take a look at your unit. The JTAG thing makes no sense, even shorting the +1 supply didn't stop that from coming up (and I spent hours doing that one fun night)

C

[Brassguy]

Just an update...

I used my own modded 1.2.9 firmware first... worked at 6.4gh... crappy chips...

Bottom heat at 100C for about 10min

Applied no clean liquid flux  (edit)

IR lamp set to 120C (pulses on and off)
about 2.5min (x2) for chip to drop (each one)

Let stand for about 10min

Assemble

Enjoying 14.7gh at about 48C (nice cold basement! )

Yum!