Diagnosing Broken BE Blade

[pmagid]

I have been overclocking my V2 BE Blades.   I did something amazingly stupid.  I applied power to the board while the resistors that are changed out to up the voltage were un-populated.   DOH!   Now the board wont even boot.   The led does not even do the boot up flash.   It looks completely dead like the fuse is blown.     My first thought was to check the fuse which is fine.   Any ideas on what else I could have fried being SOOOO STUPID.

Thanks in advance!

Paul

[dentldir]

If its anything like the blade I fried the other night, then its the 3A buck voltage regulator.  Check your 3.3V test pin.  I'll bet it doesn't read 3.3V.

I'm awaiting replacements from Digikey at the moment.

[pmagid]

Were you able to fix yours?   What did you ultimately do?

[dentldir]

Still waiting for the parts to arrive.  I'll be sure to post back my results.

[dentldir]

No go.  Swapping out the synchronous buck regulator was not the solution. 

Does anyone have a working blue back v2 blade they can measure R120 and R121 ohms in circuit while the blade is powered off?  Mine are measuring identical and one should be approximately 3 times the other.

[soy]

No go.  Swapping out the synchronous buck regulator was not the solution. 

Does anyone have a working blue back v2 blade they can measure R120 and R121 ohms in circuit while the blade is powered off?  Mine are measuring identical and one should be approximately 3 times the other.

2.31k R121
7.41k R120

Inexpensive meter.

Have any luck getting your 3.3v?

I have a Chip:xxxxxxxxxxxxxxxxxxxxxxx problem.

[dentldir]

Thanks for that.  I have not fixed the 3.3V line.

I believe there is a short past the regulator that I haven't found.  The symptom is that the both resistors in the divider (R120,R121) are reading around 1.8k as if they were in parallel.

If you measure the test points on your board in each lane, what voltage are you getting?  Should be about 1.05V not overclocked, 1.25V overclocked.  Those voltages should be good for a clock anywhere from 12Mhz to 16Mhz respectively.

[soy]

Thanks for that.  I have not fixed the 3.3V line.

I believe there is a short past the regulator that I haven't found.  The symptom is that the both resistors in the divider (R120,R121) are reading around 1.8k as if they were in parallel.

If you measure the test points on your board in each lane, what voltage are you getting?  Should be about 1.05V not overclocked, 1.25V overclocked.  Those voltages should be good for a clock anywhere from 12Mhz to 16Mhz respectively.

I'm getting 1.04v and at best it was hashing at something under 12MH/s. (Edit: I misspoke, it was getting just under 11GH/s.)

R121, across - 2.31k, IC102 side of R121 to 12V neg into board 2.31k, far side of R121 to neg 0k

R120, across - 5.64k, IC102 side of R120 to neg into board same as above as that side of the resistor pair is common; this far side of R120 is interesting in that reading resistance to neg into the board in one polarity I get 5.91k, reverse polarity I get 9.77k.

I have the far side of R120 to the in side of C127 an the far side of C127 to ground.  Check for a shorted C127.

Meanwhile I'm not making any progress clearing my ASICs xxxxxxxxxxxxxxxxxxxxxxx problem.  I think perhaps the PIC may have gotten its program scrambled when I blew it off with compressed air because not only are the ASICs showing inoperable but the board can't Switch Servers, coming up with a Current Server: :19501

[soy]

Or IC70 bad.

[dentldir]

Thanks for the extra measurements.  I'll check it tonight.

Did you try a factory reset on your blade?  Not sure if it will help any, but it might be worth a shot.  I believe you just power it up with the factory reset pins jumpered.

[soy]

Thanks for the extra measurements.  I'll check it tonight.

Did you try a factory reset on your blade?  Not sure if it will help any, but it might be worth a shot.  I believe you just power it up with the factory reset pins jumpered.

Yes.  any luck with yours?

[dentldir]

Not yet.  Caps looked ok.  I'm going to pull IC70 next and see if the divider is still shorted. 

[pmagid]

Sorry for the silence on this topic...  I have not had a minute to do some testing...

I swapped out the buck regulator and IC70 with no luck....   My symproms are: I read .6v instead of 3.3v at the test point at the top of the board and my led does not flash during boot up.   The lack of flashing led has to be a big clue....

Any thoughts?

[dentldir]

What do your R120 and R121 read?  They are the divider that supposed get the buck regulator to 3.3V.  One should be about 3 times the other.

[pmagid]

Ok so I have the exact same 1.8k reading across both R120 and R121...  I removed R121 and the part looks like it is in spec.   Does this not imply a short somewhere else?   I am sorry, I am too much of a noob at this to figure out what to do next.

[dentldir]

I believe it does.  Something is shorting 3.3V to ground.  Not sure what, but if I figure it out I'll post back here.  It could be anything that fried in the 3.3V section.  I don't think it has to do with anything in the ASIC rows even though that's where the feedback resistors were changed if you overclocked it.

[pmagid]

I am giving up...
See: https://bitcointalk.org/index.php?topic=371187.new#new