Topic: GridSeed 5-chip USB miner voltage mod - page 18. (Read 156980 times)

Thanks for the follow up; can you list the tools/hardware you used and your new hash rate? Also if possible, anyone know the page in thus thread where I find the 49.9k mod offhand?

Without my looking at the schematic, I can easily assume that there are some common connections that are completed by each chip's bridged pins. I recall there being several. Make sure, as per the schematic, that those pins are still bridged and if not, bridge them. That may take care of getting the other 2 chips working. Check and re-connect or bridge any pins that were disconnected, compromised or damaged by the bad chip or the removal thereof.
It's possible that the other two chips went bad when chip 1 went SNAP! FIZZZZZZZZZZ! and released its magic black smoke  
I think you're on the right track. Keep it up and you'll have it all figured out and working again in no time.
I wonder if we can order the actual chips ready to go now?
Good luck, mate!

I have 1 grid which I have modded which seems to be getting much more HW errors @ 1200mhz than all my other grids.. All mine are 0-5HW< per 24 hr and this one is like 100 HW per 24 Hr I have applied cooling to the top and bottom, but still happening I wonder what else there is to try... I mean its still fully functional the shares are sitting just below the others.. so I can keep running like this and just ignore it or try solve it or reduce Mhz

Well I found the dummy load.  Actually it found me ...... 

I think Gridseed numbers the chips 1,2,3,4,5
Cpuminer numbers the chips 0,1,2,3,4
Anyway, using cpuminer numbering this morning chips 2,3 & 4 were working. Chips 0 and 1 were not working.  Wattage was too high at 34w.
Chip 1 (M2 in the center I assume) is no more.  I removed what was left of it after it went up in smoke.

Now chips 0 and 4 are working. Watts are much reduced, only 14.5w at 1000MHz with the two chips working.  I feel more confident about it with the low watts. Around 170 kh/s reported by cpuminer and the pool after several hours. Very stable.

I'm assuming chip 1 connections to the board permitted chips 2 & 3 to work this morning but also prevented chip 0 from working.  Removing chip 1 allowed chip 0 to work but now prevents chips 2 & 3 from working. A 2 chip expected wattage pod is ok after that meltdown but a 4 chip pod would be better!  I need to figure out if there is some way to bring chips 2 & 3 back online. Maybe a bridge of some sort?

I did the 49.9k mod on all 14 of my gridseeds and I'm glad I did. I'm running CPUminer and all of my chips are stable at 1200mhz and up. The pool and CPUminer and both reporting very close to each other so I'm happy!!

Nervous to attempt, but an extra 100 khsh across 20 miners translates to an extra 2Mhsh. Anyone want to share any best practices aside from OP?

visit hashra site/support

So to post some good news after the lot bad from my side i got 20nF capacitors today and fixed my not working gridseed *puhh* but no vmodding for me anymore to small and tiny ..

anyway has someone info if hashra wi update to sandors cpuminer ?

Yes! It certainly will. Hashra also has an updated version of it.
Check it out.

The mod will work for any mining software, you just need to change the chip frequency.

will this mod work with hashra controla?

1/8W...0.125W ...the smallest one...

1/4 or 1/2W ...or even 2W will work...but are too BIG...

For the Vmod3 (zigmod)  is it a 1/4w 16.9kohm resistor, or a 1/2w 16.9kohm resistor?

hmm, sounds like the work of a hidden dummy load... that's WAY too high for only 3 working chips, keep searching.....something's still drawing too much current - it could be those chips pulling too much current too due to being compromised by previous problem. They still might end up being declared 'shot' and the chips will all have to be replaced. Could be a faulty support or driver component sucking too much current otherwise. keep lookin

I will try to.  Thanks for the suggestion.  I'll keep posting what I find as clues for anyone that can help me.  For example I've found that with the 3 working chips at 1000MHz single mode mining the damaged pod is drawing 34 watts with a 47.5k resistor in the R139 spot.  I think that is high so I did not let it run unattended overnight.  I ran it this morning and it still finds 3 chips.  I'll work on it this afternoon.

It just occurred to me. Shouldn't the GUI/TUI config issues and debugging stuff all be under support/turning thread?

Anyway. I found 2 of my miners dead when I checked them this morning. A comm port reset (disable/enable) took care of it, thank goodness!
But after running for weeks upon weeks without any drop out issues, this is very disconcerting. I don't need pod disconcert in my life! GO AWAY!  

So this didn't happen until I reverted to the previous version of Sandor111's latest updates.
Oh, and I see pretty much the same issue with this version where sticking at over 1200MHz happens too.
Perhaps they kicked off line after being pushed too hard for so long. Not sure. Or it's just a uP fart caused it. Who knows. I am not looking forward to having keep an eye on the drop out issue again, dammit!

I've restarted the latest non-scrolling (unless Sandor has updated it since) ver. again since the previous one has the same 'stuck at 1225+' problem too, just to get them running again.
Have you corrected this issue Sandor? I see you've been busy correcting the TUI issue.
Perhaps I missed an update while catching up on this thread.

Oh, and how do you get the GUI to resize without using all the per-defined settings?
Someone got there's adjusted so they can see the summary and scrolling output at the same time.

Can we get these two issues squared away?
It they have been, please forgive my reader lameness and show me the link for the updated Sandor-minerd.

Thanks!
Wolfman

EDIT: Just fixed the scrolling issue. Like the others that did, I simply resized the window, restarted and voila! Perfectio! Now if we can just get autotune fine tuned and the red/green color indicators back....

Does anyone have a sample config for cpuminer? I want to switch my CGMiners over to it, in hopes that I can avoid crashing every 8 hours.

I would like a mining script config with at least 3 backup pools, and I am just not sure how to setup the config file, and how to load it with running ./minerd on my pi.

Do I just make a config file and run it like ./minerd-gc3355 --config miner.conf or something? Similar to CGMiner?

Thanks Wolfey. You gave me the idea to check the bad plug for continuity. I set the multimeter to the dial position you use to calibrate it by touching the two probes together (0 ohm) to test the barrel plug. Well it looks like the "-" screw connector goes to the center of the plug ! 

The "+" screw connector does not get a reading at all on either the inside or outside of the plug.

Checked the 16 remaining screw type barrel plugs I have left over and their all fine. Guess I had 1 out of the 50 that was bad. Soooo sad

Hey everyone.

I had a Gridseed 5 chip that came from gawminers that would not hash at all, it detected, but always got 0khash no matter what I did, even trying two separate computers, swapping cables, etc.. I was ready to RMA the miner, but as a last ditch effort, I remember back to my xbox 360 RROD days and thought I would try to reflow the solder with a heatgun. To my surprise... the gridseed now works perfectly! running stock speeds for now, but I'll try the resistor mod in a few days if it still is hashing nicely.

Anyways, if you have some dead gridseeds, try a solder reflow!

For Wolfy:On another note, I have same issue with autotune. It eventually makes the pods run higher than is efficient. Even if it doesn't get hardware errors, I notice that sometimes hashrate suffers. I've resorted to manually programming mhz on alll the pods based on the data I gather using autotune, my hashrate doesn't suffer that way. Would be interested to know the autotune formula. It seems that chips can stop being efficient at higher clocks without triggering HW errors anyways.

For Sandor: Just wanted to say, now using your new miner on both windows and raspberry pi... very stable (when not using --debug or --log). cgminer on my raspberry pi would die every hour, your cpuminer is running like a champ for hours and hours.

Crap...forgot to remove the typo.  I cut and pasted from a previous post lol.