Topic: GRIDSEED G-BLADE Overclocking 7Mh/s, improvements and repair - page 3. (Read 74130 times)

You still don't have a clue LOL.  Waste your money and time on Zeus while we go with Gridseeds which now are SFarts. Cant' wait for the new chip to come into our labs.

Zeus Really ??  We got stable 400 units of Gridseed 6 MHz tuned as shown. And each unit consume less than 100 Watt.
Zeus is history and Gridseed is still kicking. SFards.com if you don't believe it check it out.
[/quote]

Used Zeus cards are GREAT in comparison to the blades, any day! Zeus gives way more bang for the buck where hash rate per $ goes right out of the box and do not require tedious and wasteful modifications to squeeze more hashes out of! PERIOD!
Boast all you want. Blades BURN UP when overclocked. They NEVER were worth the money from day one!

All I can say is ,,, JUNK! Why is anyone wasting their time on these way under performing stable overclocking wannabe not so well thought out or engineered rush to market quick made crap by GrimeSeed? Magic Black Smoke makers! POOF!
I loved the 5 chippers. Had a ton of them, very stable little things. But the blades were made to run at their recommended speed for a reason! Because they are generally not stable at all when overclocked! I mean, look at the mess of junk you have to throw on them just to keep them from going in to melt down in the first place...
If you can get your money back because the seller mus-represented it as stable, do so.
Then don't waste any more time or money on GS Blades. Try the Zeus miners. Much more bang for the buck and very stable machines, at least the ones that were made and wired right in the first place when they were brand new!
If you are really in to experimenting when them, then go for it. But I think you're wasting your time.
Good luck!

[/quote]

Zeus Really ??  We got stable 400 units of Gridseed 6 MHz tuned as shown. And each unit consume less than 100 Watt.
Zeus is history and Gridseed is still kicking. SFards.com if you don't believe it check it out.

All I can say is ,,, JUNK! Why is anyone wasting their time on these way under performing stable overclocking wannabe not so well thought out or engineered rush to market quick made crap by GrimeSeed? Magic Black Smoke makers! POOF!
I loved the 5 chippers. Had a ton of them, very stable little things. But the blades were made to run at their recommended speed for a reason! Because they are generally not stable at all when overclocked! I mean, look at the mess of junk you have to throw on them just to keep them from going in to melt down in the first place...
If you can get your money back because the seller mus-represented it as stable, do so.
Then don't waste any more time or money on GS Blades. Try the Zeus miners. Much more bang for the buck and very stable machines, at least the ones that were made and wired right in the first place when they were brand new!
If you are really in to experimenting when them, then go for it. But I think you're wasting your time.
Good luck!

OK the Part AFB26 which burned out is an 300ohm@100MHz,1.5A size 0805.  I just ordered 100+ of those so if you need any parts than E-mail me at [email protected] we are also custom rebuilding those and other Gridseed units.  Currently we replace the standard Power with a much better version to reduce heat and improve efficiency.  I can email photos .... let me try to upload one here :




And swap out the resistor down to 40K Ohm :-)   or you better also replace the mosfet.

HI guys,

ok I just bought myself a Gridseed blade and will overclock it to much more than you guys have right now.I  did work with others on overclocking the 5 chip to a now very stable 1050 I can post details if interested. I did also build a Raspberry PI solution to manage PIPe build of stacked 10 Gridseeds in a tube with 2 fan's for cooling only to maximize efficiency.

regards

Olaf

Lumanet Solar Energy LLC
We do sell too :-) 

Thanks
I look the all board but no more damages.
I hope i will get it work if i change it.

SS
LM

Hi, it's the one next to the usb plug, so something went wrong on the USB side of the board.
BTW, it's a Ferrite Bead, not a fuse.
The Zero ohm resistor you referenced should work, at least as a temporary fix to test if the rest of the board is ok..
Using a ferrite bead like one of these should be better for long term safety. (0805 size, 300 ohm@100Mhz, 1.5 or 2 or 3 amp)
http://www.digikey.co.uk/product-search/en?pv66=5&pv66=6&FV=fff40034%2Cfff802ab%2C400006%2C13440073&mnonly=0&newproducts=0&ColumnSort=0&page=1&quantity=0&ptm=0&fid=0&pageSize=25
There may still be some damages elsewhere on the USB side of the board.
As I'm not monitoring this thread, don't hesitate to send me a PM if you need more help.

Hello!!!

My gridseed g-blade lost the AFB26, I start it and it burns down.

Can i get it work if  i replace the fuse with CRCW08050000Z0EAHP.

BTRG

LM

Hi, sorry for the late answer, I only check this thread once in a while.

I don't know why it's still not recognized.

Did you try to extract the flash from the not recognized blade and make a crc comparison with the one I uploaded?

Also, what revision are your blades?
Mine were the first batch sold.
I received them the day after they were out.
That shouldn't change anything, but check it just in case.
You can use the same method to extract the firmware from one of your working blades and flash it to the defective one in case they would be different revisions.

Last thing, do you have the exact same STM32 chip?
It needs to be a STM32F103 RCT6

Worst case, I'll send you one of my chips with the firmware already flashed in it to test.

Hi J4bberwock

Thanks for your procedure! I stepped succesfully trough it, only after the programming and verify succesfull, I powertoggled the device, removed the serial and boot stuff and connected it to USB again but still get the device not recognized warning, there are no leds on.

any thoughts?
is it possible that Gridseed protected the devices, so the binary is encoded and not poratble?

When the RDP (read protection) option is active, the only command that is allowed is the "Permanent Readout Unprotect" command (which erases all of the Flash).
So while the protection is enabled, it should not be possible to download programs to RAM and run them


edit: found the answer to the protection question

And here is the firmware needed for the G-Blade

http://dl.free.fr/vLuvCPiD9

I'll make a quick detailed "how to" later with pictures.

Material needed:
-an USB to UART converter (CP2102 based is working for me)
http://www.aliexpress.com/item/Free-Shipping-1PCS-CP2102-Serial-Converter-USB-2-0-To-TTL-UART-6PIN-Module/623537804.html?s=p
-2mm spacing pin headers will make things easier
http://www.aliexpress.com/item/10pcs-2mm-40-Pin-Male-Single-Row-Pin-Header-Strip/1982700728.html

Software:
STM32 flash loader demonstrator
http://www.st.com/web/en/catalog/tools/PF257525#

Connecting the CP2102 to the Blade:

With the Blade unplugged from the USB

-Connect pin 8 (Ground) with pin 6 on the 8 pins header next to the STM32 chip.
Pin 8 is the closest to the JTAG 10 pins header
Pin 1 is the closest to the hashing chips.

-Connect pin 1 with pin 5

Now, your STM32 is configured to allow read and write of the flash.

Next, connect the pin 3 (RX) of the Blade header to TX on the CP2102 header
And connect pin 4 (TX) of the RX on the CP2102 header.

You can now plug the USB from the Blade to a computer
And the CP2102 to your computer.

The Blade will be found as an "unknown device".

The CP2102 will be assigned a port number that you will find in the hardware manager.
If it doesn't, you might need to install the CP2102 drivers from Silabs

http://www.silabs.com/products/mcu/pages/usbtouartbridgevcpdrivers.aspx

And uninstall anything you've done with Zadig earlier.

Start the flash loader demonstrator
and change the settings to match your com port
Baudrate: 115200
Parity: Even
Echo: disabled
Timeout: 10

Click "next"
You might be asked to unlock the device.
Do it, and don't bother with the note regarding erasing data since we will write a new flash anyway.

Your chip might be detected with 256k flash and say that "Target is readable"
All is good.

Select the target as 256k flash in the next window.
Click "next"

Choose the flash file to use, check the "verify" box, and proceed with the flash.

After about 1 minute, you should be good with your blade and happily hashing again.

You can send a few tips to my BTC or LTC address if this helped you.

The ferrite next to the USB plug is only for usb filtering, so your power supply doesn't matter.
It does matter for all the other ones next to the power plug.


I finally received my universal JTAG adapter, so I should be able to extract the firmware from the blade.

If I succeed, I'll post a link to it in this thread.

Only if you have a high quality ATX PSU and 18AWG power cable, (although that's what everyone should be using, not the crappy DC adapters.)

I do use an ATX PSU but I still ended up ordering a Ferrite Bead 100mhz@330ohm 1.5A of size 0805 from the bay to be safe:

http://www.ebay.co.uk/itm/181410119006

soon ill sell my gblade for EXCL

Thanks for your reply J4bberwock. Ill look at the guide when I have a moment. I unscrewed the Blade and have inspected the pcb's. Photss are below:

http://i59.tinypic.com/mwctbl.jpg

http://i57.tinypic.com/vzfo13.jpg

From looking at the pcb ls with a magnifying glass I can't see any other damaged components other than the ferrite bead at AFB26. Can anyone confirm from the pics if that appears to be the case? If so can someone provide me with the part number of the replacement or the 0ohm replacement as mentioned earlier. Also can I have part numbers for the 220f 35v cap and 39k resistor to mod other blades

Also any other recommended changes? More heatsinks, power connector mod, etc? Im using a 750w enermax psu and 18awg cables it has a cheap connector however, one of which melted so any recommendation on the plug to replace? Also is it worth changing the barrel socket to a better rather than screw terminals or directly soldering, etc?

The AFBxx and FBxx are ferrite beads. worst case, you can replace them with zero ohm resistors.
I bought 2 failed blades with the same trouble and managed to get them back to life.
The best way to solder  themthose parts is with an hot air station.
I posted a few pages ago a step by step guide to check what is failed on a blade.
Follow it, and post your results.
Obviously, you will need to replace the FB before testiong the usb part of the board.
You may also have to replace AME8805 chip.
If so, they are selling either on utsource or ebay for cheap, and I also have a few ones I can send for the price I paid them from utsource + shipping from France.

Worst thing that could have happened is the STM32 chip fried since we don't have the firmware for it.
I'll try to extract it, but not sure when.

when the STM32 chip is fried, you will get low voltage at the output of the AME8805 instead of 3.3v, and the capacitors next to the crystal will read either zero or infinite when using the ohm-meter. I can't remember which value it was.
Soldering the STM32 is not easy at all, i took me some time to get all pins correctly aligned.

If you got no reply within 2 days, just PM me, I'm quite busy at the moment.


And when/if you revive the blade, just lower the resistor from 43k to 39k, the extra power draw isn't worth it anymore.

Hi, I bought a Blade which is modified with 35v 220uf capacitor, 43k ohm resistor 0.1 +/-, 2 custom heat sinks attached on the rear of each power board, passive heatsinks on the all Mosfets and chokes.

I'm using a 750w PSU to power it as well as 2 stock unmodified Gridseed blades. I only had it running for a couple hours when the PSU shut down. I looked at the modified blade and there are scorch marks between the Barrel connector and USB port, the chip that visibly appears to be fried is the one between the USB and Barrel port labelled AFR26.

Pics of the blades below:

http://i59.tinypic.com/50i9dz.jpg
http://i60.tinypic.com/10f6ip5.jpg
http://i59.tinypic.com/2udxgcp.jpg
http://i58.tinypic.com/212sq68.jpg


As the mosfets and chokes, etc all have heatsinks on them I can't see what else is fried from a quick inspection. Would this have been caused by overclocking too much? I set the clocks between 1000mhz and 1038mhz which the seller said they run stable at.

I can do a bit of soldering so if anyone can guide me on where I can get the components I can probably fix it myself. Also do you recommend I change anything else that hasn't been modified to improve it, i.e barrel connector, heatsink, etc.

Also one of the barrel power connectors from the PSU has its plastic melted slightly but the connector itself looks fine and there are no burn marks on it.

Yes, they use the same parts as power MOSFET.
In fact, the power circuit of the blade is the same as the one from a pod.
And you can't improve HW errors unless you have better cooling and, depending on the frequency you want, higher voltage provided to the chips.

Thanks for the answer, I actually have a Gridseed 5-chip not a G-Blade, is it the same part?


No performance change, it just works again.