Topic: GRIDSEED G-BLADE Overclocking 7Mh/s, improvements and repair - page 14. (Read 74041 times)

What voltage/resistor combo were you running when this happened? Did you have active cooling on both sides of the ferrite bead? The 470uF 25v replaced the 220uF 16v?

Using a larger size cap causes somewhat higher current draw during it's charge phase, and may exacerbate the problem. If the charge phase is the same length of time and the capacity is double, it will pull twice as much current to charge it in the same period of time. You might perhaps be better off running (2) 110uF caps in parallel. Same capacitance, half current draw per cap & twice as much surface area for heat dissipation. That should cool them down considerably.

Hi,

the original ones are 0603 size.
You can use axial ones since the pads are "big", or 0805 size that is easier to solder than 0603 with a fine tip iron.
for 0805, I recommend using the alternate resistor location (that was meant for a parallel capacitor).
You can find this location on one of my first pictures, It's just over the original resistor on the board.

Replacing the power connector is a good idea too.


Last update, replacing ferrite with resistors might not be such a good idea.
On one of my blades, the power mosfet has shorted (as usual), but there were resistors.

The 3 ones for the +12V are blown, as is my 470uF 25V capacitor, and probably other components, since even after replacing the fet, resistors and capacitor, the blade isn't hashing. I'm currently tracking the damages and will report if it happens to someone else.

SMD Ferrite permanently damages the PCB when blowing becaue they are simply melting and they also melt the PCB board/tracks under them.

So for now, I'll try axial ferrite and/or fuses to protect the board.
Axial ferrite won't melt the tracks.

I'm also actively working on a replacement powerboard for the blade that could be piggybacked either to improve reliability, or easily repair a blade with damages to the power board.
I should come with a first prototype in a week or two, depending on how fast I can have all the last components I'm waiting for, and If I can etch the PCB myself or if I need to rely on a third party.

could someone tell me exactly what resistor i need to buy to replace the 33k one, i know it`s a 39.9k but there is so many variants that i wan`t to make sure i get the right one. also could i use a axiel one instead of a smd, i would connect it with 24agw solid wire.

thanks

Yeah, thanks for sharing, buddy....

WOLFEY2014

No problem, glad to help

Thanks for sharing, buddy...

ZiG

Ok than I am just lucky because I get it

regarding the sample order, I tried, and never ever got a single answer from Upi. And I believe I was polite.

There is no direct replacement pin to pin compatible for up1509 unless some other UPI products will work.

You can still try to get one from a scrap pod, maybe even the small usb with one chip can carry the up1509 and you should be able to get them for almost free, I hope.

Lack of availability of the UP1509 is why I'm going the route of a custom beefier power board. using easier to source buck controllers, either TPS40090PW (it has legs, easier to check for good soldering) or TPS40140.
After registering on TI's site, I was able to order free samples delivered in 2 days by FEDEX.

I'm currently working on a first proto board for 8 GC3355 chips I got from failed pods.

What is it with you people?
I write to company and asked them polite if they can send me a datasheet and they did it with no question why.
All you need is search for manufacturer and ask them.

Jesus, do you know how many people have been looking for this datasheet all over the internet? Where did you find it? This part is used on GPUs as well, and that community has been looking for it too. As I understand it, the part is covered by an NDA, and is part of the reason the datasheet has been hard to find.

Back up link to Datasheet

Also: http://www.upi-semi.com/sales/sales.aspx to purchase a replacement. Order a sample, or contact one of the companies somewhere in Asia. I'm sure at least one of them will do business with you.

Try to find replacaing chip for this one.
Here is datasheet.
http://filesave.me/file/39234/uP1509-DS-F0000-pdf.html

He's not there this controller.
Any other ideas?
hire: http://www.oemstrade.com/search/UP1509P/

Everything is fine until it's not. The factory need only keep it working for as long as the warranty period lasts.


That is one of their functions, yes. Switching pools, temporary disconnections, even coin switches might cause current surges. Eventually, that is going to kill some parts. With the incredibly rapid pace of profitability, I imagine these will need to be running a long time to make back their cost. That is a consideration for me, anyway.

you can find it at farnell or reichelt

I did a detailed diagnosis of my device. All power resistors are healthy.
The chip that manages power is probably blown.
It is: UP1509P
Please tell me the site where can I buy this chip and it is guaranteed quality.

Thank you in advance.
Trifon

I assume that ferrites are there because of high current and when connect it it gets a big current shock which is dumped by those ferrites, 0 ohm resistors can replace that.
But I think there are not 0 ohms but in real some 0,5 or 0,3 ohm

Not to disagree at all, though familiar with ferrite cores had not even heard of ferrite beads before modifying the 5 chip miners, but will just point out that some of the later blades did ship with 0 ohm resistors in place of the anal beads.

I soldered some wire across where the old ferrites where located and no issues so far.
Just waiting for the smaller heat sink for the FET's to add before adjusting the 10 turn
50k pot up a bit. Have a thermal epoxied heat sink on the back of the board also and
added a 220uF 35vdc capacitor and raised it up off the board with its tabs for better
cooling.

What was the issue with bridging the old ferrite locations?

There was reported some troubles bridging ferrites so I think I will put more powerfull one

I've had a soldering iron in my hand since before I was of a reasonable age to be using one. That said, I grew up working on electronics long before SMD were available, let alone mainstream. Back then, when you bought a 'computer' it was more like buying an unassembled arduino kit now. You built the whole thing, soldering your components by hand, then wrote your own programs in assembly or copied it out of some magazine.

My point is, I have plenty of experience working on circuit boards, just not with SMD components. I'm quite sure it isn't beyond me. The time I was speaking about with the SMD I was working on was a decade or more ago. Back then I couldn't have purchased new solder tips, let alone a rework station even if I had known they existed. Most jobs, including this one are tremendously easier with the proper tools -- and less caffeine.


You might consider adding a ferrite core on the supply wires if you opt with using resistors on the board. Just bridging with solder could add a considerable amount of inductance, and causing EM noise problems down the road, and I believe that to be a bad idea. This may not present as a failed blade, and may instead just limit your ability to ramp up the proc frequency because of HW errors due to unstable supply.