Hacking Hardware Picofly - a HWFLY switch modchip

[jkyoho]

Curious if anyone tried Vgs=2.5v/10v(max) mosfet like AON7524 or IRLHM630. In theory Vgs switch on Id as early as 2.5v instead of regular 8342 mosfet at 4.5v

 

[abal1000x]

nevers test this config, only

I see. Noted.

Curious if anyone tried Vgs=2.5v/10v(max) mosfet like AON7524. In theory Vgs switch on Id as early as 2.5v instead of regular 8342 mosfet at 4.5v

Thats interesting.
The gate switch happened on as small as 1.5V, with a RDS as small as 6mohm.
Based on the datasheet, everything is okay, should work as charm.
The price is 25% more than the IRF8342

 

[roxzii]

Like, solder debris or particles leaved by soldering iron can get stuck under apu or dram. Which can cause short on these components.

Is there any payload that I could use to test the DRAM's? Like the UMS-Loader can be used to "test" iRAM (and therefore the APU, kind of).

 

[jkyoho]

Is there any payload that I could use to test the DRAM's? Like the UMS-Loader can be used to "test" iRAM (and therefore the APU, kind of).

I dont think so, have you tried mounting GPP from ums-loader on PC with DiskGenius? My OLED gpp mount looks like this. Just in case your Blue screen is caused from courted eMMC partitions by DAT0 adapter.(though Hekate should still load even emmc bad IMO)

 

[Takezo-San]

Here you go, Now i just realize that i never directly posted my mosfet solder until now.. lol.
It works everytime, no drama whatsoever, at least from mosfet side.

Amazing, also does it matter too much if you just used that one cap with one wire for your drain?

 

[112c]

Im having a similar issue to @BlueBeans where it time's out after trying to glitch. All joints look clean but nothing happens. One thing I have seen is it will flash white once then continue to glitch. Any ideas?

 

[linuxares]

Is it possible to get the chip to log information that you can get out from the USB? Even if the led say a lot, a lot of people seem to have issues that doesn't correspond with the led.

 

[cgtchy0412]

Amazing, also does it matter too much if you just used that one cap with one wire for your drain?

I think shouldnt matter. I dont know about oled yet but i suppose it would be the same.
I think this method is closer to that newly backside pcb caps install, so if there is someone who failed to glitch with that then the only difference is it's solder point location.

 

[FreeLander]

@abal1000x Give this man a Nobel prize. Great discovery, it saved me plenty of time. Worked like a charm on OLED, first try. Thank you for sharing.

 

[Takezo-San]

@abal1000x Give this man a Nobel prize. Great discovery, it saved me plenty of time. Worked like a charm on OLED, first try. Thank you for sharing.

Man. This is a game changer. I'm hating opening the soc for cleaning and soldering. Will have to attempt this. Anyone tried this with a pull down resistor too?

@lightninjay whatdya think of this new method so far. Has @rehius any comments too?

 

[FreeLander]

Man. This is a game changer. I'm hating opening the soc for cleaning and soldering. Will have to attempt this. Anyone tried this with a pull down resistor too?

@lightninjay whatdya think of this new method so far. Has @rehius any comments too?

I spend 30-45 min from opening the shield to soldering two MOSFETs to the APU. This took me 5 minutes. It is a breakthrough!

 

[Takezo-San]

I spend 30-45 min from opening the shield to soldering two MOSFETs to the APU. This took me 5 minutes. It is a breakthrough!

 

[linuxares]

can someone give me the tldr on that new mosfet method?

 

[Dee87]

can someone give me the tldr on that new mosfet method?

well there is a cap on the back side of the switches u can solder a mosfet to, there are also 2 point where u can solder one mosfet to so u dont need to open tha apu shield and solder directly to the apu caps

 

[lightninjay]

can someone give me the tldr on that new mosfet method?

Proof Of Concept (POC)

This is DANGEROUS, still in a initial state. Don't replicate this experiment, without understanding the risk.

Glitching without soldering mosfet to the SoC nvidia but to the VDD_1V lines from the PMIC Max77620

I use a broken lcd V1 board, as experimentation.

View attachment 380692

Here is the IRFHS8342 soldered to the cap.
View attachment 380689

And here is the multiple time trial of glitching (power on and off).
View attachment 380690

The glitching is very fast, almost instantly.

Here is @abal1000x's original post

 

[linuxares]

Damn nice! I hope it's a good method. I hope someone manage to figure out or make a tool to make it easier with the dat0

 

[Dee87]

Damn nice! I hope it's a good method. I hope someone manage to figure out or make a tool to make it easier with the dat0

Dat0 just reball it , its not that hard

 

[linuxares]

Dat0 just reball it , its not that hard

Not for someone with experience
I'm just a fan in general of easier and more effective solutions.

 

[Dee87]

Not for someone with experience
I'm just a fan in general of easier and more effective solutions.

im mean i havent ever done reballing till i did my oled and it worked on the first run.

but yeah i know what u mean, on the other hand a there is no better way then a permanent connection

 

[linuxares]

of course not. Permanent is best. I sadly lack the knowledge, space and tools to do anything of this.