Hacking Hardware Picofly - a HWFLY switch modchip

[FreeLander]

@twins333 @FXDX

Thanks, guys. I did try it and it's ridiculously fast. I'll try it on a couple of more OLEDs before determining if it's better than the single pad.

 

[twins333]

@twins333 @FXDX

Thanks, guys. I did try it and it's ridiculously fast. I'll try it on a couple of more OLEDs before determining if it's better than the single pad.

Don't thank me, I haven't done anything to help the scene. the back of the board mosfet install was discovered/developed by @abal1000x, @deeps, @raksmey1231 and a few others. fxdx confirmed that it works quite well, I was around to help with info gathered from here.
Here is the back of the board schematic source: https://gbatemp.net/threads/picofly-a-hwfly-switch-modchip.622701/page-560#post-10196711

Also I can't see any difference in boot times between my v2 install and the oled done by fxdx. both boot really fast. are oleds notorious for slower boot times when modded?

 

[FreeLander]

Don't thank me, I haven't done anything to help the scene. the back of the board mosfet install was discovered/developed by @abal1000x, @deeps, @raksmey1231 and a few others. fxdx confirmed that it works quite well, I was around to help with info gathered from here.

Also I can't see any difference in boot times between my v2 install and the oled done by fxdx. both boot really fast. are oleds notorious for slower boot times when modded?

Not to my knowledge when doing the two APU caps, I've never waited for than 30 seconds for a glitch except in two OLEDs when I did the one-pad back mosfet. They boot somewhere between 45 sec to 2 minutes.

This one seems very promising though.

 

[FXDX]

That is pull down resistor, its not mandatory.
Without it the glitching still works.

That's right, it is. Thx

 

[revsgrow]

That is pull down resistor, its not mandatory.
Without it the glitching still works.

Take a photo on your works, if there are probable mistakes we might point it out.

Good Morning.

See images bellow.

Error code
*==

Post automatically merged: Jul 14, 2023

That is pull down resistor, its not mandatory.
Without it the glitching still works.

Take a photo on your works, if there are probable mistakes we might point it out.

Hello!
I forgot send a image of mosfet.

See image bellow.

Now the error changed.
Is ==* (long, long, small). CPU always reach BCT check. (No glitch reaction.

 

[RiotRetroGaming]

Picofly soldered on Oled with mosfet on the back pcb. I took the mosfet from the defective connection ribbon of one Hwfly. I used some thin copper wire to solder on the capacitor and loaded it with more tin. Excellent glitch time.
View attachment 383308

This is the sort of times I get with all my modded OLEDs I sell.

I have been covering my smallest soldered points with uv activated pcb resist however.
Maybe this is limiting interference?

 

[abal1000x]

Good Morning.

See images bellow.

Error code
*==

Post automatically merged: Jul 14, 2023

Hello!
I forgot send a image of mosfet.

See image bellow.

Now the error changed.
Is ==* (long, long, small). CPU always reach BCT check. (No glitch reaction.

From the picture, looks fine to me.

There are 3 possible cause:
1. The gate connection is not good enough. The glitch from rp2040-zero might not delivered to the mosfet Gate. Try to resolder the pin15 rp2040-zero and the mosfet G. And check the continuity between it, using multimeter.
2. The cmd line is not good enough. The previous error is about the CMD/CLK/Dat0. Pico will sniff the glitch status from CMD line. Maybe the cmd line problematic. Try to resolder the endpoint, then check the continuity. If necessary add the resistor to around 100 ohms. (Add another 47 ohm in series)
3. My hunch says that you use solder heat around 200-250 degrees C. Need a higher one when soldering the mosfet, especially the D part. if necessary put copper in between. Copper have a better conductivity than Tin (Sn or Pb). The mosfet purposes is to steal the power (the current) from ic max which goes to the cpu at a certain time. So need a good conductivity, lower resistance to do it.

 

[FreeLander]

Sorry, deleted.

 

[jkyoho]

a patched V1 that I think is in RCM-mode

Are you sure?

 

[Dee87]

From the picture, looks fine to me.

There are 3 possible cause:
1. The gate connection is not good enough. The glitch from rp2040-zero might not delivered to the mosfet Gate. Try to resolder the pin15 rp2040-zero and the mosfet G. And check the continuity between it, using multimeter.
2. The cmd line is not good enough. The previous error is about the CMD/CLK/Dat0. Pico will sniff the glitch status from CMD line. Maybe the cmd line problematic. Try to resolder the endpoint, then check the continuity. If necessary add the resistor to around 100 ohms. (Add another 47 ohm in series)
3. My hunch says that you use solder heat around 200-250 degrees C. Need a higher one when soldering the mosfet, especially the D part. if necessary put copper in between. Copper have a better conductivity than Tin (Sn or Pb). The mosfet purposes is to steal the power (the current) from ic max which goes to the cpu at a certain time. So need a good conductivity, lower resistance to do it.

So I say low melt is not good for MOSFET soldering?

 

[revsgrow]

From the picture, looks fine to me.

There are 3 possible cause:
1. The gate connection is not good enough. The glitch from rp2040-zero might not delivered to the mosfet Gate. Try to resolder the pin15 rp2040-zero and the mosfet G. And check the continuity between it, using multimeter.
2. The cmd line is not good enough. The previous error is about the CMD/CLK/Dat0. Pico will sniff the glitch status from CMD line. Maybe the cmd line problematic. Try to resolder the endpoint, then check the continuity. If necessary add the resistor to around 100 ohms. (Add another 47 ohm in series)
3. My hunch says that you use solder heat around 200-250 degrees C. Need a higher one when soldering the mosfet, especially the D part. if necessary put copper in between. Copper have a better conductivity than Tin (Sn or Pb). The mosfet purposes is to steal the power (the current) from ic max which goes to the cpu at a certain time. So need a good conductivity, lower resistance to do it.

Hello, I’m here again.

I will attach photos of the new installation.
I already advance that I was not successful in the glitch.

Can anyone tell me where I'm going wrong?

I used copper in the new installation of the mosfet.

Grateful for the attention.

Error code ==* (long, long, small)

 

[jkyoho]

Hello, I’m here again.

I will attach photos of the new installation.
I already advance that I was not successful in the glitch.

Can anyone tell me where I'm going wrong?

I used copper in the new installation of the mosfet.

Grateful for the attention.

Error code ==* (long, long, small)

https://gbatemp.net/threads/picofly-a-hwfly-switch-modchip.622701/post-10196711
wrong cap you soldered the mosfet on

 

[revsgrow]

https://gbatemp.net/threads/picofly-a-hwfly-switch-modchip.622701/post-10196711
wrong cap you soldered the mosfet on

View attachment 383375

Wooow,
actually soldering the wrong capacitor. But now I switched to the right capacitor and the same error code continued.
Will it never work? I am sad.

Error code ==*

See images bellow.

Thaaaanks!

 

[jkyoho]

I believe there should be a resistor there and its gone.

 

[revsgrow]

I believe there should be a resistor there and its gone.View attachment 383386

No, just the pads. There is no resistor in this location.

 

[jkyoho]

No, just the pads. There is no resistor in this location.

Look closely and you should see what you missing

 

[revsgrow]

Look closely and you should see what you missing

Sorry,
I took this image as a basis. That's why I said there wasn't.

I would like to know what the value of this resistor is and where I can find it. I have a ninteno switch lite card.

Can I remove this resistor from the Nintendo Switch Lite board?

 

[jkyoho]

Sorry,
I took this image as a basis. That's why I said there wasn't.

That's a HAC-CPU-XX v1 board, And you have a HAD-CPU-01 V2 board, clearly you have a bad reference sample.

Here is my v2 board screenshot

 

[Dee87]

Look closely and you should see what you missing

eagle eyes

 

[revsgrow]

That's a HAC-CPU-XX v1 board, And you have a HAD-CPU-01 V2 board, clearly you have a bad reference sample.

Here is my v2 board screenshot
View attachment 383393

Okay, so let's go.

Where to find this resistor to put it in place?
Is it because of him that the glitch is not working?