Hacking Hardware Picofly - a HWFLY switch modchip

[BlueBeans]

Switch Lite points are tiny as hell compared to V2 or V1.

Edit: Just figured out that SDA and SCL are on the other side of this board.

Let me know how your install goes. The switch lite has been giving me a hell of a time.

 

[abal1000x]

Let me know how your install goes. The switch lite has been giving me a hell of a time.

Lite perfectly fit if you place the pico on top of the emmc.

https://gbatemp.net/threads/picofly-a-hwfly-switch-modchip.622701/page-415#post-10167696

 

[bilalhassan341]

Let me know how your install goes. The switch lite has been giving me a hell of a time.

I have installed total 4 picofly's on v2, each of them boot into no sd on first try. I usually install picofly on v2 on avg 50min. But on this one I can tell for sure it is going to take a day or two.

Post automatically merged: Jun 9, 2023

Lite perfectly fit if you place the pico on top of the emmc.

https://gbatemp.net/threads/picofly-a-hwfly-switch-modchip.622701/page-415#post-10167696

I'm going to follow a og method (Metal Shield cut). So, I can enjoy a blue light in the dark. Like a rgb vibes, just b .

 

[BlueBeans]

I have installed total 4 picofly's on v2, each of them boot into no sd on first try. I usually install picofly on v2 on avg 50min. But on this one I can tell for sure it is going to take a day or two.

Post automatically merged: Jun 9, 2023

I'm going to follow a og method (Metal Shield cut). So, I can enjoy a blue light in the dark. Like a rgb vibes, just b .

Yeah, I've done about the same in terms of V2 and I did them with the mosfet. Booted the first time every time. I finally do a lite with the cable which I thought would be easy and its given me a run for my money

Post automatically merged: Jun 9, 2023

I have installed total 4 picofly's on v2, each of them boot into no sd on first try. I usually install picofly on v2 on avg 50min. But on this one I can tell for sure it is going to take a day or two.

Post automatically merged: Jun 9, 2023

I'm going to follow a og method (Metal Shield cut). So, I can enjoy a blue light in the dark. Like a rgb vibes, just b .

I didn't even think about cutting the shield guess it's good i saw that before putting everything together. What do you do for the V2/V1? do you cut the shield there as well?

 

[QuiTim]

Hi,
I am getting screen flickering, slow performance and black lines when I boot to the CFW or OFW on a switch oled. Overall its glitchy and slow, but works fine in Hekate.

I installed the picofly and when I was installing it I noticed that, with the multimeter in Diode Mode, I was getting ~0.355v from the CPU. I am using a V2 ribbon cable from the Hwfly modchip. Could this be the issue?

Do you have pictures of installation?

 

[bilalhassan341]

Yeah, I've done about the same in terms of V2 and I did them with the mosfet. Booted the first time every time. I finally do a lite with the cable which I thought would be easy and its given me a run for my money

Post automatically merged: Jun 9, 2023

I didn't even think about cutting the shield guess it's good i saw that before putting everything together. What do you do for the V2/V1? do you cut the shield there as well?

I always cut the shield incase if there is error, or yellow light. With the shield cut I can see blue from the vent clearly. Btw it depend on you which method you want to follow. Before opening the switch shield I can tell that emmc is the best place by looking at the metal shield.

 

[BoomBoomShroom]

Do you have pictures of installation?

Sorry for double posting.

I don't sadly and I am at work. I was just wondering what I should be checking.

This is my 7th installation and this never occurred. I can post pictures later in the day

 

[QuiTim]

Sorry for double posting.

I don't sadly and I am at work. I was just wondering what I should be checking.

This is my 7th installation and this never occurred. I can post pictures later in the day

I had a similar issue with an oled (but with hwfly).
After 2 days of banging my head on the wall I fixed it by changing 2 things:
- 1. I swaped the dat0 adapter for a old "corner" model (didn't have any other at hand at the time). It turns out the dat0 was giving good diode value but it was one of the shitty ones (could not confirm that it was shorting dat0 and dat1 but I had my doubts).
- 2. I ditched the flex cable (the one that connects clk, cmd, rst) because that flex cable was most likely causing interference on one of the lines so I wired everything manualy with 34awg and the issue fixed.

In your case, (although I didn't see the installation) I would first try to clean any residue flux on solder points, then change the wire, and reroute it as well (clean again).
If that does not work, i would reseat the dat0 (in case you did not set a permanent point by reballing)

 

[abal1000x]

Guys do you know the logical explanation why lite/v1/v2 lcd usually break after 4-5 years? While handphone lcd seems work for eternity.

 

[provato]

hey all! I've got a problem with my second picofly installation on a mariko (v2).
I've done one very successfully.
one of the drain wires of the mosfet to the furthest capacitor beside the nividia chip detached and took the capacitor's pad with it! (yeah... ripped pad).
As you can see in the picture I attached, only the capacitor's ground pad remains, and I've carefully scraped all around the ripped pad to find a possible continuation of that ripped pad. Turns out there is ground all around this pad (checked with dmm)

I tried soldering the wire to the capacitor's ground pad but I get two long yellow flashes and one short yellow flash (==*) which means check mosfet I've read.

My question is, what can I do now? Is there any alternative solder point for one of the mosfets' drain?

 

[abal1000x]

hey all! I've got a problem with my second picofly installation on a mariko (v2).
I've done one very successfully.
one of the drain wires of the mosfet to the furthest capacitor beside the nividia chip detached and took the capacitor's pad with it! (yeah... ripped pad).
As you can see in the picture I attached, only the capacitor's ground pad remains, and I've carefully scraped all around the ripped pad to find a possible continuation of that ripped pad. Turns out there is ground all around this pad (checked with dmm)

I tried soldering the wire to the capacitor's ground pad but I get two long yellow flashes and one short yellow flash (==*) which means check mosfet I've read.

My question is, what can I do now? Is there any alternative solder point for one of the mosfets' drain?

Basically you only need one. Those two cap are connected one to another.

There still capacitor on the right. The problem is you need a bigger cable, since there will be high current flows in there. I make a U-Turn to make it twice the current could flows.

https://gbatemp.net/threads/picofly-a-hwfly-switch-modchip.622701/post-10179786

 

[BlueBeans]

Basically you only need one. Those two cap are connected one to another.

There still capacitor on the right. The problem is you need a bigger cable, since there will be high current flows in there. I make a U-Turn to make it twice the current could flows.

https://gbatemp.net/threads/picofly-a-hwfly-switch-modchip.622701/post-10179786

Do you think that a single mosfet if more effective than the cable or a double mosfet?

 

[rehius]

My question is, what can I do now?

Add SCL/SDA wires to lower the glitch requierements so single cap connection could work

 

[QuiTim]

I finally got this beauty today and it looks awesome.
Rev 1.1 has 47ohm resistors in place. Just one more week to go until I get the new Zelda Oled and a chance to try this out. Maybe I even build up the courage to do a perm. dat0 by then

 

[darviral]

2 NS V2 Picofly w/ single mosfet & 2 Mosfet Done in 1hr

 

[abal1000x]

Do you think that a single mosfet if more effective than the cable or a double mosfet?

Its not about effective ness. Its about logical think.

Both cap actually connected. VDD connected to each other and GND connected to each other. Something like this (pardon me for the bad sketch)

Basically the mosfet is used as a 'switch' to steal the power that goes to the soc. As long as the mosfet could hold the current flows there, then adding another mosfet to the same point is useless. Its just the same as adding another cable to split the current, so the maximum current flows is twice than before.

IRF8342 could hold 8.5A, i assume its continuous. So if its only short time, it could hold more. If ship happened, then the mosfet will burned. Since it is not, then the mosfet could hold the high current flows to DS.

 

[BlueBeans]

I finally got this beauty today and it looks awesome.
Rev 1.1 has 47ohm resistors in place. Just one more week to go until I get the new Zelda Oled and a chance to try this out. Maybe I even build up a courage to do a perm. dat0 until then

where did you get those?!?!

 

[provato]

Basically you only need one. Those two cap are connected one to another.

There still capacitor on the right. The problem is you need a bigger cable, since there will be high current flows in there. I make a U-Turn to make it twice the current could flows.

https://gbatemp.net/threads/picofly-a-hwfly-switch-modchip.622701/post-10179786

now I get =** (=** eMMC init failure during glitch process) and it doesn't even go into OFW (black screen). Have I destroyed the switch?

 

[abal1000x]

where did you get those?!?!

In aliexpress they already sold it again. I have checked it and its correct 47 ohms. So they revised it all last week

 

[BlueBeans]

Its not about effective ness. Its about logical think.

Both cap actually connected. VDD connected to each other and GND connected to each other. Something like this (pardon me for the bad sketch)
View attachment 376943

Basically the mosfet is used as a 'switch' to steal the power that goes to the soc. As long as the mosfet could hold the current flows there, then adding another mosfet to the same point is useless. Its just the same as adding another cable to split the current, so the maximum current flows is twice than before.

IRF8342 could hold 8.5A, i assume its continuous. So if its only short time, it could hold more. If ship happened, then the mosfet will burned. Since it is not, then the mosfet could hold the high current flows to DS.

ok ill give it a try with a single mosfet. can you post a pic of your single mosfet install?