Hacking Hardware Picofly - a HWFLY switch modchip

[Viktorsilva]

Guys, does the "long long short" error only appear if there is a mosfet problem or could there be another reason? I tried the circuit with two and one mosfet, but I get this error every time.
(switch lite)

that´s the mosfet error. you can use a single Mosfet.I´ve done lots of Switch Lite, V1 V2 & Oled, and i use only single Mosfet. Never had to add 2 Mosfets to make it wok. Can you post some Pictures of your install ?

Post automatically merged: Jan 28, 2024

Before i use to install the flex cpu Mosfet, but since i run out of stock, i started to solder it directly to the capacitor and checked that i gives almost every time instant boot.

 

[linkref]

Hello,

I've modded an OLED switch with pico rp2040 board ready to solder, everything went well.
Everything is okay, but i tried like 100 boots for training, and 1 time every 20 boots the switch boots straight to OFW and error code is : short - short - long (**=)
If it happens, i just have to turn it off and turn it on and then it will glitch well.

Any ideas what to check ?

Thanks

 

[FreeLander]

Hello,

I've modded an OLED switch with pico rp2040 board ready to solder, everything went well.
Everything is okay, but i tried like 100 boots for training, and 1 time every 20 boots the switch boots straight to OFW and error code is : short - short - long (**=)
If it happens, i just have to turn it off and turn it on and then it will glitch well.

Any ideas what to check ?

Thanks

This is a dat0 error.. The reason you "sometimes" don't get a glitch is because of the dat0 adaptor. If you want a reliable constant glitch you need to either reball the emmc and solder a direct wire underneath it, or drill that point out. Both are challenging in their respective ways. I personally prefer and enoy reballing.

 

[abal1000x]

Guys, does the "long long short" error only appear if there is a mosfet problem or could there be another reason? I tried the circuit with two and one mosfet, but I get this error every time.
(switch lite)

Mosfet is the most probable one.

But of course theres another probability.
For example if the gate line goes to the rp2040 are problematic.
Or the rp2040 board itself is problematic.

Post automatically merged: Jan 29, 2024

halo apakah ada yang bisa membantu saya? saya ada kendala tidak bisa membuat emummc

Post automatically merged: Jan 28, 2024

hallo apakah ada yang bisa membantu saya? saya ada kendala tidak bisa creat emummc

Convert to english first using google translate or whatever.

From the picture it seems the 'slow mode' issue.
You might need to replace the cmd/dat0 resistor to around 100ohms.

And of course you could simply create the emummc to the microsd card and ignore that error.

 

[Jefix]

that´s the mosfet error. you can use a single Mosfet.I´ve done lots of Switch Lite, V1 V2 & Oled, and i use only single Mosfet. Never had to add 2 Mosfets to make it wok. Can you post some Pictures of your install ?

I first made a version with two mosfets, then decided to simplify and kept one. Now it looks like this.

Mosfet is the most probable one.

But of course theres another probability.
For example if the gate line goes to the rp2040 are problematic.
Or the rp2040 board itself is problematic.

Thank you for your response. I will order another rp2040 just in case.

 

[linkref]

This is a dat0 error.. The reason you "sometimes" don't get a glitch is because of the dat0 adaptor. If you want a reliable constant glitch you need to either reball the emmc and solder a direct wire underneath it, or drill that point out. Both are challenging in their respective ways. I personally prefer and enoy reballing.

I thought D0 was that error code : "=* D0 is not connected"
Are you sure it's a dat0 error ?

 

[Viktorsilva]

I first made a version with two mosfets, then decided to simplify and kept one. Now it looks like this.

Thank you for your response. I will order another rp2040 just in case.

Can you probe with a multimeter on diode mode if Source and Gate are bridged ? This would be a reason to faill the glitch on boot. I had a bunch of brand new IRH8342 Mosfet with S&G closed (bridged) and if i try to install them will have exactly the same problem (==*) . I know that is possible to open this bridge, but i didnt find the way yet.

 

[remlei]

Hello,

I've modded an OLED switch with pico rp2040 board ready to solder, everything went well.
Everything is okay, but i tried like 100 boots for training, and 1 time every 20 boots the switch boots straight to OFW and error code is : short - short - long (**=)
If it happens, i just have to turn it off and turn it on and then it will glitch well.

Any ideas what to check ?

Thanks

Also had the same issue in the past, its basically a rp2040 specific issue, changing the modchip to fpga based modchip like hwfly fixed it for me.

And nope its not a dat0 issue, because my switch already gone to dat0 adapter and now at present using a emmc interposer board for switch oled emmc. With rp2040 and that expensive install for the emmc, it still have that **= error. The installer said it maybe a samsung emmc specific issue. For bow with hwfly, no failed glitch at all.

 

[linkref]

Also had the same issue in the past, its basically a rp2040 specific issue, changing the modchip to fpga based modchip like hwfly fixed it for me.

And nope its not a dat0 issue, because my switch already gone to dat0 adapter and now at present using a emmc interposer board for switch oled emmc. With rp2040 and that expensive install for the emmc, it still have that **= error. The installer said it maybe a samsung emmc specific issue. For bow with hwfly, no failed glitch at all.

Ok so basicly no solution out of the box..
Is it okay to leave it like that ? I think it's okay to fail 1 boot on 30 boot.. ?

 

[Jefix]

Can you probe with a multimeter on diode mode if Source and Gate are bridged ? This would be a reason to faill the glitch on boot. I had a bunch of brand new IRH8342 Mosfet with S&G closed (bridged) and if i try to install them will have exactly the same problem (==*) . I know that is possible to open this bridge, but i didnt find the way yet.

I got a voltage drop of 0.65 volts between Source (red probe) and Gate (black probe).

 

[Bones00]

Hi all,

I'm new to this and to these forums but I'm hoping someone can help me... I have the hwfly installed on my V1 switch and everything seems to work fine. When it's disassembled it boots into NOSD just fine, only blue lights on the chip. But as soon as I put the metal backplate on, or put any pressure on the chip holding it in place, I get error code:

Long Long Short (or ==* CPU always reach BCT)

It consistently works just fine when the thing is disassembled and the chip is just kinda hanging there, so I do believe the soldering work and the chip itself are totally fine. But it will consistently give the code and the switch does a regular boot as soon as there is any pressure (even just using my finger to hold it in position causes this).

I put plenty of kapton tape on the metal surfaces around the actual chip (top and bottom) to hopefully prevent a short but I'm thinking maybe that still is what the issue is?

Again, I am a bit of a noob but hoping to learn through experience with this. If this is the wrong thread to post this question in, I apologize.

 

[superxoi]

Hi all,

I'm new to this and to these forums but I'm hoping someone can help me... I have the hwfly installed on my V1 switch and everything seems to work fine. When it's disassembled it boots into NOSD just fine, only blue lights on the chip. But as soon as I put the metal backplate on, or put any pressure on the chip holding it in place, I get error code:

Long Long Short (or ==* CPU always reach BCT)

It consistently works just fine when the thing is disassembled and the chip is just kinda hanging there, so I do believe the soldering work and the chip itself are totally fine. But it will consistently give the code and the switch does a regular boot as soon as there is any pressure (even just using my finger to hold it in position causes this).

I put plenty of kapton tape on the metal surfaces around the actual chip (top and bottom) to hopefully prevent a short but I'm thinking maybe that still is what the issue is?

Again, I am a bit of a noob but hoping to learn through experience with this. If this is the wrong thread to post this question in, I apologize.

your solder at the cpu are not good when you close your shield the flex got loose. redo it.

 

[Bones00]

your solder at the cpu are not good when you close your shield the flex got loose. redo it.

Ah, damn. I appreciate the quick response, I'll give it another go and see if it sticks. Thanks!

Just to verify, even just light pressure on the chip hanging off to the side of the CPU solder points would affect those? It seems like when I apply pressure to the chip and get the error there is no actual pressure being applied where the solder work is?

I'm sure you're right, just want to make sure I'm explaining myself clearly...

 

[rs1n]

Ah, damn. I appreciate the quick response, I'll give it another go and see if it sticks. Thanks!

Just to verify, even just light pressure on the chip hanging off to the side of the CPU solder points would affect those? It seems like when I apply pressure to the chip and get the error there is no actual pressure being applied where the solder work is?

I'm sure you're right, just want to make sure I'm explaining myself clearly...

I would get some clear plexiglass or something similar that you can use to simulate putting the shell back on. This way, you can also see the light to see any error codes. I have two guesses: 1) something is being shorted with the case one, but that doesn't short when the case is removed (very likely a solder joint if you used any wires for your modding, or perhaps too much solder at a joint which shorts with the metal cover, or parts of the chip itself coming into contact with the heatshield) or 2) your chip's flex cable has a break in one of the traces within the flex itself, and this works fine when the case is opened but is flexed to become disconnected when the case puts pressure on the chip/cable.

 

[Bones00]

I would get some clear plexiglass or something similar that you can use to simulate putting the shell back on. This way, you can also see the light to see any error codes. I have two guesses: 1) something is being shorted with the case one, but that doesn't short when the case is removed (very likely a solder joint if you used any wires for your modding, or perhaps too much solder at a joint which shorts with the metal cover, or parts of the chip itself coming into contact with the heatshield) or 2) your chip's flex cable has a break in one of the traces within the flex itself, and this works fine when the case is opened but is flexed to become disconnected when the case puts pressure on the chip/cable.

A short was my guess as well, I can see the error code through the bottom of the case when it's put back together, it's the same code I get when I apply gentle pressure.

Long Long Short (or ==* CPU always reach BCT)

The hwfly chip I installed only has the four solder points on the flex cable, two onto the CPU top/bottom and two grounds (I believe) to the metal edge. Again I am a beginner and I know this is a more advanced solder job but this chip seemed easier than the PIcofly so I went this route. I would not be surprised if I used too much solder and it's shorting at the flex cable solder points. I'm going to do another teardown again today and double-check. Would love if I can avoid redoing the solder, however, as that seems to be the most high-risk aspect of the mod.

 

[Doctorbeefy]

Hi all,

I'm new to this and to these forums but I'm hoping someone can help me... I have the hwfly installed on my V1 switch and everything seems to work fine. When it's disassembled it boots into NOSD just fine, only blue lights on the chip. But as soon as I put the metal backplate on, or put any pressure on the chip holding it in place, I get error code:

Long Long Short (or ==* CPU always reach BCT)

It consistently works just fine when the thing is disassembled and the chip is just kinda hanging there, so I do believe the soldering work and the chip itself are totally fine. But it will consistently give the code and the switch does a regular boot as soon as there is any pressure (even just using my finger to hold it in position causes this).

I put plenty of kapton tape on the metal surfaces around the actual chip (top and bottom) to hopefully prevent a short but I'm thinking maybe that still is what the issue is?

Again, I am a bit of a noob but hoping to learn through experience with this. If this is the wrong thread to post this question in, I apologize.

Send me a screenshot of your setup under the shielding. I bet something is touching the metal causing a short, I have seen this happen with those flex cables people leaving them on top of the metal rim instead of putting it under.

 

[Bones00]

I'll send you a PM now. I think maybe I have excess solder that may be causing a possible short.

 

[snaker]

Hello,

I've modded an OLED switch with pico rp2040 board ready to solder, everything went well.
Everything is okay, but i tried like 100 boots for training, and 1 time every 20 boots the switch boots straight to OFW and error code is : short - short - long (**=)
If it happens, i just have to turn it off and turn it on and then it will glitch well.

Any ideas what to check ?

Thanks

This is happening because the flex cables on those clones are very thin and the signals dont pass correct when glitching. If you replace the flex cable (not the cpu flex the other one) with 36awg cable you will never have this problem again.

Dont get me wrong most of the time these flexes will work but some times they wont. Ask me how i know...

 

[abal1000x]

A short was my guess as well, I can see the error code through the bottom of the case when it's put back together, it's the same code I get when I apply gentle pressure.

Long Long Short (or ==* CPU always reach BCT)

The hwfly chip I installed only has the four solder points on the flex cable, two onto the CPU top/bottom and two grounds (I believe) to the metal edge. Again I am a beginner and I know this is a more advanced solder job but this chip seemed easier than the PIcofly so I went this route. I would not be surprised if I used too much solder and it's shorting at the flex cable solder points. I'm going to do another teardown again today and double-check. Would love if I can avoid redoing the solder, however, as that seems to be the most high-risk aspect of the mod.

Maybe your solder is too much its touching the shield on top of it.

 

[Bones00]

Maybe your solder is too much its touching the shield on top of it.

Looking at it again this morning, I think that this is likely the issue. I've noticed the error code happens when the shield is full connected around the CPU and solder points, when it is loosened it works fine. I'll try removing some excess before redo-ing the whole thing.