Hacking Hardware Picofly - a HWFLY switch modchip

[Takezo-San]

Confirm that a single mosfet from backside mainboard just instantly glitch the sw oled, how sweet!!!

Wow that's fast. Will definitely try this way next.

 

[Dee87]

As far as I've read since, the bad one. This pack. But I did that sthetix trick of cutting that little lobe to the side.

Also final result. A little crooked for 2 reasons, it wanted to go in deeper like that and I read it was better for it to be a little crooked not to make contact to dat1.

okay well then i would guess u couldve have done excatly what u didnt want to do (as far as ive read this can cause a blue screen not sure about that)
but i know for sure if u do that then ull corrupt ur boot0/1
there is no real way as far as i know to check if is Bridging
i Think @abal1000x showed that if u also dont cute the circle that goes to dat0 pin u will defintly bridge dat0/1 so i would start there.

 

[roxzii]

okay well then i would guess u couldve have done excatly what u didnt want to do (as far as ive read this can cause a blue screen not sure about that)
but i know for sure if u do that then ull corrupt ur boot0/1
there is no real way as far as i know to check if is Bridging
i Think @abal1000x showed that if u also dont cute the circle that goes to dat0 pin u will defintly bridge dat0/1 so i would start there.

I screwed up both SP1 and SP2 caps on APU, so no payloads, but my objetive was to try and mount GPP, boot0/1 through ums-loader, the only payload that ever worked, on windows to see if it worked to rule it out. But now with a seemingly dead brick and all the trouble I've had to rob caps from the pi and fail at putting them in the APU not sure if it's worth it. Besides, even if I could solder at least one cap back I doubt it would help bring it from the dead after the possible short.

Gotta check those caps around the chips @cowboy619 told me and see if there really was a short and it's dead.

 

[Dee87]

I screwed up both SP1 and SP2 caps on APU, so no payloads, but my objetive was to try and mount GPP through ums-loader, the only payload that ever worked, on windows to see if it worked to rule it out. But now with a seemingly dead brick and all the trouble I've had to rob caps from the pi and fail at putting them in the APU not sure if it's worth it. Besides, even if I could solder at least one cap back I doubt it would help bring it from the dead after the possible short.

Gotta check those caps around the chips a @cowboy619 told me and see if there really was a short and it's dead.

ah okay , well then do that first, u can replace them with some from the rp2040 there are 4 of them un needed on there , where they sit are explained in the guide :-)
check that first and then report back

 

[chronoss]

Actually i've change the resistor combination for experiment like hundreds time, and its never stopped the glitch, the nosd always shows.

I have post part of the experiment data on my previous comment.
https://gbatemp.net/threads/picofly-a-hwfly-switch-modchip.622701/post-10192087

The device i use is lite and v1. I heavily experiment on the resistor.

I am pretty sure that the error occured on your case is not because of the resistor.
But who knows maybe in v2 or oled could be a problem.

I change the resistor variantly from 0 ohms to 1000 ohms, all working.

Victory with this config :

CMD = 94ohms
CLK = 47ohms
Dat0 = 94ohms

v1 console patched

It work well with the last atmo !!! Thanks for your help... and @cgtchy0412 too

 

[roxzii]

ah okay , well then do that first, u can replace them with some from the rp2040 there are 4 of them un needed on there , where they sit are explained in the guide :-)
check that first and then report back

Yeah, already managed to screw 2 of them up, I'm lucky I have another pi. I just don't think I have the right iron point for the job. I got that basic round Pinecil one and It's giving me a lot of trouble to solder them there.

By the way, is 300ºC too much to be working around the APU? I think with 250ºC the pads take too long to melt, maybe I'm doing something wrong.

 

[Dee87]

Yeah, already managed to screw 2 of them up, I'm lucky I have another pi. I just don't think I have the right iron point for the job. I got that basic round Pinecil one and It's giving me a lot of trouble to solder them there.

By the way, is 300ºC too much to be working around the APU?

the best way to replace them is with a hot air station.

for a short time its okay but not to long.

out of experience there is that die around the chip where u add the thermal past if that turns darker/Brown then ur apu has been over heated to long.

i would suggest getting thinner tips if its hard for u , u can also grind them down (what i cannot recommend) cause they dont last very long after that but they do work then

 

[roxzii]

the best way to replace them is with a hot air station.

for a short time its okay but not to long.

out of experience there is that die around the chip where u add the thermal past if that turns darker/Brown then ur apu has been over heated to long.

i would suggest getting thinner tips if its hard for u , u can also grind them down (what i cannot recommend) cause they dont last very long after that but they do work then

Okay, thank you so much for your help. I'll try all this and report back!

 

[Dee87]

no problem

 

[cowboy619]

I screwed up both SP1 and SP2 caps on APU, so no payloads, but my objetive was to try and mount GPP, boot0/1 through ums-loader, the only payload that ever worked, on windows to see if it worked to rule it out. But now with a seemingly dead brick and all the trouble I've had to rob caps from the pi and fail at putting them in the APU not sure if it's worth it. Besides, even if I could solder at least one cap back I doubt it would help bring it from the dead after the possible short.

Gotta check those caps around the chips @cowboy619 told me and see if there really was a short and it's dead.

Yes Ammeter to check if there is power draw - also check your lcd connector for damaged pins

 

[abal1000x]

Yeah, already managed to screw 2 of them up, I'm lucky I have another pi. I just don't think I have the right iron point for the job. I got that basic round Pinecil one and It's giving me a lot of trouble to solder them there.

By the way, is 300ºC too much to be working around the APU? I think with 250ºC the pads take too long to melt, maybe I'm doing something wrong.

I use 350-400 degC but my focus like 100%, cause in my heart i limit my solder maximum 5 seconds, and rest for 15seconds to cool down. And only could do it like 3-5x, more than that, the caps pad usually ripped off, cause of the heat.

 

[frozenboy]

that's 2.2uF cap. way too much value, where I'm using 0.1uF. yours are 22x capacity which takes longer time charge&discharge so LED sequence seems weird.

I ve tried 101, and 104 but not work, it seem ceramic cap is not ok

 

[FreeLander]

The new back single MOSFET discovery was done on these two OLEDs, you be the judge.

 

[jkyoho]

The new back single MOSFET discovery was done on these two OLEDs, you be the judge.

emmc type and rp2040 resistor set?

 

[FreeLander]

emmc type and rp2040 resistor set?

both SK Hynix.
All 47s.

 

[deeps]

The new back single MOSFET discovery was done on these two OLEDs, you be the judge.

they're not broken, so it looks good to me!

 

[FreeLander]

they're not broken, so it looks good to me!

“Looks good” is an understatement

 

[Takezo-San]

The new back single MOSFET discovery was done on these two OLEDs, you be the judge.

Did you lut any pull down resistors on?

 

[FreeLander]

Did you lut any pull down resistors on?

None.

 

[jkyoho]

both SK Hynix.
All 47s.

nice, Iam using 100 ohm on data0 having slow glitch (4-5s most, rarely 2-3sec).Will switch back and see