Hacking Hardware Picofly - a HWFLY switch modchip

[dibas]

Have a look at this post? https://x.com/sthetixofficial/status/1543420979426500608

Thanks a lot, I had not seen that post before while searching for similar problems. I will check the IC. The connector looks fine to me. I still wonder whether the cause might be a damaged trace close to the APU shield though, because I did not really touch anything around the Joycon handshake IC while installing the modchip.

 

[taxco]

Have a look at this post? https://x.com/sthetixofficial/status/1543420979426500608

REmove bat, spray alcohol ( alot ) on ribbon and connector.
Let it sit for a while and then do it again after blow alcohol off and trie it.

 

[Nephiel]

The only damage I have been able to identify so far is a scratch on the pcb next to where I tried to remove the cpu shield (see the image below). To me it looks like the damaged location should not cause any problems as the trace seems to be next to it. I don't even know whether that area of traces is connected to the left joy-con.

I would be glad for any help or tips on how to further debug this issue. Because of the issue the console is basically unusable as a handheld right now.

Thanks a lot in advance!

Maybe this thread can help?

 

[dibas]

REmove bat, spray alcohol ( alot ) on ribbon and connector.
Let it sit for a while and then do it again after blow alcohol off and trie it.

I did that today, cleaned up everything, unfortunately it didn't solve the problem.

This is how it looks like now:

Have a look at this post? https://x.com/sthetixofficial/status/1543420979426500608

I will try reading the resistor values next to the left joy con's handshake IC next, my multimeter broke down for some reason, so as soon as I get a new one I will post whether there is anything out of place. From just looking at them I have no idea why they wouldn't work though.

Maybe this thread can help?

Thank you, I found that thread earlier and already cut off the shield on another side to see whether any traces were damaged before, but they all look fine to me (except for the trace I posted earlier, I don't know whether it has anything to do with the joy cons though and it looks to me like only the mask came off and the trace is still intact.

I'd be glad for more tips and ideas on how to further debug this. If the IC resistor values are fine I will probably test continuity from the rail to the connector next, but I really don't know why that would be a problem as I didn't touch any of those parts.

Thanks again for all the advice!

 

[jkyoho]

I'd be glad for more tips and ideas on how to further debug this. If the IC resistor values are fine I will probably test continuity from the rail to the connector next, but I really don't know why that would be a problem as I didn't touch any of those parts.

Thanks again for all the advice!

if your right side joycon working fine, just left side cant be detect when attached, check this resistor if it still in 100~200 ohm range.

 

[taxco]

I did that today, cleaned up everything, unfortunately it didn't solve the problem.

This is how it looks like now:

View attachment 476448



I will try reading the resistor values next to the left joy con's handshake IC next, my multimeter broke down for some reason, so as soon as I get a new one I will post whether there is anything out of place. From just looking at them I have no idea why they wouldn't work though.

View attachment 476449



Thank you, I found that thread earlier and already cut off the shield on another side to see whether any traces were damaged before, but they all look fine to me (except for the trace I posted earlier, I don't know whether it has anything to do with the joy cons though and it looks to me like only the mask came off and the trace is still intact.

View attachment 476450View attachment 476451View attachment 476452View attachment 476453

I'd be glad for more tips and ideas on how to further debug this. If the IC resistor values are fine I will probably test continuity from the rail to the connector next, but I really don't know why that would be a problem as I didn't touch any of those parts.

Thanks again for all the advice!

Do that, maby the cable broke inside wich is not visible.

 

[dibas]

if your right side joycon working fine, just left side cant be detect when attached, check this resistor if it still in 100~200 ohm range.

The right joy con works perfectly fine, it is just the left joy con that is not detected. I tried measuring the resistance but it was not measurable. My multimeter cables are huge though, so maybe I just didn't make good contact to the resistor ends. I will try it again. The other resistors all gave measurements in that ballpark though. But why would that resistor just break down? Is that typical?

Do that, maby the cable broke inside wich is not visible.

Definitely still planned as soon as I get a new multimeter. I will keep you guys updated.

 

[darkgunsou86]

hello. i installed a picofly to my v2 switch and the boot time is slow compared when installed to an oled switch. i am using v2.80. when installed in an oled, the boot up time sometimes can be instantaneous. is the v2.80 not optimised for a v2 switch? which version is better for the v2 switch? TIA.

 

[Artimental]

I think the yellow flex cable is bad, I put red prong on 3.3v and black on 3.3v on the picofly while connected and it reads 1. As well as the b. I resoldered many times. A d and c read 0, when doing a from yellow ribbon to picofly etc

what do you think?

 

[Sendpie]

heyo, i installed a picofly modchip on an oled switch, (almost) everything works fine but the right joycon rail stopped working after the installation. checked if i broke the line that's right next to the apu shielding but the issue doesn't seem to be there.

i have continuity from the point next to the resistors to the connector. does anyone have any leads on where i should test?

 

[twins333]

heyo, i installed a picofly modchip on an oled switch, (almost) everything works fine but the right joycon rail stopped working after the installation. checked if i broke the line that's right next to the apu shielding but the issue doesn't seem to be there.

i have continuity from the point next to the resistors to the connector. does anyone have any leads on where i should test?


View attachment 476915

Hi. Please check the first post on this page, it has a link to a sthetix guide just for your problem.

 

[Gladuin]

Hi, I'm modding a patched V1 switch, and it glitches fine (almost instantly even), but in hekate I get the "slow eMMC error" (see screenshots).

Diode readings are as follows:
DAT0: 468
CMD: 465
CLK: 468
RST: short
CPU: open? my multimeter indicates out of range I think

Could it be that I've burnt the resistor next to the RST point? Is it causing the slow eMMC error? If yes, what value is the resistor so that I can replace it.

I also burnt the points (the pads ripped off the component) on the back where you can optionally solder the MOSFET, but I think this is not really an issue.
View attachment 476012
View attachment 476013
View attachment 476014

Seems like I was wrong. The resistor near the RST line is just fine, I was just measuring it while the battery was still plugged in. Unplugged the battery and the RST line measures around 1000mV in diode mode and the resistor value is 4.7kOhm.

I've ordered some 47 Ohm resistors that I'll put on the required lines to fix the slow eMMC issue.

On another note, how do you guys cure the UV solder mask stuff? I ordered the mechanic brand stuff from aliexpress and it doesn't seem to fully cure at all. Blasted a small blob for ~3 minutes with a 10W UV lamp, and it was still wet inside.

 

[Endracion]

Just checking here to confirm that the "CPU always reach BCT" error is indeed always due to the mosfet - first time I get this error, but I only have one mosfet left, and I'm guessing this one might just be defective? Is there any way to test it before I scrap it and use my last one?

 

[linkref]

Does someone know what pins need to be bridged in order to have CPU point ?
In the v2 flex it's the 2 pin of the middle, but what about the switch lite flex ?



Thank a lot

 

[deeps]

Just checking here to confirm that the "CPU always reach BCT" error is indeed always due to the mosfet - first time I get this error, but I only have one mosfet left, and I'm guessing this one might just be defective? Is there any way to test it before I scrap it and use my last one?

yes, that's 100% a mosfet problem. you can measure resistance between the gate and drain/source on the mosfet - it should be very high, at least 1 Mohm, possibly infinite (as in no connection). A broken mosfet will often have a lower resistance value between gate and drain/source.

 

[Endracion]

yes, that's 100% a mosfet problem. you can measure resistance between the gate and drain/source on the mosfet - it should be very high, at least 1 Mohm, possibly infinite (as in no connection). A broken mosfet will often have a lower resistance value between gate and drain/source.

Thanks for that!

So I'm reading over 1.123M Ohm on Drain to Gate and the same from Source to Drain, would that mean that it's functional as expected and I messed something else up?

To add some more context, I've done all the past 20 installs with a rear mosfet.

 

[cowboy619]

Does someone know what pins need to be bridged in order to have CPU point ?
In the v2 flex it's the 2 pin of the middle, but what about the switch lite flex ?



Thank a lot

Follow the 2 middle pins

 

[Sendpie]

another problem, i have a switch that gave proper diode readings ~800 on A C D 3v3, 1.5 on B.

after trying to boot up the chip, all readings got halved to ~400 and it seems many caps got shorted on the console, any leads on how i could fix this?

 

[Endracion]

So I'm not crazy right, this is ==* long long short, so BCT error? Mosfet removed does this too.

I did end up replacing the mosfet, even though the reading was seemingly correct and it solved the issue.

 

[abal1000x]

So I'm not crazy right, this is ==* long long short, so BCT error? Mosfet removed does this too.

I did end up replacing the mosfet, even though the reading was seemingly correct and it solved the issue.

AFAIR, the BCT is about payload uploaded to the emmc, its before the mosfet glitching.
So that could happened.

Post automatically merged: Dec 24, 2024

another problem, i have a switch that gave proper diode readings ~800 on A C D 3v3, 1.5 on B.

after trying to boot up the chip, all readings got halved to ~400 and it seems many caps got shorted on the console, any leads on how i could fix this?

The mosfet jobs are literally short the vcc to the gnd.
So thats maybe the first thing to check.

Post automatically merged: Dec 24, 2024

Seems like I was wrong. The resistor near the RST line is just fine, I was just measuring it while the battery was still plugged in. Unplugged the battery and the RST line measures around 1000mV in diode mode and the resistor value is 4.7kOhm.

I've ordered some 47 Ohm resistors that I'll put on the required lines to fix the slow eMMC issue.

On another note, how do you guys cure the UV solder mask stuff? I ordered the mechanic brand stuff from aliexpress and it doesn't seem to fully cure at all. Blasted a small blob for ~3 minutes with a 10W UV lamp, and it was still wet inside.

Using cheap purple laser and transparent resin usually use for glass art or nail.
Just search on here. I also got advised from people in here (@LogicalMadness , @lightninjay) . Use it until now, never fail.
https://gbatemp.net/threads/picofly-a-hwfly-switch-modchip.622701/post-10193981