Hacking Hardware Picofly - a HWFLY switch modchip

[QuiTim]

The soldering it's ok.
The scheme is correct?
There are no schematics in the guide for the MOSFET on the back of the logic board. I found images a few pages behind but not being in the guides or even in the main pages, I ask for confirmation.
Also I understand they are experimental installations, right?

If it's a v2 switch then the schematic is correct.
abal100x came up with this and it seems to work very well, but this is still experimental so it is still not included in any guide and as with everything on this mod, no risk no fun

 

[Switxh]

Resetting the statistics happened when all the glitch trial failed. It will followed by === led light.

For me this error occured when the dat0 adapter are not stably connected.

Okay thanks! I haven't seen it give any error lights for me though. It just does the white/yellow light that you see after training data is reset then it retrains with blue light then flashes white/yellow again on glitch completion. I was suspecting the dat0 adapter though. I just install them to take a firmware backup and then I reball the chip anyway. I don't trust myself to reball without a MMC backup.

 

[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.

thanks, I will get the 0.1uF tmr, btw, is there any kind 0.1uF bigger, your pic seem so small and hard to soldering

 

[Takezo-San]

.

Post automatically merged: Jul 8, 2023

The AON6554 has only one G terminal the other 3 below it are S so as long as you did not bridge the 3 and 4 together you should be OK as far as wiring goes (see picture).
I am assuming that this is the orientation, so 1st pin bottom right (please check since I cannot see the mark on the mosftet from the picture)
Anyway, I think there is something else at play here.
The mosfets that work (tested by me) example IRF8342 has a total gate charge Qg of 4.2; AON7506 Qg 4.3; AON7518 Qg 6.9; IRF8714 Qg 8.1 while the one you are using AON6554 has the Qg of 21.3
@abal1000x what are your thoughts about this?

The orientation was right. The big circle was at point 1. Maybe the QG value is too high for this sort of install. Well, that's the last time I take @cgtchy0412 advice. Thanks @QuiTim. You the man

 

[FreeLander]

yep 0201. if you need to replace them, you can steal from the rp2040-zero


2.74 is stable on 95% of rp2040's, so it should (probably) be fine.

Thanks for this. Which rp2040 is known to be stable with 2.74? WaveShare?

 

[cicci084]

If it's a v2 switch then the schematic is correct.
abal100x came up with this and it seems to work very well, but this is still experimental so it is still not included in any guide and as with everything on this mod, no risk no fun

OK.
I can add my testimony, for now it works.
Thanks for now.

 

[QuiTim]

The orientation was right. The big circle was at point 1. Maybe the QG value is too high for this sort of install. Well, that's the last time I take @cgtchy0412 advice. Thanks @QuiTim. You the man

I don't know man, I mean that's the only difference that I saw and it was just my theory but on the other hand if cgtchy0412 says that he has 10+ installs with the same mosfet (although none of them Oled) we should probably look at some other factor for this not working...
Edit: @cgtchy0412 did you do any installations on the back of the pcb on the APU?
Placing the mosfet on APU or on back of the pcb might have an effect on the requirement of the speed of the mosfet.
And since: "Qg in its purest sense, helps define the switching times at any given gate drive current and gives a practical comparison of switching performance between MOSFETs" from here, than maybe this is it.

 

[cgtchy0412]

The orientation was right. The big circle was at point 1. Maybe the QG value is too high for this sort of install. Well, that's the last time I take @cgtchy0412 advice. Thanks @QuiTim. You the man

LOL, this was AON 6554. And its already belong to end user, not in my possesion anymore.

https://gbatemp.net/threads/picofly-a-hwfly-switch-modchip.622701/page-554#post-10195198

Post automatically merged: Jul 8, 2023

I don't know man, I mean that's the only difference that I saw and it was just my theory but on the other hand if cgtchy0412 says that he has 10+ installs with the same mosfet (although none of them Oled) we should probably look at some other factor for this not working...
Edit: @cgtchy0412 did you do any installations on the back of the pcb?
Placing the mosfet on APU or on back of the pcb might have an effect on the requirement of the speed of the mosfet.
And since: "Qg in its purest sense, helps define the switching times at any given gate drive current and gives a practical comparison of switching performance between MOSFETs" from here, than maybe this is it.

Yup, this specific yellow is on the back. You can see it from the previous posting before that video, i posted pictures start from kaptoning the lcd and board.. lol.

I tried many AONs before, but since setled with 6554 and you can see i still have like maybe 15 AON 6554 to install.
If its not working on other install then there shoud be other reason beside the mosfet.

 

[deeps]

Thanks for this. Which rp2040 is known to be stable with 2.74? WaveShare?

not sure if there's any pattern. it works fine on my rp2040-zeros which are noname from aliexpress

 

[Takezo-San]

I don't know man, I mean that's the only difference that I saw and it was just my theory but on the other hand if cgtchy0412 says that he has 10+ installs with the same mosfet (although none of them Oled) we should probably look at some other factor for this not working...
Edit: @cgtchy0412 did you do any installations on the back of the pcb?
Placing the mosfet on APU or on back of the pcb might have an effect on the requirement of the speed of the mosfet.
And since: "Qg in its purest sense, helps define the switching times at any given gate drive current and gives a practical comparison of switching performance between MOSFETs" from here, than maybe this is it.

Yeah I think it is because the dmm showed no response on that line. There were no shorts but no response. Probably meant gate was closed.

 

[chronoss]

Another code error

 

[Takezo-San]

LOL, this was AON 6554. And its already belong to end user, not in my possesion anymore.

https://gbatemp.net/threads/picofly-a-hwfly-switch-modchip.622701/page-554#post-10195198

Post automatically merged: Jul 8, 2023

Yup, this specific yellow is on the back. You can see it from the previous posting before that video, i posted pictures start from kaptoning the lcd and board.. lol.

I tried many AONs before, but since setled with 6554 and you can see i still have like maybe 15 AON 6554 to install.
If its not working on other install then there shoud be other reason beside the mosfet. Can you see any thing wrong with my install?

You have pics of your mosfet Install? Any installs done on Oled with this specific mosfet? Can you see anything wrong with my install?

 

[cgtchy0412]

Another code error

Atmos version? . Please try older 1.5.3 if its not.

Post automatically merged: Jul 8, 2023

You have pics of your mosfet Install? Any installs done on Oled with this specific mosfet?

Nope.. never done oled until now.

 

[chronoss]

Atmos version? . Please try older 1.5.3 if its not.

latest

 

[cgtchy0412]

latest

Try to run 1.5.3

Post automatically merged: Jul 8, 2023

Pics.

 

[chronoss]

Try to run 1.5.3

Post automatically merged: Jul 8, 2023

Pics.

samething

 

[cgtchy0412]

samething

This issue is with new EMMC image (initial cfw)? as i found the same issue yesterday when upgrading the firmware, i cannot run 1.5.4 for some reason.. but 1.5.3 can.. i think it was from atmos 11.4 in firmware 15.1

 

[chronoss]

This issue is with new EMMC image (initial cfw)? as i found the same issue yesterday when upgrading the firmware, i cannot run 1.5.4 for some reason.. but 1.5.3 can.. i think it was from atmos 11.4 in firmware 15.1

Can u give me your file for test ?

 

[QuiTim]

LOL, this was AON 6554. And its already belong to end user, not in my possesion anymore.

https://gbatemp.net/threads/picofly-a-hwfly-switch-modchip.622701/page-554#post-10195198

Post automatically merged: Jul 8, 2023

Yup, this specific yellow is on the back. You can see it from the previous posting before that video, i posted pictures start from kaptoning the lcd and board.. lol.

I tried many AONs before, but since setled with 6554 and you can see i still have like maybe 15 AON 6554 to install.
If its not working on other install then there shoud be other reason beside the mosfet.

Sorry, what I ment to ask was "any installation on the APU same as Takezo", just to compare notes, since he used it in the APU and I know your installation on Lite was on the back, but I was wondering if you used this same mosfet on the "normal" APU installations.

 

[abal1000x]

.

Post automatically merged: Jul 8, 2023

The AON6554 has only one G terminal the other 3 below it are S so as long as you did not bridge the 3 and 4 together you should be OK as far as wiring goes (see picture).
I am assuming that this is the orientation, so 1st pin bottom right (please check since I cannot see the mark on the mosftet from the picture)
Anyway, I think there is something else at play here.
The mosfets that work (tested by me) example IRF8342 has a total gate charge Qg of 4.2; AON7506 Qg 4.3; AON7518 Qg 6.9; IRF8714 Qg 8.1 while the one you are using AON6554 has the Qg of 21.3
@abal1000x what are your thoughts about this?

I think The mosfet already an okay . The rds is around 3-4mohms which already suitable for the glitch.
I agreed, from the picture i can't confirm wether the 1st pin is the right bottom or left top.

Also because the transistor flipped down, we couldn't solder to the center pad. Its important for the D to be soldered as wide as possible.