Broken MicroUSB idea: Creating a new connection - Page 4 - maemo.org

joerg_rw | # 31 | 2011-06-09, 18:55 | Report

Originally Posted by abill_uk View Post

PS and if i am not wrong this mod will bypass the charging circuit and not a wise move at all.

it doesn't bypass the charging circuit, but all the protection components in charging path, incl the 2A fuse that could stop your device from catching fire

/j

Last edited by joerg_rw; 2011-06-09 at 18:58.

The Following User Says Thank You to joerg_rw For This Useful Post:
abill_uk

joerg_rw | # 32 | 2011-06-09, 19:07 | Report

Originally Posted by lorul2 View Post

OK I have recently joined the broken USB club (

) . I understand your point about the circuit protection, however I would like you to answer these 2 questions for me.

1. If I developed a circuit board which had the protection your talking about to use these easily accessible points as my new USB connection. Can you detail the entire circuit from the USB cable back to the test points.

2. If there are "easier ways to fix those pads" please detail how you would do it for someone who may not have access to the proper tools, or the small circuit expertise to attempt such a fix.

[...]

The protection is to isolate $random_cable introduced mess from N900, as

Code:

$random-cable -> connector -> protection -> N900-internal

there's little use in doing sth like

Code:

your-protection -> $random-cable -> your-connection-to-testpoints -> N900-internal

Also without proper tools and experience/expertise you'll fail on that approach as much or even worse as/than you'll fail on the suggested proper fix of USB receptacle, as shown in the thread linked to in a former post of this thread

For a general note and advice regarding these testpoints: they are fine for using a properly designed service jig mount or sort of cradle to contact them. They are also ok for an emergency recovery setup, given you make sure there are no wrong signals, static charge, wrong voltages and/or polarities and so on. Most important: Alsways make sure to first connect the GND of your auxiliary contraption and the N900 GND, to avoid static discharge smoking up N900's chips

/j

Last edited by joerg_rw; 2011-06-09 at 19:14.

The Following User Says Thank You to joerg_rw For This Useful Post:
mrsellout

Texrat | # 33 | 2011-06-09, 19:31 | Report

It may be useful at this point to "reboot" this thread by consolidating the best practices, pictures, diagrams and advice into a single post, based on the back-and-forth here. Anyone willing?

The Following User Says Thank You to Texrat For This Useful Post:
abill_uk

dr_frost_dk | # 34 | 2011-06-09, 19:53 | Report

The most important ting here is ESD protection, i blew the fuse in mine, but this was due to experimenting late at night and then reversing polarity of the homemade USB charging system........

So i would say put an ESD protection on the data lines and that should be enough.

NOTE: I fixed my USB with removing the fuse and soldering a thin wire in there in stead, and yes this is not the best solution at all, but it has worked for some months now, of course i only use the USB connection once a week on average due to my charging system.

I bet i will have the N900 that lasts the longest in the USB area

The Following 3 Users Say Thank You to dr_frost_dk For This Useful Post:
4352134, joerg_rw, Texrat

mrsellout | # 35 | 2011-06-09, 19:53 | Report

Originally Posted by joerg_rw View Post

The protection is to isolate $random_cable introduced mess from N900, as

Code:

$random-cable -> connector -> protection -> N900-internal

there's little use in doing sth like

Code:

your-protection -> $random-cable -> your-connection-to-testpoints -> N900-internal

Also without proper tools and experience/expertise you'll fail on that approach as much or even worse as/than you'll fail on the suggested proper fix of USB receptacle, as shown in the thread linked to in a former post of this thread

/j

How about building a flashing/usb connection rig? I've been thinking about this idea for a while. The schematics show a FS-94 flashing rig, but there seems to be very little on the internet about them.

They can definitely be used for the purpose of flashing though, as this blog shows.

I bought one of these of ebay. The box is plastic and the pins are springed so that when you plug in the box the contact is firm but in the wrong place for our purposes. But it does provide the dimensions of a would be flash box made out of metal, or perhaps it could be lined in foil and grounded? The protection circuit could then be inside the flash box, and a battery would be fixed beneath the box.

So from bottom up, you would have battery>>flash box>>n900.
You would then have a microusb connector mounted on the box.

This approach would have the benefits of allowing phones with pcbs too far damaged for repair to be repaired, it would also save you from having to open your phone, and stop it from looking like a bomb when going through customs!

Obviously this couldn't be carried around in your pocket, but for flashing/usb transfers purposes it would do great.

If someone could make a properly fabricated pcb that was 5-10mm thick then maybe a thin box could be made that would sit between a battery and the phone and could sit inside a mugen battery cover modded with a microusb port. This could be the complete solution.

Attached Images

The Following 5 Users Say Thank You to mrsellout For This Useful Post:
4352134, fahadj2003, fffffred, joerg_rw, SSLMM

vertigodragon | # 36 | 2011-06-10, 03:56 | Report

Originally Posted by joerg_rw View Post

The protection is to isolate $random_cable introduced mess from N900, as

Code:

$random-cable -> connector -> protection -> N900-internal

there's little use in doing sth like

Code:

your-protection -> $random-cable -> your-connection-to-testpoints -> N900-internal

Also without proper tools and experience/expertise you'll fail on that approach as much or even worse as/than you'll fail on the suggested proper fix of USB receptacle, as shown in the thread linked to in a former post of this thread

For a general note and advice regarding these testpoints: they are fine for using a properly designed service jig mount or sort of cradle to contact them. They are also ok for an emergency recovery setup, given you make sure there are no wrong signals, static charge, wrong voltages and/or polarities and so on. Most important: Alsways make sure to first connect the GND of your auxiliary contraption and the N900 GND, to avoid static discharge smoking up N900's chips

/j

so if i understand this correctly...the safest way to provide the +5V is from the original usb pad since it routes it through the circuit protection..BUT we can also rebuild the circuit protection but then we deal with the issue of the connection from the circuit protection to pad..or we can just connect directly to the pad and suffer massive consequences if something goes wrong

bottom line: the only viable way to provide power is through the original route?

The Following User Says Thank You to vertigodragon For This Useful Post:
joerg_rw

joerg_rw | # 37 | 2011-06-10, 04:03 | Report

Originally Posted by vertigodragon View Post

so if i understand this correctly...the safest way to provide the +5V is from the original usb pad since it routes it through the circuit protection..BUT we can also rebuild the circuit protection but then we deal with the issue of the connection from the circuit protection to pad..or we can just connect directly to the pad and suffer massive consequences if something goes wrong

bottom line: the only viable way to provide power is through the original route?

Exactly. And same applies for D+/- to a slightly reduced degree
Though, as mentioned, a properly designed connector jig similar to the one 2 posts above with a few improvements, that also cares about ESD and protection at large, will work as that's what the pads are made for. Soldering some wires there and routing them to a USB receptacle won't pan out in the long term.

[edit] Thanks to suggestion 2 posts above, and of course thanks to those nice fellows donating to h-e-n, I ordered some of those connectors, let's see what we can do with some modding. Just in case you feel like donating as well, here's some Donate button

/j

Last edited by joerg_rw; 2011-06-10 at 05:12. Reason: ordered jig, added "Donate"

The Following 4 Users Say Thank You to joerg_rw For This Useful Post:
Estel, mrsellout, SSLMM, vertigodragon

SSLMM | # 38 | 2011-06-24, 14:39 | Report

Originally Posted by joerg_rw View Post

Exactly. And same applies for D+/- to a slightly reduced degree
Though, as mentioned, a properly designed connector jig similar to the one 2 posts above with a few improvements, that also cares about ESD and protection at large, will work as that's what the pads are made for. Soldering some wires there and routing them to a USB receptacle won't pan out in the long term.

[edit] Thanks to suggestion 2 posts above, and of course thanks to those nice fellows donating to h-e-n, I ordered some of those connectors, let's see what we can do with some modding. Just in case you feel like donating as well, here's some Donate button

/j

Keep us informed about this please

I will donate when i receive

luketanti | # 39 | 2011-07-25, 22:32 | Report

Hello. I tried to solder the usb port how the other tutorial described but the data pins chip's 2 connectors broke as well. Can I use the two mentioned pads for data lines and still use the old +5v and GND. If I understood well I will be keeping the protection for the 5v but loosing only the esd protection.

joerg_rw | # 40 | 2011-07-25, 22:50 | Report

Originally Posted by luketanti View Post

Hello. I tried to solder the usb port how the other tutorial described but the data pins chip's 2 connectors broke as well. Can I use the two mentioned pads for data lines and still use the old +5v and GND. If I understood well I will be keeping the protection for the 5v but loosing only the esd protection.

Basically yes, though I'm rather sure your soldering and wiring skills won't result in a proper USB2.0 compatible signal path either. Anyway you're probably relatively safe, as each USB connector, on mating, has a make-GND-first property, so unless you plug in open-ended USB cables (I.E. cable not connected at other end when plugging it to N900) you probably can't create a critical ESD event on datalines by using the D+/- testpads for alternative connection of a mounted USB receptacle - only if you properly route the wires from receptacle to testpads away from case and flat on PCB surface all the time.

I'd rather suggest you send photos, I give advice for a better way to connect D+/-, and you find somebody skilled enough to do the soldering without breaking components again.

cheers
jOERG