libera/#maemo/ Saturday, 2022-01-01

siceloLooking at page 4 of N900 schematics, and testing my broken USB n900, where the USB Vcc/Vbus pin shows a short to ground17:13
siceloThe curious thing is, there's no such short to ground when I put multimeter on F530017:14
siceloF5300 itself isn't blown. Now I'm wondering - can I 'safely' ignore the mysterious short to ground on the actual usb pad and instead solder a new port's vbus pin to F5300? Unless I'm missing something here, at least schematic and my F5300 findings suggest it should be safe17:17
sicelojoerg, and others, ^17:17
joerghow did you detect that "short"?17:19
joergand where's my N900 schematics?17:19
siceloIf i place multimeter on pads that correspond to 1 and 5 on microusb, it shows continuity17:24
joergwell, that's what is supposed to happen since this is a power-in rail, no? :-)17:26
siceloThere was a previous failed attempt to re-solder the usb port back on. That said, I really can't understand how it could lead to this short, and more weirdly,why it isn't 'visible' on f530017:26
joergmeasure true ohms17:27
siceloHehe, well, if i place dmm on 1 and ground, i have continuity. Exactly same thing with pin 5 and gnd17:27
siceloWi test ohms now17:27
joergor duh, just connect a USB charger and measure voltage from F5300 to GND17:27
siceloTrue ohms between 1 & 5 is exactly 017:29
siceloFound your schematics?17:31
joergsure :-)17:39
joergyou say you don't find that short from F5300 to GND? smells like a failure in probing17:40
joergor you soldered VBUS pin to ... sleeve? instead of PCB pad17:41
joergmeasure ohms between both sides of F5300. When you are confident to always get a proper contact and instant result (~0 ... 1 ohms), measure again F5300 (either side) to GND or pin2 of batt connector17:44
siceloNew port isn't soldered in yet ... i won't do it right away17:44
joergso what is it you're measuring? USB PCB pad to F5300?17:46
joergan what's >>between 1 & 5<<?17:47
siceloUsb pad 1 and pad 517:51
joergaccording to schematics there is absolutely nothing other than a plain simple trace from VBUS pad of USB socket footprint to F5300. I don't see how that could possibly have a short to GND at one end and none one the other17:51
siceloYes, beats me17:51
joergthere must be a glitch in probing somewhere17:52
joergmeasure continuity from pad to F530017:53
joergwhat means >>F5300 itself isn't blown<<?17:53
siceloPad to f5300 = 47kOhm17:55
siceloF5300 not blown means i get 0 ohm when dmm leads are on each side of f530017:56
joergyou actually managed to damage PCB so the trace from USB,pin1 to F5300 is cut and instead shorted to GND. Congrats :-)17:57
sicelo:-p17:57
joergyou of course can connect USB receptacle pin1 VBUS "free air" yellow wire to "right hand" pin of F5300. Make sure the receptacle pin1 does never touch the PCB footprint pad17:59
siceloRight hand pin of f5300 means?17:59
joerghowever carefully check the other 4 footprint  pads before you proceed17:59
joerg"right hand" in schematics, the one supposed to connect to USB18:00
joergthe fuse can't be 0 Ohms, must be at least fraction of an Ohm18:00
siceloD+, D- are intact. Continuity seems okay. Only the vbus is borked18:00
sicelo0.5ohm18:01
joergcheck for short to GND too18:01
joergon D+-18:01
siceloI did. They're okay18:01
joerg0.5R sounds absolutely correct. So you can determine which end of Fuse is real 0 ohms to N|L|C 530018:02
joergto the other end you may solder some thin wire and run that to the "aired" pin1 of receptacle18:03
siceloNLC means? :-)18:03
joergN5300 diode, L5300 coil, Capacitor18:04
siceloRight. Let me locate those18:04
joergC5300 is a 4pin capacitor, prolly looks odd18:04
joergN5300 diode also looks like it's supposed to be 4pin18:05
siceloYes, so i can't probe that. Even the cap is a weird 4 pin thing :-p18:09
siceloAt least all are accessible. The diode's pins are underneath (balled is the term?)18:09
joergsicelo: the capacitor like the diode have 2 connected pins that are VBUS and 2 connected that are GND18:10
joergBGA package, yes18:11
siceloI guess my T235H dmm's resolution isn't the most accurate around. But looks like the vbus side of f5300 is the one on left when camera module is on top18:11
joergcan't duplicate that now. you could send a 1A current through the "short" from F5300 to L5300 and measure the voltage across F5300. If there are something like 0.5V its ok, if there's 0.00V then you feed the current to the wrong side of F530018:13
siceloMakes sense. I'll try that18:15
joergwhen you use a lab PSU, watch out for damn output buffer capacitors that defeat current limit completely for a curent surge on making contact!18:26
joergkeeping that probelm in mind, you may just get away with two tips and lab PSU: adjust current limit to a few 100mA; connect L5300 and "wrong" (L5300) side of F5300; ramp up voltage until PSU signals current limited mode, add a 50 or 100mV on top ; now switch off PSU and short output to discharge buffer capacitor inside PSU; connect to "right side" of F5300 and L5300 and switch on PSU, current limit shouldn't kick in thanks to the 0.5R of fuse you have now in18:30
joergcurrent path18:30
siceloUnfortunately I don't have it (lab psu)18:34
siceloBut can do that test later at a friend's18:35
joerggood luck!18:41
joergand happy new year, better than damn 202118:42
joergto all of you18:42
joerg*sigh* topic is vasty meaningless without infobot :-/18:43
joergmaybe I could find infobot factoids for  ~lf (or ~tabletsdev ~combined ~emmc ~flasher) sdk: ~sb in my chanlogs and set up an autoresponder for them18:46
joerghttp://reisenweber.net/irclogs/libera/_maemo/_maemo.2021-01-17.log.html#t2021-01-17T13:24:06 etc18:48

Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!