[Gta04-owner] [Tinkerphones] GTA04A5 and News from the Kernel

H. Nikolaus Schaller hns at goldelico.com
Mon Nov 7 10:32:01 CET 2016


> Am 03.11.2016 um 18:58 schrieb H. Nikolaus Schaller <hns at goldelico.com>:
> Hi,
>> Am 03.11.2016 um 18:18 schrieb Ларионов Даниил <scumcoder at yandex.ru>:
>> Dear NIkolaus,
>>> Here they finally are
>> Thanks a lot for the x-rays, but I didn't get this part:
>>> A first measurement of the boards shows shorts in one case and some missing connections in others.
>> Does this mean that one of the boards *does* have a short-circuit, but it didn't show up on the x-ray?
> At least the first impression was so. I have to verify if it is really the case or was a wrong measurement.
> I just did it quickly after getting the X-ray images and writing this mail.

finally I did the planned measurements and it can be confirmed that there are short circuits on one
board which isn't visible in X-Ray. Either they are very tiny whiskers or we have mixed up the X-Ray
screen shots.

Anyways there is a very clear pattern.
If we have (almost) no flux there are many unconnected pads.
If we have (too) much flux there are shorts in the middle.

Therefore the most promising approach appears to reduce the amount of flux in the middle.

This either means using a stencil with smaller openings in the middle or close them completely.

I will discuss that with the production line and then I think the risk of building some more
broken boards is no longer high. Wen can then use this result for final fine-tuning.

BR and sorry for the slow progress. It is unlike software where we just have to run the
compiler, download the result to the device and test. Within minutes...

It is more like old FORTRAN style with punched cards where they were handed over to the
operator who did run the task over night and next day you got a pile of computer printouts
showing errors or results. Then you did punch some replacement cards to modify the card deck :)

BR and thanks for your patience,

More information about the Gta04-owner mailing list