Close

Version 6

A project log for µGame

A handheld game console programmable with (Micro/Circuit)Python.

dehipudeʃhipu 12/13/2017 at 22:137 Comments

The PCBs for version 6 arrived about a week ago, but I didn't have time to work on it, between the day job and being sick. This evening I finally tested it properly:

The good things:

There are still some rough edges, though, that I only found out after actually testing a physical prototype.

The bad things:

So I scrapped the version 7 which was basically just a cleanup of version 6 ready for ordering a larger number of them, and prepared version 8. I moved practically all parts around, including the microcontroller and the flash chip, but now:

I know I said that about the previous version, but this is really to be the last one I'm making (at least with this microcontroller and display) — if everything works, that is. I'm going to order this one at Aisler again, to see how they improved (and to get a stencil).

Discussions

Frank Buss wrote 12/14/2017 at 07:15 point

There shouldn't be a problem with a finger on CS. Do you drive it with a normal output port? Do you have the circuit diagram in PDF format? Couldn't open the fzz files and looks like the last version is 5 in your github repository.

  Are you sure? yes | no

deʃhipu wrote 12/14/2017 at 08:22 point

I don't want to push the changes before I have tested the PCB, so that people don't get surprises. But I added the schematic to the download files: https://cdn.hackaday.io/files/276291182079136/output.pdf

  Are you sure? yes | no

Frank Buss wrote 12/14/2017 at 10:29 point

The pin names are wrong at the display connector, but I assume it is ok. Maybe check if the CS pin is configured as push-pull and not open-drain or something like this. A finger shouldn't change it much. But you could try to add a small delay after pulling CS low, a finger could change the signal form a bit, which would be an indication that something is on the edge. I would measure the timings with a scope and verify that everything is within the spec of the display, like setup and hold times, max clock speed etc.

BTW, I don't see a via for the CS signal. In your circuit diagram it is pin 12, display CS. Isn't this the 4th pin on the top pin row, counting from right, because pin 1 is bottom right?

  Are you sure? yes | no

deʃhipu wrote 12/14/2017 at 10:53 point

Yes, because that's the new, redesigned pcb, without any vias on signal lines (and with cs and dc swapped, so that they don't cross).

  Are you sure? yes | no

deʃhipu wrote 12/14/2017 at 16:43 point

Oh, and the display indeed has wrong labels, because I edited a footprint for a different display to adapt it, and for some reason Fritzing didn't update the labels on the schematic graphic. I guess another bug.

  Are you sure? yes | no

Clara Hobbs wrote 12/14/2017 at 01:09 point

Regarding the problem with the via on the CS line: why not just tent your vias?

  Are you sure? yes | no

deʃhipu wrote 12/14/2017 at 01:14 point

Same reason why I keep the PCB size under 5cm — so that you can make it cheaply at any PCB fab out there. Also, it's more nitpicking now, because there would be a battery taped there anyways... So I just consider this practice.

  Are you sure? yes | no