My new parts created and tested :)


#1

Hello, I’ve made some parts and tested them at prototype,
crystal for 2520,2016,1612 ; 3535 LED package for apa102, and a PTH micro-USB slot
I focus only on footprint and not breadboard views

my1612.fzpz (6.2 KB)
my2016.fzpz (6.2 KB)
my2520.fzpz (6.2 KB)

1,2,3 -
crystal footprint for 2520,2016,1612 :
I modified them from the 3225 part available,
and the part print was easily hand-soldered,
so far check with electrometer show no unwanted short happened,
(only ground pads did short)
not sure if they did actually contact, but I’ve once tried placing my working 3225 crystal (on my custom arduino) to 2520,2016,1612 crystal with the same 24Mhz, and all did work on 3225 footprint (so I actually don’t need smaller footprint to use the part, if I had room LOL

APA102-3535.fzpz (13.7 KB)

4.- apa102 footprint for 3535 version - chained and worked fine, though little bit hard to hand-solder but still pretty feasible, so should be fine for solder-paste.

USB_MICROB_PLUG_PTH_A.fzpz (9.1 KB)

5.-
micro-USB plug with PTH slot, the holes seemed a bit big, and buggy as it don’t do copper/mask block at bottom-side, still usable as they’re typically grounded

I did another version with oval holes, not working as the oval were ignored and not drilled :frowning:

here are my tested board, there are also some other testing lipo charger components as well, worked too if anyone is interested

USBcharger-apa3535T-PMOS_2032.fzz (174.2 KB)


#2

By and large look fine, the svgs are not at the scale specified in the parts file format document, but functionally that doesn’t affect anything (although it is desirable to do.) It would be preferable to have a url field that points to the part this applies and/or was based on. Other than the scales the only thing FritzingCheckPart bitches about is that silkscreen is below copper1 which means that Fritzing will select silk before copper1. The reverse order is preferred.

Same two complaints (scale and pcb layer order) as above plus missing bus members. That would be because the breadboard and schematic svgs are 8 pins and pcb is 6 pin (and the bus definitions refer to the 8 pin part.) It is preferable to do all three views (even if you don’t need them) when publishing parts to reduce confusion. Again a url to what part this was derived from would be useful.

Footprint may be ok (without a datasheet url it is difficult to tell), but there is no connection to schematic or pcb (the data connections go to the mounting holes) which makes it less than useful as a part other than the footprint potentially:

it also lacks a label field (thus the part1 label.) Assuming you are using Inkscape the hole size can be set by hole diameter = pad diameter - 2* stroke width. As an example for a standard 0.038in hole for a .1 header connector assuming a stroke-width of 20 thou, the pad diameter is 0.078in which gives a 0.038in hole (0.078 - (2 *0.02)) = 0.038.

Non round holes are typically a problem in both Fritzing (which needs path manipulation, see forum articles on making pcb slots) and I think (being far from an expert on pcb manufacture) problems finding a board house willing to deal with non round holes. Hope this helps.

Peter


#3

Thanks for reply, will look into improvements

Actually,
I didn’t find the suggested layout/footprint from datasheet
I think I can link them to mouser or digikey, like those
https://www.digikey.com/en/product-highlight/n/ndk/nx1612aa-series-smd-crystal
but instead scale the bigger prints 5050 and 3225 smaller, and just worked LOL
I know that’s not generally a good idea :stuck_out_tongue:
some pads should have bigger width in order to met hand solder reach.

and I just realized that selecting connector in the editor don’t take much time, and image doesn’t need to look as real :slight_smile: will do that in my future parts

the micro-USB had problem in PCB view is because due to unknown overlap, I cannot pick the right pad in the editor
and yes i’m doubting how the drill size was calculated, as the inkscape can only see circles and stroke in the xml layers.

I do multi-hole/vias for slots and never get a complaint from my PCB house yet,
it sometimes looked jagged if the holes were not dense enough.

Po Ting
regards,