deskthority

you need two 2.5mm holes (you need some error margin) or one 10x2.5mm. but should be okay

edit: I tried, the buttons hit the plate with minimum flex

There is already 0,5 mm margin in the 2 mm holes, but I guess having more would not matter much aesthetically.

edit: more would be better here though as you thought. We cannot assume that people will be able to align the Elf board with sub 0,5 mm accuracy.

Note also that the prototypes have been hand soldered and the button sits a bit higher than they will on the production models.

pomk wrote: edit: We cannot assume that people will be able to align the Elf board with sub 0,5 mm accuracy.

exactly!

pomk wrote: Note also that the prototypes have been hand soldered and the button sits a bit higher than they will on the production models.

that is good to know.

well, then... all I have to do is hand $2000

Have you checked that the USB socket fits neatly under the plate as well? It should have a similar height.

pomk wrote: Have you checked that the USB socket fits neatly under the plate as well? It should have a similar height.

yes, the usb port fits by a whisker. if the plate flexes it hits the port, but I guess no harm could happen there, maybe you can add some insulation on the plate in case of bad grounding.

The USB shield is not connected to anything, and by design (if connected to anything) it should be connected to the casing of the device, so in this case the plate maybe. No harm should come from having just intermittent connection from the shield to the case.

I just checked the tolerances as well, and the button housings maximum height is 1,35 mm. 1,6 mm PCBs have usually 10% thickness tolerance. 1,6 mm * 1,1 + 1,35 mm = 3,11 mm. So if the holes match with the button heads and the plate is not unusually thick there should not be any problems.

matt3o wrote: before placing the order I was reviewing the board. I noticed that if you place the board between the PCB and the plate the buttons are really close to the plate and if the plate flexes I feel like the buttons could get pressed. Any comment?

what about using a low profile button like this

http://www.kr4.us/mini-push-button-swit ... oCLKLw_wcB

this one is huge but you get the idea.
worst case just leave jumpers that have to be shorted like the pro micro

manufacturers are closed until February 4. I'll place my order then.

Will this support QMk fw?

Dan wrote: Will this support QMk fw?

I believe so

The reason i'm asking is because i've recently discovered the QMK fw builder site (http://qmk.sized.io/) and all the neat things it can do for a noobie like me (including the wiring diagram!). Is the firmware updating process as easy as connecting the board to the computer and dropping the .hex file there?

the firmware has been ported to the whitefox, so if it works there it should also work on our controller.

To update the firmware all you have to do is to flash the hex file.

Do you have a quick summary of the final features, like the number of i/os, dimensions, pin spacing and such?

that is definitely something I'd need to do, together with documentation, website and all... I really have too many thing to do

I can try to find time this weekend to make a kicad module and a footprint of the device.

What do you guys think of this?

http://www.digikey.com/product-detail/e ... ND/5270318

Would it be a drop in replacement of what we already have?

Also, what do you think about using through holes pads instead of the buttons (or one of the buttons)? This way you could easily short them with a wire or solder some headers or even a cherry switch if you wanted.

Well almost drop-in at least. I would not change to that exact one, as it is no longer in production, but I can try to find alternatives. I would just need to know the reason for changing the design. Is it height, price or something else?

The USB connector is already taller than the button housings, so height should not be an issue. I chose the high quality omron switches, because it is beyond frustrating if the switches fail, as has happened to many similar development boards in the past.

If the issue is price, I suggest just removing the reset button completely and I can try to find a cheaper alternative to the remaining omron switch. Panasonic should have one with a similar footprint and it would be a couple of cents cheaper while still being a relatively high quality part.

mostly cost. on a $9 board, $2 is just for buttons.

Then lets remove reset and I'll try to find slightly cheaper alternative for the other.

pomk wrote: Then lets remove reset and I'll try to find slightly cheaper alternative for the other.

roger that!

These should be less than 50 cents a piece: http://www.digikey.com/product-detail/e ... ND/6125110
The current footprint should work, but for aesthetic reasons I will tweak it a bit (these are 0,5 mm thinner).

great! let's give it a shot! Please let me have the new files when you are ready!

what if you just use a jumper and be able to set it in bootloader mode with a keystroke. so flash it with a generic. build the board. then kepress to bootloader mode?

I believe a button for the bootloader is rather important (fail safe)

i dont disagree. i really hate jumping the pro micro's but to save cost and space on the board its a thought. just my 2 cents
i agree buttons are nice to have. what about the ones they put on the teensys.

http://www.digikey.com/product-detail/e ... ND/6125106
http://www.digikey.com/product-detail/e ... ND/5209088
http://www.digikey.com/product-detail/e ... ND/5209087

hope this helps. as many hand wired boards as i do something like this would be handy

the first one is actually cute I really don't know these kind of buttons so for me one is like the other. I trust your judgment guys.

Just remember for a BOM cost of ~US$3.7ea@200pcs(components only), that a button at $0.5 although it looks cheap is +13% of the total component cost for a single part that I would argue is non-essential.
It is also at least another $100 that Matt3o will have to put down when purchasing.

If its a price vs functionality argument, it really depends on the priority, that is up to those leading the design.

Personally as nice as the button is, you can have relatively similar functionality by having 2 through hole pads that you can short using jumper/shunt/wire/whatever.
If you individually require easier access to that function, you can easily solder a header on.

My question to put out there I guess is, as this is designed primarily for the use in custom keyboards, how many times is a typical person who would be buying this be using this board going to need to press this button? And is it worth adding that cost to the board?

Sythe wrote: My question to put out there I guess is, as this is designed primarily for the use in custom keyboards, how many times is a typical person who would be buying this be using this board going to need to press this button? And is it worth adding that cost to the board?

as a custom keyboard maker myself I can tell that you use the button very often the first days while configuring the keyboard, then probably once a year. It is also true that you can probably put the keyboard in bootloader mode via a key-combo (soft-button), but I still believe at least one physical button is required.