[opensource] GH80-3000 | Cherry G80-3000 replica

[Rayndalf]

How much would your boards cost?

I'm still waiting on USB C connectors . I currently have a couple left though. They'd be $25 a piece + shipping flashed and ready to go.

JLCPCB is pretty much idiot proof (by some miracle even I figured it out), so I'd totally recommend going that route if you don't mind getting some extras.

I have waited ages to see if stuff comes back in stock please can I buy some boards from you? I will then just get the USB C connector. How do you want to go about it?

Feel free to PM me. The C connectors are on their way (They've been in the US since the 16th, so I think they'll be arriving any day now).
I also have some green or blue LEDs that I can throw in too for the lock lights.

[twiddlemeister]

Hey folks,

This is my first post actually, so hello for starters!

I've been following (lurking around) this project pretty closely and took the plunge on some PCBs with black solder mask to fit into a case I had lying around.

I've got three left (one went disastrously, I'll stick to through-hole from now on thank you very much), so thought I'd offer on here. They aren't assembled, but the component cost only came in at around £4 per PCB at LCSC.

If you're interested in one, fire us a message. I'm in the UK by the way.

Cheers all, hope to see you around.

P.S. This project is awesome. Good job

[TalkingTree]

I can safely say the switch alignment holes are painfully tight

I looked closer into the matter and found out that they were actually smaller than spec. I'm fixing it soon.

Edit: fixed.

[Rayndalf]

I think I may have killed one (read several ) in the process of soldering the USB C connectors, any idea what would have been damaged if I shorted the connector? What do you think I would need to replace to get it working again?

I was able to flash one just gluing the USB C connector down, but after the connector popped up it no longer works, and the other, I originally hardwired, but then tried to attach a connector, and it also no longer works... did I short something out? These connectors have been surprisingly difficult to attach.

[TalkingTree]

any idea what would have been damaged if I shorted the connector?

Likely none.

What do you think I would need to replace to get it working again?

Depends on how bad the connector is now. Can you show a picture?

Where are you in the world?

[Rayndalf]

I'm in the US

[TalkingTree]

That connector looks in a very bad shape on the sides.

[Rayndalf]

That connector looks in a very bad shape on the sides.

I think it's slightly rotated, and the plastic part got a bit melted.

Would the hardwire point work with a JST daughter board or something? I have at least one thats pretty messed up.

[TalkingTree]

Would the hardwire point work with a JST daughter board or something?

Definetely. Even a different connector. That's why I put a breakout there.

[Rayndalf]

Would the hardwire point work with a JST daughter board or something?

Definetely. Even a different connector. That's why I put a breakout there.

Is there a specific header spacing/connector that fits those pins? I remember seeing something in the BOM, but there are two wiring points, one by the atmega and one under the USB C connector.


Any idea what components might have failed if my soldering iron was too hot or I shorted something?

[TalkingTree]

Is there a specific header spacing/connector that fits those pins?

Anything with 2,54mm spacing.

Any idea what components might have failed if my soldering iron was too hot or I shorted something?

Iron was definetely too hot. Some flux would also help.

[Rayndalf]

Is there a specific header spacing/connector that fits those pins?

Anything with 2,54mm spacing.

so 2.54mm spacing for the header that is near the Atmega32u4? The header below the USB connector appears to have narrower spacing

Any idea what components might have failed if my soldering iron was too hot or I shorted something?

Iron was definetely too hot. Some flux would also help.

I actually did use flux, but its the thicker paste meant to be used for sweating pipes (I really need some liquid flux and a pointier soldering iron tip). I found it was really easy to use too much solder and short all the connections, so I had to go back and remove extra solder.

Is is possible I burned the atmega32u4 or one of the capacitors in the process?

What iron temperature do you use?

[TalkingTree]

so 2.54mm spacing for the header that is near the Atmega32u4?

The six pads header? That's the ISP to flash the bootloader. Don't mess with that.

The header below the USB connector appears to have narrower spacing

Could be 2mm then. I forgot, but you can solder live wires to it.

Is is possible I burned the atmega32u4 or one of the capacitors in the process?

I don't think you could do that without actually burning the PCB as well.

What iron temperature do you use?

Between 250 and 300 °C works fine for me.

[Tobai]

Hi, thanks for your work! This is one my favorite projects on GH and I'm eager to build some boards using your PCB.
What is the difference between rev 1 and rev 2? Is the alignment-hole sizing fixed in both revs?

[TalkingTree]

Rev. 2 gets RGB LED indicators, an additional switch support and USB type-C socket.
Side pins holes are fixed in Rev. 2 only.
I've ordered a batch of PCBs for myself so I'll be testing them soon. I've also changed the matrix but I'm yet to update the code.

[h9n9n3]

I can't believe I just found out this thread...
Thank you so much to OP. This is a huge contribution...

[SloothNap]

First of all, thanks for putting work into this project and for posting it.  I was looking for a 100% PCB, and this is exactly what I was looking for.  I ordered and have received a few Rev. 2 PCBs.  I'm just waiting on some of the components.

I have a question.  In looking at the schematic, I don't see any current limiting resistors on the RGB LEDs.  Are they not necessary?

[TalkingTree]

I don't see any current limiting resistors on the RGB LEDs.  Are they not necessary?

In my understanding the driver inside the package will handle that. You do need a 0.1uF capacitor for each one of the LEDs.
Be aware that the new revision is made out of a different matrix and the code on QMK github won't work. I've ordered myself some PCBs but they're suffering a long delay for some reason.
I'll try to update the code as soon as possible.

[nehrajsa]

Hi all. Anyone with rev2 pcbs for sale? Ideally 2 pieces within Europe.
 

[SloothNap]

My Rev 2 PCB is soldered, with the exception of the switches and LEDs (I bought the wrong ones).  I have QMK up and running, but I think I may have found a bug.  The switch positioned between S101 and S102 is wired to column 5, so it registers as a right bracket.  I admit that I could be wrong that this is a bug because this is my first time building a keyboard.

[TalkingTree]

My Rev 2 PCB is soldered, with the exception of the switches and LEDs (I bought the wrong ones).  I have QMK up and running, but I think I may have found a bug.  The switch positioned between S101 and S102 is wired to column 5, so it registers as a right bracket.  I admit that I could be wrong that this is a bug because this is my first time building a keyboard.

As stated in my last post. That matrix is for Rev. 1 and I've not uploaded a Rev. 2 code yet because I was waiting for my PCBs to arrive but they were greatly delayed and yet to arrive.
I'll try to provide an untested code this weekend. Have a little patience please.

[SloothNap]

I didn't explain myself very well.  I didn't use the Rev1 firmware that comes packaged with QMK.  I built my own QMK firmware using kbfirmware.com.  Like you, I thought that I had simply assigned the keys to the wrong pins on the microcrontroller, meaning that I had a software issue.  However, after a lot of troubleshooting, I opened up the schematic and board on EasyEda, and I think there is a mistake in the matrix.

I'm referring to the switch physically positioned between S101 and S102 on the board.  On the schematic, it's called "6.25u Space."  I'll call it S101.5.  S101.5's row pin is wired in parallel to S101 and S102.  However, it's column pin is connected by a copper trace to S83, meaning that its position in the matrix is exactly the same as the right bracket key (S51):  row 7, column 5 in the schematic.  On the PCB, S101.5 and S51 are connected to the same column and row pins on the microcontroller, meaning that any software-defined keybinds will be shared by S101.5 and S51.  I think this is a hardware issue, not a software issue.

If this was the intention when the board was designed, I apologize.  But I think that 101.5 should be routed to column 2 or 3 to allow it to be used as a "space" and S51 to be a right bracket key.

[TalkingTree]

I think there is a mistake in the matrix.

I'll look into it when I have some boards handy. I should be getting mine in a few days now.

Thanks for the heads up.