Hey guys!

I am having an issue with it being stuck at "Waiting for drive to deploy" when I run "make cheapino:default:flash". not sure why this would be happening. I see the rp2040 being listed when I run "lsusb".

Thanks!

Hey, is it possible to turn off the LED or change the color on the rp2040?

Thanks!

Need to recreate on breadboard, but avoiding reusing rows across sides should fix it.

Con: will need to introduce shorting pads which needs to be soldered on non-mcu side
Pro: reversible rj45 will only need one extra hole(as opposed to 3 now)

Should probably also use extra rows instead of extra columns...

Using 5 pin switch will give a more stable, wobble free keyboard without using a plate. Please change the switch footprint to support it.

To allow routing, reversible rj45 leads to displacements of rows/cols when going through the rj45.

Still, I should name the reverted lines correctly at the receiving mcu-pins, to make it clear where they are coming from.
(today, its just fixed in qmk, but misleading in the schematics)

Hello!

Excellent project, I'm very close to having built mine.

I'm thinking of working on a model for a case, and would like to suggest considering mounting holes for the board in the next revision.

Hello,
When i set home row keys to be shift, control etc. on hold it registers as held even while normally typing. I tried to change tapping term but it doesn't change anything, even when i set it to 1000 it was still same. Maybe qmk settings doesn't take effect?

hi!

so i ordered all the parts and the pcb for cheapino v2 and things mostly work!

the index finger columns and the pinky columns on both sides work. and the encoder works!

however, strangely, the middle finger and ring finger columns are completely dead. the weird thing is that they're completely dead on BOTH SIDES!

so that doesn't sound like i wired the switches incorrectly or there are bridges or anything.

any clue what this could be?

also, i've compiled my own QMK setup. but the keys are kind of crazy. the pinky row activates windows menus on the left hand and on the right hand, it brings up emoji stuff. and stuff like home row shift, ctrl, etc don't work even though they work on the default uf2. any thoughts?

thanks!

P.S. on the cheapino v1, it looks like the rp2040 microcontroller is attached on the left and right side and not the bottom edge. but the cheapino v2 had connectors for the bottom edge so i've soldered that on to. is that correct?

just wanted to offer a tip that's made my experience much much smoother - get an ethernet cable crimper!
they just don't make them THIS short and this has really perfected my setup!

Built V2, and really enjoyed the process. However, when using the Vial keymap from releases, I have encounter the oddity that upon leaving the editor (in one case I switched to a different keyboard, in another test, plugged/unplugged).

On layer 1, upon returning to Vial, the keymap was shifted down one row, leaving the top row empty, and layer 2 shifted two rows.

Is this something I've done wrong during build, or something else do you think?

Very new to this, so I've been using the firmware files found here, rather than trying to build my own.

Thanks in advance for any insight.

Firstly, thank you so much for you help and you work on this project! This is my second cheapino I've made, and I love my first one.

I am having a strange issue that I can't reason about, maybe two separate issues. All three keys on the ring finger of the left hand do not register. Also, when I press a key in the index column on the left hand, it registers three key presses. It will register the inner index (farthest right), middle finger, and pinky key in the same row as the initial index key that was pressed. It is always in the order of inner index, middle, pinky as well.

Triple key register example:

results in:

I have checked all the diodes on the board and they seem to be oriented correctly and registering the correct voltage drops. I have checked the switches and they are working correctly. I have checked that all pads have continuity to the MCU and they do.

I can't think of what else to do or check. If you could help me understand what might be causing this I'd really appreciate it.

I will probably be checking over the diodes again tomorrow.

Hi, the project is super interesting. Did you plan to add the hotswap feature in the future and to release the gerber file?

I have two builds going and can't identify any soldering defects but on both boards the X, V, U and O keys (in standard QWERTY layout) don't work. I'm guessing the mirrored problem keys are due to the Japanes matrix layout. Any suggestions for where I should have a closer look?

Hello!

I've been breaking in the Miryoku layout with the encoder programmed to be arrow keys. I've noticed that using the encoder sometimes causes the backspace key to stop working. Using the encoder again will get it.

I see in the code that the encoder disables some lines due to a hardware bug, is there some kind of debugging I can help perform to see what part of the code needs adjustment?

I like this board, but I prefer the low profile choc switches, any plans for adding choc support?

Good Morning,

Thank you for the hard work and getting the keyboard down to be as cheap as possible (I was able to build a his/her set due to minimum order quantities :D)

I was just wondering if there was any plan to port the Vial firmware to the V1 (I didn't realize V2 was so close to release back in Oct!)?

If not, no worries, I wouldn't be so upset having to build another!

Its not really necessary to use a duplex matrix on the left side, and for wireless, to achieve deep sleep, it best to avoid it.

Should be possible to route the "reverse" col+row to separate inputs on mcu(possibly same as "other side" with yet another optional solder pad...).

Hi, @tompi.
I've finally soldered everything together but the keyboard works very strange. I can't get what is wrong with it, because this is my first custom build, so I don't have any experience in the process. But I've tried to do everything according to the build guide and in the correct order.
Unfortunately the left half is not working (not a single key). When I connect the right half then some of the keys on the right side are working, but the keycodes are VERY wrong.
I've tried to flash it by using my custom layout, and then tried to flesh it by using default one:

qmk flash -kb cheapino -km default

So, the left half is not registering any keypresses.

The right half prints the following (in order of the rows from left to right):
top row: nothing is printed (no keys working)
second row prints: wesdx
third row prints: <tab>q<space>a
thumbs row: <nothing>cz

Here are some photos of my build, maybe you will be able to find something wrong.

Thanks for this awesome project, anyway.

The holes where the headers for the RP2040 Zero measure 0.762 mm in diameter (according to Kicad). It seems that some square-y pins measure 0.62 mm (per side), so there should be plenty of room for the pins to fit into the holes.

A PCB (v2) recently manufactured by JLCPCB, however, proved difficult to have the MCU headers inserted, probably by extra thick pins and by manufacturing tolerances.

A simple solution here would be to increase the size of the holes for the MCU so that they are wider. For comparison, other footprints seem to use 1.6 - 1.7 mm pads with 1.0 mm holes.

Cheapino v1 btw.

Pretty consistently the two columns on the right board stop working for like 30 seconds or so. I've checked all of the solder points and I think they're all fine. I've also used both QMK and Vial, results are the same.

I've removed the keys that stop working. (Also, they all stop working and start again at the same time)

Increasing spacing on the thumb cluster downwards by about 3 to 4 mm to allow 1.5u keys will significantly enhance the reachability and comfort of keyboards for medium and smaller hands. The closest key to the main keyboard is too close to put a 1.5u key on the board, lowering the stretching needed to reach the keys. Also, using DSA keycaps or other sculpted keycaps makes it less likely for the thumb to sit on the rim and dig in over log typing sections.

I need to curl my hand 70-ish degrees to reach all the essential centres, or my thumbs barely reach the edge of 1u keys on the thumb cluster. It makes my wrist hurt after a medium to long tying session due to stretching.

The default keymap has two spaces on key cluster, and default keymap is a c file instead of json.

TODO: make default keymap a qwerty json file, and update build guide with reference to qmk configurator. Also update default encoder action to volume up/down, since undo/redo is mac specific...

There is also two new builds that should be on the page:
https://discuss.tchncs.de/post/852362
https://discuss.tchncs.de/post/410101

Oh, and add a note about the tenting by pixelprimer

I've ordered all the pieces for building my own cheapino, and while I wait, I've been thinking of what I have extra, and what I find helpful about my current ergodox. One such thing is a visual indicator of what layer I'm currently on. I currently have an OLED display that I've used with a Pi Pico 2040. Looking at the suggested MCU, it looks like the pins along the bottom include those that would be required to drive the OLED display. As such, I'm curious if anyone has suggestions for using an OLED display with this setup.

hi. so installing qmk msys is very straightforward.

within qmk msys, typing

qmk setup

results in a directory qmk_firmware. that's great. and that seems to setup the environment correctly for compiling.

however, i have no idea how get the cheapino branch now.

git clone -b doesn't work neither does git clone --recurse-submodules ....

and not doing qmk setup and doing git clone -b instead results in an environment where i can't compile. so i'm ending up in a situation where either i can compile and can't see cheapino or i can see cheapino but can't compile!

for those of us who are complete noobs to git, can we please get a step by step instruction just to the point where
qmk compile -kb cheapino -km default
works?

thanks.

Hi

I'm having a problem with my RP2040. When I use a usb-c port (from my ThinkPad) plugged into my cheapino, I can enter DFU mode, and type. But when I go to Vial (in the app or the website) vial cannot find my keyboard. I've read all the troubleshooting steps, and the firmware guide. However, when I use my usb-a port (on my ThinkPad), I get a green light on my rp2040 but I cannot enter into DFU mode nor can I type, and I still cannot get vial to recognize the cheapino. I made sure all the diodes are the right way, not connections are bridged on the MCU, the only questionable thing is the detached piece of solder or two that is soldered onto the hotswap solder pads, but that is not connected to anything else so I don't think that's the issue. I can send you ample pictures if needed, just let me know. I've tried switching laptops and cables. The only cable I did not switch was the TRS cable, but I don't think that's the problem and I don't have another cable, as I bought the TRS cable from the ordering guide. I've built the V2 non hotswap version and I've bought all the parts from the ordering guide links minus the switches and keycaps.

Hi,

this is more of a suggestion than an issue. USB-C supports 24 pins, so more than enough to connect the halves and the cables are usually more flexible and have a smaller connector. Maybe not as cheap though (cable and port), that I did not yet check.

disclaimer: I did not build or try the cheapino with RJ45 yet.

Hello!

I've been getting the hang of using my Cheapino with Miryoku, and found a few things that I wanted to tweak. I have my own repo for these, and wanted to share those tweaks here:

Default behavior for encoder has been modified to left/right arrow for cursor control. Seems reasonably universally useful.

I brought everything (QMK and Miryoku) up to upstream current (as of 2023-11-21).

I made what I consider to be a very useful change to default Miryoku's behavior. The right modifier keys are actually the right modifier keys now, so the left split has LCTRL and the right split has RCTRL. For the most part, this change should be invisible, but for programs like VirtualBox that define hotkeys based on CTRL position, this will match expected behavior on a more traditional keyboard.

I have not touched the default keymap, only compiled it with current QMK sources.

I also re-enabled the LED color changes, mostly because I wanted to :-).

The release page is here: https://github.com/ggppjj/qmk_firmware/releases

I can absolutely put a PR in to add these notes, but just wanted to get the information out of my head here first:

The RP2040-Zero linked... If you get the one with headers thinking that you can just use them with the sockets... you will be wrong... and if you then think "oh well, I'll just solder it to the board, no big deal", you will be wrong a second time in the same day. The pins on the headers are too big for either the board or the sockets suggested.

The EC11 encoder linked is a listing with about 8 "sizes" or "styles" or however you might refer to it... You will absolutely want to order the actual encoders, and not just the knobs 😂

Now if you'll excuse me, I need to go desolder some headers 😬

Hi, first of all I want to say this project is awesome. I have built two sets of the Cheapino v2 with a friend and I really appreciate all the details that have been considered in the ordering and build guides.

Apologies if this is not the right repo to ask about this, but would it be possible to add "n-key rollover" support to the Cheapino v2, i.e.: a way to bypass the software 6 key limit? I would like this to be possible as it is important for rhythm games requiring up to 10 fully independent keys (or more!).

I am currently using the Vial firmware, and I can see that the home row and thumb keys are independently and correctly registered in Vial's matrix tester regardless of how many are pressed simultaneously, but when using other keyboard testers, typing etc. a limit of 6 keys is imposed. I have tried the NKRO toggle in the QMK tab of the Vial client and the "NKRO On" keybind but to no avail.

Hi, I stumbled upon this project from some post wanting to do a GB for the PCB as a low cost minimal build.

I was wondering what your helper "fork" looked like to help with building this. I designed some switch alignment rails that should work here I think, but unsure if it is overcomplicating it if a simple fork works.
Scaled down to 3U size should allow the rails to fit perfectly for each column. I did notice your spacing is 19mm instead of the typical 19.05mm. The difference is small enough that the original interrow locks should still work but small differences like that could stack up to be quite the problem over a few columns, so there is a 19mm version too now.

Hello,

Thanks for the great design. I have been looking at the ordering guide (doc/orderingguide.md), and it seems the link to the silicone bumpers is dead (https://www.aliexpress.com/item/1005002619943801.html) . I am not entirely sure what they are supposed to look like and where exactly they are supposed to go below the PCB, which is why I wanted to ask if it would be possible to update the link to a similar model on AliExpress to make it easier to prevent a missbuy.

All the best

Hi, I've been trying to build miryoku to be able to use QWERTY with it, but so far to no avail. I'm able to build the default cheapino firmware just fine, then I add the miryoku branch and merge tompi/cheapino. After that I run qmk compile -kb cheapino -km manna-harbour_miryoku but I get Invalid keymap argument., then I try make cheapino:manna-harbour_miryoku and get

QMK Firmware 0.20.6
Making cheapino with keymap default and target manna-harbour_miryoku                           [ERRORS]
make[1]: *** No rule to make target 'manna-harbour_miryoku'.  Stop.
Making cheapino with keymap default_split_3x5_3 and target manna-harbour_miryoku               [ERRORS]
make[1]: *** No rule to make target 'manna-harbour_miryoku'.  Stop.
Making cheapino with keymap lars and target manna-harbour_miryoku                              [ERRORS]
make[1]: *** No rule to make target 'manna-harbour_miryoku'.  Stop.
Making cheapino with keymap tompi and target manna-harbour_miryoku                             [ERRORS]
make[1]: *** No rule to make target 'manna-harbour_miryoku'.  Stop.
Make finished with errors
make: *** [Makefile:415: cheapino:manna-harbour_miryoku] Error 1

Other than that I tried forking miryoku_qmk to use github actions and add tompi/qmk_firmware/cheapino to Merge QMK Branches, but that seems to fail because of merge conflicts. I'm new to qmk so I'm not sure what else to try. P.S. let me know if it's better to ask in the miryoku repo.

First of all, thank you so much for the work you've put into this project and the continued support! I'm incredibly excited to get this up and working.

I put together my first attempt this weekend, but I ran into a small problem. On the default key map, the . and h keys show as always pressed like the picture below. When I actually do press them there is no change.

I am submitting an issue only because I only see other issues about keys not registering, but not about keys always registering. It has been a long time since I soldered, so I wouldn't be surprised if I did a bad job. I checked with a multimeter and the switches are constantly shorted with no resistance between the leads. I have tried to remove excess solder and to reflow the solder for quite a while, but nothing has changed.

Here's how the solder joints look now. I think I might have done a little bit of damage to the board and switches by trying for too long to fix it, which is why I've stopped.

One thing that seems weird to me is that if I press the stem of the switch towards the top (parallel to keyboard pcb plane in direction of finger extension) of the keyboard, the key no longer registers as pressed and this is consistent.

I have a feeling that the switches could be broken, but I wanted to check first before I bought new switches or tried to redo anymore work. I don't know what else to try besides new switches

If there's anymore info you need to help, obviously just let me know. I get the feeling this is probably a quick fix, I'm just at a loss.

If this isn't the correct place for this request, I apologize. If you could point me in the right direction that would be appreciated.

add-on:
As I've continued to mess with the kb, I've found that the p key gets stuck down but isn't always firing. Makes me more suspicious of the switches.

A bit inconvenient when trying to unplug cable...

Hello, first of all thank you for your work on this project 👏

I have a problem with qmk, I just checked out qmk repo. Then I followed the instructions, in the firmware guide (because I have a mac, I started on step 3) here

Unfortunately I am always getting the error keyboard.json not found.

Does anyone know what I am doing wrong?
Thanks in advance.

Hi!

Cheapino is such a nice project! Congratulations, and thanks a lot for sharing it with the community. :-)

I'm about to build a Cheapino V2, and I'm having some problems figuring out how the MCU pins are used for the right side of the keyboard.

Here is what I tried. I just received all the materials, including the RP2040-Zero. I dropped https://github.com/tompi/cheapino/releases/download/v2.0/cheapino_vial.uf2 on it, and I started tinkering with a pair of tweezers and Vial to understand how the matrix works: I wanted to see which pin couple is used for each switch. When I short pins, I see two switches detected at the same time: I understand that that happens because I don't have the circuit with all the diodes and Cheapino uses a Japanese matrix.

When I short a couple of pins for the left side of the keyboard, everything works as expected. But when I try with the pins for the right side (i.e. pins from 0 to 6) I something different from the schematics that I opened in KiCad. As an example, if I short pin 3 and pin 6 I have the K1 switch on the right side (so the inner one on the top row) and the one beside it detected on the matrix checker: according to the schematics, I was expecting to see those switches detected by shorting pin 0 with pin 3, but that's not the case. Could you help me understand what I'm doing wrong?

I get this error when i try to compile it, no alteration, only uncommented the lines that make the led change color per layout

Ψ Compiling keymap with make -r -R -f builddefs/build_keyboard.mk -s flash KEYBOARD=cheapino KEYMAP=default KEYBOARD_FILESAFE=cheapino TARGET=cheapino_default INTERMEDIATE_OUTPUT=.build/obj_cheapino_default VERBOSE=false COLOR=true SILENT=false QMK_BIN="qmk"


arm-none-eabi-gcc.exe (GCC) 12.2.0
Copyright (C) 2022 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Compiling: keyboards/cheapino/ghosting.c                                                           keyboards/cheapino/ghosting.c:50:9: error: unknown type name 'ushort'; did you mean 'short'?
   50 |         ushort row_num_with_possible_error_cause,
      |         ^~~~~~
      |         short
keyboards/cheapino/ghosting.c:52:9: error: unknown type name 'ushort'; did you mean 'short'?
   52 |         ushort row_num_with_possible_error,
      |         ^~~~~~
      |         short
keyboards/cheapino/ghosting.c: In function 'fix_ghosting_column':
keyboards/cheapino/ghosting.c:69:9: error: implicit declaration of function 'fix_ghosting_instance'; did you mean 'fix_ghosting_column'? [-Werror=implicit-function-declaration]
   69 |         fix_ghosting_instance(matrix, i, possible_error_cause, (i+1)%3, possible_error, error_fix);
      |         ^~~~~~~~~~~~~~~~~~~~~
      |         fix_ghosting_column
cc1.exe: all warnings being treated as errors
 [ERRORS]
 |
 |
 |
make: *** [builddefs/common_rules.mk:373: .build/obj_cheapino_default/ghosting.o] Error 1

Hello.
Thanks for your time.
Have you any intentions to re-route encoder a bit?
Is it possible to use GP7 and GP8 for A and B pins?

If you rip off a hotswap pad that has row or column routed through it, you will need to bridge manually.

Fix this by routing through-holes only(more difficult to lift), or both(probably best).

I understand that V2 is still on Beta and not fully stable yet. But I've made ordered it and already made the PCB to test it out. As there's quite a difference in pinout on the encoder from the V1 (and the matrix too on the right side if I'm not mistaken).

I've copied the qmk matrix, and encoder code and changed some part of it. I've got the matrix to work already but I couldn't get the encoder to work. Tried with the original code, and it doesn't work. I also try to directly check through the matrix by using bitwise and, and still no result. Is there a pinout I'm missing or should it be handled differently?

The code is at (the latest attempt is the one that reads directly from the matrix) https://github.com/fauh45/qmk_firmware/blob/550ae30a64828afa28911ce23fae2954f7aadcc1/keyboards/cheapino/v2/encoder.c#L67

Hi,
am having a problem with the Cheapino right pinky keys.
The top pinky button is sent middle pinky button in my case O(Colmak) but when using qwerty ;
middle right pinky is sending M
while M key is sending correct key.

I have tried both keyboards with a single rp2040
I have tried default and custom keymaps
I have tried different ethernet cables
I have tried different USB cables
visually both my keyboards have all diodes pointing the same direction (black marks to square)

To be able to make cases with trackball/oleds etc without changing the PCB, consider adding support for VIK: https://github.com/sadekbaroudi/vik

This would be for left side only, and completely optional.

I'm trying to test the switches before installing the encoder on the right side but the first two columns of switches aren't working, I would like to know if the encoder is necessary for them to work or if I should troubleshoot/change the PCB.

Hi there! This project is awesome, congratulations!

Just finished building my first one (V2, great guide by the way) and flashed the vial config, and only "cheapino_vial.uf2" worked for me.
Am I doing it wrong? All the other are install3d by boot starting and then drag and drop the file?

Thank you very much!

Hi!

I am having trouble with the cheapini v2 build. I soldered all diodes with the MCU and the ethernet port. After that i decided to solder one switch and test if everything I did works before soldering everything else, but I don't get any output after pressing the switch.

I tryed the default software and the miryoku.uf2 as well, but none of them work.

If i go to vial.rock website it detects the pcb as Cheapino, but when I try the matrix test the button doesn't work.

Do I need to solder all the components before trying?
I am sending pictures as well so you can see the PCB.

I tryed the troubleshooting points, but I can't seem to make it work.

I also used a tester to check the diodes and they seem fine. I also soldered two switches in order to check if one of them is faulty.

If you need more information please let me know. Thank you!

Hello and thank you for designing such an awesome keyboard. I ordered the parts for a V1 months ago when V2 wasn't out, but only recently came around to actually building it. Everything went fine, and I installed the firmware with Miryoku preconfigured and started learning Miryoku.
However, as you probably know, it only includes English letters. I wanted to create two new layers for some German letters in Vial, but I couldn't because the firmware only allows for 7 layers currently (and all of them are used by Miryoku). If I have read this part of the Vial documentation correctly, the amount of possible layers is hard coded when compiling the firmware.
Are there any hardware or software constrains that wouldn't allow the Cheapino to run with e.g. 9 layers?
And somewhat unrelated, would it be possible for you to release the keymap files used to implement Miryoku in the Miryoku firmware for V1, as I am sure I have done something wrong when trying to implement it in Vial?

Thanks again for this awesome project.

Should support via for configuration with web interface.

I'm testing the Cheapino v2 with the cheapino_vial.uf2 firmware.
I currently have only the left half connected.

When I hold two keys from the same row on column 1 & 2 then pressing a key on row 3 will trigger an additional key on the same row but column 1.

e.g. with QWERTY
Holding down Q & W and then press D will also activate A.
Pressing A will not activate D.

The same ghosting happens when holding two keys from row 3 & 4 and than pressing a key in row 5.
e.g.
Holding C & V and then press T will also activate E.
Pressing E will not activate T.

I have checked the direction of the diodes and they are correct.

Is this an issue with my soldering work, the vial firmware or a hardware bug?

On step 3 it should be
git remote add tompi https://github.com/tompi/qmk_firmware
(just need to remove the two dots at the end)

and before that you should add that we need to
cd qmk_firmware

I really like your build. I want to get my own split keyboard but all of the assembled solutions are so god damn expensive. Unfortunately, I'm still a noob when it comes to hardware electronics. Things like: diodes (both smb and throug hole are supported) don't make much sense to me. That's why you would help me a lot when you would share the links to the products you ordered somewhere. I don't think anyone would mind if these are affiliate links, but that's up to you.

Thank you for the time. I really appreciate your work!