[IC] Bluetooth Planck - a Wireless 40% Ortholinear keyboard

[mco]

Hey.

I don't have any fancy images yet just an idea.
I'm really loving my 40% ortho keyboard, but am missing a wireless function, so I can connect it to iPad or just use with my laptop without all the messy cables.

I've started designing it and wanted to check if anyone is interested in a GB.

For the start it would probably only be a raw 40% Ortholinear PCB (a Planck clone) with Bluetooth and hotswap sockets for MX switches. I would make it compatible with BM40 or NIU Mini cases and plates, so it wouldn't be hard to get a compatible case (NIU Mini probably, as it has more space for the battery, and bigger battery simply means more time without wires )

What do you think?

FAQ:

What will be the layout?
- It will be a full grid layout if I can fit antenna somewhere else, if not, MIT only.

Is it all in one or do I need a BT ProMicro?
- It will have an integrated MCU and antenna. You just need the li-po battery, a case and switches/keycaps

What firmware will it use?
- ZMK, unfortunately QMK cannot be used on nRF52840 due to licensing issues

[Darth_munkee]

So this is a good idea for people that don't want to build a Choc 40% ortho like the Pancake for wireless, but you need more of an idea if what you're going for. Is it going to have an integrated MCU and BT package or a pro micro footprint to use a Nice!Nano, or something else entirely like an Adafruit Feather? Are you going to use QMK or ZMK for firmware? What layout options are you going to have i.e. Grid, MIT, 2x2u, etc?

[mco]

Hi. Thanks for the questions.

All stuff will be on the PCB - MCU, antenna and battery charging chip.
You only need to add a li-po battery, a case and switches/keycaps.

Right now I'm designing it using nRF52840 chip directly on the PCB.

If I can fit the antenna somewhere else, it would have a full grid layout along with MIT.
If not, antenna will go between space and next 1u switch, so no full grid layout, but I think it should be enough space for it at the top of the PCB.

For the firmware - it would have to be ZMK. QMK cannot be baked into nRF52840 chip due to licensing issues.
I've seen a fork of QMK somewhere with support for it. I will test it when the prototype is ready, but you would need to flash it yourself.

Hope it answers your questions.

[Darth_munkee]

Thanks for the answers, sounds like a good plan for the board. And I'm familiar with that branch of QMK you're talking about. I'd say stick with ZMK honestly, it's way easier.

[keybo_hobo]

Yes, i'd like to replace my bm40 with this. I kinda liked poseidon40 case from kprepublic, but was not happy with bm40.

[mco]

What problems did you have with BM40?

[el_murdoque]

I'd be super interested. I'm using Preonics at home and in the office, but write on my laptop keyboard when I'm working remote.
A wireless Planck would be just the thing to add to my backpack for anything more than just writing a few lines.

[Technofrikus]

Yes please That would be great to have. Also as a basis for a printed case I am working on with some serious flex and thermal inspired design. Could even make a collaboration if you are interested.

[learntoletgo]

What type of battery will it have?

Sent from my Pixel 5 using Tapatalk

[mco]

Idea was to simply include a LiPo battery charger chip on the board and allow for any reasonable LiPo battery you can get on the market (at least 500mAh).
Something like this: https://www.vatsbattery.com/product/303450-3-7v-500mah-high-temperature-battery/

After some work with the project I've found that there is a bit of an issue with the BM40 case. There's simply no space for a battery this large in the case if we use hotswap sockets. You could use it with tiny 160mAh LiPo battery, but it would require recharge once a week or so. If you have a bigger case you can connect even a 2000mAh battery, but it would need few hours to charge (fast charging is 500mA so around 4h, slow charging is around 100mA - 20h).

I'm now experimenting with the design that has 2x CR2032 batteries and a LiPo port. I think that for my personal use case (mostly on cable with few wireless hours per week when I'm working outside of the office or connecting it to iPad) it could be enough to just use batteries and replace them every few months instead of a rechargable LiPo battery. The LiPo charger will still be on board, so you can use that instead of the batteries.

[athhh7101]

I don't really have much knowledge about this, but consider me interested! Keeping an eye for this project

[learntoletgo]

keep us posted!

Sent from my Pixel 5 using Tapatalk

[mco]

Hi.

Sorry for being quiet. Project moves slowly forward.

I've dropped idea to power it from CR2032 batteries for now.
Power section became a lot more complicated with 3 power sources, each with different voltage and safety requirements.
For now it will be only USB + LiPo battery. If it works fine I'll revisit the idea, but for now just want to make something that works and can iterate from.

Regarding progress, I'm a bit busy recently and have very limited time to work on this.
Nevertheless I'm finishing up schematics and will move to PCB layout somewhere this week.
It's a bit hard to estimate, but I hope to order first prototypes in 1-2 weeks.

Will try to post a bit more often from now on

[keybo_hobo]

Glad that it is moving forward))

[mco]

Just wanted to let you all know that I have finished the design and ordered parts for first prototype. Should receive it all before end of the year.
I’m pretty happy with what I have managed to fit onto this tiny board - support for mit and full grid layout and 0-4 rotary encoders (all four corners).
Probably there will be some issues to fix, but hope to get to final design somewhere before march next year 🎉

[keybo_hobo]

Great news. Can't wait. Have you thought about venders or how you going to sell it?

[jayndoodle]

Do you think there would be enough space in the Bm40 case if we try to mill some material off?

[mco]

Haven't thought about sales yet, need to get it all up and running first

About the BM40 case: I'm using one right now and hope I can fit a small battery without any modifications.
(ex: https://www.vatsbattery.com/product/80℃-high-temperature-551235-3-7v-160mah-battery/).
I've intentionally reversed hotswap sockets in the middle row to make some room for it.
If it's not too thick, it should fit. Will test it out after assembling the prototype.

Of course larger battery won't fit the case. I was also thinking about designing my own case, similar to alu BM40, but with space for battery.
No idea when I'll start on it though.

[longluong]

I love this idea!

[Vush]

 

up ^^

[ClicketyClacks]

LMK when you would like to take my money.

[hulster]

I would be massively interested as well. Do not understand anyway, why still all "high quality" custom keyboards like GMMK Pro, Keychron Q1 still only cable connected.
Especially for the Keychron Q1 I do not understand while all other Keychron boards are BT.
My reason - when I see this wonderfull Ad videos for the Planck EZ from Ben Vallack and see him in front of an iPad cable connected I am lol.
I do not no ANY Apple user, neither for IOS devices nor the Mac who likes to connect peripherals via cable.
Currently I have a Keychron K3 connected to my business windows laptop and in parallel to my iPad. So even if you say OK it works to connect a Planck to the iPod via cable, but who likes to switch between the devices by re-connecting the cable?
Maybe originally for the older BT standards Gamer may have an issue. But for me not anymore.
@mco - do you have any idea, why the still building keyboard without. Just for my interest.
Further - maybe I did not read careful enough here - is your plan to create a fully new PCB or as an add-on to the current version?

[mco]

Hi. It’s a full pcb with everything on it. A 40% keyboard with hot swap is really tight on space so I’m not sure I would be able to squeeze any addon there. Also I don’t think there’s any connector I could use on BM40 or Planck to connect a slave controller.

Also while I’m already here: I’ve assembled first prototype and it seems to work. Have not made a proper firmware yet, but I can connect it via BT to my Mac, so it looks like it will be fine. There will be another revision required for sure though because it does not work without battery connected. Don’t know why yet, but I have to made some error with charging controller wiring. So I’m slowly moving towards a final version.

[hulster]

Hi. It’s a full pcb with everything on it. A 40% keyboard with hot swap is really tight on space so I’m not sure I would be able to squeeze any addon there. Also I don’t think there’s any connector I could use on BM40 or Planck to connect a slave controller.

Also while I’m already here: I’ve assembled first prototype and it seems to work. Have not made a proper firmware yet, but I can connect it via BT to my Mac, so it looks like it will be fine. There will be another revision required for sure though because it does not work without battery connected. Don’t know why yet, but I have to made some error with charging controller wiring. So I’m slowly moving towards a final version.

All ok - I was just interested. It is not a must to update a Planck. Just it would be good if it would fit in existing cases.
Did you plan to apply per key LED?

[mco]

It should fit BM40 case. There might be some problems to fit a reasonable battery there. But I’ll test it out and try to find one that fits and has some capacity.

Unfortunately there are no per key leds in current version. There are some space issues preventing me from adding them, and also they would eat up battery pretty fast. Of course I could make them work only when cable is connected, but I assume nobody wants to use his wireless keyboard on cable

[hulster]

It should fit BM40 case. There might be some problems to fit a reasonable battery there. But I’ll test it out and try to find one that fits and has some capacity.

Unfortunately there are no per key leds in current version. There are some space issues preventing me from adding them, and also they would eat up battery pretty fast. Of course I could make them work only when cable is connected, but I assume nobody wants to use his wireless keyboard on cable

Regarding the case - I missed out the Drop Planck case is quite minimal. So no space for a battery.
You are right for the battery drain. My Keychron K3 with permanent light effects on drains quite quickly, but at least never last less then 1 day.
But for the Planck I was not thinking about light effects. More about indexing layers and active keys on it. Would be helpful in learning phases and when you changed something to your layout.
So using LEDs for this purpose should not drain battery that much. The cabling the board for longer time slot should not be a general issue.
Question is about the intention of using BT. For me it is to re-connect to other devices easily. So permanent power connections could be ok. Just connecting it easily to other device by BT would be MY main need.
About the case - is the general size of BM40 the same as the Planck. Thinking about an acrylic case from P3 store. We may can discuss with them to get a cut out in the bottom layer above the feet layer to get some space for the battery.
Feel free to have a look:


It should fit BM40 case. There might be some problems to fit a reasonable battery there. But I’ll test it out and try to find one that fits and has some capacity.

Unfortunately there are no per key leds in current version. There are some space issues preventing me from adding them, and also they would eat up battery pretty fast. Of course I could make them work only when cable is connected, but I assume nobody wants to use his wireless keyboard on cable
[/quote]

Regarding the case - I missed out the Drop Planck case is quite minimal. So no space for a battery.
You are right for the battery drain. My Keychron K3 with permanent light effects on drains quite quickly, but at least never last less then 1 day.
But for the Planck I was not thinking about light effects. More about indexing layers and active keys on it. Would be helpful in learning phases and when you changed something to your layout.
So using LEDs for this purpose should not drain battery that much. The cabling the board for longer time slot should not be a general issue.
Question is about the intention of using BT. For me it is to re-connect to other devices easily. So permanent power connections could be ok. Just connecting it easily to other device by BT would be MY main need.
About the form factor of the board - the outer size for a Planck or BM40 would be the same? Thinking about an acrylic case from P3 Store. Not a cheap solution, but maybe we could talk with them about a cut out in the bottom layer above the feet layer to get some space for a battery.
You like to have a look:

[url]https://p3dstore.com/collections/40-or-smaller-acrylic-cases/products/planck-acrylic-gasket-mount-keyboard-case]

It should fit BM40 case. There might be some problems to fit a reasonable battery there. But I’ll test it out and try to find one that fits and has some capacity.

Unfortunately there are no per key leds in current version. There are some space issues preventing me from adding them, and also they would eat up battery pretty fast. Of course I could make them work only when cable is connected, but I assume nobody wants to use his wireless keyboard on cable

Regarding the case - I missed out the Drop Planck case is quite minimal. So no space for a battery.
You are right for the battery drain. My Keychron K3 with permanent light effects on drains quite quickly, but at least never last less then 1 day.
But for the Planck I was not thinking about light effects. More about indexing layers and active keys on it. Would be helpful in learning phases and when you changed something to your layout.
So using LEDs for this purpose should not drain battery that much. The cabling the board for longer time slot should not be a general issue.
Question is about the intention of using BT. For me it is to re-connect to other devices easily. So permanent power connections could be ok. Just connecting it easily to other device by BT would be MY main need.
About the form factor of the board - the outer size for a Planck or BM40 would be the same? Thinking about an acrylic case from P3 Store. Not a cheap solution, but maybe we could talk with them about a cut out in the bottom layer above the feet layer to get some space for a battery.
You like to have a look:

https://p3dstore.com/collections/40-or-smaller-acrylic-cases/products/planck-acrylic-gasket-mount-keyboard-case

I know it is not a cheap solution - but what is cheap in the custom keyboard world. 

[mco]

Yeah, it's a good idea to have leds just for layers. I'll think about some low power leds on the top layer of the PCB when I fix up all errors and prepare firmware for the current revision.
Unfortunately there's no way to fit the common programmable leds. BT + MCU module just fills whole space between first 2 keys of the middle layer



BM40 case I'm using for one of my keyboards is also really tight on space. I was also thinking about designing my own case, but this is a bit more involved.
To be honest I'm most happy with a simple 3d printed case I've designed some time ago and use it at home: https://www.thingiverse.com/thing:4937726.
It's really low profile and adds a bit of angle. Would love to get something like this but made from aluminium.

I'm also still thinking about a crazy idea to make the same keyboard but with simple 2xAA batteries power. I was using Apple keyboard that was built this way and really liked it.
It also makes design much simpler and there's no headaches with blowing up cheap batteries from china getting fire, overcharging, undercharging, etc.
You also don't need any cable then. Out of power? Just swap batteries and you can work wirelessly all the time. The MCU I'm using should work months on pair of AA batteries.

Anyway, need to finish up firmware and flash it somehow to the current board. There's already at least 1 issue to be fixed with the current design and new revision to make.
Will keep you posted.

[mco]

Just wanted to give you an update on the keyboard.

I've spent few hours yesterday on the firmware and got it up and running.
It works pretty well - got full key grid working, 3 BT channels and running on battery.
It also connects via cable and disables BT when connected.

Had no time to test rotary encoders and indicator RGB led that's on board.
BT range looks pretty good, at least in 3d printed plastic case.
I could get at least 5-6 meters of distance when testing in the office.

There's still some work to do in power supply regarding the electronics part and quite a lot of work on firmware side:
- rotary encoders support (easy)
- led indicator for battery empty, bt connecting, etc. (medium)
- programmable charging power, aka. fast & slow charging - there's hardware for that (medium)
- programmable power input (to comply with USB standard and allow it to work on iPad using cable - 100mA vs 500mA consumption)

Will keep you posted.
Have a great week.

[hulster]

Yeah, it's a good idea to have leds just for layers. I'll think about some low power leds on the top layer of the PCB when I fix up all errors and prepare firmware for the current revision.
Unfortunately there's no way to fit the common programmable leds. BT + MCU module just fills whole space between first 2 keys of the middle layer

Show Image

BM40 case I'm using for one of my keyboards is also really tight on space. I was also thinking about designing my own case, but this is a bit more involved.
To be honest I'm most happy with a simple 3d printed case I've designed some time ago and use it at home: https://www.thingiverse.com/thing:4937726.
It's really low profile and adds a bit of angle. Would love to get something like this but made from aluminium.

I'm also still thinking about a crazy idea to make the same keyboard but with simple 2xAA batteries power. I was using Apple keyboard that was built this way and really liked it.
It also makes design much simpler and there's no headaches with blowing up cheap batteries from china getting fire, overcharging, undercharging, etc.
You also don't need any cable then. Out of power? Just swap batteries and you can work wirelessly all the time. The MCU I'm using should work months on pair of AA batteries.

Anyway, need to finish up firmware and flash it somehow to the current board. There's already at least 1 issue to be fixed with the current design and new revision to make.
Will keep you posted.

Having a 3D printed case at the first stage would be ok for me. More keen on getting the PCB quickly.
What I don't like is the idea of the batteries. This would be not very sustained design. There should be a huge amount of LiPo batteries with std connector. Just looks for something flat as possible.
I am currently quite busy, but I will have a closer look at the P3 case. They are talking about a gasket mount design. So maybe fitting more easy then you think, as you may not need to take care of the screw holes. Depends if the BM40 is designed to be a Planck replacement. So size of the BM40 or at least yournew would be interesting. Just to see if it could fit from outer size and secondly the plate regarding the switch position.

[hulster]

Just wanted to give you an update on the keyboard.

I've spent few hours yesterday on the firmware and got it up and running.
It works pretty well - got full key grid working, 3 BT channels and running on battery.
It also connects via cable and disables BT when connected.

Had no time to test rotary encoders and indicator RGB led that's on board.
BT range looks pretty good, at least in 3d printed plastic case.
I could get at least 5-6 meters of distance when testing in the office.

There's still some work to do in power supply regarding the electronics part and quite a lot of work on firmware side:
- rotary encoders support (easy)
- led indicator for battery empty, bt connecting, etc. (medium)
- programmable charging power, aka. fast & slow charging - there's hardware for that (medium)
- programmable power input (to comply with USB standard and allow it to work on iPad using cable - 100mA vs 500mA consumption)

Will keep you posted.
Have a great week.

Great news - thx a lot.

[mco]

Just another quick update.

I've found and fixed issues with power supply.
Indeed I have messed up charging controller wiring, so it was not initialized properly and was not supplying power when battery was not present.
(without battery MCU was not initialized and had no chance to configure charging chip).
Also found issue with charging not working. It was a bit similar to the no-battery case.

Done a quick & ugly wire hack on prototype and it worked, so I have this one figured out.
Hopefully I'll have some time next week to work on other hardware issues and will be ordering a new PCB in 2-3 weeks.

Will keep posting any new updates.

[hulster]

Just another quick update.

I've found and fixed issues with power supply.
Indeed I have messed up charging controller wiring, so it was not initialized properly and was not supplying power when battery was not present.
(without battery MCU was not initialized and had no chance to configure charging chip).
Also found issue with charging not working. It was a bit similar to the no-battery case.

Done a quick & ugly wire hack on prototype and it worked, so I have this one figured out.
Hopefully I'll have some time next week to work on other hardware issues and will be ordering a new PCB in 2-3 weeks.

Will keep posting any new updates.

thx for the update

[mco]

Some time has passed, but here's another quick update about the keyboard:
I've finished revision 2 of the hardware. I'm using it as my main keyboard for around 3 weeks now and it works pretty well.
I was able to fit one of the boards into BM40 aluminium case with small 300mAh battery (had to use a bit of force, but was able to fit it all in).
There are no BT issues in the aluminium case. I can walk around the room and nothing wrong is happening.
With this tiny battery it lasts for a week or week and few days (full day of work, this is my main keyboard).
With 600mAh battery in my other prototype it lasts more than 2 weeks, so it looks like it scales more or less linearly.
Fitting it with 1500mAh should give you more than a month of battery life.

There will be third revision of the board before release as I want to still add a mechanical on/off switch, so you can take it on a plane and disable bluetooth.
Also want to fiddle with hotswap sockets a bit, so I can fit it in smaller case.

I've also started to look around for a factory to build me an alu case that will have space for the battery. Will also keep you posted on that.

Attached a photo my current setup with rotary encoder, in BM40 case.

[haplopoda]

Thanks for the update! So stoked for this board

Sent from my GM1900 using Tapatalk

[fedest]

Amazing. This is very interesting and I'm definitely down to try this.

Hope this makes it through.

Sent from my ASUS_I006D using Tapatalk

[jayndoodle]

How's it going?

[mco]

Hi.

Sorry for being quiet for a while.
Generally design is ready. Me and my friend have been using 2 prototypes each on daily basis for last 6 months or so, and it works pretty well.
With tiny battery - 300mAh it survives around a week. With something more rational - 1500mAh i would say it should survive around 2 months.

Unfortunately the chip I have used https://www.seeedstudio.com/mdbt50q-p1m-nrf52840-based-ble-module-p-3148.html is nowhere to be found.
I've been constantly checking out if I can buy at least few to build few more prototypes for new revision (I've added power on/off switch and charging indicator, updated charging power, etc.) and cannot get any.
I'm slowly loosing hope to be able to produce anything with this chip shortage on the module I've used. There's glimpse of hope as the base MCU that's used there start to show up.

I'll try in the following weeks to redesign the board to use the MCU directly (without using the BT modulee). But there might be a bit of fight with antenna calibration.
Maybe modules will show up in the meantime and I'll be able to wrap things up and get some boards running. Or the antenna will work fine and I'll be able to switch to raw chip with PCB antenna.

Anyway, current chip shortages are really making this hard to finish. Hope it will get better soon.

[hlord2000]

have you considered using a smt antenna? done very little wireless work personally, not sure if that's a good suggestion

[mco]

Hi.

I'm not really into wireless myself. Just learning on the go But from what I understand, antenna itself is not the biggest problem here.
Problem is with matching network and the ground plane. Antenna performance relies on how it's connected (impedance of matching network) and if ground reference around it is designed correctly.

I'll start with reference implementation antenna or use chip antenna as you suggested and try to calculate matching net. Then I'll just order the PCB, assemble and check where it will get me.

Thanks for the hint.

[pixls]

sorry to necro, but wondering if you've made any progress on this, or if you're sharing progress somewhere else, would love to be able to have a 4x12 board with zmk

[mco]

Hi, long time, no see  . But things started to move again. Looks like chips are getting more and more available. I was even able to build few more prototypes and it looks like I should be able to get few hundreds of them off the shelf. I've also designed a case and will be looking for some manufacturing next week. If everything goes well I should have a rough idea about final cost and how it will all look like.

Right now plan is to run a Kickstarter or something similar somewhere around end of the year or early January. Attached few renders of the case, it has aluminium top with plastic bottom (we need some plastic in the case, or wireless will work poorly).

Let me know what do you think. I hope I'll have more updates later this month.

[deukman]

I was just wondering whether you have any update. I love to see this keyboard released.

And, what about releasing two case profiles(mid, high profile) like olkb's planck? I prefer high profile since dust get stacked less in high profile.

[mco]

Hi guys. Sorry I've been quiet lately. Wanted to give you an update, as I think some of you were waiting for the keyboard and nothing new was here from my side.
So, things were a bit hard lately. I had a pretty busy time. Had to focus more on my work and family and had to put a lot of things on hold. There was also a huge change in tax rules last year in Poland (my home country). Generally impression was that event if I would finish the board I could not really sell it to anyone, or I would have to move to a different tax rate that would probably cost much more, than I could ever earn from sales.

Second part of the story is the case, which I have designed, but did not manage to find any workshop to produce it in reasonable price. Generally it's a bit hard to get any company out of China to talk with you if you don't have much experience in manufacturing and quoting thousands of parts. It got me a bit discouraged and dragged the project a bit more.

Things got a bit more positive last week, when I had a call with an accountant. It looks after "the dust has settled" that it wouldn't be that bad and I could actually sell the board without tax issues. Maybe we have all overreacted last year when tax rules changed.

To wrap this up: my idea right now is to run a last round of prototype for the PCB. Then, if everything is fine, I'll publish a small site/form to get a list of people that are interested in getting just the PCB, without the case (it's BM40 compatible, and I'll publish 2 or 3 3d printable cases I have designed for myself, so you might get some case to use it). For final version - with case, you'll need to wait a bit more. I'm not so good at mechanical engineering and finding workshop to build it.

I'm sorry if you we're waiting for the board and I know that it really takes ages from my side. I don't want to give you any false hopes, so I won't give you any specific date. This keyboard is still my main driver, and I use my prototypes every day. Hope I'll be able to share with you soon.

[justinyhuang]

Happy 2024!
Any updates on this thread?