Pointduino v2.0

Hi everyone

The Square Inch Project contest on Hackday.io made me focused on this project which I have been thinking on for a while (previously shared my opinions under an old topic).

I decided to enter the contest with a STM32duino compatible boards kit. It will be including one core board with many features. I am also planning to build some shields compatible with the core board. You must have seen such tiny board kits.

I had previously designed Arduino compatible tiny boards kit named Pointduino. The details are on my webpage if you are interested in: http://www.furkancetin.com/pointduino-boards.html

Pointduino v2.0 is about to come soon!
I will share new designs, photos and other stuffs related.

Hope you’ll like it

Furkan

You have a high output rate, Sir!
Again, a nice project, where do let produce your PCB’s? (I see you did a panelizing pcb job and not every (cheap chinese) provider accept it)

Before anyone writes, I wanted to mention.
I have seen previously Microduino set however I just saw the latest core board on which STM32F103 is used It costs $20. Here is the link

I’ll STILL continue to design my board with RTC and LiPo charging and monitoring stuff on it and make it open source. I think it will be also a little cheaper. It will be definitely compatible with stm32duino.

YESSS, another nice project!

Looks awesome! Can’t wait to watch your progress.

They’re using a resonator on the micro board?? Little low budget, huh?

Hi guys,
I have been working on PCB designs of new Pointduino, Mentol the smart watch and some other boards.
It’s a long process with continuous updates on design. I will share when they are ready to publish.

They’re using a resonator on the micro board?? Little low budget, huh?

Yes, but…!! For STM32 Core of Microduino, I have reached the image below. It has no 8mhz crystal or similar! It only includes a 32.768k resonator for RTC obviously! As far as I know, they use a clone of Maple Mini bootloader. Do they use internal oscillator or what? Can anyone explain?

The crystal is probably on the bottom of the pcb.

i.e as far as I know, people are using the libmaple core with this board, and only supports boards with external oscillators.

So they must have a crystal somewhere.

Because they are stackable, they avoid putting parts on bottom. My best guess was that 8mhz is wrongly written as 32.768kk Also RTC capability is written on nowhere!

Thanks for your input. I was thinking they were using it as the external oscillator. I’ve used a very similar one before with the ATmega328.

FurkanCetin wrote:Because they are stackable, they avoid putting parts on bottom. My best guess was that 8mhz is wrongly written as 32.768kk Also RTC capability is written on nowhere!

Is it really possible? This is an official document from wiki page and it is there for months. Yeah, maybe.

Anyway, microduino team know what they are doing. I appreciate their work. Absolutely the best(I think) board set with a number of extension boards. So I assume, Stm works well with those compact resonators (that was an issue we talked here last month) I will definitely use 8mhz one on my compact boards.

I’ve been watching the microduino STM32 board for many months.
https://www.microduino.cc/module/view?i … 00004fb0c3

perpetual back-order, suggesting they are in no rush to build more.

For $20, one can get the Particle.io Photon board which is WiFi + a substantial ARM processor and a good IDE. (I have one).

perpetual back-order, suggesting they are in no rush to build more.

Let me tell you what I think about their being-in-no-rush; Creating a core board with STM32 chip is a great idea but they are probably thinking about problems with bootloader. MiniMaple bootloader (and our stm32duino) has issue of pressing reset button, drivers and not-recognised-error… etc. It’s not so straight forward to upload code for beginners.

Photon is revolutionary useful and compact. On the other side, boards of Microduino allows you to build prototypes faster and progress on coding quickly if you are using many peripheral modules.

Hey guys, can anyone see any capacitors anywhere on the board? Only one next to the voltage regulator!
There must be four of them at least between VDD and VSS pins, right? Or they are not essential for minimal system?
Also we see too many resistors around. Two must be for BOOT1 and BOOT2, two for USB, one for LED, one for reset. Others?

It’s beneficial for us to analyze such development boards and see different hardware layouts.

might be an idea to remove the solder bridges(s); 8-9 ( maybe last pin to (?) as well), 1 being left most pin.
of the facing side

stephen

Are you sure that there are no components on the bottom side? Because, as you can see, there is a black plastic on the back side, like in blue o red pills, and there is much space for smd components. In blue pills, for example, the crystal caps are in the bottom side with other components.
One of the resistors is named R19 but I can see only 10 resistors.

Since there is insufficient info and media about this board, I am not sure whether there are components on the bottom side or not.

One of the resistors is named R19 but I can see only 10 resistors.
Thats good point @zoomx. But then, why using at least 19 resistors? My bluepill has 8 resistors(2 for usb, 2 for BOOTpins, 2 for LEDs, other 2 for reset and SWD I think) and 13 caps. Ok If they are taking Maple Mini as a source and convert it, even MapleMini doesn’t have this much resistors.

I am also curious about the mini black chip with 6 legs on Microduino STM Core. Anyone knows it or I am the only noob here?

I really appreciate open source projects!

Hi guys,

Here is fresh news from my new development board. I post here the current PCB view and the schematics. They are not the final versions!

I appreciate your comments and suggestions. This is my first STM32 based board experience. I am not sure about the USB connection, can anyone confirm whether is it correct or not? By the way, for BOOT pins, I have soldering pads on the back side of the pcb to configure as HIGH or LOW.
(Sorry for huge images. No options to put as thumbnail)

Looks good. One suggestion. Solder pads or a jumper to connect/disconnect the power LED. In low power devices, the LED can consume a significant amount of juice.

Is it normal to have 10k in series with one of the USB lines?
Normally I see 22 ohms in series with both usb lines

Also if you want the generic bootloader to work, you will need to pull up the appropriate USB line using a 1.5k resistor.

Yes, the 22R or so resistors are there to reduce ringing in the data lines and the USB_DP line definitely needs to be pulled up. Also looks like you have USB_DP and USB_DM swapped. USB_DM should be PA11 and USB_DP should be PA12. Here’s how I did it after these guys let me know we could omit the extra transistor/GPIO pin for disconnect.

My boards should be here very soon.

EDIT:Here’s my completed schematic for the development board. I’m only using this for prototyping for my next project, but it’s been fun learning about the STM32f1xx so far.

I tried to do USB connections by analyzing my bluepill by a multimeter.

I thought someone had posted a link to the BluePill schematic in another thread, but unfortunately I can remember where.

But there are plenty of other generic STM32 schematics online, so it may be worth looking at one or two of them to inform your design

RogerClark wrote:I tried to do USB connections by analyzing my bluepill by a multimeter.

I thought someone had posted a link to the BluePill schematic in another thread, but unfortunately I can remember where.

But there are plenty of other generic STM32 schematics online, so it may be worth looking at one or two of them to inform your design

Thanks Andy

I should probably put them all in a dedicated repo, and link via a wiki on that repo

That info is also in the data sheet. Don’t forget about those data lines being switched.

Bingo! Thanks Andy, Thanks Roger!

How did I missed it? That’s a shiny piece of work to save in wiki and repo.

I will work on an update for my boards soon.

STM32Noob wrote:Also looks like you have USB_DP and USB_DM swapped. USB_DM should be PA11 and USB_DP should be PA12
…
That info is also in the data sheet. Don’t forget about those data lines being switched.

My first design is also absurd now when compared to the actual connections It is as if I wanted it to be that simple to fit it on my micro pcb

It’s fresh in my mind, as I’m working on a similar project. Mine is working towards a different project, but I designed basically the same thing with the F103T8 recently. Boards should be here tomorrow. I’ll let you know if I notice anything else.

EDIT:Not noticing something right away is OK to me. Noticing it after I send the design off to the fab house is when it’s a bit frustrating.

Noticing it after I send the design off to the fab house is when it’s a bit frustrating.

DirtyPCBs let you upload revised files while your board is still waiting to be manufactured, which can be handy.

I had to do this on one board I designed, as i forgot some clamping/limiting zener diodes on one board and had to reupload.
However as I was in a rush, I didn’t realise that I had removed a track while adding the 2 new diodes.
So when the boards arrived, there was no connections to the MCU from one of the main inputs

Fortunately I only got 10 boards made, and the missing track was from through hole to through hole, so i was able to put a strap ( small wire) on the back of the 10 PCBs

Cutting and Strapping is not unheard of, on production boards… at least for Version 1 boards.
I have taken apart VCRs etc which have obvious additional wires on them, and when I worked for a company that manufacturer electronics, there was a dedicated section in the factor to perform these changes.

However perhaps with better CAD etc nowadays, and the ease of changing designs and putting fixes into later batches, it much less likely to see this in most consumer Electonics in mass production.

We have OSHPark here. Super cheap boards, but they take about 2 weeks. I paid < $5 shipped for 3 of the little STM32 boards. This place sends them off to the fab house so quickly that uploading a revision might not work. Luckily they’re so cheap that it doesn’t really matter.

DirtyPCBs let you upload revised files while your board is still waiting to be manufactured, which can be handy.
A little warning! The backside silk mask is mirrored in the dirtypcb preview but normal on the PCB itself. (I got confused about that the first time I ordered there)

Hi guys,

On PCB production, there is no local producer here (for small amount of production), so I use Seeedstudio, Itead Studio or Elecfreaks. Therefore, I should have finished all of my board designs to make it more efficient. Anyway, working on them.

For bluepill, we have an issue. My bluepill has direct connection (without any component) between PA12 and USBDP + . Could you check yours please? I think I should verify the schematics and the board connections one by one.

My schematics had a small problem of swapping PA12 and PA11, I noticed, thanks. I will update it.

I’ve used PCBway with success as well. The only other STM32 board I have is the blue maple mini 2 layer clone. It definitely has the two 22R resistors for the USB data lines. The resistors are to reduce ringing and should really be there.

I built 3 of my boards that are very similar to what you’re working on. I can confirm 100% that the schematic I posted is good. Just run a 22r resistor in series with USB_DM and USB_DP and pullup USB_DP with a 1.5k resistor to 3.3v(VCC).

Thanks MarkM. This was great verification of the schematics. I should follow as you told and purchase a maple mini clone asap. However, I am 100% sure on direct connection of PA12 to USBDP on bluepill board. I would be glad if anyone can check it.

Furkan

Here is the updated version(v0.5) of the schematics.
Updates:
– PA11 and PA12 are swapped to correctly connect with USB lines. (Thanks to STM32Noob)
– 22R resistors are addded for both PA11 and PA12 serially. (Thanks to RogerClark, MarkM, ahull)
– 1.5K resistor is added to pull-up USB DP line (Thanks to RogerClark, MarkM, ahull)
– Pad is added to deactivate power led for power saving applications. (Thanks to ahull)

ahull wrote:FurkanCetin wrote:Because they are stackable, they avoid putting parts on bottom. My best guess was that 8mhz is wrongly written as 32.768kk Also RTC capability is written on nowhere!

Furkan
even if you’re not planning to use the adc, would it not be better to use the low noise circuit from the mic5205 data sheet?
as it’s a development board it’s about capabilities and the adc might be used at some point, it seems to be a false economy not to add an extra cap and a change of value for a second.
stephen