Red Pill or Blue Pill?

If you look on the ebay site there are some red boards and some blue boards. I didn’t really look closely at the differences when I ordered. I just picked the least expensive one. I ended up with one of the red stm32f103c8t6 boards. Well the fun has ended. The usb connector came loose and now it only works if I power it externally.

Looking at the differences between the red board and the blue boards …

The red board:

Good things about the red board:

• 1.5k pull up resistor on the PA12 pin (USB D+) which you can programatically drag down for automated USB reset.
• large power capacitors and LDO power (AMS1117 3.3v).

Problems with the red board:

• Silk screen is barely readable, the text is chopped off on some of the pins
• USB connector only has two anchor points and it is directly soldered on the surface
• Small reset button with hardly any resistance

The blue board:

Good things about the blue board:

• Four soldered anchor point on the USB connector. What you can’t tell from this picture is that there is a notch in the pcb board and the USB connector sits down inside it some. This provides some lateral stability that takes some of the stress off the solder points. +2
  (Actually there are now blue pills that have a two anchor point usb connector)
• It has nice clear readable silkscreen printing. +1
• It also a larger reset button.+1
• Often supplied with a 300mV regulator (rt9193-33gb)

Problems with the blue board:

• Probably won’t work as a USB device if it has a 10k or 4k7 pull-up on PA12. You have to check the pull up on PA12 (USB D+). If it has a 10k or 4k7 pull-up resistor, you will need to replace it with a 1k5 one to use the native USB.

Nice features common to both:

• SWD pins broken out and easily connected (VCC, GND, SWDIO, SWCLK)
• Multiple GND and 3V3 pins
• USB 5V is broken out with easy access.
• User LED on PC13
• Power LED
• You can probably use more flash (128k) than officially documented for the chip (stm32f103c8t6 64k), I was able to load 115k of flash on mine and it seemed to work.
• Both boards have the same pinouts, you can use them interchangeably

Problems with both boards:

• No preloaded maple bootloader * to me this isn’t really a problem as the entire 64k of flash is available for use. Also, every chip provides a factory ROM based serial boot.
• No user button

If people have any more comparisons or comments, please post on this thread.

I have two blue boards where the USB is like on your red board, but the silk screen is like your blue one.

I din’t noticed how the USB connector was soldered but searching for differences in the photo that I have posted here
viewtopic.php?f=28&t=37#p221
it seems that if you have a bad USB connector you have also a big LDO underneath.

I believe that big LDO meah high current.

But pinout seems to be the same. Compared it with the Board that i got today and it seems like it uses the same pinout. Maybe the boards were manufactured by different manufacturer and they changed it because they have f.ex. the usb port with 4 ground pins or the bigger button already in stock?

Is there a difference on the bottom side between the red and blue one? I think i’ve already seen some with ams1117 as an ldo instead of the sot235 one (DEA1D if i can read it right, prop not)

~Straw

See the link that I have posted for more photos.

Red and blue seems identical in functionality.

I have the red board, and got the same problem as Rick.

I had to reflow solder the connector, both its electrical connections and also the 2 pads that hold it down ( or doesn’t )

But as I knew the repairs wouldn’t last, I got out my hot glue gun and completely smothered the connector and the end of the board in hot glue. I suspect epoxy resin glue would be even better, but it takes a while to set, ( even the 5 minute epoxy seems to take somewhat longer to fully set)

I’d advise anyone who has one of theses boards to put a load of epoxy over the USB connector, but make sure you don’t let it ooze inside the connector.

A trick I have used for years is DeoxIT by Caig. Cheap USB connectors often ‘stick’ metal-to-metal and cause strain on the reflow joints holding the connector to the PC board. This product is the best I have ever used – not cheap, but one only uses a small amount, so it is economical over time.

I still think the epoxy or Goop overcoat is not a bad idea. If you decide to use the DeoxIT product – “small amount” means to use very little! Whatever is in this stuff is concentrated. Using a lint-free lens cleaning cloth, one can actually see the dirt and oxide come off connectors! A film applied to DIP components assure easy insertion, excellent contact low-resistance, and ease of future removal, if necessary.

Ray

Ray

Connector failure because of bad design seems common loads of boards that I’ve seen, not just stm32

Just fixing into 2 small PCB pads causes the copper pads to be pulled off the board.

Actually it could also be the quality of the whole PCB. I managed to accidentally dislodge a reset button on a Zigbee programmer board a few months ago, and had to hand solder some tracks back into place.
So perhaps the copper is not well bonded to the fibreglass

The progressively smaller and smaller features on modern PCBs don’t help. There are mechanical limits to how much strain the bond between a very small track and a very small pad can take. Smaller means more fragile. Potting things in hot glue, silicone or epoxy is sometimes the only option it you want it to survive. On the plus side, potting a tiny board is far easier and cheaper than potting a large one.

Good idea @ahull. I’d like to see what one of those look like.

I think I can only classify this as “semi-potted”

But once I have everything ready to go into Mr Pig, I suspect there will be a little more “hot snot” applied, particularly round the USB connector and anchoring the other wires.

Generally I find the connectors that are “through board” don’t tend to be the problem.

But… Whatever works for you……

RogerClark wrote:Generally I find the connectors that are “through board” don’t tend to be the problem.

But… Whatever works for you……

ahull wrote:I think I can only classify this as “semi-potted”

Sometimes I’m glad NOT to know what you planning to do with the ST32 boards
(earth closet level indicator is still the most harmless thing that comes into my head at the moment….)

madias wrote:Sometimes I’m glad NOT to know what you planning to do with the ST32 boards
(earth closet level indicator is still the most harmless thing that comes into my head at the moment….)

Now we just need to put an Estees rocket motor up thr pig rump and have a flying pig-o-scope.

Ray

So now that I’ve spent more time with the “Blue Pill” I’m not sure it is a such a good choice. The voltage regulator on them is of some unknown variety and it didn’t seem to last long given my messing about with the PA12 USBDP+ pin.

Each board seems to have its strengths and weaknesses. The one thing I really like about the Red / Blue pill boards is the easy access to the SWD pins. However, I can’t really recommend any of those at the moment. I’m going to give one of those Baite Maple Mini’s a try next.

-rick

I resurrected my “Blue Pill” board that died. I think the problem was that the voltage regulator failed. I also did some surgery on the PA12 pin and added a resistor to see if I could get it to work properly with native USB.

This is the underside of the problematic “Blue Pill” board:

I unsoldered voltage regulator?, marked (U1). I also unsoldered the 10k resistor connected to PA12 marked (R10). I have no idea why that was on PA12. Temporarily for testing, I just grabbed a through hole 1.5k resistor and just put it in place from PA12 to the 3.3v pin (see pic below). I then supplied external 3.3v power to the board from another “Red Pill” board. I had loaded up some blinky software on the blue pill so I plugged it in and pressed reset. It all worked. Feeling brave, I plugged in the USB connection on the Blue Pill. Even though it isn’t supplying the 3.3v the USBD+ and USBD- pins are connected and I watched it enumerate as /dev/ttyACM2 ( I had another board plugged in and it was using ACM0 and ACM2). When I pressed reset on the blue pill board, it dropped the USB connection and then enumerated itself again.

It seems like you can use this board with the native USB if you do some minor surgery. Just now, I loaded the ASCIITable example and it sent the data output to the USB Serial port. No chance I could confuse that as I’ve not soldered any pins on this board yet, so all it has is native USB. Worked fine. If I press reset on the “Blue Pill” board it drops the /dev/ttyACM2 device and my putty teminal emulator window exits. I can then connect again with putty to the new /dev/ttyACM2 and it works.

Here a “Hacked Blue Pill” is being programmed and debugged from a “Red Pill” running the BMP firmware:

hack_blue_pill.jpg (201.81 KiB) Viewed 2738 times

Rick

So.. Where does this leave things

It sounds like the actual stm32 was not damaged.

Do you think that the USB reset code damaged your regulator(s) somehow

Or do you think that in your experimentation you could have damaged the regulator etc by doing something else other than just connecting to USB and having the board re enumerate by pulling PA12 low

Just a thought, but I wonder what the smallest reliable time we can drive PA12 low

I’ve tried dropping it down considerably from our original test values, and I suspect it can go even lower.
Actually I tested it much lower and it was ok on my PC.

E.g. If a 1 uS pulse is enough, its less likely to do any damage. (But I know this is a somewhat hacky approach and only something that people who are willing to take the risk should entertain if there is a definite risk of damage)

RogerClark wrote:So.. Where does this leave things

Rick

Thanks

I will re-instate the code when I get chance (later today) but it was better to be safe and remove it for a while

Cheers

Roger

I just got a blue version of the board.
Without any program uploaded, should the board enumerate or not if plugged into an usb port? Mine does not…
I was under the impression that the board would enumerate and would be programmable over usb.
I tried both windows 7 and linux. No luck

The Red and Blue pill don’t have a bootloader unless you upload one using either USB Serial or STlink

The bootloader for that board is

https://raw.githubusercontent.com/roger … 0_pc13.bin

I use STM’s own Serial Flash loader but if you don’t have that you could use stm32flash in the tools folder

Note, you need a USB to Serial converter to flash the board. It can also be done if you have a Uno (I posted how to flash using a uno elsewhere on the forum)

Flashing the bootloader via serial you need to set Boot0 link to HIGH

see https://www.youtube.com/watch?v=G_RF0a0hrak

For using a Uno see
http://www.stm32duino.com/viewtopic.php … &hilit=uno

I do have a ftdi usb to serial 3.3V adaptor.

What STM serial flash loader are you talking about? Is it a free download?
I tried stm32flash but no luck…

Just to be sure:

Boot0 should be 1
Boot1 should be zero
press reset
flash

correct?

Also, what baudrate do i need to flash? Is the default 57600 ok? is the default 8E1 ok?

Ok, i’m one step further
I found the stm loader demo and burning the bootloader was successfull!

next… drivers

Windows drivers are in the repo

See driver/win folder

Run the bat file

If you are using Linux or OSX you need to make changes to the rules files, see the wiki on GitHub , it has information about this

it’s working!
BlinkNCount over usb is counting and blinking

No hardware modifications where necessary on my blue board.

@zatalian

OK. Some other people have already tested with Blue Pill and it works OK, it also works on Red Pill and for some people it works with the “Ugly” board, but the Ugly board doesnt work for me , perhaps mine has a fault

Just performed the “resistor surgery” on my blue pill board. There is no manufacturer marked on the silk screen, flux visible on the board is not my work. Original resistor was 10 kOhm, now it’s replaced with a 1/8W 1.5kOhm one.

bluepill.jpg (71.11 KiB) Viewed 2852 times

Rick Kimball wrote:
I hadn’t really looked at the voltage regulator on this board. * At first glance I thought it was a transistor *. I think people should replace the voltage regulator on “Blue Pill Board” (hereafter called) BPB. It seems to be under powered and leaves a lot to be desired.
-rick

cazimirb wrote:Rick Kimball wrote:
I hadn’t really looked at the voltage regulator on this board. * At first glance I thought it was a transistor *. I think people should replace the voltage regulator on “Blue Pill Board” (hereafter called) BPB. It seems to be under powered and leaves a lot to be desired.
-rick

Greetings,
Been following the Blue Pill USB saga … have one of those with the 10K on PA12 as well.
I did the R10 (10K) resistor swap for a 1.5K but not much luck. Maybe the 3.3V supply on the Blue Pill?

Installed the “_PB13” bootloader via DFU & USB serial dongle, all OK. The Blue pill’s LED starts off flashing fast after reset and after a while flashes at a slower rate.

XP does recognize something gets plugged into USB but unfortunately the XP driver install tells me the install program is not a valid .EXE, so no drivers gets installed and I’m stuck there. On Linux, nothing gets enumerated (nothing listed with “lusb”). This is with and without the 10K resistor swap. Made no difference at all on Linux (Mint distro).

I’ve been playing with similar arrangements using ATTiny85 (VUSB), also DigiSpark that implements similar hardware for their Arduino bootloader. There the USB lines are limited using 3.3V zeners, also use the 1.5K resistor on D+ to VCC. With that hardware there is no issue regarding driving USB low (maybe Atmel chips have more sink capacity on I/O pins?). DigiSpark’s USB bootloader has an interesting enumerate workaround though… unlike Arduino modules, one do not connect the USB hardware until the Arduino downloader instructs one to plug it in. At that moment, and only then, the magic happens. DIgiSpark then enumerates and shows up as /dev/ttyACMx. Casually plugging DigiSpark into USB does not enumerate at all.

The USB feature is really not a big issue for me as the USB serial dongle works fine for me.

Thanks for all the efforts, all much appreciated.

Regards.

ahull wrote:
From the full bluepill schematic here… BluePill-6-STM32F103C8T6原理图.pdf

Hi Rick,

Thanks for the suggestion … indeed had my 1.5K connected to 5V and fixed now.
Shows up as “Bus 002 Device 006: ID 1eaf:0003” on Linux and Arduino download works fine too.

Regards.

I’m hereby cross-posting a link to a related topic about USB + BMP, at Roger’s suggestion – well, I think this is the thread he was referring to anyway!

http://www.stm32duino.com/viewtopic.php?f=3&t=699

Would appreciate input on this, as it had me quite confused (which isn’t that uncommon BTW 🙂

jcw wrote:I’m hereby cross-posting a link to a related topic about USB + BMP, at Roger’s suggestion – well, I think this is the thread he was referring to anyway!

http://www.stm32duino.com/viewtopic.php?f=3&t=699

Would appreciate input on this, as it had me quite confused (which isn’t that uncommon BTW

I guess the question is…

Do BMP users want to use the BMP virtual serial as Arduino “Serial” or do they want to use USB Serial though the USB connector on the target board

If we enable USB serial in the sketch, it becomes “Serial” so the UART 1 connection to the BMP would become Serial1

Hence any debug text would go to a different place

However if BMP users would be willing to plug in 2 connectors, thats fine.

Personally, If I was using the BMP, I’d want “Serial” to be via the BMP as it would mean that the comm port was always open, as the BMP USB connection is not reset ever time the sketch is loaded.

RogerClark wrote:…
Personally, If I was using the BMP, I’d want “Serial” to be via the BMP as it would mean that the comm port was always open, as the BMP USB connection is not reset ever time the sketch is loaded.

I guess it really depends on whether you see the BMP as a debugger or as a USB replacement.
I treat the BMP as a debugger, and expect it to be minimally invasive. A “probe”, as the name says.

But the inability to use USB while debugging with the BMP is what actually threw me off.
When compiled with BMP, the code shrank by 5..6 KB, and using USBSerial gave me compiler errors.

Very confusing. Especially since it all worked when I was using an ST-Link, i.e. it included USB as expected.
Now try explaining to a newcomer what “Serial.print(123)” does. And how it depends on the “upload method” menu.
It seems to me that more people will start to use an ST-Link or a BMP once h/w debug becomes convenient.
Shouldn’t the behaviour be consistent, at least between an ST-Link and a Black Magic Probe?

I rest my case, pedantically 🙂 – part of the confusion is no doubt just me, but part of this is bound to trip up others.

-jcw

PS. Unfortunately, there also appears to be a re-build bug: switching to BMP first builds with USB, and then without, after an IDE restart. IOW, I see two different build results, without changing any settings – just an IDE restart. There seems to be an missed build dependency after a variant change, despite a message in the IDE to that effect.

Edit – fixed my comment above about trying to use USBserial.

@jcw

Edit – fixed my comment above about trying to use USBserial.

There is a lot of history around USBSerial

We (well actually I ) got rid of its usage in the sketch, as it was non-standard.

When Leaflabs wrote the libmaple core, none of the AVR boards had onboard usb serial, so LeafLabs for some reason chose to call the serial over usb by a different name.

But in 2014 when I looked at it, in the light of the AVR Leonardo and the Due, it made more sense / more compatibility to map the Serial USB to Serial for the Maple mini.

I agree it can be confusing, that the upload method changes this, but I assumed that if you are using an external device to upload that inherently has serial i.e a Serial to USB converter or the BMP, then it makes sense to assume you are doing this for a reason and want to not use the onboard USB

The confusion is that the BMP does both. Its both a debugger and also has Serial USB.

STlink does not have USB Serial, so it made sense to enable it for STLink.

So, yes it can be confusing, but I suspect its because you are using the BMP if you were using STLink you’d not have noticed a conflict

RogerClark wrote:
<…>
So, yes it can be confusing, but I suspect its because you are using the BMP if you were using STLink you’d not have noticed a conflict

Ray,

I’ve flashed a Maple Mini with the 2.0 boot loader, and I can see the upload progress bars, but the code either doesn’t start, or the next upload doesn’t start, or the IDE loses track of the serial port (I’m on Mac OSX). This is after overcoming my confusion with BMP. So it’s been anything but smooth sailing for me so far. I’d really love to just be able to upload over USB.

But USB and the boot loader don’t seem to want to be friends with me!
I do have some hope that maybe November will be better…

-jcw

@jcw

Try running the reset script manually, and see if the Maple mini resets

Ah. Another possible issue was the name of the TTY port. I can’t quite remember the details, but there was an issue with the TTY driver on OSX which we had issues with.

I recall needing to change something in STM32Flash as the TTY port was not getting called the correct name.

I will try to find a Maple mini with the new bootloader on it, and test with my 2 different Mac’s later, in case we now have a problem which we didnt use to have

I’ve found that most people use Serial for debug output. If you use a debugger you can often dispense with Serial all together. Even when you do want serial output, directing it to a UART makes your code smaller and does not use CPU cycles as does the USB UART.

RogerClark wrote:So, yes it can be confusing, but I suspect its because you are using the BMP 😉

No that was not the intention.

Rick…. Would you mind if we enabled USB serial for BMP ?

Or can we just add an option for this via a menu, and default it to Off

Would you mind if we enabled USB serial for BMP ?
I’m fine with that.

Thanks Rick

We’ll find out what @jcw wants when he wakes up in about 8 hours

@jcw

I’ll PM you.

But what board are you using ??

A Maple mini or a Generic STM32F103C

If you are using a Maple mini board, but have selected the Generic STM32F103C option, its not going to work, because the USB reset stuff will not be correct for a Maple mini.

I have a feeling that what you are really asking for is the BMP option on the Maple mini.

But as we have tried to keep things as simple as possible for new users, the Maple mini does not that option.

If you want to use BMP on a Maple mini you’d need to add the upload menu from the Generic F103C to the Maple mini board section, in boards.txt, but you’ll need to remove the define for “Generic Bootloader” as this means actually tells the build process that your board doesnt have the Maple hardware and that it should reset the board via GPIO on the USB pin.

I know this name is somewhat confusing, and perhaps we should called this internal define by a more meaningful name, (and it could be changed), but what it really means is “Does not have maple USB reset hardware”

However thats a really long name.

I’m sorry for complicating things and bringing up both BMP and boot loader issues in this same thread.

For the BMP, I propose to enable USB for all upload combinations, at least across the Generic F1 boards – that’s what this GitHub pull request does: https://github.com/rogerclarkmelbourne/ … 2/pull/133

The Maple Mini entry assumes a boot loader, always. Most logical that way, IMO. If you want to debug it, I think there’s a way to do so with the generic F1 entries, which pulse PA12 for D+ control: at the start of setup(), add code to permanently enable the USB transistor on the Maple Mini via PB9, before calling “Serial.begin()”. Then, PA12 pulsing should work on the MM as well. This trick should work with all active USB control circuits (including the HY-TinySTM103 I’ve been doodling with, which uses PA0), now that PA12 pulsing has been invented and shown to work.

In short, for all generic F1’s: enable USB “Serial” regardless of upload method (github PR #133) and make PA12 pulse (already the case?). For individual boards with their own menu entry: keep options minimal for clarity (already the case?). Probably would be good to extend this to other Cortex types as well, I haven’t yet looked at those.

I’ll investigate further why the USB boot loader isn’t working for me, and report in a separate thread. It may well be Mac OSX related. BMP only works with gdb when using the /dev/cu.usb* devices and not /dev/tty.usb*, for example.

Thanks for all the comments and arguments. I apologise for the number and tone of my posts, if any sharp edges came through – it was definitely not out of disrespect, nor did I want to muddle this thread with unrelated issues!

-jcw

I recomend using this board : http://ru.aliexpress.com/item/stm32f103 … 86246.html
+cost 4$
+USB (Mini-USB better connection but rare cable)
+already soldered
+USB OTG (via connecting ID to gnd (need soldering))
+Good reset button (better than blue pills one)
+More Power Pins
+Good voltage regulator (LM1170 From 5V to 3.3V but 5V can be connected to any voltage)
+SWD + Serial pins on one side
+U can choose to have led or not, u can choose do u have USB connection of D+ and D-
+Easy to find docs (link on the board )

-No LED on PC13 (on another pin or broken because of my try to turn it connecting 3.3v to it)
-can’t be installed on Breadboard/PCB
-No bootloader
-No 5V Regulator

Does anyone have a nice pinout for the blue board ?
Similar to http://www.stm32duino.com/download/file.php?id=17

Thanks !

armleo wrote:I recomend using this board : http://ru.aliexpress.com/item/stm32f103 … tem-board-

I still think, that maple mini is the best lowcost board for STM32F103….

Slammer wrote:I still think, that maple mini is the best lowcost board for STM32F103….

michaeliv wrote:Does anyone have a nice pinout for the blue board ?

jcw wrote:michaeliv wrote:Does anyone have a nice pinout for the blue board ?

mrburnette wrote:Slammer wrote:I still think, that maple mini is the best lowcost board for STM32F103….

Looking at the STM32duino code, I noticed something strange.
The blue pill and the Maple Mini have basically the same micro on them (the extra 64k of flash is not relevant), with the same pinout and the same functionality, correct ?
Yet from the code I am seeing that if you do a digitalWrite(0, 1), different physical pins on the micro will be affected on the blue pill vs the Maple Mini. Am I correct about this ? Do the digital pin numbers map to different physical chip pin numbers on the blue pill vs the Maple ?

The “Blue Pill” is a generic term, there isn’t a product called a “Blue Pill”.

Hence there isn’t a board variant called “Blue Pill”, just “Generic STM32F103C”

The Arduino style pin numbers are only relevant to the Maple Mini and AFIK, Not to any other boards

If you have a Blue Pill (Generic STM32F103 ) board, you should use the real STM pin designations for GPIO, like PC13 etc
The real STM pin name is generally what is written on the generic boards

OK I see. Any reason not to keep the numeric pin numbers consistent between the Maple Mini and other STM32F103C boards ?
I would think this would make porting code from the Maple Mini to other boards easier, whilst having no disadvantage that I can see.

The generic boards got renumbered so that the STM codes don’t appear in a random mixed up order (this was via user request consensus)

e.g. if you look at the enums they start with PA0 and work up to PC13 and above

I’ve no idea why the Maple mini has such a screwed up numbering order, I guess it was to aid the hardware PCB layout, as apart from that it doesn’t make a lot of sense.

But I see why Leaflabs originally use pin numbers and not pin names, as they were selling to buyers who were used to the AVR pin numbering, years before there were any third party Arduino boards, and even before boards like the RPi.

I actually find it hard to use the Maple mini, because it doesnt have STM pin numbers on it, and its harder to work out what pins have what functionality

michaeliv wrote:OK I see. Any reason not to keep the numeric pin numbers consistent between the Maple Mini and other STM32F103C boards ?
I would think this would make porting code from the Maple Mini to other boards easier, whilst having no disadvantage that I can see.

Rick Kimball wrote:michaeliv wrote:
what exactly do you mean by consistent?

I was saying that since there was already a numbering system(the one in use by the Maple), why create a new numbering system.

Its the other way around

The Maple has the non standard numbering system, all the generic boards use the STM official numbering system.

Can we slice off questions into a new post regarding the xtal? This thread is way off topic most of the time

I guess the pin numbering is also off topic

I’ll see if I can figure out how to split postings into a different topic

the pin number conversation is useful

Rick Kimball wrote:I also unsoldered the 10k resistor connected to PA12 marked (R10). I have no idea why that was on PA12. Temporarily for testing, I just grabbed a through hole 1.5k resistor and just put it in place from PA12 to the 3.3v pin (see pic below).

Since that post, I’ve ordered some surface mount 1k5 resistors. They look much neater.

-rick

Rick Kimball wrote:Since that post, I’ve ordered some surface mount 1k5 resistors. They look much neater.

WereCatf wrote:
<…>
But oh well, having the two resistors hanging a bit outside the board doesn’t bother me, and my board was never even meant to enter any beauty contests in the first place anyways!

The Blue and Red pill boards have been added to the new wiki

http://wiki.stm32duino.com/index.php?ti … 103_boards

http://wiki.stm32duino.com/index.php?title=Blue_Pill

http://wiki.stm32duino.com/index.php?title=Red_Pill

Note. This is a work in progress, and not all details have been added for this board (e.g. no schematic or photo)

I will add the photo with pinout from this posting, but I’m not sure if they all have the same schematic.
I found a schematic for one STM32F103C8T6 board that had been posted to the forum, and it seemed to have the 32khz RTC crystal, but as far as I can tell my Red pill doesn’t have the 32khz crystal

PM me if you want to edit the Wiki, as I have to manually create accounts for the wiki as its totally separate from the forum and has its own database and user login etc

Edit.

The top 20 posters to this forum have been added as editors to the wiki, just do a password reset request with your login and you should get sent an email with instructions on how to reset / assign a password

I have one question about the 10k resistor on Blue Pill board.
Using a cheap/ebay st-link V2 and texane tool, I flash the bootloader in the Blue Pill, then with arduino IDE I upload a new program using USB. But I can do it only once. If I want to upload a new program, I must before do an erase with texane st-flash tool and flash again the bootloader. Then I can upload with usb again but only one time.
This is related to the 10K resistor or not?

If I need to replace the 10k resistor by a 1k5 one, which surface mount resistor size must I use?

The easiest way to fix the issue with the resistor is to solder a normal 1.8k resistor in parallel.

Put it between PA12 and 3.3v

Please see the page in the wiki.stm32duino.coming the boards section

Where can I find an explanation why an incorrect resistor value allow to use the USB only one time to upload a program

fired my geda pcb up and discovered its 32 bit… …arrggghhhh, moving swiftly on.

it seems that the dimensions of 1206, 0805 & 0603 are such that a 0603 will mount on 0805 pads and a 0805 will mount on 1206 pads, additionally 0603 will just mount on 1206

so it depends on what’s already there as well as what’s too hand.

stephen

denis wrote:Where can I find an explanation why an incorrect resistor value allow to use the USB only one time to upload a program

zmemw16 wrote:it seems that the dimensions of 1206, 0805 & 0603 are such that a 0603 will mount on 0805 pads and a 0805 will mount on 1206 pads, additionally 0603 will just mount on 1206

i just pulled up the footprints from geda gaf library and over laid them

i didn’t say it was neat, but 0805 overlaps on to 1206, as does 0603 ‘just a tad of overlap.!

think butt weld with solder.

if they’re 0603, replace or just stack a 1k8 on top.

stephen

Thanks all.
I did some tests. The issue does not comes from the resistor. It is cause by usb connection. If I plug the usb cable to the computer screen, I can only upload one time. If I plug the usb cable to the computer box, no issue, I can upload many times.
It is the same with 10K or 1k5 resistor.

I am not sure if this is the issue, but the usb connections are often not soldered well

Also, did you select the serial com port for the board in the IDE and can you see serial data e.g from println()

Serial is used to reset the board, so if serial is not working it wont reset.

That being said, Some people have issues with specific usb ports on specific PCs and going through hubs etc.

Do you have another PC or Mac you can try it on

I try with 4 different cables, on another computer (a macbook 2007), connected to the keyboard usb port and in the usb port where the screen is connected. On all of them, the serial com port is available.
I also write a program with Serial.println(value++);, I was able to see in serial monitor the result.
But when connected to a usb port of the screen, no serial port is available. This confirm that issue comes from my connection to the screen.
I checked with an ohmmeter the resistor between PA12 and 3V3. The value is 10k04 exactly.
I was lucky, I can use my Blue Pill without modify it.

The 10k resistor is far too high.

The USB spec value for this resistor should be 1.5k

This is a design fault on these boards

You can just solder a 1.8k resistor between PA12 and 3.3V as it will be in parallel with the exiting 10k that is on the board

As 1/(1/10000 + 1/1800) = 1525 ohms, which is close enough.

In fact, even 1.5k or 2k in parallel may fix this

But..

Running the board via a hub or a keyboard etc, is still likely to be problematic

Hi STM32duino.

Fist i want to say thanks for the information here and the wiki :
http://wiki.stm32duino.com/index.php?title=Blue_Pill

I have purchased the blue board and pinout is the same as in the wiki page.
Sadly I’m a beginner and it’s the first Arduino/ARM hardware i use and i have some problem.

I use Win7 as main PC when i connected the board in USB in my computer it display as unknown USB.

i have installed these repro driver :
https://github.com/rogerclarkmelbourne/ … er/drivers
now the device is named µPSD DFU STMicroelectronics Extension Application Demo with a yellow warning.
But DfuSeDemo.exe display nothing in the available DFU Device.

I have tried to add the 1,8K resistor in // but error is the same ( i have the 10K CMS resistor soldered).

After reading your discussion here , i think it’s normal beacause it don’t have the bootloader inside , so my computer can’t drive it yet.

i have downloaded “generic_boot20_pc13.bin” and seach a way to send it into the board.
I don’t have Serial USB cable but i have a JTAG USB Blaster clone.
Mayube i can build a DIY cable in RS232 ?

Is it possible to use OpenOCD and a JTAG USB Blaster for flashing the bootloader with these board ?
I use OpenOCD in my Raspberry for flashing Altera CPLD.

After that i think i will be able to upload new code directyly from USB into the flash ( after bootloader of course)

Thanks in advance

for information here is what dmesg display ( seems linux tells more information than windows ) :

X-death wrote:
<…>
for information here is what dmesg display ( seems linux tells more information than windows ) :

Look in the FAQs page

http://www.stm32duino.com/viewtopic.php?f=2&t=873

There is a section on windows problem, and a possible solution using zadig

Thanks both for your reply.

Here is result with Zadig and Usb View!

Is it because they are no bootloader inside ?

i have buy this for send the bootloader.

http://www.ebay.fr/itm/Emulateur-Downlo … SwstxVOzP-

look for those on aliexpress, much cheaper, bit longer to wait; i’d suggest getting at least 2 or 3
very useful if you need to program both ends of an experiment.
these are just first couple for the search string ‘stlink programmer’

http://www.aliexpress.com/item/Hot-Sale … dfd54a9515

http://www.aliexpress.com/item/Hot-Sale … dfd54a9515

both £1.61

i’ll get corrected, but they use a protocol hardwired into the device, no bootloader.
if its just a small target board you can supply 3v3 at low milliamps, that’s 4 wires.
i usually power the target with a usb cable and connect just 3 wires, ground and the data/clock signals.

stephen

@X-death

The board (MCU) is blank, you will need to flash the bootloader using a USB to Serial adaptor (or STLink etc)

I plan to swap the F1 CPU on a bluepill with an F3 CPU. Is there anything I should take into consideration when doing this?

Second, are the pins on the blue pill board 5V tolerant? Is there a list/schematic with those 5V pins?

Here it is: http://www.olliw.eu/uploads/stm32f103c8 … t-wp01.jpg
The red “FT” indicates 5V tolerant pins.

Arise wrote:I plan to swap the F1 CPU on a bluepill with an F3 CPU. Is there anything I should take into consideration when doing this?

Second, are the pins on the blue pill board 5V tolerant? Is there a list/schematic with those 5V pins?

Oh, sorry for confusing things.

I don’t really look into using the board for Arduino (F1 boards are just perfect for that), I’m more interested as using it as a flight controller with a custom firmware. I would like to swap the F1 CPU with an F3 for the floating point, and while I’ve read that the swapping should be pretty much pin compatible I decided to give it a try.
My only concern was some special cases I should look into with the swap.
But since from your answer looks like nobody did that I assume I’m kinda alone here and I’ll have to learn things the hardway…

Anyway, I’ll keep you up with the progress, thanks anyway, the info on this forum is awesome for me as a beginner.

Another question, can I power this board from 2 separate sources?

Usually this board should be powered separately (from 5V or 3.3V) but I also want to connect over USB in order to access the board. Could this cause issues?

Arise wrote:Another question, can I power this board from 2 separate sources?

Usually this board should be powered separately (from 5V or 3.3V) but I also want to connect over USB in order to access the board. Could this cause issues?

ahull wrote:Arise wrote:Another question, can I power this board from 2 separate sources?

Usually this board should be powered separately (from 5V or 3.3V) but I also want to connect over USB in order to access the board. Could this cause issues?

Not sure I understand how to connect the schotty diode, I’m not really good at electronics theory/physics.

But I’m looking at the following documents:
https://storage.googleapis.com/google-c … 2_rev4.pdf
https://github.com/TauLabs/TauLabs/blob … y_v1.1.pdf

Both documents, top left corner. Is that the way I’m supposed to wire them?

Hello,
after some search I found 3 different version of Blue Pill.
Version 1: 2 Anchor for USB, big components.

Version 2: 2 Anchor for USB, small components.

Version 3: 4 Anchor for USB, small components.

Does any one know the difference between them?

@arise, Yes, those schematics are showing schottky diodes, so the USB_VBUS and VCC_IN are not in conflict.

@denis, quickly, I can see 2 differences :

– The LDO of the first seems to be an AMS1117, with huge quiescent current, while the second use small LDO, probably such as SPX3819, with lower quiescent current.
– The MicroUSB connector of the first has only 2 mounting pad while the second has four, which make it more solid.

I’ve received a blue pill today and connected it to PC and got an USB error about malfunctioning, not about missing drivers.
Initially there was a 10k ohm resistor.
I’ve replaced that resistor with a 1500ohm resistor (152) and I still get the same error (about malfunctioning).

When I supply power to the board (over USB) or via 3.3V then the PC13 led is blinking fast.

I’ve checked the resistance between PA12 and 3.3V and I have 1.5k now.

Do I have to write a bootloader first in order to be able to connect directly over USB to the board? Or the jumpers must be set in some other position?

Nice job soldering. You can use the USB just fine without a bootloader. It really comes down to how you want to program the device. You have at least 3 choices:

1.) Use the built-in serial uploader with an usb->serial dongle (FTDI, CP210x, CH340 .. etc) and stm32flash
2.) Use Serial Wire Debug (SWD 2 wire JTAG) with an stlink programmer with openocd
3.) Load a bootloader using either method 1 or 2 and then use the bootloader upload. http://wiki.stm32duino.com/index.php?ti … bootloader

Rick Kimball wrote:Nice job soldering.

I must be using the wrong sized SMD resistor mine is slightly larger and has room for 4 digits (1501) vs your 3 digit (152) I had to look that up to see if 152 was really 1k5

Ultrasonic cleaner?
Dont know what is that…
Unsoldered the old chip with a tiny bit of bismuth alloy, cleaned then tinned the pads, cannibalized a 1500ohm resistor from another dead usb donor device, and soldered it with a light touch of the tip while holding the resistor on board with tweezers.

I also had a very big SMD for 1.5k resistor, but after I took the original resistor out, I feared the big sized smd will not fit on the tiny pads, so I started looking for a board for donors.

Anyway…

Using the Flash Loader Demonstrator, I was able to upload (boot0 set to 1, boot1 set to 0) a file generic_boot20_pc13.bin. I assume this is the right??? name for my board? Used a USB to Serial adaptor to do that.

After uploading it and moving the jumpers back (boot0 set to 0, boot1 set to 0) my computer is recognizing the board as a Maple DFU when I connect it over USB.

Now I’m stuck again, how to make it to look like a Maple Serial and not a Maple DFU?

If I try to rewrite a sketch, I can only do that on USB To serial Adapter, not on USB cable directly.

Help?

I can’t really help there. I’ve never loaded the bootloader. I bought my blue pills because the swd pins are broken out and it makes it very easy to use with an stlink programmer. If you are really bent on using the bootloader it is possible. However, the right board to use for bootloader operation is the Baite Maple Mini clone as that is what most people who want to do bootloading are using.

The bluepill must be connected to your computer USB socket.
Select:
* Variant: STM32F103C8 20k RAM 64k flash
* Tools -> Upload method -> STM32duino bootloader
* keep pressed the reset button on bluepill board.
* Compile a blink sketch and click on arduino IDE “upload” button.

When you see the following message release the reset button.

arduinoIDE.JPG (34.9 KiB) Viewed 1114 times

Have you watched any of Roger’s videos?

https://www.youtube.com/watch?v=0jdJp3TQuJY

Ok, I managed to do it.

With the USB to Serial adapter, and proper jumpers(boot0 set to 1, boot1 set to 0) I uploaded file: generic_boot20_pc13.bin with the Flash Loader Demonstrator.

Then disconnected the USB To Serial cable and reconnected over the pure USB cable with default jumpers(boot0 set to 0, boot1 set to 0).
Now my blue pill was found as Maple DFU.

Then following Vassilis steps:

The bluepill must be connected to your computer USB socket.
Select:
* Variant: STM32F103C8 20k RAM 64k flash
* Tools -> Upload method -> STM32duino bootloader
* keep pressed the reset button on bluepill board.
* Compile a blink sketch and click on arduino IDE “upload” button.

When you see the following message release the reset button.

I managed to get then new USB Id device (Maple Serial).
Warning: platform.txt from core 'STM32 Boards (STM32duino.com)' contains deprecated recipe.ar.pattern="{compiler.path}{compiler.ar.cmd}" {compiler.ar.flags} {compiler.ar.extra_flags} "{build.path}/{archive_file}" "{object_file}", automatically converted to recipe.ar.pattern="{compiler.path}{compiler.ar.cmd}" {compiler.ar.flags} {compiler.ar.extra_flags} "{archive_file_path}" "{object_file}". Consider upgrading this core.

Sketch uses 13,536 bytes (20%) of program storage space. Maximum is 65,536 bytes.
Global variables use 2,560 bytes of dynamic memory.
maple_loader v0.1
Resetting to bootloader via DTR pulse
Reset via USB Serial Failed! Did you select the right serial port?
Searching for DFU device [1EAF:0003]...
Assuming the board is in perpetual bootloader mode and continuing to attempt dfu programming...

Found it!

Opening USB Device 0x1eaf:0x0003...
Found Runtime: [0x1eaf:0x0003] devnum=1, cfg=0, intf=0, alt=2, name="STM32duino bootloader v1.0 Upload to Flash 0x8002000"
Setting Configuration 1...
Claiming USB DFU Interface...
Setting Alternate Setting ...
Determining device status: state = dfuIDLE, status = 0
dfuIDLE, continuing
Transfer Size = 0x0400
bytes_per_hash=270
Starting download: [##################################################] finished!
state(8) = dfuMANIFEST-WAIT-RESET, status(0) = No error condition is present
error resetting after download: usb_reset: could not reset device, win error: The system cannot find the file specified.
Done!

Resetting USB to switch back to runtime mode


Arise wrote:Ok, I managed to do it.

<…>

Thanks guys!!! You are so awesome!

Oh, so happy I managed to make it look like serial but now the new issue, I can’t upload on that serial.

When I switch from STM32duino bootloader to serial and I try to upload the sketch I get this:

stm32flash 0.4

http://stm32flash.googlecode.com/

Using Parser : Raw BINARY
Interface serial_w32: 230400 8E1

Failed to init device.

Arise wrote:Oh, so happy I managed to make it look like serial but now the new issue, I can’t upload on that serial.

When I switch from STM32duino bootloader to serial and I try to upload the sketch I get this:

stm32flash 0.4

http://stm32flash.googlecode.com/

Using Parser : Raw BINARY
Interface serial_w32: 230400 8E1

Failed to init device.

Yes, exactly that.

Don’t tell me is not possible for whatever reason like “that serial is running and can’t be erased during a reprogramming of a new sketch”.

Oh yes. These are two different upload methods.

If you use the usb cable then you have to select the stm32duino bootloader and the virtual COM port that is appeared on you device manager when you connect the bluepill to your computer.

If you want to send the compiled sketch to bluepill through your USB-to-serial adapter that is connected to the bluepill Serial 1 interface, then you select upload method : Serial.

So is not possible to just upload a sketch using the simple USB cable selecting Tools>Upload Method> Serial (not STM32duino bootloader)?

Arise wrote:So is not possible to just upload a sketch using the simple USB cable selecting Tools>Upload Method> Serial (not STM32duino bootloader)?

Opinion:
Rapid Development w/ ArduinoIDE should utilize the bootloader – if 8K is an issue, you have the wrong uC.

After all fluid recompiles have been completed, you can remove the bootloader and go into a full-blown test cycle; noting numerous defects; correcting all issues in one edit session; then flash and retest. Do not recompile & flash again for something like a mispelled word – document.

We are not using coding sheets or punched cards: but IMO recompiles/reflashes just to get a lcd.locate(x,y) correct is an abuse of the methodology. Folks doing massive numbers edit/compile/reflash cycles sbould maybe investigate a PC based emulator.

Ray

I want to use it as a test base/replacement for a flight controller, like naze32.
I will also try to replace the CPU with a F3 instead of the current F1, but until I do that I was trying to familiarize with the F1 platform first, to know the weak points and the quirks, etc.
F1 is not really a big concern regarding the memory, however once I’ll upgrade the CPU on the board then the memory size will be probably my last concern. Later, if all is good I will try to develop probably my own board, but based on F4, in case nobody else comes with another F4 bluepill at a nice price or some other variant.

The only reason I want USB is because is one cable, one connector, I don’t have to mess with 4 wires like I do in the USB-to-Serial adapter. This is useful in a couple of situations

Arise wrote:
The only reason I want USB is because is one cable, one connector, I don’t have to mess with 4 wires like I do in the USB-to-Serial adapter. This is useful in a couple of situations

Arise wrote:I want to use it as a test base/replacement for a flight controller, like naze32.
I will also try to replace the CPU with a F3 instead of the current F1,
<…>
Later, if all is good I will try to develop probably my own board, but based on F4, in case nobody else comes with another F4 bluepill at a nice price or some other variant.

mrburnette wrote:… We are not using coding sheets or punched cards: but IMO recompiles/reflashes just to get a lcd.locate(x,y) correct is an abuse of the methodology. Folks doing massive numbers edit/compile/reflash cycles sbould maybe investigate a PC based emulator.

Ray

It is a fucking nightmare to get USB Serial working…
I tried it some months ago, with bluepill #1 and stopped trying after reading everything about it on this forum and others.
Some days ago I picked up another bluepill from a friend and did the following:
– I removed the wrong resistor and soldered the proper 1,5kOhm in place
– The bootloader was already flashed on from an earlier attempt so
– I switched the jumpers and connected the pill with my computer via USB
– ENUMERATION WORKED, i was happy as fuck and flashed some sketches with Serial.print stuff in it and was happy that even the USB Serial thing worked!
– Then suddently enumeration stopped working and I was forced to use the DFU stuff instead. I remembered myself that lately Roger changed something in the bootloader in order to get the enumeration stuff working better, so I flashed the recent bootloader via Serial and from this point on not even DFU is working anymore and Windows keeps saying that something is wrong with the USB device…
– Flashing the “old” bootloader back on it didn’t help either.
– After that, I uninstalled and installed the maple drivers in every possible way and tried the Zadig approach – nothing.
Personally, I will just stop trying and use a Serial adapter instead.

Rick Kimball wrote:
<…>
you probably used silver tape to cover the holes in your punch cards instead of sending it back downstairs to data entry when you had errors

I dont know if here is the proper thread to post it, but I am starting a hardware project using BluePill.
The project includes a custom PCB and for this reason I designed a schematic part for BluePill. For designing I am using KiCAD (I am pretty familiar with it)
The part is included in _stm32.lib (hopefully more parts will added).
For PCB I am following DIP40 numbering, with this way the part can be used with the generic module, but latter I will design a more precise BluePill pcb module.

mrburnette wrote:… As the weeks progressed, everyone in class went back to the prof and asked for an extension of CPU allocation… everyone but me….

mrburnette wrote:
The seniors & grad students had access to the TSO room… I knew the work-student lady that managed access Once I punched the JCL and had run them into the batch system, the line printer would print the job-scheduling number. That number was accessible from the TSO.

Ray

zmemw16 wrote:
<..>
is a blue/red/maple a deterministic(correct word??) system ? i did mech eng, not software eng at uni.
where might be the documentation describing the start up flow / mechanism of the stm32f103c8t6 be found?

stephen

hi,
i have received my ST link programmer.
With it i can sucessfully reprogram the Blue pill with the file : generic_boot20_pc13.bin
Now device is named : Mapple DFU in my device list.

Now i will try to use USB Dfu for reprogramming devices.

i launch DfuSeDemo but i can’t see any DFU device listed.
Is it because Mapple dfu use custom bootloader instead the ST one or maybe device can only be rewrited by serial ?

Thanks

you should look using the dfu-util.exe (or linux dfu-util, depending of your plateform) provided in the tools directory, the “-l” option will list all discovered devices.

Thanks for your fast reply.
here is the result

Good !
This mean the devices is been seen and showing the STM32duino bootloader.
Enjoy your new toys !

i remember having to use
alt=1

Thanks for your help.
Now i can sucessfully upload over USB Dfu.

Sadly it seems that don’t works well.
I have made a simple blink led program.

-When i send it into adress 0x08000000 with ST link program works fine after a reset ( but rewrite bootloader).
When i send it with dfu-utils :

dfu-util.exe -d 1EAF:0003 -a1 -D led.bin

Bootloader launch ( led blink in high speed and after that turn off).
So it seems my user program don’t be launched.

Maybe it’s an adress problem ?
Sorry for question each day ^^

If you want to use the DFU you must select as upload method: STM32duino bootloader
That way the IDE compiles the sketch at VECT_TAB_ADDR=0x8002000
Then you can use dfu-util.exe -d 1EAF:0003 -a2 -D led.bin to send the bin to the board.

I suppose you use Board: Generic STM32F103C series

There is something I don’t know exactly how to interpret.
http://www.st.com/content/ccc/resource/ … 161566.pdf

On the page 33/117 it is written:
PC13, PC14 and PC15 are supplied through the power switch. Since the switch only sinks a limited amount of current (3 mA), the use of GPIOs PC13 to PC15 in output mode is limited: the speed should not exceed 2 MHz with a maximum load of 30 pF and these IOs must not be used as a current source (e.g. to drive an LED).

However, all bluepills are using a led on PC13, is this a mistake from chinese designer/producer or I’m reading it wrong?

Considering there is a 510R resistance in series with LED, it should be Ok, although it is border line with around 4mA instead of the 3mA mentioned.

Amazing work, @RogerClark and others, in getting Arduino working for Generic STM32 boards (and especially the Blue pill) along with the USB Bootloader (DFU) upload.

I had a blue pill with 10K pull-up. I uploaded the bootloader via USB-Serial adapter. After the 1.8K addition (in parallel), the USB enumeration and upload works flawlessly!

Initially, I followed the press-the-reset-button,initiate-upload,release-reset-button routine on Arduino. But it didn’t work and I spent some tense moments agonizing over UDEV rules in Linux. Then I tried uploading without pressing the reset button and it worked!

It seems the automatic reset through USB (through DTR?) is working now. It it a recent development?

Automatic reset works for most people, but not all.

It seems to depend on the USB hardware in your PC / Linux machine, and whether you go though a hub.

But you definitely need to put the 1.8k in parallel with the 10k, as 10k is far far higher than specified in the USB spec (which I think requires 1.5k – hence 1.8k in parallel, gives the correct resistance)

I attached a nice and big pin diagram of blue pill, I think is better than anything we have.

(Taken from: https://developer.mbed.org/users/hudakz … i/Homepage )

nice!

Cool thing, Slammer, I like those diagrams for every MCU/dev board.
I really miss a blue pill in my collection and since they are 1.72 Euros (cheaper than a Nano 3.0) they is really no reason not to buy one
http://www.aliexpress.com/item/1pcs-STM … 20209.html

The only thing I see on the pics is the sub optimal minimal LDO (5>3.3V). Somebody knows how much mhA you can draw with it?

Trying to power anything outside the board with this LDO may cause problems. There was a version of bluepill with better LDO, but most latest boards are using the small type with small caps.
Another weak point of bluepill (except the error on USB resistor) is that there is no way to separate the USB supply from 5V plane (no diode attached)

I attached the photo of the type with the better LDO and big caps. Unfortunately you can’t ever know which type you will receive from chinese sellers…

Slammer wrote:Trying to power anything outside the board with this LDO may cause problems. There was a version of bluepill with better LDO, but most latest boards are using the small type with small caps.
Another weak point of bluepill (except the error on USB resistor) is that there is no way to separate the USB supply from 5V plane (no diode attached)

I attached the photo of the type with the better LDO and big caps. Unfortunately you can’t ever know which type you will receive from chinese sellers…

While I don’t have any maple mini clone, I think it is the best small board for stm32duino development for beginners. The weak point of maple mini is the lack of SWD interface (while you can connect stlink directly on the pins, I would like a small header)
I personally prefer plain boards without bootloader, (I prefer STLink for download) and for this reason (and the low price) I am still using BluePills. I also like all Nucleo boards from ST, as development boards for any STM32 variant (and get a Stlink V2.1 free).

The ST-Link connector is a big “+”.
I started with the STM32F103 nucleo board about 2 (or 3?) years ago, I have totally forgotten I own one This is an upload method we miss in STM32duino: Just file drag&drop as it’s done with ST-Link 2.1 (I’ve done it manually every time). And the best about Nucleos: You can break away the ST-Link section

I fried the regulator on one of my bluepill boards. To fix it, I just unsoldered it and then dropped in a replacement TO-92 MCP1702. Seems to work nicely.

mcp1702_hack.jpg (68.95 KiB) Viewed 870 times

Perfect! Well…. add a bit if insulation to that middle leg to stop it shorting out when you squash the legs together by mistake, and it will be perfect.

The middle leg is a wooden leg..

Ok, I’ve ordered 2 PCS for burn down ….testing.
Rick: I know this one use the small LDO, but take this as cure for everything:
http://www.aliexpress.com/item/S1065-Fr … 69056.html
0.03 Euros for one fire cracker is cheap enough

heh! .. no wonder they use those!

madias wrote:Ok, I’ve ordered 2 PCS for burn down ….testing.
Rick: I know this one use the small LDO, but take this as cure for everything:
http://www.aliexpress.com/item/S1065-Fr … 69056.html
0.03 Euros for one fire cracker is cheap enough

At this price you CAN be sure, that they are fakes, but you have 100PCS as fire crackers. (Meanwhile I’m really fast in de- and solder the LDO’s)

Mmmmm…
I got these
http://www.aliexpress.com/item/50pcs-fr … 86579.html
0.04€/piece, I will experiment too!

I’m going to build a little pcb with a bare minimum AMS1117 and a (variable) load resistor, to check this crackers out, so something like this (AMS1117 3.3v) – > Input 5V load 500mA, 750mA, 1A , – > Input 9V load 500mA, 750mA, 1A and so on.

If you strart @500mA, you will finish your tests very fast…..

Filming the results might be fun.

and the proto board cost how much?

And you had the camera handy as well

stephen

I did a STM32F103 pinout diagram, comments are welcome.

http://reblag.dk/wordpress/wp-content/u … iagram.pdf

Regards
Rasmus Friis Kjeldsen

Very nice! Thank you

rasmus wrote:I did a STM32F103 pinout diagram, comments are welcome.

http://reblag.dk/wordpress/wp-content/u … iagram.pdf

Regards
Rasmus Friis Kjeldsen

I think the SWD pins might be mislabeled. Isn’t SWCLK PA14 and SWDIO PA13? They are labeled PC in the pdf file.

Very nice. I like this.

How about a webpage where you can search pins by e.g. port name? Simple javascript would highlight corresponding pin in the picture.

michael_l wrote:Very nice. I like this.

How about a webpage where you can search pins by e.g. port name? Simple javascript would highlight corresponding pin in the picture.

mrmonteith wrote:michael_l wrote:Very nice. I like this.
How about a webpage where you can search pins by e.g. port name? Simple javascript would highlight corresponding pin in the picture.

Rick Kimball wrote:I think the SWD pins might be mislabeled. Isn’t SWCLK PA14 and SWDIO PA13? They are labeled PC in the pdf file.

Rick Kimball wrote:mrmonteith wrote:michael_l wrote:Very nice. I like this.
How about a webpage where you can search pins by e.g. port name? Simple javascript would highlight corresponding pin in the picture.

Hi guy’s, I replaced the R10 from 10k to 1k5 (0603 is original resistor but 0805 fits ok) but on the schematic I see that R10 goes between D+ (after R11) and 5V but on my board it goes to Vcc3V3 and not to 5V rail ?! … how many variants of these boards are there?!

I took a short video with microscope of mine (with replaced R10) if anyone care to look ..
https://youtu.be/eAvqU8QzRwU

The Maple mini pulls D+ to Vcc, not 5V so it should work OK.

We have noticed that some “BluePill” boards have 10k resistors fitted, some have other values, so it looks looks like a “BluePill” is not always the same as the other BluePill boards that people have.

I don’t have any BluePill’s but I have some of the rare RedPill boards; however your board is different from my RedPill as well, (not just the colour).

It looks like the USB connector is different. It appears to be a micro USB but has 4 fixing points and also they seem to have a cutout in the PCB under the USB socket, so I presume its slightly recessed into the board.

arhi wrote:how many variants of these boards are there?!
https://youtu.be/eAvqU8QzRwU

New version spotted (I suppose, I’m new on this platform). It doesn’t have any components on the bottom side, has screw mounting holes, and none of the headers are soldered. I don’t know what chip is between oscillator and USB port – an LDO? I don’t see RTC, unless it’s packed under the oscillator. Shame it has weak USB port.

Mounting holes with same size => it has less power pins than the “standard” pills…

€ 1,45 !!

That is the voltage reg between the usb conn and the big crystal. 32kHz resonator is the yellow piece between the big crystal and the mcu. Most probably
It has got an 1k5 (the resistor close to B8) as the usb d+ pullup.
There is none 5V available on the headers..
BlackPill

Looks interesting.

I think I better order a few for testing.

stevestrong wrote:Mounting holes with same size => it has less power pins than the “standard” pills…

Meanwhile the “regular” blue pill price is about 1.38 EUR (less than AVR pro mini!)
https://www.aliexpress.com/item/Free-Sh … 44057.html
and 1.63 EUR if you buy 10pcs (registered mail – strongly recommended on ali!) : https://www.aliexpress.com/item/10pcs-l … 93738.html

marekic wrote:stevestrong wrote:Mounting holes with same size => it has less power pins than the “standard” pills…

i have a bunch of these boards some of them even GD32 and would say i defenitely dont like the red one, for the usb is loose soldered and consumes 1 of my pull up resisters, choose the blue one

ahull wrote:You can of course get the STM core to shutdown completely before you reach the lower limit of the RTC retention voltage, and run the whole thing on one battery rather than having a separate supply for the RTC.

The workings of the STM32 internal clock/calendar are detailed here.

The VBAT voltage preserves not only the clock but also some ram, which they call backup registers.

On boards where VBAT is hard wired to VCC then you will need to preserve VCC in order to preserve the clock and backup registers.

In a battery operated device, this can be done by putting the main processor to sleep, whereupon the power consumption drops to less than 1mA (considerably less if you implement this correctly and there are no other parts of your circuit, such as LEDs running).

This is not the best way to set things up, having a separate supply like a coin cell for VBAT is a better arrangement, but this is not possible on some boards due to the way they have been designed. If VBAT has not been hardwired to VCC, then you can add a coin cell to preserve the RTC and backup registers. If VBAT and VCC have been hardwired by the board design, then you must use other tricks.

My black pill board arrived today, net time was a little under 3 weeks to Japan, which is more than usual.

Pin row to pin row it is 0.1″ wider than the blue/red pill boards. Overall it is 4mm longer and 2mm wider.
I’ll solder headers on and try to bootload it tonight if I can get the baby to sleep.

ahull wrote:The workings of the STM32 internal clock/calendar are detailed here.

The VBAT voltage preserves not only the clock but also some ram, which they call backup registers.

On boards where VBAT is hard wired to VCC then you will need to preserve VCC in order to preserve the clock and backup registers.

evildave_666 wrote:My black pill board arrived today, net time was a little under 3 weeks to Japan, which is more than usual.

Pin row to pin row it is 0.1″ wider than the blue/red pill boards. Overall it is 4mm longer and 2mm wider.
I’ll solder headers on and try to bootload it tonight if I can get the baby to sleep.

I was also able to bootload the blackpill today and got usb upload and usb-serial working (on ubuntu though).

My Black Pill arrived today

It seems to have a better USB connector than the BluePill in that it has through hole soldered pins holding it in place.

The downside of this board, is that the rows of pins are even wider apart than the BluePill so make it harder to use on a breadboard.

I installed the bootloader using STLink, but need to build a version that flashes PB12 as we don’t currently have one with that config

Unfortunately at the moment I seem to have some build issue, I’m not sure if its a compiler version problem, as my default compiler is 4.9.3, but the IDE uses 4.8.3, so I think I better try going back to that version and confirm why any new bootloader I create doesn’t seem to work correctly

Update.

Looks like the bootloader will not compile correctly with GCC 4.9.3 . Its probably the optimisation which is causing the problem
(I should have read my own ReadMe, as its listed in there that 4.9.x doesnt work. I am sure it can be bug fixed, as its probably one of the hardware addresses for the USB registers needs to be declared as volatile, but I don’t have time to bug fix this at the moment)

So I’ve changed my default arm compiler back to 4.8.3 and compiled a new version of the bootloader for the BlackPill

The bootloader can be downloaded from github, here https://raw.githubusercontent.com/roger … 0_pb12.bin

If I get time I’ll also make a bootloader for the HYTiny STM32F103T, as I suspect the reason I could not make the bootloader work on that board may also have been the compiler version issue.

RogerClark wrote:
The downside of this board, is that the rows of pins are even wider apart than the BluePill so make it harder to use on a breadboard.

Hi all,
I request your help about the STM32 blue pill board.
I the beginning of this thread i have buyed one and you helped me for installing the bootloader (page 9).
with this board i have sucessfully completed my project an USB HID Sega Megadrive Cartridge Reader :

https://github.com/X-death25/STM32_Proj … adrive_HID

But i have buyed a pack of ten STM32 Blue pill on Aliexpress and when i solder these board on my PCB the USB enumeration fail…
i can hear the “usb fail sound”.
it works just well with the same binaries on another blue pill board..
Strangly it’s works fine with an old computer under Windows XP

i have tested a 1,8K resistor in parallel of PA12 & VCC3 but without sucess…
here is a picture of usb view

http://imgur.com/a/U9qLN

I don’t understand were is the problem , the pcb and binaries is OK because it works with my first blue pill…
Could you help me ?

Check the soldering of the USB connectors.
Also, you should first erase the whole flash and then burn the STM32duino bootloader.
It should work after these steps.

STM32duino bootloader seems to works but when it’s load my program usb failed.

Which method do you use to upload?
Usb serial should work if you use srm32duino bootloader method.

When the bootlaoder is active i can use it for upload my own bin , it’s works fine.
But when it’s switch onto my own program the USB HID don’t works…

Got myself a couple of Black Pills, I can confirm that https://raw.githubusercontent.com/roger … 0_pb12.bin works nicely.

I have noticed that the CDC Serial does not wait for connection like it does on the Leonardo though, if I do the usual:
while (!Serial)
;

Try
while(!Serial.available());

For the USB serial we have:
while ( !Serial.isConnected() ) ;

@Pito: Those are different things

@stevestrong: That works fine, thanks! But then this is a deviation from the Arduino API, why?

SukkoPera wrote:@Pito: Those are different things

@stevestrong: That works fine, thanks! But then this is a deviation from the Arduino API, why?

I have not seen any platform on which while ( !Serial ) ; actually works as expected (wait till serial connection is established).

It actually works perfectly well in my experience with Leonardo and Micro boards. I can’t remember if I ever used it on the Due, though.

PR submitted: https://github.com/rogerclarkmelbourne/ … 2/pull/270

What do you think about this board? Is a ZET6 for ~11€ a good bargain?

http://s.aliexpress.com/Nfe6vqYB

For 11€ I would rather buy a black F4 board or the F4 mini version.
It is faster and has more RAM.

But the F4 core is not as mature as the F1 one, is it?

It depends on what do you need.
USB serial, GPIO, SPI (with DMA) is working out of the box.

There are a bunch of threads devoted to the F4 questions. Please move your discussion over there.

* rick wishes he had admin capability on the thread he starts …

Right ! Core maturity is depending of what you wish to do.
Many libs can be shared between F1 and F4.
For example, almost 2 years ago, I’ve done lot of things with Netduino2Plus.
Networking, SdFat, LCD shield, etc…

OK, you convinced me, I’ll get the F4. I’ll post in the relevant section when I get it. Thanks .

Go for the F4 mini version – I am very pleased for it.

New contender in the war:
STM32F103C8T6, STM32 ARM Cortex-M3 Minimum System Development Board
http://s.aliexpress.com/vyAfIjUB
(from AliExpress Android)

I have an arduino mega ( 2560) from them and it is well made, the price is too high compared to the others boards

Interesting

[aster – Sun Jul 30, 2017 10:50 pm] –
New contender in the war:
STM32F103C8T6, STM32 ARM Cortex-M3 Minimum System Development Board
http://s.aliexpress.com/vyAfIjUB
(from AliExpress Android)

I have an arduino mega ( 2560) from them and it is well made, the price is too high compared to the others boards

as noted in this thread about somewhat pricy stm32f103 source prices (from st itself)
http://www.stm32duino.com/viewtopic.php?f=3&t=2359
there is this lingering speculation that perhaps even bluepill itself may no longer be as cheap as it used to be at some point

I have yet to see a significant price hike, and if there is one, GD32 may become the more cost effective option

So it has been over 2 years since I first started this thread. The debate over Red Pill vs Blue Pill is over. Sadly, I can’t find any place to buy a red pill even though its availability would have saved so many from the pullup resistor headache.

Yesterday, I received an order of 5 Blue Pills recently purchased from ebay. The price is still about the same ~$2USD. The latest ones look exactly like the ones I purchased more than 2 years ago. They still have the wrong resistor on R10, the smallish voltage regulator, and the notch for the four pad usb connector nestled in the notch on the pcb. I still don’t see a schematic online that actually represents the board I purchased. This one comes closest, however it shows a 4k7 pullup resistor connected to 5V for PA12/USBD+ The pullup should be 1k5 and connected to 3v3.

All in all still pretty amazing, that we can buy such a powerful ARM chip and thanks to all the effort by people here it is useful with the Arduino IDE. Thanks everyone!

I don’t know if it’s partially because all the work we have done to get Arduino onto the BP, or if it’s just because I tend to spot a BP if I see it in a blog post or YouTube video etc, but…
It does seem or be cropping up more and more in videos and blog posts and other articles.

In the last week, 2 posts on Hackaday have used the BP, neither of which were using it with the Arduino IDE.

The week before, I noticed someone using it to replace a cypress chip, to make an open source signal generator , which goes up to 4GHz.

I was contacted a few months ago about doing a special version of the core for the ZumSpot HAM radio system ( just needed a minor flow control change on USB serial )

And the list goes on and on.

I still have 5 pcs of red pill , was planning to make a MMDVM Zumspot digital radio but notice ALL the projects was using the blue pill with carrier PCB made for the blue pill pinouts..

Only until I read this post that I realised that all the red pills are gone from the market..

I will probably use up all the red pill with point to point wiring and move forward to go with the blue pill after that …

[Rick Kimball – Sun Aug 06, 2017 3:16 pm] –
All in all still pretty amazing, that we can buy such a powerful ARM chip and thanks to all the effort by people here it is useful with the Arduino IDE. Thanks everyone!

It’s also amazing that you picked that name to differentiate them, and has pretty much become a brand for the board, even ebay sellers use that name:
http://www.ebay.com/itm/STM32F103C8T6-A … 2145343898

Had you trademarked it, you could be get rich just with royalties and fees

[stanleyseow – Sat Sep 16, 2017 1:00 pm] –
I will probably use up all the red pill with point to point wiring and move forward to go with the blue pill after that …

The red pill and blue pill use the same pinouts. You shouldn’t have to do anything. They are completely interchangeable.

The red pill and blue pill use the same pinouts. You shouldn’t have to do anything. They are completely interchangeable.

Thanks .. I just notice that they are the same pinout, so for the Blue Pill, I will have to do the “R10” operation like before ??

If you are asking do you have to worry about the PA12 pullup resistor with the red pill board, then no. The red pill boards I have ordered all had the proper 1k5 resistor. Just try it. I’m guessing it will just work.

We are able to order like 30 pcs of red pill from taobao easily …

Here is the picture of the red pill with my version of MMDVM HAT …

SVT MMDVM HAT by stanley_seow, on Flickr

Can you please post link to where you got them? I cant find the red pill on Taobao for the life of me.

Thanks a bunch

I’m not sure why you would want a Red Pill, apart from the colour.
AFIK they are functionally equivalent to the Blue Pill

A better option now seems to be the Robot Dyn version which has a black coloured PCB, and has the correct USB pullup resistor (unlike the Blue and Red pill which both have the incorrect resistor resulting in USB connection issues for some people)

roger the pull up on the red pills are fine.

[Rick Kimball – Sun Nov 12, 2017 9:56 am] –
roger the pull up on the red pills are fine.

Thanks…

All my red pills have broken USB connectors, so I am still not sure if they are worth buying unless they now use a stronger usb connector

There are BluePills with 2 pads- or 4 pads-style microusb sockets.
I’ve been ordering only the 4 pads ones.

The 4 pads ones come in two variants:
1. soldered to the pcb – sometimes not well enough
2. weakly soldered, the mechanical connection to the pcb done by pressing the fins into the mounting holes and soldered with a tiny amount of solder.

As the microusb connectors create a much bigger insertion force on the socket (in comparison with mini usb used in MM) the 4 pads in both 1. and 2. need to be resoldered with a large blob of tin such the pads and the adjacent body (the walls) of the socket are covered by the tin.
That improves the mechanical and electrical durability.

Sun Nov 12 13-07-03.jpg (56.58 KiB) Viewed 604 times

[RogerClark – Sun Nov 12, 2017 10:48 am] –
All my red pills have broken USB connectors, so I am still not sure if they are worth buying unless they now use a stronger usb connector

That is exactly the problem. It also the reason I penned this redpill vs bluepill post in the first place. I have 5 red pills with perfectly good pullup and usb connectors that have snapped off : (

@Pito

I do the same thing… Solder the BluePill through holes on the connector to make it stronger

For comparison, on the black pills I’ve just looked at, there is no pad on the PCB on the top-side for the fins to be soldered to, but the fins go through the PCB and are soldered to the inside of the hole. They fins are almost exactly flush with the holes (I think they stick out just the tiniest amount).

Naked-eye I couldn’t see any solder, but under a microscope it looks like the hole is well-filled with solder around the fins.

innerpin.jpg (226.47 KiB) Viewed 985 times

Which Black Pill ?

Robot Dyn or someone else ?

[RogerClark – Mon Nov 13, 2017 1:04 am] –
Which Black Pill ?

Robot Dyn or someone else ?

Someone else. It looks just like this one.

[RogerClark – Sun Nov 12, 2017 9:02 am] –
I’m not sure why you would want a Red Pill, apart from the colour.
AFIK they are functionally equivalent to the Blue Pill

A better option now seems to be the Robot Dyn version which has a black coloured PCB, and has the correct USB pullup resistor (unlike the Blue and Red pill which both have the incorrect resistor resulting in USB connection issues for some people)

Roger, I was just interested to see how many vendors were selling these and what the quality looked like on them compared to the blue and black pills. So now it looks like the black pill from RobotDyn is the best quality one out there? I appears to have the same pinout as the BP, is that correct? That would be nice as it would give 2 different sources for these. What are your thoughts about this Black Pill vs the Baite Maple Mini clone with respect to quality and long term viability as I am still contemplating using one of these in my commercial product?

Thanks again

[RogerClark – Sun Nov 12, 2017 9:02 am] – A better option now seems to be the Robot Dyn version which has a black coloured PCB, and has the correct USB pullup resistor (unlike the Blue and Red pill which both have the incorrect resistor resulting in USB connection issues for some people)

The one downside of the Robot Dyn black pill over the other options (black pill, red pill, PB12-LED black pill) is that you can’t avoid paying shipping, so if you’re ordering just one, it costs about a dollar more. If you’re going for larger quantities, it’s not a big deal. I do like how it exposes more pins than the PB12-LED black pills.

I ordered 4 of them and they got shipped today.

I also paid for ePacket (which worked out to be about a dollar extra on each one.

I still think they are good value, and at least I’ll get them before Christmas

I ordered two of RobotDyn blacks: https://www.aliexpress.com/item/STM32F1 … 56794.html
The shipping fee to Germany (china packet plus) is only total 40 cent extra.

BTW.

Robot Dyn will be offering 2 additional versions of this board from next week

One has our bootloader on it and I think the other version has some other firmware or bootloader pre-installed.
They did not say if the price was different

[RogerClark – Tue Nov 14, 2017 2:47 am] –
I still think they are good value, and at least I’ll get them before Christmas

Right after 11.11, I kind of doubt it.

[evildave_666 – Tue Nov 14, 2017 11:23 am] –

[RogerClark – Tue Nov 14, 2017 2:47 am] –
I still think they are good value, and at least I’ll get them before Christmas

Right after 11.11, I kind of doubt it.

Hopefully, as I specifically paid for postage, I my packet my be put a bit higher on the pile that get scooped onto into a container ship