Search results for:

@possmann sorry I got distracted by other things and never finished this module properly, but as Gordon said it shouldn't be too difficult now - unless you want to handle alarms as events, in which case I'm not able to help :)

As a side note - if you don't need alarms and don't mind hacky solutions then you can use write/read scratchpad functions to persistently store your data in these unused registers.

Exactly.

Also it may be a good idea to specifically say that it happens ONLY when there is a USB cable connected to a PC which is not listening (i.e. not when running on batteries or mains). It really looks bad otherwise, like you couldn't use espruino in standalone mode.

EDIT: never mind, it's already there :)

Cool. Turns out it's just a storm in a teacup then.

Got it from this site http://www.espruino.com/Serial+Bootloader
Is it different instructions for other boards (I have HY-Mini too) or does it need to be updated to include -k?

I'm using the latest version of Google Chrome (stable).

Don't know how I managed to skip the -k flag, I will try it the next time. Thanks.

Maybe it's a daft question, but just to double check - when Espruino is powered via USB with a mobile charger or a USB battery then it's like USB was not connected? Meaning it's not blocking, right?

Success! After countless tries, fiddling with cables, changing ports, pushing espruino buttons in frustration etc finally it went through with no errors. After that every consecutive try so far works fine.

I don't have a clue what was the problem as I'm still using the same cable and USB port. But I guess if you get the same problem then just keep trying, it will come back to life :)

@Sacha thanks...

Did anyone else have troubles with updating to 1v62? It failed quite badly in my case http://forum.espruino.com/conversations/1250/#comment19230

Hi,

I was trying to flash a new version to my 1.3b board but that failed and now I don't really know what to do with my Espruino board as it's kind of bricked - when I plug it in to my linux box it's not recognized at all and blue and red lights are lit very dimly. All I can do is switch into bootloader mode (blue light is pulsing) at which point device pops up in the system as /dev/ttyACM0

When I try to flash with python script then I get this:

python stm32loader.py -p /dev/ttyACM0 -evw ~/Downloads/espruino_1v61/espruino_1v61_espruino_1r3.bin 
Reading data from /home/graf/Downloads/espruino_1v61/espruino_1v61_espruino_1r3.bin
Bootloader version 0x30
Chip id 0x430, STM32F1, performance, XL-density
Global mass erase; this may take a while
Writing 214728 bytes to start address 0x8000000
Write 256 bytes at 0x8000000
    WRITE FAIL - try and recover
    WRITE FAIL - wait
    WRITE FAIL - retry
    WRITE FAIL - try and recover
    WRITE FAIL - wait

and so on for ever (well I gave up after ~15 minutes). During this time the green light is blinking from time to time.

When I try to flash with WebIDE then it's a bit different. After download is complete green and red LEDs are blinking very quickly for a couple of minutes. Then only the red LED is lit for about a minute and then I get this error message:

Can't find STM32 bootloader. Make sure the chip is reset into bootloader mode by holding down BTN1 while pressing RST

I was flashing without problems with the previous versions and also Espruino was working fine before that failed attempt to update today. Now I can't even flash 1v61.
Any ideas, please??

@mgg1010 that's assuming you always have a working internet connection at startup and you want to expose your device to the internet. Both assumptions are not an option if you ask me, but I guess that depends on the project. Thanks for the module BTW :)

Nice thing about these modules is that they have a tiny battery, you can't really go without it if you need a time and date (instead of a counter).

Surely you're just dumping the memory usage before you remove the timer?

Ahhh good catch, thanks! But it was also before I add another timer..

EDIT: That's it! On the first console.log the memory usage was lower because timer was not even set at this point. Then on the second console.log the timer was finally set and memory usage was higher. Thanks for helping me understand this :)

Now it's only the ID numbering that puzzles me.

as you hadn't yet freed it with clearTimeout?

And this is that part that I don't understand :) I AM doing clearTimeout and led behaves like it was removed (timeout is reset) but memory usage and ID numbering suggests otherwise.

One more question because this is driving me nuts and I can't find the reason why ledTimer gets incremented at all.

What happens when I clearTimeout(ledTimer) ? Is it set to an empty function so that when it comes to the given time it does nothing, or is it really removed?

Also when I add console.log(process.memory()); at the beginning of watch function then it shows memory usage goes up quite a lot when I press the button before the timeout passed.

The output now looks like that:

{"free":1722,"usage":78,"total":1800,"history":44,"stackEndAddress":536909512,"flash_start":134217728,"flash_binary_end":134432224,"flash_code_start":134443008,"flash_length":262144}
before: null
after: 2
{"free":1695,"usage":105,"total":1800,"history":44,"stackEndAddress":536909512,"flash_start":134217728,"flash_binary_end":134432224,"flash_code_start":134443008,"flash_length":262144}
before: null
after: 4
{"free":1695,"usage":105,"total":1800,"history":44,"stackEndAddress":536909512,"flash_start":134217728,"flash_binary_end":134432224,"flash_code_start":134443008,"flash_length":262144}
before: null
after: 6
> 

Lol now that's reassuring.

Turns out setWatch and setTimeout IDs are shared so that's where value 2 is coming from (watch ID is 1).

If I add another watch for btnPin (falling) then the timeout ID has value 3 and it's growing +3 every time. That means the ID's value can grow very fast if I have a bunch of timers/watches that I want to add or remove dynamically.

And that's just the ID's value growing, the number of timers is the same as far as I understand.

It's awesome to have such an easy-to-use tool and to have results in minutes.

Yeah second that. Probably it's just the nature of the forum that you hear mostly about problems - and that's the purpose of this thread :) - but some praise can't be bad. For example I was playing with switches recently and was blown away by setWatch(). On Arduino doing timers and debouncing never felt intuitive to me, but on Espruino it's a pleasure. Thanks!

OK so what is the maximum ID this function can return? Is it some huge number in tens of thousands?

Could someone please explain me how are the IDs allocated? Or even better why it's not like I thought? :)

In this code below I would expect the ledTimer to get the same value (ID) over and over again but that doesn't happen.

var btnPin = BTN;
var ledPin = LED1;
var ledTimer = null;
digitalWrite(ledPin, 0);

var btnWatcher = setWatch(function () {
  if (ledTimer) {
    clearTimeout(ledTimer);
    ledTimer = null;
  }
  digitalWrite(ledPin, 1);
  console.log("before: " + ledTimer);
  ledTimer = setTimeout(function () {
    digitalWrite(ledPin, 0);
    ledTimer = null;
  }, 4000);
  console.log("after: " + ledTimer);
}, btnPin, {repeat: true, edge: "rising", debounce: 70});

Now when I press the button every 5 seconds I get what I expected:

before: null
after: 2
before: null
after: 2
...and so on

But when I press the button every 3 seconds then it's not what I expected:

before: null
after: 2
before: null
after: 4
...and +2 every time

@mgg1010 Thanks! If it was up to me I would get rid of all the print statements, personally I find them a bit annoying (and cryptic) and it takes up memory too :)

Quite a weird setup though. It will be permanently tethered to your PC?

Not a desktop PC, I was thinking about something small like Alix board or BeagleBone Black if/when ARM is supported.

Thanks a lot for your suggestions and ideas, I hope one day I will find more time to play around with them :)

@gadicc Any thoughts on meteor+espruino over the serial connection (i.e. no wifi)?

Edit:
To be more specific. I'm thinking about data logger where data is continually collected by sensors connected to Espruino and sent to Meteor. All the data would be stored and presented by Meteor.

This is great, thanks for sharing!

+1 for putting it where the pinout diagram is. I would say it's not a big deal at all to have problems like that, as long as you know beforehand and can work around it.

Yes I'm sure about the wiring as I checked and tried again so many times I've lost the count now :)

It's a new LCD so I don't know for sure, maybe I was out of luck and got a broken one. I'll dig out my Arduino board and see if it's working there.