the esp is running fine on 4.2V lipo
maybe a bit worm
but running stable, not hot at all
well, that's what happened to mine who took 30v
after that it was only making heat xD
i'll let it drain it
also the adc seem to have a reference voltage inside
just like arduino ?
better
the arduino one that's inside depends on vcc
this one does not
isnt the one in arduino 1.01v whatever the vcc ?
i thought it did, but i guess my other battery is almost empty
yes, the mega has 2 ref voltages
but they fluctuate with vcc
i think...
what is there by default?
whatever is usualy used fluctuates with vcc
so analog reads on analog sensors are constant
like thermistor for exampe
i'll probably make my own :D
i'll thrown in a ldo
point is, with divider, you can measure the vcc pretty accurately
true
as long as you have precise resistor and not 10% ones
yup
time to charge my lipo
think 250mAh will be enough ?
this will not do if you want to run it on lipo
why ?
because this is simple LDO
it requires higher voltage to give 3.3
more like 4.4-4.5 V
the least
it's has .18v drop
have not seen such LDO yet
i use up/down switching for such things
but i doubt you need one at all
The minimum VIN must meet two conditions: VIN 2.3V and VIN VR + 3.0% VDROPOUT.
as I said, running strong here
protection
hmm, didn't print as i wanted, but it's vin >= 2.3v and vin >= (Vr + 3%) +Vdropout
if it really does 0.18 drop
3.5 is a safe spot for lipo
so it would need >= 3.579V
250mAh should be plenty
100 :D
ok
flash at 80?
there was one more reason
cpu's at 80, i never changed the flash speed
uninitialized interrupt handler pointer being called before I thought it would.
that was the other reason for that same error
hmm, is it my fault or SDK error ?
hmm, seems like lipo is going fine here too
function mesure 4.05V when it's showing 4.2 on my multimeter, but it's not meant to go that high so i suppose it's normal
@Makuna thanks @igrr thanks, i did see wire.pins in the library before, but i think it was marked as deprecated " void pins(int sda, int scl) attribute((deprecated)); // use begin(sda, scl) in new code
"
"
maybe it s worth keeping for cases like this
I'm getting frequently wdt resets with reset cause 4 running for example the westerner example code without any change. After a few requests from the browser it stops responding for a while and the wdt triggers, with reset cause 4. Any idea for the cause? I've already guarantee a good power supply with capacitors never vcc and gnd of the esp-01 connector
I failed googling "westerner example code", can you share a link?
Sorry, dumb auto correction!!! I mean webserver example :-)
right :) works fine on an esp12 here, but I'll check if there's any problem with esp01
can you do Serial.setDebugOutput(true); in your setup method after Serial.begin?
and then we should see if there's anything strange happening
Ok, thanks, I'll do it later today, I don't have the boards with me at the moment. My main concern is that I'm checking it's stability and have always problems like that (for example exception 28 running sming examples). I've already made a change in the power supply, now I'm using a 5v 1a adapter with a 3.3v 1A regulator, and capacitors near the esp-01 pins, with very short connections, but it is still not stable. Could be a problem with the esp-01 modules? I have also some esp-12 in my desk, I could try them to see if the behavior is simila, just have to build some adapter. I've already ordered some modules from olimex including the development board to be sure that there is no hardware issue
exception 28 in my experience was always related to power supply issues or lack of strapping resistors on gpio0 and gpio2. when I switched from esp01 to esp12 mounted on a yellow carrier board (the one sold with battery compartment attached), I no longer have any of those issues...
seeing that your power supply is okay you may want to make sure all the resistors are in place (GPIO 0 and 2, chpd, rest)
perhaps replacing the module is also an option.
@igrr , thanks for your help! What do you recommend for the pin connections, having them with a pull-up resistor to Vcc and use a jumper to connect them to ground when necessary? What resistor value are you using? Currently I have the reset pin "floating", could this be an issue? I was convinced it had some internal resistor, read it somewhere.
they dont :(
i use 10K and 4.7K
seen 3.3 used as well
olimex uses 12K on their pullups
soanything above 3Kohm will be fine
CHPD need lower value, i think Espressif docs have it at 2k
others can be around 10k
I've the CHPD pin directly connected to Vcc, do you think that is a problem? Olimex is using 12k as pull-ups for RST and CHPD, but 2K pull-up for GPIO14/SCL. And uses 2k resistors for GPIO0/GPIO2 jumpers. I could use this as a reference to see if it improves the ESP-01 stability: I'll try to put 12k for RST and CHPD and 2k for GPIO0 and GPIO2, do you agree?
Sorry, you have refered 2k for CHPD, so I'll also use 2K for CHPD, so only 12k for RST, all other with 2k?
i think that would be a good configuration
the conf really does not matter much
as long as they are above 1K they will draw small enough current
lower value pullups require more time, energy to be pulled in the other direction (if not done by a switch)
i've ordered an oscilloscope and will look at those thigs exactly, as some I2C slaves have issues when lots of other things are going on in the chip
Yes, specially for I2C, it will be a tradeoff between speed and power. For now I just wanted a 'safe' (specially 'stable') hardware setup for testing the ESP-01 boards I have, until now I wasn't able to use them in a reliable way, even using the supplied examples from the Arduino porting (or the Sming examples, I get the same non stable behavior with Sming also). I am considering them for a product, but have to be sure of its stability and then to choose the more adequate framework to use (Arduino, Sming, Espresiff SDK...)
@chadouming yes i switch my flash chips to 4MB
you think these : http://www.aliexpress.com/item/10pcs-lot-Winbond-SPI-Flash-Memory-W25Q32BVSSIG-W25Q32-W25Q32B-25q32-SOP-8-Free-shipping/32259577223.html would be good ?
yup
any W25Q above 256K will do
yes have W25Q32 on my board :) they are little big bigger then the orginal one.
fit nice on the ESP-12E, but harder on some older models
i run an 8MB chip
had to bend the legs a bit on the 201 board
no point to have above 4MB though
same here but working well
only chip erase works above 4MB mark
seems the bootloader can not address the flash above that
8 MB seems a lot excessive
could not find 4MB chips here
only 1 and 8
and I plan second partition there maybe
ota ? :D
since espressif stays away
could be used for anything
store sketches for esp or arduino
or whatever
as said, only chip erase works there and it's a hardware command
the bootloader can't address that space
there are settings for larger flash to be set by the compiler
but espressif does not care
btw, adc is pretty accurate :)
is acces by the flash functions possible?
still running and after some observations and calculations, what i see is what my meter shows as well
@Links2004 there is something that could be done but it needs understanding first
a small program can be uploaded in the memory and run from there
we can control the UART through the registers
we need to learn how to talk properly to the flash (I'm on it)
if we go down to SPI level i see no problem
then that could maye used to reflash from uart or spiffs stored image
yeah, generally should be fine
some SPI and some support lib for the flash commands themselves
yes esptool py has an example for ram upload and run inside
but then have to know how to load it and so on...
yes, exactly
but that's like nothing compared to what we are talking about
so i'm not quite sure how to get it all done
but know @igrr will help
anyone can
:D
for the bootloader its the same. only more data
so the ram load shut work until the ram is full.
yeah
32K there
for the program I think
should be plenty for uart to flash
no clue of the spiffs footprint yet
yes
but other options are also available
spiffs is actually not that hard to read
i can prolly hack a function to just read a file
https://www.onlinedisassembler.com/odaweb/cejB6EW3/0 the asm from the embadded stuff in the py tool
that online disassembly ide is awesome
yeah!
yes its very usefull and can handle many core types
eh, world is kinda small. I thought i knew that name ficeto. I had darky rom on my SGS
:D
that is still kinda running
i'm the crazy dude that flashed the GT-I9000 bootloader on my T959D
So i could flash international rom on my telus phone xD
i'm not actively developping,but there are still folks that do, and i'm still hosting the forum and OTA and whatnot
hahahahah
eh, I know have a nexus 5
i had a very close friend of mine that did the same thing
way easier
me too :D
and using optipop rom
well... and an iPhone
no rom
devil
xD
samsung got me out... way too shitty phones and security on top
when the SGS3 came out.. selinux came
so lots of gymnastics to get root going...
later they put some crp in the bootloader that screwed the phone if hacked
well, before my N5, i had the Galaxy S II (Telus version), galaxy nexus, Optimus G (That i kinda transformed into a nexus 5), nexus 5 (Mic broken), Xperia Z II, Nexus 5
too much fight and not enough fun
remember the flash reset device
flash counter reset device*
yup
i still have my SGS around
i sold mine
like to hear Voodoo on boot :D
only phone i've sold
my mom has my SGS2
I had lots of phones... worked on many
gave them to friends
can not sell a phone
o yes still have all my phones xD
forget phones :D MCUs talking here :D
i keep reading about voip support in those
saw some Sip functions in the rom
any info on those?
yes but no documentation
my Z2 is derped too
i tried to make root happen, but bootloader didn't like
that broke my TA partition
and it won't boot anymore
fun...
yeah, well it was fun, until it didn't boot xD
i got so fed up with it that I no longer write any root apps for android
try not to write at all for it.. :D
lol
i hate java
only app i made is my apps to read data from the sensors i am making with the esp8266
java is not bad, but writing apps for android is
gazillion screen sizes and capabilities
you can never know on what phone and OS your app will endup on
i'm writing for my N5, that's about it
sticking to the SDK is generally fine in means of functionality
UI is a different story
send a shot better
phone is not hooked
flashing a new rom, i'll send a shot when it's done
:D
have not heard that in a while
i'm helping a friend trying to find why play services are crashing
adb logcat can help with that
lol, i know xD
might be more worthy than the app
same data, different format :P
just change for today's date
which DHT library are you using
adafruit dht sensor library
works fine out of the box
will give it a go
still running on the lipo.. 3.64V
found something esp8089 looks simulare to the esp8266 and has the sip stuff some code: https://github.com/george-hopkins/esp8089-spi/blob/master/ http://espressif.com/en/products/esp8089/
still running here too and it's over 4v xD
will fill it agan and take a nice discharge graph
i do not sleep :D
mine is online 100%
that indeed make a huge difference
mine does : wake up, take temperature and humidity, send data, get time, sleep until next XX:X5 or XX:X0 (i.e. : 15h20)
@Links2004 ours are different
PROVIDE ( sip_alloc_to_host_evt = 0x40005180 );
PROVIDE ( sip_get_ptr = 0x400058a8 );
PROVIDE ( sip_get_state = 0x40005668 );
PROVIDE ( sip_init_attach = 0x4000567c );
PROVIDE ( sip_install_rx_ctrl_cb = 0x4000544c );
PROVIDE ( sip_install_rx_data_cb = 0x4000545c );
PROVIDE ( sip_post = 0x400050fc );
PROVIDE ( sip_post_init = 0x400056c4 );
PROVIDE ( sip_reclaim_from_host_cmd = 0x4000534c );
PROVIDE ( sip_reclaim_tx_data_pkt = 0x400052c0 );
PROVIDE ( sip_send = 0x40005808 );
PROVIDE ( sip_to_host_chain_append = 0x40005864 );
PROVIDE ( sip_to_host_evt_send_done = 0x40005234 );
hmm, so if this thing has sip, we could make wifi chip to spy on people :D
you can make really small device with that MCU
yup
(>:D)
this is not that SIP (VoIP one), it's a shorthand for serial interface protocol
D:
and what is that serial interfece protocol?
probably still possible to hook a microphone on these thing and stream everything xD
i imagine SIP is not that hard to implement... it's UDP and taxt based controll
esp8089 is used in a slave mode by the host processor (in some rockchip-based tablets)
then data is udp stream on different port
it's the protocol that is used to communicate between the host and the slave
why do we have it?
oh.... oh...
i think i know why...
esp8266 supports it as well
i wanted to bood from sd card
while that SDIO is actually the other way arround
the esp acts as SDIO slave
since there are such WiFi adapters
it can boot in sdio mode, being a SDIO slave
pin15 high
yup
i wandered why i was getting "waiting for host" in boot
that explains it
btw I noticed that my ESP-12Es reboot and flash better when EN pin is used for reset
and my ESP-12s do better using the RST pin
so i tied both together and pulled them up with single resistor and reset at the same time
anything wrong with that?
would it break sleep if I want it?
well, you need the rst pin tied to gpio 16 to sleep
is it a problem if they do trisom with the EN pin?
don't know
you mind pasting that sleep command so i test?
i'm new to sleep :D
ESP.deepSleep((secondToSleep * 1000000), WAKE_NO_RFCAL);
and tie the pins
that's it?
wake_no_rfcal is optionnal
yup
tie the pin to a pull=up
pull-up*
what does WAKE_NO_RFCAL do?
oh yeah
no calibration
faster boot
take less battery
which is useful for me xD
i'm still getting a few WDT with reset cause 2 every now and then
good news everyone
very good news in fact
cool
lol they created a new github account for their app
bad app? xD
essentially, this says : Fuck you all developers who want to use our tools on other device like the emw3162
hmm, seems like they fixed readvdd33 with station only mode :D